Abstract:Optical frequency combs consists of a series of discrete and equidistant coherent lasers just like combs in the frequency domain, which has good application prospects in many fields such as optical clocks, ranging, spectral detection, coherent communication and other fields due to their wide spectrum, high coherence, and low phase noise. Microcavities provide a small and integrated platform for the generation of optical frequency combs, which makes microcombs attractive for researchers in recent years, especially for the highly-coherent dissipative Kerr soliton combs. The implementation of the vast majority of microcombs relies on homogeneous continuous wave driving. However, the inhomogeneous driving, such as pump in the presence of phase or amplitude inhomogeneities, has come into view owing to its specific advantages in soliton repetition control and energy conversion efficiency. In this article, the research progress of micro combs is reviewed. Specifically, an overview focusing on the generation and dynamics of dissipative Kerr soliton under inhomogeneous driving fields is provided. Future development trends and perspectives are also discussed. The inhomogeneous driving can improve the controllability and energy conversion efficiency of dissipative Kerr soliton, which gives the way for more microcomb applications

Abstract:The performance of the inertial sensor directly determines the accuracy of the inertial navigation system. The quantum inertial sensor based on the atomic system is expected to achieve the performance of the traditional inertial sensor with a smaller volume and lower cost, and theoretically can obtain higher measurement sensitivity and long-term stability than the existing technology. In recent years, with the rapid development of the field of quantum precision measurement, the practical and engineering research of quantum inertial sensors has made remarkable progress. In the future, by replacing traditional accelerometers and gyroscopes, it is possible to form a highly integrated, low-power, and low-drift quantum inertial navigation system. This article briefly introduces the basic principles of quantum inertial sensors based on atomic systems, and summarizes the current research status of quantum inertial sensors mainly including atomic interference gyroscope, atomic spin gyroscope, atomic interference accelerometer, atomic interference gravimeter, and gravity gradiometer. The article also reviews and analyzes key technical issues that need to be solved. It provides valuable insight for the development of quantum inertial sensors.

Abstract:Gravity assisted underwater inertial navigation has been a hot and cutting-edge issue in the research of underwater vehicle navigation and positioning in recent years, which is expected to become an important direction for the development of the next generation of high-precision underwater navigation systems. Firstly, the importance of underwater gravity information for correcting inertial navigation system error is introduced, and the basic principles and technical connotations of underwater gravity assisted inertial navigation are elaborated. Then, the research status and development trend of underwater gravity assisted navigation based on traditional relative gravimeter are summarized from different development stages such as unmapped matching and mapped matching. Furthermore, the demand of high-precision absolute gravimetry technology for next-generation underwater autonomous navigation system is analyzed, and the latest development and application status of atomic interferometric gravity measurement technology are reviewed and discussed. The application prospects of atomic interference gravity measurement technology in the field of underwater inertial navigation are forecasted and the key technologies that still need to be solved are summarized. Finally, the shortcomings and development suggestions of gravity assisted navigation research in China are given.

Abstract:Diamond nitrogen vacancy (NV) color center has the advantages of excellent quantum properties at room temperature, long decoherence time, compatibility with micro and nano processing, etc. Thus, it becomes a new development direction in the field of highprecision on-chip sensor technology. The sensitivity of magnetic, electric, temperature, angular velocity, and other physical quantities based on the quantum effect of diamond NV color center has gradually broken through the limits of existing sensing and measurement technology. It has been increasingly applied in aerospace, deep space exploration, life science, and other disciplines. This review mainly introduces the mechanism of diamond NV color center quantum sensing, the research progress, and future development trends of sensors with different physical quantities.

Abstract:With the rapid development of laser cooling technology, cold atoms have become one of the key platforms for emerging quantum technologies, including quantum sensing, quantum precision measurement, quantum simulation and quantum computing. The temperature of cold atoms is an important parameter of the cold atom experiments. Ultra-cold atoms have a low collision rate, long coherence time, and reduced Doppler broadening. Therefore, temperature measurements on cold atoms are of great scientific significance and application value. In this article, the research progress of cold atomic temperature measurement methods is reviewed. Firstly, the design and implementation of a simple method to quickly and nondestructive measure the temperature of selected regions in the cold atomic ensemble. Secondly, the colder atoms with temperature less than 1 μK are filtered out from about an ensemble with 20 μK, and the original form of Maxwell′s demon model is demonstrated in the experiment.

Abstract:With the continuous development of information technology, the issue of information security has become increasingly prominent. In the military, finance, government, enterprise and other fields, the security and reliability of information is of paramount importance. Traditional communication technologies have been widely used, including the Internet, optical fiber communication and radio. Meanwhile, the progress and application of quantum computing related technologies have made the security of traditional communication face severe challenges. Based on the quantum non-cloning theorem and Heisenberg′s uncertainty principle, quantum communication can theoretically achieve provable security in information theory. Therefore, ultra-long-distance quantum communication and space-based platforms have become the focus of research. This article reviews the research progress of space-based platform quantum communication in the past ten years, introduces the emergence of space-based platform quantum communication, important development nodes and main achievements, and summarizes the research status of long-distance quantum communication based on free space quantum communication and space-based platform quantum relay. Based on the current problem of high gap in quantum relay, the idea of airship as quantum relay is proposed.

Abstract:Mobile gravity measurement is important in metrology, earth science, national defense, etc. The atom gravimeter has good technical potential in mobile and high precision gravity measurement. However, it is still difficult to achieve both high integration and high sensitivity. In this article, a vehicle gravity measurement system based on the atom interferometer is developed. Based on the miniaturized and highly integrated instrument design, the high precision gravity measurement in the field could be achieved. The sensitivity is 520 μGal / Hz for outdoor gravity measurement and the measurement repeatability is better than 20 μGal Then, we carry out a rapid gravity measurement on a field survey line with a length of 10 km and an elevation change of 100 m. The gravity measurement results are compared with the LG-1 relative gravimeter. The debugging time of each point is less than 5 min, and the effective measurement time is less than 15 min. In the general field environment, the measurement residuals of two gravimeters are less than 100 μGal, and the residuals of the designed vehicle gravimeter are less than 15 μGal. Therefore, this vehicle-mounted gravity measurement system greatly improves the efficiency of field gravity mapping. It is convenient for transportation and has fast and accurate measurement. In general, the instrument provides a reliable technical solution for the field of mobile gravity measurement.

Abstract:The generation of high-precision time-frequency signals and the transmission of ultra-high-precision time-frequency signals are the foundation of modern physics, astronomy, and metrology. The high-precision time-frequency scientific experimental system of the Chinese space station has been successfully launched with the " Mengtian" cabins, which marks a crucial step in the construction of Chinese national time-frequency system with integration and interchange of space and ground. This article provides an overview of the high-precision time-frequency scientific experimental system project and the research progress of microwave time-frequency transmission links, with focus on the signal system design of the microwave transmission link in high-precision time-frequency systems. Based on the proposed signal system, two microwave time-frequency transmission link comparison modes of the same-frequency and three-frequency are designed, and a clock difference comparison model is provided. Finally, simulation and in-orbit measurement data analysis show that the designed microwave time-frequency transmission link can achieve sub-picosecond pseudo-range measurement.

Abstract:To solve the problems of high noise and obvious drift in long-term observation of proton vector magnetometer, obvious temperature effect of fluxgate magnetic field meter and large investment in thermal insulation magnetic room, a seismic geomagnetic vector measurement system based on the optically pumped cesium magnetometer is developed. In this article, the measuring principle and instrument structure are introduced in detail, the mixed-component coil device is designed, and the stability is analyzed. The prototype is tested at the Mengcheng seismic station. The results show that, during the experimental period, the instrument records well, the sampling rate of the experimental instrument reaches 1 Hz, the monthly noise of the total intensity, the horizontal component and the magnetic declination in January are 0. 01 nT, 0. 02 nT and 0. 003′, respectively, which are obviously superior to the best noise level of proton vector magnetometers of seismic geomagnetic observation basic network. They are also better than the noise level of two fluxgate magnetic field meters at the same venue and the average noise level of the seismic geomagnetic observation reference network. The maximum drift of the horizontal component and the magnetic declination are 1. 3 nT and 0. 16′, respectively, which is less than the baseline drift of the pedestal-clock proton vector magnetometer, so good stability can be obtained.

Abstract:A high sensitivity cesium atomic magnetometer with wide temperature characteristics is developed to meet the magnetic anomaly detection application requirements on an airborne platform. The problems of existing sensors such as small operating temperature range and easy magnetic field lock-out at low temperature. The temperature feedback mechanism is utilized to compensate the excitation source of the cesium atomic lamp in real time, and the atomic magnetic sensor′s stability and working temperature range is improved. The stability of the output optical power of cesium atomic lamp is increased from 2. 10 ( standard deviation) to 0. 62, and the operating temperature range of the sensor is increased from - 20℃ ~ 60℃ to - 50℃ ~ 70℃ . The design and parameter optimization of cesium atomic magnetic sensor are presented, which are based on low-noise cesium atomic lamp and a cesium vapor cell with a narrow bandwidth. A prototype of high sensitivity cesium atomic magnetometer is developed. Test results show that the cesium atomic magnetometer prototype′s measured sensitivity in the geomagnetic background is around 140 fT/ Hz@ 1 Hz, which is better than that of comparable worldwide commercial items (Geometrics G-824A cesium atomic magnetometer).

Abstract:Laser-cooled 87Rb atoms are loaded from free space into anti-resonant hollow-core fibers ( AR-HCF) using a dipole trap generated by an 852 nm laser. We propose a theoretical model of the optical depth of atomic ensemble inside hollow-core fibers. The resonant optical depth ( ODres ) is obtained by fitting the transmission spectrum with the theoretical model. Time-of-flight ( TOF) measurement is used to estimate the radial temperature of the atomic ensemble and the number of atoms inside the fiber. The influence of the trapping depth, the initial temperature, and the position of the atomic cloud on the number of guided atoms are systematically studied, which guide the parameters optimization direction of atoms inside HCF. This work provides the foundation for the development of atom interferometers inside HCF.

Abstract:The entangled optical quantum imaging efficiency is mainly affected by the sampling time cost of the digital micromirror device (DMD). The existing DMD sampling methods are to scan all pixels point by point, which result in the low imaging efficiency. To address this problem, this article uses the compressed sensing algorithm to sparsely sample the reference light based on the temporal and spatial correlation properties of the entangled light, and also uses the orthogonal matching pursuit (OMP) algorithm to obtain the target image from the coincidence counting result. In addition, this article analyzes the impact of different number of bundled pixels, DMD sampling block size, and image sparsity on imaging quality and efficiency. The effectiveness of the proposed method is evaluated by establishing an actual entangled optical quantum imaging system.

Abstract:In the 87Rb- 129Xe spin exchange system, the accurate measurement of the relaxation times of 87Rb and 129Xe is crucial for applications related to nuclear magnetic resonance gyroscopes and magnetometers. In response to this requirement, the influences of pump light and excitation magnetic field on 87Rbrelaxation time, as well as the spin exchange processes of 87Rb and 129Xe in the dark state, are analyzed. Based on theoretical analysis, a dark state sweep frequency measurement method is proposed. The measurement of longitudinal relaxation time of 87Rb and 129Xe is achieved. The transverse relaxation time and longitudinal relaxation time of 87Rb measured using this method are 1. 36 and 5. 18 ms, respectively. The longitudinal relaxation time of 129Xe is 519 s, whose R-squared is 0. 999 9 which has a very high goodness of fit. Compared with previous measurement methods, the dark state sweep method can completely eliminate the influence of magnetic field gradient caused by pump light, and has the advantages of high accuracy and easy operation. This article has high reference value for the performance analysis and calibration of nuclear magnetic resonance gyroscopes and magnetometers.

Abstract:Superconducting quantum interference devices ( SQUID) are superconducting devices that use two properties to resolve very weak magnetic field variations, including flux quantization and Josephson effect. They are used in a wide range of applications for sensitive detection of magnetic signals. Therefore, to ensure the accuracy and quality of its output mapping, it is essential to calibrate it periodically during use. In this article, the calibration system optimization of SQUID second-order magnetic gradiometer in use is discussed. A Cartesian coordinate system is established with the center of the calibration coil as the center point, and the relative position in the Y-axis direction is fixed. The calibration coil is moved from the Z-axis direction to find the maximum position range of current sensitivity. Then, the position is further calibrated precisely to find the relative position in the X-axis direction that is least sensitive to the movement in the Z-axis direction. , which provides a larger tolerance range for possible human errors in calibration. The analytical model and the finite element simulation model are validated against each other to provide a theoretical basis and precedence for subsequent experiments. The calibration coefficient of SQUID second-order magnetic gradiometer in use is determined to be 1. 107 through the analytical model, the finite element model and the measured data. The uncertainty generated by the proposed calibration method is analyzed to provide greater robustness for the calibration of the gradiometer under low-noise environment conditions.

Abstract:Firstly, this article introduces the composition, instrument layout, and hardware connection of the low-temperature superconducting full tensor magnetic gradient measurement system. Then, the test results of each module in the system development process are analyzed, including the noise test and SQUID sensitivity test of the measurement and control system. Secondly, the ground static and dynamic experiments of the system are described in detail. The dynamic range, measurement accuracy, and stability of the system are tested. Each component of the magnetic gradient measurement accuracy of the system static experiment is better than ±30 pT/ m. Finally, the flight test is implemented in Danyang, Jiangsu Province. According to the DZ/ T 0142—2010 Technical Specifications for Aeromagnetic Survey, the measurement results show that the coincidence accuracy in the repeated lines of each tensor component is better than ±25 pT/ m, and the average dynamic measurement sensitivity of each measuring line is better than ±30 pT/ m. This system has reached the level of engineering prototype, greatly improving the practicality of the low-temperature superconducting full tensor measurement system. It provides a solid foundation for future research work in the field of low-temperature superconducting navigation magnetic measurement in China.

Abstract:The superconducting / magnetoresistive (MR) composite magnetic sensor is capable of greatly improving the sensitivity of the MR sensor due to thousands of magnetic field magnifications of the superconducting magnetic amplifier. It is expected to achieve the magnetic field resolution of fT level. Thus, the demands of new weak magnetic detection could be satisfied, such as underwater target detection and biological magnetic field detection. After decades of research, the related works of superconducting / MR composite sensors have made important progress and a series of achievements. In this article, the working principle, highly sensitive, and triaxial research status of the composite magnetic sensors are systematically reviewed and summarized. The magnification of the superconducting magnetic amplifier has exceeded 1 000 experimentally, and the thermal noise at low temperature (4. 2 K) is only 2 fT/ Hz. Furthermore, the next research work is also briefly prospected in view of the existing problems. The related works in this article can provide significant guidance for improving the performance of superconducting / MR composite sensors and promoting its popularization and application.

Abstract:Negative stiffness vibration isolation platform (NSVIP) is widely used in quantum precision measurement equipment such as atomic interference gravimeter. Its performance is significantly influenced by low-frequency performance and environmental adaptability of the isolator. To improve overall performance of the vibration isolator, this research performs a semi-active transformation of a standard commercial NSVIP. Firstly, the force-displacement relationship is analyzed theoretically, focusing on the determinants of negative stiffness characteristics and bearing capacity. Then, the motion model is formulated and analyzed, and the parameters are accurately identified in the time and frequency domain, respectively. Moreover, the semi-active vibration isolation ( SAVI) is simulated and analyzed, and the Bayesian optimization is used to quickly determine the optimal control parameters. Finally, the scheme is further verified on the actual system. The result shows that SAVI can attenuate the low-frequency resonance peak by 357 times. The system can achieve good vibration isolation effects in the frequency band below 0. 3 Hz and above 8 Hz. The low-frequency vibration isolation performance is significantly improved. The proposed solution can be widely used in related vibration isolation equipment for quantum precision measurement.

Abstract:Cable is widely used in engineering due to the good tensile performance. The cable tension is an important factor affecting the safety, which is necessary to be tested. Currently, the testing of cable tension mostly is calculated by the frequency method, which is to measure the cable frequency based on accelerometer or computer vision. Recently, the φ-OTDR has been developed, which can measure the frequency of several cables with just one low-cost communication fiber. It has a great prospect in the testing of cable tension. However, the coefficient of effective optical elasticity used to determine the sensitivity of the distributed fiber optical system is not unique in practice. To ensure the reliability of the data collected by φ-OTDR in the testing of cable tension, this study establishes a performance calibration platform for the distributed fiber optical system for cable monitoring by attaching the sensing fiber to the cable, and the midspan displacement of cable under different testing conditions are measured by using the digital image correlation method. The mid-span displacement is converted into the axial displacement of cable by analyzing the dynamic model of cable. Finally, based on the converted axial signal of cable and the phase collected by φ-OTDR, the performance calibration of distributed fiber optical sensing system is realized.

Abstract:There is an urgent demand for high-performance sensors in the fields of aviation, aerospace, etc. Graphene materials, due to their excellent mechanical and electronic properties, are expected to further improve sensor performance and have attracted extensive research. Graphene resonators, as a new sensitive unit, have very high sensing sensitivity due to their extremely thin thickness. However, currently, the anti-interference ability and stability of graphene resonators at room temperature are generally poor. The antiinterference ability and stability of resonators are closely related to their quality factor. The low quality factor of graphene resonators at room temperature has become a key factor that restricts the performance improvement and application of graphene resonators. Improving the quality factor of graphene resonators has become an urgent research issue. This article focuses on the problem of low quality factor of graphene resonators at room temperature. By formulating an energy dissipation dilution theoretical model of graphene resonators, the energy dissipation distribution characteristics of graphene resonators are analyzed. Guided by the theoretical model of energy dissipation dilution, a new mechanism for suppressing energy dissipation in graphene resonators based on the phononic crystal soft support structure is proposed. Optimization design, preparation, and characteristic testing analysis of graphene phononic crystal resonators are carried out, which provides a new technical way for improving the quality factor of graphene resonators.

Abstract:The driving and detection modal resonant frequency difference (Δf) of MEMS gyro structure is the main factor that determines its mechanical sensitivity. When Δf≈0, the gyro is in the frequency tuning state, and the mechanical sensitivity of the gyro reaches the maximum value. In this article, a capacitive fully symmetric S-shaped elastic beam silicon ring wave gyro is presented. In the process of frequency tuning with frequency modulation voltage, there is a certain stiffness coupling between the modes. This article analyzes that the reason for the stiffness coupling between modes is the structural error. The structural error of ring harmonic oscillator is mainly reflected in two vibration parameters. One is frequency, and the other is damping. Since the main parameter of frequency modulation voltage change is the stiffness coefficient, this article only conducts modeling analysis on the frequency error. Firstly, the structure of ring gyro is introduced, and the principle of electrostatic negative stiffness is analyzed based on this structure. Secondly, the causes of frequency error are analyzed, and the influence of frequency modulation voltage on the two operating modes is deduced by modeling. Finally, by comparing the theoretical model with the experimental results, the correctness of the theoretical derivation is evaluated, and the sensitivity of ring gyro is increased by 2. 7 times under the condition of frequency tuning.

Abstract:The transverse flux sensor is widely used in the magnetic bearing system due to its compact structure and high detection accuracy. With the development of magnetic bearing technology, higher requirements are put forward for the detection performance of the transverse flux sensor. To further improve the performance of the transverse flux sensor and meet the high-precision displacement monitoring requirements of the magnetic suspension rotor, this article designs and analyzes the sensor for the sensitivity index. By formulating the mathematical model of the sensor and the finite element analysis of the electromagnetic field, the relationship between the excitation frequency and the coil parameters is studied. The correlation between the number of turns of the sensor coil and the sensitivity is numerically studied. The influence of the skin effect on the sensitivity of the sensor is analyzed from the perspective of detecting the rotor. The sensor signal processing circuit is designed to realize the conversion from displacement signal to voltage signal, and an experimental platform is established to measure the output characteristics of the sensor. The experimental results show that when the sensitivity of the sensor is 20 mV/ μm, the detection range is ± 500 μm, and the linearity is 0. 69% . It has good dynamic characteristics, which is suitable for high-precision radial position detection of the rotor of the magnetic bearing system.

Abstract:The optical radiation temperature measurement is of great importance in the field of characterizing the spatiotemporal evolution characteristics of combustion field temperature due to its advantages of high spatiotemporal resolution and wide temperature measurement range. The testable space in the combustion chamber of the solid rocket motor is limited, and the three-dimensional temperature field reconstruction method based on multi line of sight projection is ineffective due to light obstruction. To address this problem, the object image spatial mapping relationship model between the combustion cross-section and the image plane is formulated based on the Fourier optics theory. A dynamic radiation tomography thermometer with the integrated opto-mechatronics is designed, which achieves multiple image sensors sharing the same optical axis and being able to synchronously focus on combustion cross-sections at different spatial positions. The experimental results show that, by convolution and deconvolution processing of the combustion radiation sampling data of solid propellant strips in a certain type of windowed combustion chamber, the stackable combustion sections can be separated from each other in the direction of the optical axis of the instrument. The relative error of temperature measurement is less than 8% in the process of measuring the temperature distribution of the separated combustion sections using Planck′s radiation law based on the calibration of the photoelectric signal relationship. This instrument analyzes the dynamic three-dimensional combustion temperature field in a tomographic way on a single projection.

Abstract:To address the problems of conventional LiDAR simultaneous localization and mapping (SLAM) in dynamic environments with large cumulative errors in pose estimation and dynamic object error point clouds in the map, this paper presents a tightly coupled LiDARinertial SLAM (DM-LIO) method for real-time removal of dynamic objects based on the visible point method. This method by utilizes the IMU measurements to provide a priori poses for the dynamic object removal module based on the visible point method, and also introduces a point cloud clustering method based on curved voxel space to solve the problem that dynamic points of viewable point method cannot be fully captured at low resolutions, which enables the rejection of dynamic objects in laser point clouds at the front end of algorithm. The performance of algorithm is evaluated by both building a real indoor experimental platform and using a public dataset. The results of real-world experiments show that the proposed DM-LIO method is able to remove multiple dynamic objects as well as non-a priori dynamic objects online; the test results based on the public dataset of Urbanloco show that the absolute trajectory error of DM-LIO is reduced by more than 60% compared to LIO-SAM in highly dynamic environments, which verifies that the algorithm possesses good positioning accuracy in a highly dynamic environment.

Abstract:In practical engineering, the rotating machinery fault data set is prone to noise labels due to human labeling or data preprocessing, which could lead to the degradation of fault diagnosis model performance. Therefore, the attentive feature Mixup method of rotating machinery fault diagnosis is proposed. First, a residual neural network ( ResNet) is established to extract time-frequency features from the samples. The correct label sample groups, partially noisy label sample groups, and noisy label sample groups are constructed through random grouping and feature interaction. Secondly, an attention mechanism is introduced to calculate the correlation between samples within each sample group, and assign weights to each group of samples. The different weights that can distinguish noisy label samples within partially noisy sample groups are achieved. Then, Mixup is performed on each group of samples according to their weights, which can interpolate noisy label samples and update the attention layer parameters during backpropagation to reduce the proportion of noisy label samples. Finally, the online label smoothing (OLS) is used to update the model′s prediction information by reducing the negative impact of noisy label samples on the model loss update and further suppressing the effects of noisy label sample groups. Experiments on the rotating machinery fault dataset with label noise interference show the effectiveness of the proposed method.

Abstract:Accurate monitoring of tool wear during machining helps to avoid product quality problems caused by tool failure. To formulate tool wear monitoring models for different working conditions, it is necessary to adjust the parameters for each group of working conditions to ensure the accuracy. To reduce the number of parameter adjustment and ensure the prediction accuracy, the advantages of deep forest are combined, such as few hyperparameters, parameter insensitivity to the model and adaptive training process. A tool wear state prediction model with multi-sensor signals and autonomous feature selection for multi-conditions is established by using deep forest. The multi-sensor and wear data of TC18 milling process under three sets of different process parameters, and the open data in the predictive and health management (PHM) society 2010 high-speed CNC machine tool health prediction competition are utilized. For the three sets of working conditions, the prediction accuracy values of deep forest are 95. 35% , 96. 63% and 97. 06% , respectively, and 98. 95% on PHM data, which evaluate the high accuracy and applicability of deep forest for tool wear prediction under multiple working conditions. It provides a strong guidance for online monitoring technology in intelligent machining technology.

Abstract:Remaining useful life prediction of metal oxide semiconductor field effect transistor ( MOSFET ) can prevent gradual degradation or lose efficacy of devices due to long-term conduction. However, the traditional prediction models are difficult to extract the detailed characteristics of nonlinear changes in MOSFETs degradation parameters, resulting in poor prediction accuracy. To address this issue, a remaining useful life prediction method for MOSFETs is proposed, which is based on variational mode decomposition and nonlinear auto-regressive model with exogenous inputs (NARX) neural networks with external inputs. Firstly, the degenerate parameter sequence is decomposed into multiple sets of characteristic components containing nonlinear change information using the VMD method. Secondly, the NARX prediction model is optimized by using Bayesian regularization and Levenberg-Marquardt algorithms, respectively. Finally, integrating multiple sets of feature component prediction values to obtain the remaining life prediction results of MOSFETs. The experimental results show that the root mean square error of the proposed method is less than 0. 003, the mean absolute percentage error is less than 5% , all of which are better than the comparison method. The average error of remaining useful life prediction is less than 5 min, which evaluates the effectiveness of the method.

Abstract:Aiming at the enormous threat that railway intrusion obstacles pose to train operation safety, and the existing railway obstacle detection models have difficulty balancing detection accuracy and speed and poor multi-scale object detection robustness in complex railway environments, this article proposes an all-weather high-precision real-time multi-scale railway obstacle detection model. The model improves the feature extraction speed of the model by using dual-branch structure and linear operation. By modifying the Transformer structure, the lightweight model can model global contextual information. By designing high richness feature fusion structure and lightweight attention mechanism, the model′s multi-scale object detection ability is further improved. In addition, we embed the model and develop an intelligent detection system. The experimental results show that the proposed model has a detection accuracy and speed of 94. 93% and 132 fps in the dataset collected from actual railway scenes, respectively, which is 3. 09% higher than YOLOv5s. It can meet the application requirements of high-precision real-time detection of multi-scale obstacles in complex railway scenes.

Abstract:The occurrence of inter-turn faults in the electric motor of the fuel pump in new energy vehicles cannot guarantee fuel supply, pressure control, lubrication, and cooling, which poses a threat to driving safety. To address this issue, this article proposes an online monitoring method for winding inter-turn faults by combining electromagnetic parameters and vibration signals. Firstly, the electromagnetic force model containing fault current harmonics is formulated according to the Maxwell tensor method. Then, a multisensor motor signal acquisition circuit is designed. Finally, the improved adaptive empirical mode decomposition method is applied to adaptively decompose the denoised vibration signals, and a set of intrinsic mode functions is selected and reconstructed by using the correlation coefficient method. The comprehensive evaluation of the kurtosis-to-root mean square ratio and envelope spectrum feature factor results in 52. 3% improvement in the fault characteristic indicator. This indicates that the reconstructed signal has higher sensitivity. The consistency between the reconstructed signal and fault characteristics is further evaluated through analysis of current waveforms. This research holds important engineering significance for the fault diagnosis and state prediction of oil pump motors.

Abstract:Traditional milling stability analysis has relatively low prediction accuracy under real working conditions for using the static tool tip frequency response functions (FRFs) and average cutting force coefficients. Therefore, a milling stability prediction method based on a small number of experimental samples is proposed by introducing transfer learning. First,the tool tip FRFs at idle state and the average cutting force coefficients are measured to generate multiple series of random values within the spindle speed range. An optimal series is determined by comparing the limited experimental stability limits and their related predicted values, and it is used to further construct sufficient source stability data close to the real data. On the basis, a multi-layer perceptron model for predicting the stability limits is formulated by the source data, and it is globally fine-tuned by the limited target experimental samples for adapting to the real machining scene. Forty groups of chatter experimental samples are used to develop a validation case study. The prediction accuracy of the proposed method is 32% higher than that of the model constructed only using the 40 samples. In addition, accuracies of different types of prediction models trained by different target data sizes are also compared to evaluate the effectiveness of the proposed method.

Abstract:In this article, a new operating performance assessment method based on adaptive fusion of multi-source heterogeneous information (AFMSHI) is proposed for magnesium melting process. First, the data pre-processing is performed for the multi-source heterogeneous information (MSHI) in the process of magnesium melting, and deep learning methods are used to formulate performance assessment sub-models based on different types of information. Secondly, to fully consider the impacts of MSHI on the assessment results under different melting states, the attention mechanism is used to establish an adaptive fusion network for the assessment results of each sub-model. Finally, the fused assessment results are input into a SoftMax classifier, and the magnesium melting process assessment model is formulated. The simulation results show that, comparing with the assessment model established by a single type of information or the existing deep learning MSHI assessment methods, the assessment accuracy of AFMSHI based on simulation platform data and actual production data reached 99. 5% and 98. 44% , respectively, which is higher than the compared methods by comprehensively considering the roles of MSHI. The effectiveness and the superiority of the proposed method are verified.