Abstract:Wind power is crucial for the energy transition. Wind turbine blades, which capture wind energy, require effective defect detection to ensure reliable operation. The integration of drone aerial photography and machine vision can efficiently detect surface defects on these blades. This paper reviews recent developments in drone-based wind turbine blade defect detection. It begins with an overview of blade characteristics and defect types. Four detection methods are compared, highlighting the advantages and technical processes of drone-visual inspection. Traditional image processing and machine learning methods for image stitching, defect segmentation, and feature extraction are summarized, alongside deep learning approaches for defect classification, recognition, and segmentation. Relevant datasets and performance metrics are organized, and the paper concludes by identifying challenges and discussing potential solutions K . eywords:UAV aeri

Abstract:The cylindrical roller bearing (CRB), as a key part of the rotating machine, has a significant influence on the reliability, stability, and safety of the mechanical system in operation. Dynamic modelling of CRB based on kinematics and physics is beneficial to make it clear that the intrinsic excitation mechanism of the defect, revealing the mapping mechanism between the evolution mechanism of the internal failure excitation mechanism and the dynamic behaviors of the CRB. It could provide a theoretical foundation for promoting the accuracy and reliability in the condition operation and maintenance of the CRB. Existing models in the dynamic modelling of the CRB rarely consider the coupled excitation mechanism caused by the composite defect located on the inner and outer raceways, respectively. In view of this, a new dynamic model for the coupled excitation mechanism caused by the composite defects of the CRB is proposed in this article. The models for representing the time-varying mechanical parameter evolution mechanism due to the contact between the roller and the inner or outer raceway defect and between the roller and both the inner and outer raceway defects are deduced and formulated. The mapping mechanism between the excitation mechanism of the composite defect and the evolution law of mechanical parameters such as contact displacement and contact force of the CRB is studied. In addition, the variations in the vibration behaviors of the CRB with the different shaft speed scenarios are investigated. Comparisons between simulation and experiment show that the simulated vibration responses of the CRB are in good agreement with the measured results, and the characteristic defect frequencies calculated from the simulated signal are very close to the measured results. As a result, the effectiveness and accuracy of the proposed model is evaluated. These results have the confidence to provide certain theoretical significance and practical application value for the improvement of the accuracy and reliability in the condition monitoring and maintenance of the CRB.

Abstract:In electric vehicle circuit systems, DC series arc faults frequently occur at loose contact points or damaged line connections, leading to hazards such as fires and explosions. To study series arc faults in electric vehicles, an experimental platform specifically designed for electric vehicle arc faults was established. The causes and patterns of changes in the main current waveforms under various operating conditions were analyzed in detail. Given the stringent real-time requirements for arc fault detection, this study employed a lightweight convolution operation known as depthwise separable convolution to develop an arc fault detection network. This network achieved the detection of arc faults and the identification of fault lines in electric vehicles. To address the limitations of depthwise separable convolution in feature extraction in low-dimensional spaces, this study made improvements and proposed a convolutional operation with superior feature expression: group separable convolution. Ultimately, a progressive ladder structure was implemented, where the number of convolutional kernels within each group of the group separable convolution gradually decreases from the shallow layers to the deeper layers of the network. This approach streamlined and optimized network architecture while ensuring detection accuracy. Further enhancements involved optimizing the convolution kernel size within the detection model and integrating a lightweight attention mechanism into the architecture. A dynamic learning rate adjustment strategy was also applied during the model′s training process. Through these optimization measures, both operational efficiency and detection accuracy were systematically improved. The model achieved a detection accuracy rate of 96. 76% and exhibited good generalization and anti-interference capabilities.

Abstract:Considering that the difference in spatial motion characteristics is more direct to the mechanical property degradation, and the vibration signal contains rich mechanical state information, a circuit breaker health state identification method is proposed using vibration signals to characterize spatial motion properties. First, the displacement signal is used to obtain the motion characteristic parameters that can reflect the mechanical state of the key mechanism. Secondly, the AFF-AAKR is utilized to construct motion-characteristic health indicators offline. Then, multi-domain feature parameters are extracted based on the characteristics of the three-dimensional vibration signal in the key action phase. The features with higher correlation are selected for hierarchical clustering and the mutual information with the motion characteristics is calculated to achieve the key degradation feature vectors with strong characterization ability of the motion characteristics. Finally, the degradation feature vectors are used as the input and the health indicator of motion characteristics is used as the output to construct a 1D-CNN performance degradation regression model. In this way, the health state identification of the energy storage mechanism is realized. The example validation shows that the three-dimensional vibration signal fits the motion health indicator better than the one-dimensional vibration signal, and the regression analysis RMSE is 0. 018 6 and MAE is 0. 011 2, which can accurately identify the health status of the circuit breaker.

Abstract:The steer-by-wire system is a key part of the development of autonomous driving technology, and its reliability directly affects driving safety. The randomness of intermittent faults and the complexity of imperfect maintenance pose significant challenges to the design of fault prediction methods. Therefore, how to solve the intermittent fault prognosis problem of the steer-by-wire under imperfect maintenance is of great significance for driving safety. This article innovatively proposes an intermittent fault prediction method for the steer-by-wire system in the presence of imperfect maintenance based on the variable-size window and the operation condition-dependent compound degradation models. Extracting two features of intermittent fault through the concept of variable size window, the operation condition-dependent compound degradation models incorporating the effect of imperfect maintenance are formulated to predict the remaining useful life of the intermittently faulty component. The experimental results show that the developed method can achieve the relative accuracies of remaining useful life prediction with 95. 48% and 96. 14% under two operation conditions, which is obviously superior to the methods used in the existing literatures for comparative experiments. Keywords: steer-by-wire system; intermittent fault; imperfect maintenance; variabl

Abstract:To assess the yield strength of in-service pipes, this paper investigates the impact of steel microstructure on mechanical properties and its effect on electromagnetic characteristics. A method for detecting pipe yield strength using electromagnetic techniques is proposed. The study examines how steel′ s electromagnetic properties influence the impedance of a detection coil and develops a simulation model to explore how detection frequency affects magnetic flux density and induced current. The method′s effectiveness is validated through experiments, and the impact of temperature and surface corrosion on detection is analyzed. The results indicate that the electromagnetic-based yield strength detection method performs effectively for Q345 and Q235 steels. At a detection frequency of 10 kHz, the yield strength of both steels shows an approximately linear relationship with the detection coil′s impedance, with Pearson correlation coefficients of 0. 94 and 0. 87, respectively. Impedance values increase with yield strength, providing both theoretical and experimental support for pipeline yield strength detection

Abstract:Aiming at the overreading problem of vortex wet gas metering, a new overreading correction and wet gas flow measurement method based on vortex meter-disturbance wave frequency dual-modality detection system were proposed. The conductance ring sensor was developed to obtain the liquid film flow parameters, the excitation module, acquisition module, demodulation module, and host computer program were designed, respectively. The sensor's sensitivity and linearity were enhanced through the optimization of parameters such as excitation frequency, electrode width, and electrode spacing. The real flow tests were conducted on various carried gas conditions (gas flowrate and pressure) and liquid loading conditions, analyzing how vortex over-reading and disturbance wave frequency change with different two-phase conditions. Finally, the meter overreading equation was established with disturbance wave Strouhal number and gas Weber number, and the wet gas measurement model was developed combined with Newton iteration algorithm. The model gives a prediction error of gas flow within ±1. 5% (97. 5% confidence interval) with uncertainty of 0. 75% . Compared with the error up to 12. 1% before overreading correction, the accuracy of wet gas metering is largely improved. The model utilizes the disturbance wave frequency for meter overreading correction without the need for liquid film thickness calibration, due to the low requirement for medium conductivity and fewer calibrated coefficients, thus enhancing the model′s applicability and scalability effectively.

Abstract:To address the problems of poor flexibility, low load capacity, and insufficient structural rigidity of the existing in-situ inspection robots for aero-engine blades, a rope-driven adsorbable continuum robot is designed to solve the contradiction between high flexibility and low rigidity of the continuum structure. Based on the characteristics of the leaf environment and the mechanism of octopus tentacle adsorption, the robot consists of multiple mortise-and-tenon flexible joints connected in series with pneumatic pressure adsorption units, which enhances the structural stiffness and prevents deformation and destabilization through active adsorption. The kinematic model is formulated based on the improved D-H method and the Euler transformation principle. The mapping relationship between its bending deformation and structural parameters is analyzed. The structural stiffness and deformation characteristics are analyzed by using finite element ANSYS. The experimental platform evaluates the proposed kinematic model and control performance. Compared with the traditional continuum structure, the results show that the maximum load of the robot reaches 4. 42 N, and the end position deviation under the same load is reduced by 78% and 57. 5% , respectively. Thus, the correctness and validity of the proposed adsorbable continuum structure and the proposed model are verified. Keywords:continuum robot; aero-engine blade inspection;

Abstract:Aiming at the control requirements of maximum efficiency and constant voltage output of dual-channel wireless power transfer ( WPT) system under load and mutual inductance variation, a maximum efficiency control method based on equivalent load tracking is proposed in this paper. First, a rotary-coupled WPT system with a double-layer quadrature DD ( DQDD) coil as the transmitting mechanism and a crossed dipole (CD) coil as the receiving mechanism is constructed. Subsequently, the parameter configurations of the LCC-S network in dual-channel WPT system are carried out considering the cross-coupling and the internal resistance of the coil, and the expressions for the transmission power and efficiency of the system are derived. Then, a composite control method is introduced to regulate the output voltage of the system by using a Buck circuit at the DC input, while a Buck-Boost circuit is applied at the output load to achieve impedance matching. Finally, simulations and experiments validate the accuracy of the derived expressions for loop current, input / output power, and efficiency, as well as the feasibility of the resonance component configuration when accounting for coupling circulating currents. The results demonstrate the accuracy of the mutual inductance identification results, showing that the system can realize the maximum efficiency tracking and constant voltage output, with the transmission efficiency maintained above 90% .

Abstract:Aiming at the precise positioning issue of the battery pack locking mechanism during the battery swapping process of new energy electric vehicles, a pose estimation method based on point cloud segmentation and principal component registration is proposed. The method first utilizes an instance segmentation network to extract the locking mechanism instances from the scene images. The depth data corresponding to the locking mechanism instances are projected into point clouds, and statistical filtering and voxel filtering are applied to denoise and downsample the point clouds. Secondly, by embedding the SGE attention mechanism module into the feature extraction layer of PointNet++ network, the spatial semantic features of points in the point cloud are enhanced, and the lock head point cloud is segmented. Finally, the spatial pose for locking mechanism is obtained by aligning the locking mechanism head point cloud with the target point cloud through principal component registration. Experimental results show that the pose estimation algorithm proposed in this paper has high accuracy and certain anti-interference capability, achieving a locking mechanism point cloud segmentation accuracy of 98. 02% , a translation error of 2. 239 mm, an angular error of 1. 822°, and an RMSE of 1. 495 mm, meeting the positioning accuracy requirements for battery swapping robots.

Abstract:To address semantic inconsistency in multi-state associated feature extraction and balancing model performance with complexity in most multiple perspective view-based bird′s eye view (BEV) generation method, a light-weight Transformer-based BEV generation model is proposed. The method utilizes an end-to-end one-stage training strategy to establish a mutual association between dynamic vehicle and static road information in traffic scenes, effectively filtering out noise in the generated BEV. A Transformer-based recurrent cross-view transformation module for multi-scale features is introduced to perform image encoding and representation learning. This module improves the robustness of the extracted BEV features by capturing the location-dependent relationships in the perspective view (PV) feature sequence. Additionally, a multi-state BEV feature fusion module is designed to address semantic inconsistencies, extracting correlated information between dynamic vehicles and static roads, thus enhancing the performance of the generated BEVs. Experiments on the NuScenes dataset show that this method achieves advanced BEV generation performance with low model complexity, achieving 43. 2% and 82. 0% semantic segmentation accuracy for dynamic vehicles and static roads, respectively.

Abstract:Three-dimensional (3D) thermography technology combines the advantages of 3D reconstruction and infrared thermal imaging, enabling simultaneously acquisition of both the temperature distribution and geometric structure information of an object’s surface. It is an effective method to visually present the surface information of industrial materials. However, in the single view imaging environment with dust interference, the conventional 3D thermography method is difficult to apply. Therefore, this paper develops a 3D thermal imaging system using an infrared thermal camera and a depth camera, and proposes a single-view 3D thermal imaging method for the surface of industrial materials under dust interference. Firstly, this paper establishes a joint external parameter calibration model based on virtual imaging to solve the challenge of spatial synchronization in single-view imaging devices. Secondly, based on the principle of radiation temperature measurement, an approximate compensation method for infrared temperature measurement is proposed to obtain the temperature distribution with small error under dust interference. Then, to solve the problem of depth information loss and view occlusion in single-view 3D thermography, a single-view 3D thermal imaging method based on depth map rendering and viewpoint optimization is proposed to achieve fusion of temperature and topography data of the measured object under the optimal view. Experimental results show that the proposed method can effectively realize single-view 3D thermography of the surface of objects under dust interference, retain more temperature and depth information, and reduce infrared temperature measurement errors

Abstract:Defect image detection is an important technical tool for substation equipment operation and maintenance. However, due to the scarcity of defect samples, the traditional deep learning model based on massive data training faces the challenge of few-shot defect detection in practical applications. Therefore, this article introduces the idea of meta-learning and proposes a deep learning model for few-shot defect image detection of substation equipment. The core of the model is the optimization of front-end network weights and model fine-tuning for few-shot testing tasks. The former enables the model to quickly adapt to new tasks through a task generation strategy based on semantic information, while the latter fine-tune the model through a network optimization method based on meta-learning. Therefore, the model can obtain excellent performance on new tasks. The experimental results show that the improved method can enhance the model′s overall detection accuracy by 7. 33% and the detection accuracy of the new categories by 11. 48% , which significantly improves the detection performance on few-shot defects and defects of new categories.

Abstract:Temperature field measurement plays a crucial role in industrial production and manufacturing. As a visualization technique for flow fields, the Schlieren method enables non-contact measurement of temperature fields. However, traditional background Schlieren methods require precise measurement of the center coordinates of the flow field when calculating the angle of light deflection,which greatly limits their application in certain industrial testing fields. To address this issue, this paper proposes a method for adaptive reconstruction of axisymmetric temperature fields based on binocular background Schlieren. Firstly, temperature field Schlieren imaging is obtained from two directions using a binocular background Schlieren imaging system. Then, the Perspective-n-Point (PnP) algorithm is employed with monocular visual imaging technology to calculate the position parameters of the background scatter plate relative to the binocular camera, thereby determining the center coordinates of the temperature field. Finally, the temperature field distribution is reconstructed using the Schlieren method. Experimental results show that this method enables precise reconstruction of temperature fields with unknown center coordinates, significantly expanding the application scope of the Schlieren method.

Abstract:Accurate measurement of the kinematic parameters of payload swing is crucial for lifting and transporting dissipative swing control. However, the existing single-image projective correction methods are susceptible to the interference of background and other factors. It is difficult to automatically extract the coplanar straight-line constraints, thus restricting the measurement accuracy. To address this problem, a projective rectification method is proposed that utilizes dynamic line constraints from temporal image sequences. Firstly, dynamic straight lines are extracted from specific temporal images through the integration of background differencing and color thresholding methods. Subsequently, clustering is performed using an improved random sample consensus algorithm, thereby establishing reliable constraint conditions. Secondly, the Levenberg-Marquardt iterative method is used to estimate the vanishing point and the absolute quadratic curve based on the linear constraints. Then, the correction matrix was calculated. Thirdly, the coordinates of the marker points are rectified, which are employed to calculate the swing angle and trolley speed. In the measurement of kinematic parameters of the payload swing, the aforementioned method achieves a root mean square error of 0. 124 7° in comparison with angle measurement by the inertial measurement unit, and a root mean square error of 0. 003 6 m/ s in comparison with the set trolley speed. Compared with the inverse perspective correction method, the projective correction method with coplanar linear constraints and noncoplanar linear constraints, the angular accuracy is improved by 69. 05% , 76. 35% , and 93. 91% , and the velocity accuracy is improved by 42. 85% , 47. 06% , and 59. 55% , respectively. The proposed method utilizes the swing string in the payload swing as constraint for projective rectification, aiming to obtain richer and more precise constraint conditions, thereby enabling superior measurement precision of kinematic parameters in the payload swing.

Abstract:As oil and gas drilling advances into deeper and more complex formations, traditional methods for monitoring overflow accidents show delays and offer multiple potential solutions. The challenging formation conditions, along with high temperature and pressure underground, result in flow measurement deviations, which significantly affect the accuracy of overflow detection. To address this issue, a flow correction model was developed, accounting for the thermal expansion of mud to correct these deviations by mitigating non-overflow influencing factors. Furthermore, based on the corrected flow data, an overflow identification and warning model was established using multivariate data fusion and a temporal neural network. This model eliminates missed overflow reports and significantly improves the timely detection rate compared to conventional methods. It can provide warnings up to 5 minutes in advance, offering substantial application value and potential in drilling operations K . eywords:thermal expansion effect; overflow identification; correct

Abstract:To address the issues of poor accuracy and reliability in existing underground pipe diameter measurement methods, this paper proposes a new approach that combines laser ranging with geometric relationships. A laser device is positioned inside the pipeline, and by strategically placing the laser, it emits beams in three directions towards the inner wall of the pipe. The distances are measured, and a formula is derived to calculate the pipe diameter based on the laser measurements. This method enables accurate measurement of urban underground pipelines, and a technique for reducing measurement errors through alignment is introduced. Partition experiments were conducted to verify the correctness and feasibility of the theoretical derivation. Prototypes were developed, and experiments were performed using actual pipelines in a simulated laboratory environment. Results show that, in an ideal laboratory setting, the absolute errors for pipe diameters of 400, 600, and 800 mm are within 10, 9, and 6 mm, respectively, with relative errors of 2. 5% , 1. 5% , and 0. 75% . The measurement accuracy remained high even in partition experiments. Although the error in real-world environments is slightly higher than in the lab, it still meets practical measurement requirements. The system enables fast and accurate measurement of underground pipe diameters with high reliability and a simple measurement structure, making it suitable for field applications.

Abstract:In radiation therapy, the respiration may cause lung deformation. The subsequent deformation and spatial movement of tumors on the lung adversely affect the accuracy and safety of radiation therapy. By combining computed tomography (CT) image 3D modeling technology and respiratory mechanics principle, a new method of lung dynamic deformation modeling and deformation prediction based on a massless spring mechanism is proposed in this article. First, a lung model is formulated, which is based on medical image information before radiotherapy. Then, the deformation of the lung and the movement of the tumor on it are predicted based on respiration. The experimental results show that the lung deformation obtained by this method is consistent with the deformation observed under continuous medical imaging. The error between the predicted value and the observed value is within a reasonable range. This study can accurately and efficiently predict lung deformation and mechanical properties under the influence of respiratory movement under conventional medical conditions. It provides theoretical and method support for subsequent research on tumor displacement compensation.

Abstract:Ocean electric field measurement can be applied to the exploration of marine geological structures, exploration of marine mineral resources, monitoring of marine organisms, and discovery and tracking of moving targets in and at sea. Due to the complexity of the marine environment and the working mechanism of the modulated ocean electric field sensor, the sensing output contains complex noise components. Only by suppressing these noises and obtaining a high signal-to-noise ratio, reliable electric field signals can be extracted from the sensing output. A single noise suppression method is difficult to effectively reduce multiple types of noise. Therefore, this article proposes a composite denoising method targeting the noise characteristics in the output signal of modulated ocean electric field sensors. The noise is divided into three categories, including broadband random noise, noise within the useful signal frequency band, and noise outside the useful signal frequency band. Utilize the uncorrelation between broadband random noise and useful signals, and use adaptive filters to suppress broadband random noise. A bandpass filter is adopted to suppress noise outside the useful signal frequency band. Finally, the useful signal is extracted from the sensing output through variational mode decomposition, and demodulation is achieved through envelope detection. Compared with using bandpass filtering denoising, empirical mode decomposition denoising, and wavelet denoising alone, the sensor output signal-to-noise ratio processed by the proposed method is improved by more than 12 dB, ensuring reliable demodulation of modulated ocean electric field sensing signals.

Abstract:To address the issue of low testing accuracy caused by noise spectrum aliasing in ultrasonic testing, a data-model-driven method for high-quality extraction of ultrasonic detection signals is proposed. This approach combines empirical mode decomposition (EMD) with a component clustering index to pre-process the ultrasonic signal, reducing the impact of noise on signal extraction. Using the Gaussian echo model, time-frequency transformation, spectral Gaussian fitting, and the artificial bee swarm algorithm, the model parameters of the pre-processed signals are accurately estimated. The Gaussian echo model is then reconstructed to achieve high-quality signal extraction. Simulation results demonstrate that this method can extract ultrasonic signals with an SNR as low as 4. 56 dB, improving the mean SNR of the extracted signals to 28. 71 dB—significantly outperforming common methods such as EMD ( SNR = 9. 82 dB) and variational mode decomposition ( SNR = 11. 07 dB). Furthermore, ultrasonic testing experiments confirm the method′s ability to extract high-quality signals even under noise spectrum aliasing interference.

Abstract:Acoustic imaging is a key technology for applications such as noise source localization and abnormal sound diagnosis. Since the acoustic signals are non-modulated broadband signals, existing acoustic imaging methods divide microphone array data into several subbands and then perform acoustic imaging on each sub-band separately. However, the energy distribution of the acoustic signals across different frequency bands is uneven, leading to potential estimation errors in some sub-bands due to low signal-to-noise ratios, significantly impacting the accuracy of acoustic imaging. To address this issue, research was conducted on band-weighting methods based on complex Gaussian mixture models. By jointly utilizing data from multiple frequency bands to assign weights to each sub-band, the impact of subbands with erroneous estimates on the accuracy of acoustic imaging is reduced. To validate the effectiveness of the proposed method, experimental verification was conducted, measuring the accuracy of acoustic imaging using indicators such as the false alert rate, miss detection rate, and root mean square error. Experimental results demonstrate that the method effectively improves the accuracy of acoustic imaging, particularly reducing the false alert rate by more than 2. 1% under conditions where the signal-to-noise ratio below 10 dB.

Abstract:To meet the temperature measurement requirements in a confined high-temperature environment, a longitudinal mode ultrasonic guided wave temperature measurement method based on a high-purity graphite rod is proposed. By considering the temperature-dependent dispersion equation, the main factors affecting the wave velocity due to temperature are analyzed, and suitable excitation frequency and modes for temperature measurement are selected. A prediction model for the variation in the guided wave signal transit time with temperature in a waveguide rod with non-uniform temperature distribution is established, allowing for the calculation of accurate echo transit times in the graphite rod. Temperature measurement experiments using guided waves within the range of 1000℃ were conducted to verify the model’ s accuracy. The experimental results demonstrate that the temperature measurement accuracy of L(0,1) mode guided waves with an excitation frequency of 30 kHz is within 10℃ , with good repeatability across multiple trials. The research in this paper proves the high-temperature measurement capability of ultrasonic guided wave, providing a feasible solution for temperature measurement in extrem high-temperature environments.

Abstract:In the application of passive towed linear array sonar in deep sea environment, aiming at the problem of false target discrimination in the non-end-fire direction of tugboat noise, a method for identifying the false target of seabed reflection of deep-sea tugboat noise is proposed. Firstly, based on the ray theory model, the process mechanism of the deep-sea tug noise reflected by the seabed to form a false target is given. Then, the interference characteristics of beam output power of each target at different frequencies are extracted. Finally, according to the interference characteristics, the frequency intervals of each target interference structure are analyzed to realize the false target discrimination. The analysis results of simulation data and sea trial data show that the accuracy of the false target decision is more than 95% when the interference structure is stable and the energy ratio of the reflected sound energy to the background noise energy in the beam domain is more than 8 dB. It effectively solves the problem of false target discrimination formed by tugboat noise in non-end-fire direction, providing a reference for subsequent target judgment.

Abstract:To address inaccurate positioning caused by non-line-of-sight ( NLOS) and multipath effects, we propose an enhanced cooperative ultra-wideband (UWB) positioning algorithm based on the full-centroid-Taylor approach. In scenarios involving three or more positioning base stations, the stations are grouped into sets of three. For each group, ranging data is initially processed using the bilateral bi-directional method, and positioning coordinates are then estimated using the full-centroid algorithm. The coordinates of the ‘pseudocentroid’ point are subsequently optimized using a simulated annealing algorithm. This optimized result serves as the initial value for a Taylor series expansion, which provides a refined initial estimate. Particle filtering is then applied using this refined value to obtain precise positioning coordinates during carrier movement. Simulation and experimental results indicate that, compared to traditional UWB positioning algorithms based on Chan-Taylor and WLS-Taylor methods, the proposed algorithm significantly improves positioning accuracy in both static and dynamic scenarios. Furthermore, the application of particle filtering enhances positioning accuracy in complex environments, demonstrating the robustness of the algorithm.

Abstract:When the traditional error-state Kalman filter algorithm is used for aircraft attitude estimation without heading reference, significant errors can occur due to inaccurate linearization. To address this issue, this paper proposes an error-state Kalman filter algorithm based on the navigation coordinate system (NCS-ESKF). An aircraft attitude estimation system is designed, and both indoor static turntable experiments and DA40 airborne flight experiments with a general aviation aircraft are conducted. Experimental results show that, compared to three traditional algorithms, the proposed NCS-ESKF algorithm produces smaller errors, with mean absolute errors (MAE) for roll and pitch angles of just 0. 809° and 0. 934°, respectively. During the taxiing and flight phases of the airborne experiments, a segmented threshold method was employed to set different horizontal maneuvering acceleration thresholds, resulting in MAEs of 0. 954° for roll and 0. 867° for pitch. This effectively improves the accuracy of aircraft attitude estimation. The NCS-ESKF algorithm significantly reduces estimation errors and enhances aircraft attitude estimation performance, contributing to improved stability in flight control for general aviation aircraft.

Abstract:A high-precision and small-blind magnetostrictive displacement sensor with a magnetoelectric coupling sensor configuration was proposed for the displacement measurement scenarios where the installation space is limited, such as in electro-hydraulic power system, hydraulic cylinder and precision machinery, and where measurement accuracy is required. The signal detection element of the sensor adopts Co70Fe30 / PZT-5/ Co70Fe30 magnetoelectric composite as the core sensitive material, which operates in a longitudinal magnetizationtransverse polarization operating mode. An adjustable experimental platform was set up for pulse current and axial bias magnetic field, using a 0. 5 mm diameter Fe46. 5Ni 48. 5Cr2Ti 2. 5Al 0. 5 waveguide to verify the accuracy of the model. The optimal excitation pulse current range of the sensor is 15~30 A and the optimal bias magnetic field range is 15~25 kA/ m. Using Fe-Ni as waveguide wire material and Co70Fe30 / PZT-5/ Co70Fe30 magnetoelectric composite material, a prototype magnetoelectric coupling magnetostrictive displacement sensor was designed, and the measurement and calibration platform of the displacement sensor was built. The measurement accuracy of the prototype was ± 0. 02 mm, and the distance of a small blind area was 15 mm. Compared with the traditional coil-sensing magnetostrictive displacement sensor with the same parameters, the precision is increased by 33% , and the blind zone distance is reduced by 85% .

Abstract:When the core of an LVDT displacement sensor deviates from its central position, a non 180° phase shift is induced, causing a mismatch in the amplitudes difference of the output signals from the two secondary coils. This, in turn, leads to a linearity deviation in the signal conditioning circuitry. To address this issue, this paper proposes a digital conditioning and phase compensation method based on discrete Hilbert transform. This method integrates digital moving average filtering and Butterworth low-pass filtering to achieve multilevel signal filtering and processing, ensuring smooth signal output while effectively reducing noise. It accurately compensates for the non 180° phase shift that occurs when the iron core of the LVDT displacement sensor deviates from the central position, and realizes the digital conditioning of its output signals. Experimental results indicate that the theoretical non-linearity of the proposed digital method is ±0. 076% , while the experimental non-linearity is ±0. 093% , showing a significant improvement compared to the non-linearity of ±0. 2% in conventional analog conditioning circuits for LVDT sensors. This method not only demonstrates theoretical validity but also exhibits superior performance in practical applications.

Abstract:IEEE 802. 11-2016 defines the fine time measurement protocol, which uses signal round trip time (RTT) to achieve indoor WiFi positioning accuracy at the meter level. However, in non line of sight or multipath environments, the accuracy of RTT ranging decreases, which seriously affects the positioning performance. To improve the accuracy of RTT positioning, this article proposes a method to convert the WiFi RTT ranging sequences measured by multiple access points into the multi-channel image, and uses an efficient channel attention-convolutional neural network to learn the relationship between the ranging data and the target position based on the multi-channel image. The experiments show that the positioning error of the proposed model is about 1 m, and 31. 03% , 16. 78% , and 10. 68% less than the conventional deep neural networks positioning, the single-channel-image-based CNN positioning, and the single-channel-image-based ECA-CNN positioning, respectively. Keywords:indoor positioning; attention mechanism; convolutional