Qr Code
-
WeChat
- Current Discuss
- First Published
- Album Paper
-
Development of the Ultra-low temperature cesium atomic fountain clock microwave cavity test system
Zhang Ze, Wang Xinliang, Nie Shuai, Guo Wenge, Zhang Shougang
Abstract:
Cesium atomic fountain clocks are standard frequency signal generators based on quantum transitions in atoms, widely utilized in timekeeping systems and other precision measurement applications. The ultra-low-temperature cesium atomic fountain clock represents an advanced version of this technology, operating with cesium atoms in a liquid nitrogen (80 K) environment. Reducing the surrounding temperature from 300 to 80 K leads to a 187-fold decrease in the blackbody radiation frequency shift and a 79-fold reduction in the associated uncertainty. Additionally, improvements in microwave cavity phase frequency shifts and background collision frequency shifts are observed. The microwave resonant cavity, a critical component of the ultra-low-temperature cesium atomic fountain clock, requires tuning and testing to align its resonant frequency with the cesium atomic transition frequency. Although the cavity is tuned under atmospheric, room-temperature conditions, it functions in a vacuum and ultra-low-temperature environment, where thermal expansion and contraction cause significant parameter variations. To validate the tuning process, it is crucial to replicate the actual working conditions of the cavity experimentally. This paper describes the development of an ultra-low-temperature resonant cavity testing system. Using microwave resonant cavity design theory, the system′s working parameters were calculated, and a finite element model was created to simulate the temperature distribution of the cavity in the ultra-low-temperature environment. The testing system meets all necessary requirements for uniform temperature, vacuum level, and insulation performance. Specifically, it achieves a vacuum level of 10-2 Pa, a temperature tuning range from 78 to 86 K, and a temperature control accuracy of 0.02 K.
-
Research on the calibration method of low temperature rise airflow
Zhao Yijun, Wang Xiaolu, Li Zetao, Jing Zhuoyin, Zhao Jian
Abstract:
The accurate calculation of the efficiency of the aero-engine compression system affects the design and planning of the compression system. At present, the most commonly used method to calculate the efficiency of a compression system is the temperature rise method. However, the uncertainty of the existing temperature measurement method is close to 1 K when the air temperature rise is less than 100 K and the Mach number is less than 0.5. In this case, the accuracy of temperature measurement is seriously affected, and the efficiency accuracy of the compression system calculated by the temperature rise method is insufficient. In response to the current problems of insufficient calibration accuracy and missing calibration methods in low temperature rise airflow calibration, this article proposes a low temperature rise airflow calibration method based on energy conservation and heat dissipation correction. Using a second-class standard platinum resistor with higher accuracy as the reference airflow temperature sensor, it is moved forward from the wind tunnel test section outlet to the stable section and installed on the same axis. The wall heat dissipation temperature loss of the reference airflow temperature sensor is corrected, resulting in low temperature rise calibration uncertainty of the reference airflow temperature sensor for 0.08 K. The low temperature rise airflow calibration method is used to calibrate the airflow temperature sensor to be calibrated. The calibrated platinum resistor airflow temperature sensor has a temperature measurement deviation between ( 0.075~-0.031) K at Mach number 0.398 and a temperature rise of about 50 K in the test after calibration. In accordance with the results of the uncertainty analysis, the calibration device, the calibration method, and the uncertainty evaluation method of low temperature rise airflow are verified. The calibration method of low temperature rise airflow improves the ability to indicate the true airflow temperature of the airflow temperature sensor under the low temperature rise environment, providing strong support for the accurate calculation of the efficiency of the aircraft engine compression system.
-
Non-contact voltage measurement method based on parameter identification of dual reference excitation signals
Chen Xiaotao, Zhang Wenbin, Suo Chunguang, Tan Xiangyu
Abstract:
In practical applications of capacitive coupled non-contact voltage sensors, differences in wire diameter, insulation layer thickness, and the relative position between the wire core and the induction plate can affect the coupling capacitance between the wire core and the induction plate. Additionally, the presence of the wire influences the structural capacitance between the induction plate and the grounding plate, causing edge effects that alter the size of the structural capacitance. As a result, conventional LCR meters cannot accurately measure non contact voltage, resulting in uncertain sensor gain and limited accuracy in non contact voltage measurement. To address this issue, a non-contact voltage measurement method based on dual reference excitation signal parameter identification is proposed to achieve self-calibration of sensor gain during the measurement process. Firstly, an equivalent model with parasitic parameters is presented, and its transfer function is analyzed. Multiple internal parameters of the sensor are simplified into two lumped internal parameters. Through simulation, the influence of the measured wire on the edge effects of structural capacitance and its variation is revealed. Subsequently, a sensor parameter identification method is proposed to obtain the internal parameters of the sensor considering parasitic parameters and wire influences as a fixed parameter for voltage reconstruction to improve voltage measurement accuracy. A sensor prototype is developed, and an experimental platform is constructed to perform parameter identification and conduct multiple validation experiments. Experimental results show that the amplitude error in the amplitude accuracy test is within 1%, and the phase difference in the phase accuracy test is 0.13 °. The wire diameter adaptability experiment confirms that the method accommodates wires of different specifications, with a maximum error of only 0.15%. The interference signal shielding ability tests validate that the coaxial probe with a shielding cavity has good anti-interference performance. This provides an effective solution to improve the accuracy of non-contact voltage measurement.
-
Prediction of tunneling speed of shield machine under varying geological conditions
Wang Yucan, Yuan Haiwen, Sun Qi, Yang Lei, Xiao Changshi
Abstract:
TThe tunneling performance of shield machines is greatly influenced by varying geological conditions. This study investigates an electrically-driven earth pressure balance shield machine, analyzing 3,761,006 tunneling data points across 961 rings. The dataset includes six geological combinations, such as sandy cohesive soil layers, along with corresponding tunneling parameters. Through correlation analysis, key features strongly related to tunneling speed, including total thrust, sync grouting volume, and foam pressure, were identified. To address the issue of uneven data distribution in practical tunneling projects, Gaussian resampling was applied, resulting in a dataset with 19 950 valid samples. A tunneling speed prediction method for shield machines based on the Kolmogorov-Arnold Network (KAN) was then proposed. The KAN model approximates nonlinear relationships by combining multi-level composite functions, breaking down the complex nonlinear interactions into simpler univariate function combinations. This approach ensures high prediction accuracy while significantly improving computational efficiency. Using the Shenzhen-to-Daya Bay Metro Shield Tunneling Project as a case study, experiments showed that the KAN model outperforms CNN and LSTM models in handling high-dimensional data and nonlinear coupling relationships. The prediction results align closely with measured data, with prediction errors ranging from 5.12% to 7.02% in simpler geological conditions (such as completely weathered mixed granite and strongly weathered mixed granite). In mixed geological layers, the prediction errors are higher, but the overall average error remains below 15%. This method offers strong decision support for optimizing shield machine operations under complex geological conditions. In the future, geological spatial distribution data will be incorporated into sequential modeling, and cutterhead wear prediction will be added to provide a more comprehensive intelligent management solution for shield tunneling.
-
A novel dynamic weighing method integrating improved stationary wavelet denoising and extended Kalman system identification
Long Baoxin, Teng Zhaosheng, Sun Biao, Lin Haijun, Liu Tao
Abstract:
During the operation of a checkweigher, its weighing signal is affected by vibrations rising from the mechanical transmission systems, impacts from the measured object and other random disturbances. As a result, the weighing signal is severely contaminated by noise, making it difficult to meet the requirements of national standards. To address this issue, a novel dynamic weighing method based on improved stationary wavelet denoising with a shrinkage soft threshold and extended Kalman system identification is proposed. First, leveraging prior knowledge of the weighing signal and the ideal signal, a seven-layer stationary wavelet transform is applied to the weighing signal for multi-scale decomposition. Next, the high-frequency noise-dominated detail coefficients d1,k~d4,k are set to zero, while a soft-threshold function with a shrinkage factor is applied to process the detail coefficients d5,k~d7,k that contain both useful signal and noise components. Then, the inverse stationary wavelet transform is performed using the processed detail coefficients and the original approximation coefficients to reconstruct the weighing signal, effectively suppressing various interference noises. Following this, the extended Kalman algorithm is employed for system identification to determine the model parameters of the checkweigher system, which are subsequently utilized to calculate the mass of the samples. To validate the effectiveness of the proposed algorithm, experiments were conducted using five samples of different masses at speeds of 30, 45, 60, 75 and 90 m/min, with multiple loading tests performed at each speed, and the results were analyzed and compared. The results demonstrate that the proposed algorithm achieves superior weighing accuracy compared to the time-variant low-pass filter (TVLPF) algorithm, identification-based approach with signal-adaptive prefiltering (AID) algorithm, and signal-adaptive prefiltering with extended Kalman system identification (AEKSI) algorithm. Furthermore, it meets the accuracy requirements for class XIII checkweighers as defined by the national standard “GB/T 27739—2011 Automatic Checkweigher”.
-
Robot path planning by fusion of AIP-RRT* and DGF-APF in dynamic environments
Song Junhui, Liu Yuting, Guo Shijie
Abstract:
To address the problems of slow convergence speed, numerous redundant points in paths, unsuitability for dynamic environments, and the lack of an effective coordination mechanism to integrate global and local planning results leading to a significant increase in path length in mobile robot path planning, a path planning fusion algorithm based on the Adaptive Improved Potential Function Rapidly-exploring Random Tree* (AIP-RRT*) and the Dynamic Gravity Field Artificial Potential Field method (DGF-APF) was proposed. a path planning fusion algorithm based on the adaptive improved potential function rapidly-exploring random tree* (AIP-RRT*) and the dynamic gravity field artificial potential field method (DGF-APF) is proposed. Firstly, an adaptive goal bias probability strategy is constructed, generating new nodes through a heuristic function to improve the search efficiency of the path planning algorithm. Secondly, an adaptive step size function is developed to enhance path exploration capabilities and accelerate the convergence speed of the path planning algorithm. Thirdly, a pruning optimization strategy based on goal backtracking is employed to remove redundant points in the global path, thereby improving path quality. Finally, a fusion algorithm of AIP-RRT* and DGF-APF path planning for dynamic scenarios is proposed to realize the path planning of AIP-RRT* and DGF-APF fusion algorithms by using the global key nodes as local subgoal points for local path planning in dynamic environments, and a synergy mechanism based on the dynamic gravitational field strategy is constructed to synthesize the global and local path planning results to shorten the path length. The results of the combined simulation and real experiments show that the path planning fusion algorithm has better global path planning capability as well as local path planning capability, which enables the robot to better adapt to static as well as dynamic environments. In the real environment, the improved fusion algorithm reduces the path length by 6.34% and the running time by 10.71% compared with the traditional algorithm.
-
Path planning of robotic arm based on RTSR-RRT* algorithm
Liu Xiaosong, Kang Lei, Shan Zebiao, Su Chengzhi, Liu Yunqing
Abstract:
In response to the issues of poor random sampling bias, low path search efficiency, and slow convergence speed in the path expansion of the traditional RRT* algorithm, a redefined sampling region RRT* (RTSR-RRT*) algorithm is proposed. Firstly, a target bias strategy is introduced into the RRT* algorithm to reduce the randomness of sampling and increase the bias of sampling points. Secondly, the offset angle between the expansion node and the target point, along with the density of surrounding obstacle distribution, is converted into an angle based on the duty cycle. This angle is then superimposed, and the expansion node is used as the vertex, with the line connecting to the target point as the bisector, to bisect the sum of the two angles, thereby redefining the sampling region. This redefinition narrows the sampling space and enhances the efficiency of path search. Furthermore, a secondary sampling is conducted within the redefined sampling region. By leveraging the fixed gravitational force of the target point and the variable gravitational force of the sampling points, the growth direction of new nodes is optimized, further increasing the bias of path expansion and accelerating the convergence speed of the algorithm, ultimately generating the planned path. To validate the superiority of the proposed algorithm, comparisons were made with the RRT*, informed-RRT*, GB-RRT* and AEC-RRT* algorithms. The results indicate that compared to the RRT* algorithm, planning time is reduced by 35%, and the number of sampling points is decreased by 58%; compared to the informed-RRT* algorithm, planning time is reduced by 40%, and the number of sampling points is decreased by 50%; compared to the GB-RRT* algorithm, planning time is reduced by 29%, and the number of sampling points is decreased by 54%; and compared to the AEC-RRT* algorithm, planning time is reduced by 31%, and the number of sampling points is decreased by 53%. Finally, the planned path was tested on a robotic arm platform, further verifying the effectiveness of the proposed algorithm.
-
Path planning of mobile robot based on the dynamic optimization ant colony algorithm
Abstract:
The path planning algorithm is a key component in the research of mobile robots. The ant colony algorithm is indeed a relatively mature algorithm. To address the problems existing in the path planning algorithm of mobile robots, such as slow convergence speed, numerous turning points, and poor stability, an improved dynamic optimization ant colony algorithm (IDOACO) is proposed. First, heuristic information with directional guidance is introduced to enhance the purposefulness of path planning through the angle guidance factor. Secondly, an obstacle exclusion factor and safety factor are incorporated into the pseudo-random state transition probability to improve path safety. Furthermore, a multi-objective evaluation function is proposed to balance the path length and energy consumption to achieve global optimization of path planning. Finally, a dynamic obstacle avoidance adjustment module is formulated to assess and adjust the path in real time, enabling instant dynamic obstacle avoidance functionality. Simulation experiments are implemented to compare the IDOACO algorithm. Compared with the existing algorithms, experimental results show that, in a complex map environment, the IDOACO algorithm improves the average path length by approximately 4.63% and 11.78%, and the standard deviation of the convergence speed is increased by 55.21% and 66.27% respectively. Experiments show that the shortest path generated by the IDOACO algorithm not only converges faster, the number of turns is less, but also has higher stability and convergence accuracy. Then, the dynamic obstacle avoidance effect is successfully verified. Finally, the improved algorithm is applied to the ROSMASTER-X3 mobile robot, and different target points are set for actual path planning. Experimental results show that the algorithm can effectively solve the problems faced by the mobile robot in path planning, and has certain practical application value.
-
Research on robot path planning based on dual-resolution grid maps
Abstract:
To enhance terrain adaptability and storage efficiency in mobile robot path planning on non-flat terrains, an improved A* path planning algorithm based on the dual-resolution hierarchical grid map is proposed. This map includes a high-resolution obstacle layer using binary representation for rigid obstacles and a low-resolution elevation layer quantifying terrain undulations via a digital elevation model. On this basis, the A* algorithm is improved by reconstructing its dynamic weighted composite cost function. The improved algorithm introduces three optimization dimensions into the mobility cost function, including slope constraints, energy consumption weight, and safety factor. The heuristic function is extended to a multimodal evaluation metric that integrates spatial distance, root mean square slope, and terrain risk values. A distance-sensitive dynamic weight adjustment strategy is designed, and the Sigmoid function is utilized to achieve a smooth transition between global heuristic search and local path optimization. Experiments show that within a rectangular mapping range of 700 m×700 m, the dual-resolution hierarchical grid map structure reduces storage load by 61.7% compared with a three-dimensional grid map. Compared with traditional A* algorithms, this method reduces the standard deviation of elevation fluctuations in planned paths by 38.9%. Real robot experiments demonstrate that this method effectively avoids steep slopes and obstacles. Engineering application experiments indicate that this method reduces memory usage by 62% in large-scale unstructured scenarios such as oilfield inspections, with path planning response times under 6.9 s and the planned paths exhibiting gentle low undulation characteristics.
-
A laser SLAM algorithm for unmanned vehicles based on depth map in dynamic environment
Sun Mingxiao, Wang Xinyuan, Luan Tiantian, Wang Xiao, Liu Pengfei
Abstract:
To address the problem of accuracy degradation and poor system robustness of laser SLAM in a dynamic environment, a laser SLAM algorithm for unmanned vehicles based on a depth map in a dynamic environment is proposed. Firstly, aiming at the problem that some ground point clouds are misjudged as dynamic obstacles when the laser incident light is close to the ground, the laser point cloud is divided into ground point cloud and non-ground point cloud by angle threshold, excluding the influence of ground point cloud. Secondly, the non-ground point cloud is projected into a depth map, and the point cloud in the 3D is projected onto a two-dimensional image. The complexity of point cloud data processing is reduced by using the image domain method. Then, to address the issue that the sparse laser point cloud cannot accurately reflect the real environment, the range information of the LiDAR is used as the weight of the adaptive adjustment depth map to set the resolution of the depth map. The depth map in different distance intervals has different resolutions, and the dynamic obstacles can be accurately identified in different distance intervals, which improves the efficiency of the system in identifying dynamic barriers. Finally, the local map is subtracted from the depth map formed by the LiDAR query frame, the dynamic obstacles are removed by the obtained disparity map, and the static points that are mistakenly deleted are recovered by reducing the resolution of the depth map to get a high-precision static map. Experimental results show that the dynamic obstacle removal score of the proposed algorithm is 94.13% in the simulation environment. The dynamic obstacle removal score of the proposed algorithm is 95.22% in the KITTI dataset. The dynamic obstacle removal score of the proposed algorithm is 96.43% in small scenes and 93.30% in large scenes. The algorithm can efficiently remove dynamic obstacles and improve map accuracy and system robustness.
-
Dynamic factor-influenced surface roughness prediction in robotic grinding
Guo Wanjin, Hao Qinlei, Xu Mingkun, Cao Chuqing, Zhao Lijun, Wang Li
Abstract:
The robotic grinding process is affected by both dynamic and static factors and has dynamic characteristics such as complex coupling and high time-varying nonlinearity. To solve the problems of difficult feature selection of dynamic factors and low prediction accuracy of surface roughness caused by only considering static factors, a prediction method of surface roughness of robotic grinding considering the influence of dynamic factors is proposed by combining deep learning technology. Firstly, the convolutional neural network is used to automatically extract the spatial features of dynamic factors in the grinding process, and capture complex dynamic behaviors of robotic grinding. The temporal features are extracted from the obtained spatial features through the bidirectional long short-term memory network to characterize the dynamic changes of robotic grinding. The attention mechanism is introduced to realize the automatic weight distribution of spatial features, temporal features and static factors. The improved whale optimization algorithm is used to adaptively optimize the hyperparameters of the bidirectional long short-term memory network to improve the convergence speed and adapt to the dynamic changes of robotic grinding. Secondly, according to the proposed prediction method, an IWOA-CNN-BiLSTM-Attention surface roughness prediction model considering the influence of dynamic factors is formulated. Thirdly, the robotic grinding experiment is carried out. The spatial and temporal characteristics of the extracted dynamic factors, the collected static factors, and the measured values of surface roughness are normalized to construct the experiment dataset. Finally, the experimental dataset is input into the prediction model for model training, and the surface roughness prediction of robotic grinding considering both dynamic and static factors is realized. The effectiveness of the proposed method is evaluated by comparative experiments. The mean absolute percentage error, root mean square error, and coefficient of determination of the corresponding prediction model are 0.027 6, 0.029 5, and 0.998 8, respectively. Compared with the comparison prediction model, the prediction accuracy is improved by 17.14%, 13.65%, and 21.35%, respectively. compared with the comparison prediction model.
-
Human-computer interaction follow-up control based on sEMG
You Bo, Liu Jiaqi, Cheng Chenchen, Liu Yufei, Li Jiayu
Abstract:
To solve the problems of poor follow-up in the gripping task of the manipulator under the condition of human-computer interaction, resulting in low operation fluency and unsatisfactory trajectory tracking, a human-computer interaction follow-up control method based on surface electromyography (sEMG) is proposed. Firstly, the angle of shoulder and elbow joints are obtained by using the IMU data of dual gForcePro+ arm rings. By combining this data with the feature extraction of sEMG signals, the angle of wrist joints is estimated by the PSO-GRNN model, establishing a mapping relationship between the human arm and the robotic arm to realize the follow-up control. Experimental results show that the Root Mean Square Error (RMSE) of the PSO-GRNN model in wrist joint angle estimation is reduced by 62-39%and 55-18%, respectively, compared with the traditional GRNN method, effectively improving the control accuracy. To further enhance the control accuracy in the grasping task, a gesture recognition method based on a CNN-LSTM network is proposed to realize the real-time control of the gripper. At the same time, a stiffness estimation algorithm for the human upper limb is constructed by leveraging the mapping relationship between sEMG signals and actual stiffness. The stiffness adjustment information is then introduced into the adaptive RBF-NFTSMC controller to realize the compliant control of the robotic arm. Experimental results show that the optimized RBF-NFTSMC method reduces the trajectory tracking error by about 30.2% compared with the traditional NFTSMC method, enhancing the anti-interference ability of the system. In addition, in order to verify the effectiveness of the sEMG variable stiffness control strategy, an experimental platform based on dual gForcePro+ arm rings and UR3e robotic arms was built. Experimental results show that the end trajectory of the manipulator based on sEMG variable stiffness control was closer to the target trajectory, with trajectory tracking error reduced by 24.6% compared with the fixed stiffness control method. Furthermore, the flexibility of the manipulator in object interactions was improved, leading to improved stability and adaptability.
-
Research progress on visual detection methods for bolt/rivet faults
Liu Chuanyang, Wu Yiquan, Liu Jingjing
Abstract:
Bolts and rivets serve as essential fasteners in engineering applications, including transmission lines, railway transportation, bridges, and aircraft. However, exposure to external environmental factors makes them susceptible to various faults, such as missing pins, loose nuts, corrosion, and structural damage. Accurately detecting these faults is crucial for ensuring the safe and stable operation of transmission lines, railway systems, aircraft, and other related infrastructures. Leveraging large-scale data, deep learning-based bolt and rivet fault detection employs convolutional neural networks (CNNs) to automatically extract deep image features through hierarchical learning. By optimizing network parameters, these methods enhance feature extraction and generalization capabilities, yielding superior detection performance compared to traditional image processing techniques. This paper provides a comprehensive review of vision-based bolt and rivet fault detection research over the past decade. It begins by outlining common fault characteristics and the challenges associated with visual inspection. Next, deep learning-based detection approaches are categorized into three main types: two-stage algorithms, one-stage algorithms, and cascaded detection models. The paper then explores visual fault detection methods in key application scenarios, including line-type, box-type, and component-type bolts and rivets. Finally, it discusses challenges in machine vision-based fault detection, such as dataset limitations, sample annotation, and small target detection. By integrating existing deep learning technologies with the latest research advancements, this study presents an in-depth analysis of future development trends in deep learning-based bolt and rivet fault detection.
-
High-precision localization and robust control for single-steering-wheel AGV based on 2D laser SLAM and pure pursuit method
Abstract:
Automated guided vehicles (AGVs) play a crucial role in intelligent logistics, where localization accuracy and motion robustness directly impact system efficiency. A critical challenge in AGV research is achieving accurate localization and robust control using simultaneous localization and mapping (SLAM) while avoiding the reliance on fixed markers. This paper proposes a high-precision localization and robust control method for single-steering-wheel AGVs based on two-dimensional (2D) laser SLAM and a pure pursuit tracking approach. To enhance environmental perception, a laser sensor is mounted on top of the AGV, and a 2D occupancy grid map is constructed for localization in structured indoor environments. Motion control is achieved through a preview-distance-based pure pursuit algorithm combined with flexible acceleration-deceleration strategy, ensuring smooth trajectory tracking. Experimental validation in real-world indoor logistics environments demonstrates the system′s accuracy, achieving a localization precision of ±5 mm over 2 000 laser-based positioning instances, a straight-line trajectory tracking accuracy of 25 mm, and a repeatability precision of ±6 mm across 120 task executions. The proposed system provides a high-precision, marker-free localization and control solution, supporting cost-effective AGV upgrades and advancing intelligent logistics.
-
Research on coal gangue elemental composition analysis algorithmbased on CT imaging
Lyu Ruihong, Li Dawei, Shen Hongbo
Abstract:
To overcome the challenges presented by conventional coal gangue images, which often lack significant textural and color differences and are prone to noise interference (such as dust), making it difficult to extract critical information about their internal structure and elemental distribution through standard imaging methods, a high-resolution CT imaging dataset of coal gangue with strong anti-interference capabilities has been developed. This dataset allows for a more detailed analysis of the internal structure of coal gangue. Furthermore, to address the issue of low classification and recognition accuracy of traditional coal gangue images using deep learning algorithms, a new algorithm for analyzing the elemental composition of coal gangue based on CT images is proposed. The algorithm utilizes an enhanced Res-Unet semantic segmentation model to segment the elemental regions within CT images and analyze their proportions, enabling effective classification and recognition of coal gangue. The model incorporates an efficient local attention (ELA) module within the Res-Unet encoder, allowing it to focus more on important features. Additionally, improvements to the skip connections in the Res-Unet model facilitate better information fusion across different scales, significantly boosting segmentation performance and ensuring accurate delineation of elemental regions in coal gangue CT images. Experimental results demonstrate that the enhanced Res-Unet model successfully segments elemental regions, achieving an mIOU of 84.48%. By calculating the proportions of elemental regions for the final classification of coal gangue CT images, the improved model achieves a classification accuracy of 94.4%, outperforming other models. These results confirm the effectiveness of the proposed algorithm for analyzing the elemental composition of coal gangue based on CT images. This algorithm provides a novel approach and methodology for coal gangue image classification, offering valuable technical support for intelligent coal sorting in factories and promoting the advancement of intelligent and automated systems in the coal industry.
-
Research on workpiece size detection method with binocular vision system carried by robot
Wang Jindong, Xie Chengsheng, Zhang Xingjian, Zheng Peng, Tang Leiyu
Abstract:
As the manufacturing industry rapidly advances, the demand for precise workpiece size measurement continues to grow. Efficient and accurate three-dimensional measurement of workpieces is crucial for ensuring processing quality. This paper proposes a detection method for the three-dimensional size of machined workpieces using a robot equipped with a binocular vision system. A flange plate is chosen as a typical detection object, and a visual detection system is developed along with a corresponding workpiece size detection algorithm based on binocular vision. To address the issue of highlight and noise interference in flange images, which leads to pixel value contamination, a gray-level aggregation algorithm is introduced. This algorithm improves the robustness of stereo matching cost calculations by detecting and reconstructing contaminated pixel values. Additionally, to tackle the challenge of large matching errors caused by weak or repeated image features in the flange, a weight adaptive calculation algorithm is proposed to enhance stereo matching accuracy by effectively characterizing pixel features. Building on this, an AD-Census stereo matching optimization algorithm is developed, combining gray-level aggregation and weight adaptive calculation, with its effectiveness validated through flange size detection experiments. Furthermore, by analyzing the transfer process of parallax errors during flange visual inspection, an evaluation model for camera measurement pose is established, allowing the determination of the optimal measurement pose. Flange size detection experiments under different poses confirm the effectiveness of the proposed pose optimization method. The results show that the proposed method significantly improves workpiece size detection accuracy and offers a new technical approach for three-dimensional size measurement of machined workpieces.
-
Small sample foreign body detection in power lines based on double coding and meta-learning
Chen Zhexuan, Gao Xuelian, Song Jiayu, Liu Yi
Abstract:
Foreign body detection is a critical component of power grid inspection and maintenance, as it plays an essential role in power transmission. However, detecting foreign objects in transmission lines with small sample data under complex environmental conditions remains a challenging task. This paper proposes a Meta-Learning-based Double Coding Target Detection Network (ML-DCTDN), combining a Swin Transformer Network and a Convolutional Neural Network (CNN). The innovation of this network lies in two key aspects: firstly, the Swin Transformer network enhances its generalization feature extraction ability through a two-stage meta-learning process. In the first stage, it learns transmission line features, while in the second stage, it focuses on foreign object features, improving performance for target detection tasks on small sample datasets. Secondly, the double coding network uses both RGB and grayscale images as inputs, and employs a Layered Fusion Module (LFM) and a Feature Pyramid Network (FPN) to achieve multi-modal feature fusion. This approach leverages the rich color and texture information of RGB images while also utilizing the robustness of grayscale images against lighting variations and fine details. The model′s anti-interference and detection capabilities are thus strengthened in complex backgrounds. Ablation experiments reveal that the meta-learning strategy significantly improved the Mean Average Precision (mAP), with grayscale image input increasing the mAP by at least 4%. Comparative experiments with SSD, Faster RCNN, YOLOv5, and YOLOv8 algorithms demonstrate that the proposed meta-learning strategy and double coding network structure greatly enhance detection accuracy in foreign body detection tasks for transmission lines with small sample datasets. The mAP50 and mAP75 values achieved were 98.6% and 64.7%, respectively.
-
Enhanced hierarchical multi-scale feature fusion model for metal defect classification
Li Jitong, Liu Jie, Yang Na, Wang Zining
Abstract:
Metal defect detection, as a critical component of industrial quality control, directly determines the advancement of intelligent manufacturing. To address existing issues in feature fusion modules including feature information loss, insufficient cross-scale interaction, and low recognition accuracy, a hierarchical multi-scale feature fusion-based classification model is proposed. By integrating complementary advantages of Swin Transformer and ConvNeXt architectures, a hierarchical perception-enabled feature learning network is constructed. Specifically, the Swin Transformer employs shifted window mechanisms and multi-stage self-attention to effectively capture global features, while ConvNeXt utilizes depth separable convolution and efficient convolutional operations for precise local feature extraction. To achieve efficient global-local fusion, an innovative adaptive hierarchical feature fusion layer is designed, incorporating channel attention mechanisms, spatial attention mechanisms, and multi-scale fusion strategies to enable effective multi-level feature integration. Additionally, a multi-layer inverted residual fusion module is incorporated to dynamically adjust feature extraction, ensuring precise and reliable feature fusion. Experimental validation on public NEU-DET and GC10-DET datasets demonstrates superior performance with accuracy rates of 99.6% and 96.9%, respectively. To verify generalization capability, evaluations on a self-constructed dataset achieve an accuracy of 99.8%, outperforming mainstream models including ConvNeXt, Swin Transformer, VGG16, and ResNet34 by 3.4%, 2.3%, 4.3%, and 2.7% respectively. The results confirm that the HMFF model exhibits enhanced classification accuracy and robustness in metal defect detection, providing a novel methodological framework for high-precision industrial defect inspection.
-
Research on intelligent diagnosis of high resistance faults in permanent magnet synchronous motors based on electromagnetic multidimensional characteristics
Wu Zhenyu, Zhang Jie1, Wang Hui, Hu Cungang, Cao Wenping
Abstract:
Permanent magnet synchronous motor (PMSM) is subjected to frequent electric-magnetic-force-thermal shocks for a long period of time, which accelerates the aging of winding insulation and leads to the occurrence of high-resistance connection (HRC) faults. HRC faults further induce more serious damages to the PMSM, making their accurate diagnosis highly significant. At present, based on the evolution law of PMSM operating voltage and load current, it can provide reference for accurate identification of HRC. However, the above are invasive methods, potentially causing interference with the normal operation of the motor. Since HRC faults significantly alter the electromagnetic field distribution within the motor, the leakage signals in the motor space can serve as an alternative, non-invasive source of state information. By acquiring these leakage signals, a non-invasive approach to diagnosing HRC faults can be realized. To this end, an intelligent diagnosis method for PMSM high-resistance faults based on electromagnetic multidimensional spatio-temporal characteristics is proposed. This method establishes the correlation relationship between spatial leakage signals and the motor′s differentiated state, enabling the intelligent motor state evaluation by a joint intelligent algorithm. First, the electromagnetic signal evolution law under fault is analyzed based on the equivalent circuit model of the motor winding, identifying the optimal electromagnetic test point. Secondly, the feature image conversion and dimensioning method based on the array of leakage signals are proposed to diagnose motor faults with GoogLeNet network. Finally, the proposed method is verified by simulation model and experimental platform. The experimental results show that the feature image upscaling and intelligent assessment method based on the leakage signal array can accurately identify and locate HRC while also assessing fault severity, achieving an accuracy rate of up to 97%, which verifies the effectiveness of the proposed method. The method has the advantages of non-invasive and high accuracy, and has a wider application prospect for PMSM.
-
Wind turbine condition monitoring based on improved multilayer self-organizing map
Jin Xiaohang, Yang Yuchen, Yu Xuanang
Abstract:
This paper proposes an improved multilayer self-organizing map (MLSOM) network-based method for wind turbine condition monitoring, addressing the lack of consideration for the interrelationship and information transmission between the unit and its key components in existing methods. Initially, the Pearson correlation coefficient is employed to select features from the supervisory control and data acquisition (SCADA) system, with the feature information serving as the input for the wind turbine′s tree structure′s bottom node. Recognizing the nonlinear and time series nature of wind turbine data, long short-term memory (LSTM) models are developed and trained using historical data to predict SCADA feature values. The prediction residuals replace the feature information as input to the bottom node of the self-organizing map (SOM) network within the MLSOM model, creating a normal behavior model for each component. The minimum quantization error serves as an indicator for assessing component health using the trained SOM model. Monitoring models are established for key components such as the generator, gearbox, and converter. These component health indicators are then integrated as top-level node information and used as input for the top-level SOM model in the MLSOM to form a normal behavior model for the entire wind turbine, yielding a comprehensive health indicator for condition monitoring. Case study results of two wind turbines demonstrate that the proposed method effectively transmits and aggregates component information step by step, enabling the condition monitoring of the entire wind turbine.
-
Reconfigurable topology for electric vehicle wireless charging systems tolerating 400 V or 800 V battery
Liu Chao, Zhou Mingzhu, Chen Xiaoying, Zhang Yiming
Abstract:
As more electric vehicles (EV) companies jump on the bandwagon of wireless power transmission, interoperability issues are becoming increasingly apparent. In addition to the common problems of coupling structure, compensation network and communication protocol interoperability, battery voltage (BV) interoperability issues are also gradually emerging, that is, different battery voltage levels of various models. For the BV level of the existing EV power battery, the rated voltage can be mainly divided into 400 and 800 V. Considering the interoperability of public radio energy transmission facilities, EV radio energy transmission systems must be effectively compatible with 400 and 800 V BV levels at the same power level. A reconfigurable topology with the BV interoperability is proposed. The device uses two completely stacked unipolar coils as magnetic coupling structures, enabling two different output voltages under the same power level by switching. The proposed system is analyzed and modeled by establishing a mathematical model, followed by the development of a 1.3 kW downscaled experimental prototype to verify the function of the proposed reconfigurable topology. The results show that the system can switch between 200 and 400 V output voltages while maintaining an output power of 1.3 kW. The maximum system direct current-direct current (DC-DC) efficiency is 90.97% and 95.51%, respectively. Finally, to assess the migration performance of the proposed structure, additional testing was conducted using the experimental prototype. The results show that the DC-DC efficiency of the proposed system is higher than 90% in the migration range of ±100 mm.
-
A predictive maintenance decision-making method for AC contactors considering failure threshold randomness
Abstract:
AC contactors are widely used in control systems. An efficient maintenance strategy is the premise to ensure the safe and reliable operation of the system. Due to the inevitable differences of three-phase contacts of AC contactors, the failure threshold of each phase contact is random. However, the existing maintenance decision-making methods do not consider the randomness of the failure threshold, which are not suitable for AC contactors. To solve this problem, this article uses the cumulative arcing Joule integral to characterize the performance state of each phase contact of AC contactor and formulates a performance degradation model considering the randomness of failure threshold, competing failure, and degradation correlation of three-phase contact. A method for estimating the parameters of the degradation model, updating the distribution of the three-phase contact failure threshold and residual life based on maximum likelihood estimation and conditional probability formula, is proposed. A predictive maintenance strategy considering the randomness of the AC contactor failure threshold and the update of the maintenance strategy is proposed. The predictive maintenance strategy takes minimizing the expected cost rate as the optimization goal and adaptively optimizes the preventive replacement threshold of AC contactor three-phase contacts at each predictive maintenance strategy updating time. Finally, the effectiveness of the proposed maintenance strategy is evaluated by case analysis. The results show that, under the predictive maintenance strategy, the expected cost rate of AC contactors can be approximately divided into fluctuation stage, stable stage, and rising stage. The actual cost rate is reduced by about 10% compared with the existing maintenance strategy, indicating that the strategy can make more efficient use of AC contactors and has a better economic effect.
-
Research on variable voltage drive control method for high-speed on/off valve considering temperature feedback
Tian Zuzhi, Fan Rongrui, Guo Yangyang, Xu Chunjie, Xie Fangwei
Abstract:
Good dynamic characteristics are the key to accurate flow control of high speed on/off valve. However, the existing drive control strategy of high speed on/off valve ignores the influence of temperature rise generated by variable voltage excitation on the dynamic response characteristics of high speed on/off valve. In order to improve the dynamic response and improve operational reliability, a variable voltage drive control method based on temperature feedback is proposed in this paper. Firstly, the electromagnetic thermal coupling simulation model of high-speed on/off valve is established based on Maxwell and thermal, and the influence of coil turns, driving voltage and ambient temperature on coil temperature rise are analyzed. The results show that an increase in drive voltage exacerbates coil temperature rise. Secondly, the influence of temperature change on the dynamic response characteristics of high speed on/off valve is further explored. The results show that with the increase of temperature, the opening delay time increases, while the closing delay time decreases. Based on this, a variable voltage drive control strategy considering temperature feedback is proposed, and a simulation model is built in Simplorer to verify the effectiveness of the control strategy. Finally, the performance test bench of high speed on/off valve is set up, and the control effect of single voltage control and variable voltage control strategy are compared through experimental analysis. The test results show that, compared to traditional single-voltage control, the proposed variable-voltage control strategy based on temperature feedback reduces the closing lag time of the high-speed on/off valve by 5.55 ms and lowers the steady-state coil temperature by 24.5℃. These improvements effectively improve the dynamic response characteristics and operational reliability of the high-speed on/off valve.
-
Research on the prediction method of milling force for micro-textured tools under thermally-assisted conditions
Tong Xin, Wang Baiyi, Li Xinyu, Yang Shucai
Abstract:
The texturing of the tool surface can significantly improve the cutting performance of the tool. However, laser processing is characterized by rapid heating and quenching, which can lead to problems such as remelting layer stacking and microcracking. In order to solve the above problems, heat-assisted laser processing technology is introduced in this paper. Since titanium alloy is a difficult-to-machine material, the tool is subjected to large milling forces during the milling process, which leads to the dynamic response and vibration of the mechanical system, which in turn affects the tool life and the machined surface quality. Therefore, accurate prediction of the milling force can adjust the cutting parameters in time, ensure the machining quality at the same time, and make the milling force in a reasonable range, to improve the processing efficiency and reduce tool wear.In summary, this study takes the cemented carbide ball nose milling cutter as the research object, combines the heat-assisted process and laser processing technology, builds a milling test platform, and proposes a method based on the dung beetle algorithm (DBO) to optimize the variational mode decomposition (VMD) parameters, and combines the wavelet packet threshold noise reduction (WPT) method to denoise the original signal. The time-frequency analysis was carried out by using the Hilbert-Huang transform (HHT) to explore the variation of tool milling performance under different thermal auxiliary temperatures. On this basis, combined with Bayesian optimization (BO), convolutional neural network (CNN), bidirectional long short-term memory network (BiLSTM) and multihead-attention mechanism, a regression analysis model is formulated for real-time monitoring and prediction of milling force. Through verification, the R2 value of the model reaches 0.996 7 on the training set and 0.991 94 on the test set, which proves the accuracy of the model.This study proposes a new method for defect repair in the process of microtexture preparation and provides an effective method for the prediction of milling force in titanium alloy milling.
-
A non-contact temperature sensor for tiny-point-heat source based on micro-junction thermo-optic effect
Liu Mei, Bai Xin, Wang Jiacheng, Zhou Xiaotong, Wang Zhiming
Abstract:
Traditional temperature measurement methods for point heat sources include thermocouples, resistance temperature detectors (RTDs), and infrared thermometers. However, these techniques are often limited by their relatively large size, slow response time, or difficulty in achieving precise measurement in small areas. To address these limitations, a temperature sensor based on a photonic waveguide cross-junction and the thermo-optic effect is designed and developed, enabling non-contact temperature measurement of small-sized point heat sources. Photosensitive resin junctions with various morphologies are fabricated using 3D printing. A laser beam is directed at the cross-junction, which disperses the light into its branched structures. An alumina ceramic heating plate serves as the point heat source, providing a small-range, constant temperature, with adjustable temperature control via a variable power supply. The heat emitted from the source alters the refractive index of the micro-junction material (i.e., the thermo-optic effect), thereby changing the intensity of the light emitted from the branches. The change in light intensity is detected in real-time using a photodiode. Experimental results confirm the sensor′s temperature sensitivity and linear response characteristics. The results show that the sensor exhibits excellent measurement accuracy within a small temperature range (330℃~554℃), with good stability and repeatability. In the temperature range of 465℃~554℃, the sensor shows a strong linear relationship, with a detection sensitivity of -9.4 mV/℃, and the repeatability error (δR) ranges from 1.41% to 2.11%. This scheme presents a novel method for temperature detection of micro-nano point heat sources with a simple structure and low cost, providing a new approach for the design of miniaturized, highly sensitive, and non-contact temperature sensors. The methodology exhibits substantial application potential specifically in temperature monitoring for healthcare, laser processing, and 3D printing.
-
Frequency splitting compensation of hemispherical resonator gyro based on NSGA-III
Gao Zhongfeng, Li Pinghua, Qiao Qi, Liao Jialuo, Zhuang Xuye
Abstract:
The hemispherical resonator gyroscope (HRG) is currently the most accurate type of vibrating gyroscope. However, manufacturing defects inevitably cause uneven circumferential distribution of parameters such as mass, stiffness, quality factor, density, elastic modulus, and damping in the resonator, leading to frequency splitting and coupling errors between primary and secondary vibrations. Traditional frequency splitting compensation methods often reduce the quality factor and involve high costs and complex operations. In contrast, an electrostatic balance compensation scheme has been proposed, which applies electrostatic forces to different electrodes to adjust the resonator′s stiffness and compensate for frequency splitting. When combined with the NSGA-III multi-objective optimization algorithm, this approach optimizes the compensation parameters while considering the impact on resonator performance, power consumption, and the achievable frequency splitting compensation values. Validation results demonstrate that this method provides optimal compensation for frequency splitting in various resonators and frequency splits, achieving a 50.2% increase in compensation value. It also reduces required compensation voltages by 6.3% and 56.3%, maintaining an accuracy better than 0.5 mHz. After compensation, measurement errors decreased by an order of magnitude, with only a 2.3% reduction in natural frequency. This approach significantly enhances the dynamic performance of gyroscopes and offers valuable insight for optimal frequency splitting compensation in HRGs. It can also be applied to other types of gyroscopes, such as cup and ring gyroscopes.
-
Sensorless vector control of induction motors based on phase-locked loop
Ji Juanjuan, Cui Yanliang, Wang Kaiyun
Abstract:
This paper presents a sensorless vector control strategy that utilizes a sliding mode observer (SMO) and a phase-locked loop (PLL) for rotor flux and speed estimation in induction motors. While traditional SMOs excel in flux estimation, they tend to suffer from chattering due to their inherent switching characteristics, which can compromise system stability and control precision. To overcome this limitation, the paper introduces a super-twisting SMO for current and flux observation and enhances it by incorporating a pre-filter to mitigate high-frequency chattering. This improvement enhances the smoothness and phase angle accuracy of flux estimation, increases adaptability to motor parameter variations, and reduces the impact of harmonic disturbances. The pre-filter effectively suppresses high-frequency noise, improving flux estimation accuracy and ensuring robust dynamic performance across varying operating conditions. For speed estimation, an enhanced PLL is proposed, with an optimized structure to improve frequency tracking at low and variable speeds, while effectively eliminating steady-state errors in ramp frequency inputs. This results in high-precision speed estimation and rapid dynamic response. Additionally, the enhanced PLL improves the system′s adaptability to motor operating conditions, ensuring more stable and reliable speed observation and boosting control performance and disturbance rejection capability. Experimental results demonstrate that the proposed method reduces flux waveform distortion by about 20% compared to conventional SMO-based methods, significantly enhancing system robustness. The method shows excellent performance across various operating conditions, not only improving sensorless vector control accuracy but also enhancing motor reliability, providing a practical and effective solution for engineering applications.
-
Multi-parameter optimisation of comb capacitive pressure sensors structure based on BP+ NSGA-Ⅱ
Liang Ruimei, Li Pinghua, Liu Yang, Miao Jiaqi, Zhuang Xuye
Abstract:
To address the challenges of low sensitivity and the difficulty in simultaneously optimizing sensitivity and range in comb-type capacitive pressure sensors, this paper proposes a novel beam-membrane structured comb-type capacitive pressure sensor. An optimization approach combining curve fitting with BP (Backpropagation) and NSGA-II (Non-dominated Sorting Genetic Algorithm II) methods is utilized to enhance the sensor′s performance. By introducing anchor points and cantilever beams to the diaphragm, creating a lever amplification structure, and connecting the movable comb fingers to the lever′s output, the displacement of the comb fingers is amplified, improving sensitivity. To handle the high dimensionality and substantial computational demands of the dataset, MATLAB is employed for data fitting and quantitative analysis of the structural and performance parameters. A correlation analysis between geometric parameters (such as anchor points and cantilever beams) and performance metrics identifies key factors influencing sensor performance, allowing for the elimination of redundant variables and reduction of dataset complexity. The dimensionality reduction process decreases the dataset from 14 to 6 dimensions without compromising accuracy, thus enhancing data collection efficiency and reducing computational resource consumption. The reduced dataset is trained using a BP neural network, and the NSGA-II algorithm is applied for co-optimization of sensitivity and range, improving output reliability. The results show that within the 0~50 kPa range, the optimized sensor achieves a sensitivity of 0.106 pF/kPa, a 30.4% improvement, with a non-linearity error of 0.4% F.S. This optimization methodology provides valuable insights for refining complex structures with multiple parameters. The proposed sensor, with its enhanced sensitivity and reduced nonlinearity, offers an innovative perspective for advancing MEMS pressure sensor technology.
-
Research on the self-sensing demodulation method of magnetic bearing rotor displacement based on periodic peak current
Huang Fei, Zhang Lieping, Zhong Zhixian, Liu Peng, Wei Lanyu
Abstract:
At present, the current ripple is usually used as the demodulation signal in the self sensing shaft displacement detection technology of magnetic bearings. However, this method has a strong dependence on the quality of the ripple. The analytical formula is more complex, and the sampling requirements of the controller are higher. To improve the detection accuracy of the self-sensing shaft displacement of magnetic bearings, a periodic peak current demodulation method is proposed based on the buck chopper circuit. Firstly, the high frequency pulse width modulation voltage is input to the magnetic pole coil of the magnetic bearing. The nonlinear relationship between the peak current and the coil inductance in a single current cycle of the high-frequency pulse width signal is established. Then, the Newton-Raphson method is used to iterate the nonlinear values of the relationship. Finally, the displacement of the magnetic bearing rotor is calculated by combining the iterative results with the inductance formula of the magnetic pole coil of the magnetic bearing. Simulation and experiments show that the dynamic self-sensing rotor displacement real-time detection of the magnetic bearing controller can effectively track the rotor displacement detection signal of the eddy current sensor. The displacement fluctuation error of the two is less than the minimum error required by the magnetic bearing suspension control. In the displacement demodulation experiment of 0.8 mm air gap, when 5~15 kHz high-frequency pulse width modulation voltage with different voltage amplitudes is used as the self-sensing detection signal of the magnetic bearing rotor, the error between the static self-sensing rotor displacement demodulation value of all detection signals and the static demodulation displacement value of eddy current sensor is within the controllable range. When 10 kHz high-frequency pulse width modulation voltage is used as the detection signal, the maximum error between the static self-sensing displacement demodulation value of magnetic bearing and the static displacement demodulation displacement value of the eddy current sensor is not more than 24.7 μm, and the minimum is 0.9 μm.
-
Online extraction of acceleration of gravity while drilling based on MIGJO
Yang Jinxian, Yang Xiaojian, Lin Yuke, Zhang Ying
Abstract:
To extract gravitational acceleration during drilling, the Magnetic Inertia Golden Jackal Optimization (MIGJO) algorithm is employed. Initially, the vibration characteristics during drilling are analyzed, and a gravity extraction model is established by categorizing non-gravity accelerations into solution vectors. Then, based on the output characteristics of the magnetic inertial sensor during drilling, an objective function for the ideal gravitational acceleration is defined, along with constraint conditions such as the gravity angle and tangent Pearson coefficient of the drilling tool diameter. Utilizing the Golden Jackal Optimization (GJO) algorithm, the solution vector from the previous step initializes a dynamically scaled random walk population, reflecting the random variations of non-gravity accelerations during drilling. A gravity factor balance algorithm is developed to perform global search and local refinement using the relative error of gravity modulus and trigonometric functions. Additionally, an attack-defense coefficient is introduced to manage the magnetic inertia golden jackal′s behavior, optimizing both attack and defense strategies to improve gravity extraction accuracy and speed. The attack strategy, based on the positions of the optimal and suboptimal solutions, enhances accuracy, while the defense strategy, utilizing upper and lower bounds and mutation points, helps the algorithm avoid local optima. The similarity between the current gravity solution and local gravity design is used to dynamically adjust the solution vector′s position, further refining the accuracy of gravity extraction. Simulated and real-world drilling experiments demonstrate that MIGJO significantly improves the accuracy of gravitational acceleration extraction, with average absolute errors in inclination and tool face angle controlled within 0.63° and 0.8°, respectively. This method effectively enhances the precision of gravity acceleration extraction during drilling.
-
A method for demodulating weak signals in optical fiber hydrophones based on Sampson distance parameter estimation
Liang Qiumin, Wang Min, Yang Ping, Wang Wenxia, Zheng Huifeng
Abstract:
The interferometric fiber optic hydrophone based on a 3×3 coupler has reached the stage of engineering application in marine target monitoring. Typically, the phase shift signal demodulation employs ellipse fitting based on the least squares method to estimate the parameters of the interference signal, thereby addressing the demodulation bias caused by non-ideal 3×3 couplers. However, when the fiber optic hydrophone receives weak signals and the phase shift generated by the interferometer is small, the Lissajous ellipse formed by the interference signal becomes incomplete. In such cases, the high curvature bias issue of the least squares method leads to significant deviations in the demodulated phase shift signal. Additionally, although the orthogonal distance fitting method can effectively fit incomplete ellipses, its computational complexity and time-consuming nature are unsuitable for real-time demodulation. To address these challenges, this article proposes a weak signal demodulation method for fiber optic hydrophones based on Sampson distance parameter estimation. By utilizing Sampson distance to fit the incomplete Lissajous ellipse formed when the 3×3 coupler fiber optic hydrophone receives weak signals, the parameters of the output interference signal can be accurately estimated. This approach not only improves demodulation accuracy but also significantly enhances computational efficiency, outperforming the orthogonal distance fitting method. Through numerical simulations, the demodulation results of Sampson distance, the least squares method, and the orthogonal distance are compared and analyzed. The results show that the proposed Sampson distance demodulation method exhibits smaller fitting and demodulation errors under weak signal conditions compared to the least squares method and requires significantly less computation time than the orthogonal distance method. Experimental comparisons of the fiber optic hydrophone demodulation method are implemented within the frequency ranges of 10 to 30 kHz and 20 Hz to 2 kHz using a laser interferometry free-field calibration system and a vibrating liquid column low-frequency calibration system, respectively. The effectiveness of the proposed Sampson distance demodulation method is validated.
-
Adaptive bandwidth segmentation method improved by variable iteration mechanism
Lin Sen, Zhang Weihao, Yi Cai, Tao Ye
Abstract:
The operational condition of high-speed train axle box bearings has a direct impact on both train safety and dynamic performance. However, under complex working environments, bearing fault signals are often contaminated by strong noise interference and random impacts, making it challenging to effectively extract fault impulses and leading to reduced diagnostic accuracy. To address this challenge, this paper proposes an improved adaptive frequency band optimization strategy based on a variable iteration mechanism, aimed at enhancing fault diagnosis accuracy and robustness. The method first leverages the cyclostationarity of fault impulses to enhance the harmonic prominence index, enabling precise identification of the fault resonance band while effectively suppressing noise and random disturbances. Additionally, to overcome the limitations of fixed iteration step sizes, a variable-step iteration adjustment mechanism is introduced. By integrating energy spectrum trend analysis, the approach facilitates rapid localization and dynamic adjustment of the iteration step size, improving fault resonance band identification accuracy while reducing computation time and enhancing efficiency. This fault-driven adaptive frequency band division method addresses the shortcomings of traditional data-driven techniques, proving to be effective and superior in dealing with random impacts and strong noise interference. Simulation and experimental analyses show that the proposed method can quickly and accurately identify the fault resonance band under complex working conditions. Compared to fixed band division methods, improved power spectral density methods, and fixed-step adaptive division techniques, the proposed method offers significant advantages in signal-to-noise ratio enhancement, fault feature extraction accuracy, and computational efficiency.
-
Propagation characteristics of longitudinal L(0,2) mode guided waves in bend pipeline
Wang Xiaojuan, Ma Xuze, Zhao Kai, Zheng Yi, Gao Heming
Abstract:
The bend pipeline is a common form of piping in the pipeline industry, and the elbow section is the weakest part of the entire bend. Due to bending stress and material erosion, various defects can easily occur. Ultrasonic guided waves can effectively detect defects in pipeline structures, and numerous studies have validated and advanced the application of this technology in straight pipe inspections. When ultrasonic-guided waves pass through the elbow of a bend, they undergo complex changes, which can affect the defect detection performance of the guided waves in the pipeline. To address this issue, the finite element method is used to quantitatively study the propagation characteristics of the longitudinal L(0,2) mode-guided waves in the elbow region and the straight pipe section after the elbow through full-wavefield simulation data. The study analyzes and discusses the influence of the elbow structure on the guided wave propagation, wavefield distribution, and interaction with defects, as well as the detection of defects at different positions in the bend. The findings are evaluated through experiments. The results show that the elbow structure in a bend causes significant attenuation of ultrasonic-guided wave energy, and the distribution of the guided wave field changes in both the axial and circumferential directions, showing different energy focusing and diffusion characteristics before and after the bending point of the elbow. The propagation characteristics of ultrasonic guided waves in the elbow of a bend are closely related to the axial position of wave arrival, the bending radius of the elbow, and the excitation frequency. The guided waves after the elbow introduce asymmetric modes, presenting additional complexities. This research contributes to a deep understanding of the propagation characteristics of ultrasonic-guided waves in bends and provides a theoretical foundation for further utilizing ultrasonic-guided waves to achieve comprehensive defect detection in bends.
Precision Measurement Technology and Instrument
机器人感知与人工智能
Visual inspection and Image Measurement
先进感知与损伤评估
传感器技术
Information Processing Technology
About
Organizer：China Association for Science and Technology
Governing Body：China Instrument and Control Society
Chief editorial unitf：Zhang Zhonghua
Address：23rd Floor, Building A, Horizon International Tower,No.6 Zhichun Road, Haidian District,Beijing, China
Zip Code：100088
Phone：010-64004400
Email：cjsi@cis.org.cn
ISSN：11-2179/TH