Abstract:In order to reduce the adverse effect of environmental temperature change on the reliable measurement of piezoresistive pressure sensor, a compensation method combining cubic spline interpolation and Hermite interpolation is proposed in this paper. Firstly, the calibration data are obtained through the calibration experiment, and the cubic spline interpolation is used to establish the functional relationship between the ambient temperature and the output voltage of the sensor and the pressure to be measured to compensate the temperature error of the sensor. Then the Hermite interpolation is used to construct the mapping relationship between the output voltage and the pressure to be measured to describe the measurement characteristics of the sensor. The results of temperature compensation experiment under two different working conditions show that the maximum qualified error are 2. 414×10 -4 and 6. 129×10 -4 respectively, the average qualified error are 2. 353×10 - 5 and 1. 313×10 -4 respectively, and the variance of qualified error are 1. 751× 10 -9 and 1. 613×10 -8 respectively, zero temperature coefficients are 2. 780×10 -7 and 8. 862×10 -7 , sensitivity temperature coefficients are 1. 952×10 -6 and 3. 672×10 -6 ,which verifies the consistent effectiveness of the compensation performance under different modeling data conditions.

Abstract:Geomagnetic map is an important factor affecting the performance of geomagnetic matching navigation. The random errors of the geomagnetic map seriously affect the accuracy of the matching and positioning, even lead to mismatching. In order to improve the geomagnetic matching performance, a method for modeling and compensating the random errors of geomagnetic data is proposed. This method establishes a non-stationary time series model of the data based on the analysis of the random error characteristics of the geomagnetic map data, uses Kalman filter, which the state equation is the times series model and the measurement is the real-time data, to filter the geomagnetic map data and compensate the random error. The effectiveness of the filtering method is indirectly proved through the navigation and positioning experiments based on the geomagnetic data before and after filtering. The actual geomagnetic map data processing results show that the geomagnetic map data filtered by the method in this paper can improve the positioning accuracy by 54. 7% when it’s used for geomagnetic navigation.

Abstract:The interference field caused by the airborne platform is one of the main sources of error in aeromagnetic measurement. In order to improve the accuracy of the aeromagnetic measurement of the geomagnetic vector, a vector aeromagnetic compensation algorithm based on component constraints is designed, and a compensation model including fixed, induced and eddy current magnetic interference is established. By introducing the attitude information of the carrier obtained by the inertial sensor, the problem of solving the compensation coefficient is linearized, and the parameter is solved by the least square method. The simulation results show that the algorithm can accurately estimate the compensation coefficient and reduce the root mean square error of the magnetic field vector to within 3 nT. The flight verification test was carried out using the helicopter aeromagnetic measurement system. After compensation, the root mean square error of the total field was reduced from 716. 97 to 14. 27 nT, and the root mean square error of the three components was reduced from 422. 86, 240. 68, and 676. 21 nT to 54. 21, 52. 34, 38. 61 nT, respectively. The results show that the algorithm can significantly improve the dynamic measurement accuracy of the aeromagnetic vector.

Abstract:With the development of disciplines such as climate diagnosis, climate change and weather forecasting, the measurement accuracy of the sounding temperature sensor has increased to the order of 0. 1℃ . Due to the interference of factors such as solar radiation, lift-off speed, cloud in and out of the cloud, the sensor measurement error can reach 3℃ or even higher, which has become the main obstacle restricting the improvement of meteorological detection accuracy. Aiming at this problem, firstly, the optimal design scheme of the temperature sensor is obtained through three-dimensional modeling and fluid mechanics analysis. The measurement error is minimized from the sensor morphology design. Then, by analyzing and summarizing historical meteorological observation data, the first domestic high-altitude observation dataset containing 900 000 detection records based on the real environment was constructed to solve the problem of deviation between the simulated environment and the real environment. Finally, Morlet wavelet is used as the activation function of the deep neural network. The support vector machine, XGBoost, deep neural network, and linear regression are combined to construct a prediction model for the measurement error of the sounding temperature sensor. After the error prediction model proposed in this paper, the average error is reduced from 0. 817 to 0. 008, the root mean square error is reduced from 0. 878 to 0. 068, the standard deviation is reduced from 0. 458 to 0. 204, and the fitting coefficient R 2 is 0. 93. The measurement accuracy of the temperature sensor has been significantly improved, which is more conducive to the development of relevant content of the meteorological discipline.

Abstract:Aiming at the problem that the power interference in the sensor transient response is difficult to remove due to frequency swinging, harmonic and frequency aliasing, a joint anti-power interference method of iterative ICA and LEVKOV is proposed. For the proposed method, the iterative ICA method is combined with LEVKOV method; iterative ICA method is first used to reduce the power interference with frequency swinging based on constructing reference signal of power interference, and then LEVKOV method is used to further remove the weakened power interference and its harmonic interferences by the means of mean filtering for linear segment and noise template filtering for nonlinear segment, so as to improve the filtering accuracy of the power interference mixed in the sensor transient response. The simulation and experimental data verification results show that, compared with the separated iterative ICA method and LEVKOV method, the proposed anti-power interference method of combining iterative ICA and LEVKOV has a better suppression effect on the power interference in the sensor transient response.

Abstract:Aiming at the problem that the input excitation of the traditional gravity field calibration method can only be within ± 1g, which cannot meet the calibration requirements of the high-range accelerometer in the full range, a calibration method suitable for the high-range accelerometer is proposed. First of all, analyzed the calibration principle and error sources and the accelerometer error compensation model was established, secondly, designed the twelve position calibration scheme that is suitable for high range accelerometer and calculation method for the calibration parameters is deduced, finally, through the precision centrifuge experiments, verified the feasibility of calibration scheme, the validity of the calibration parameters was verified by data processing. The results show that the calibration method is feasible and effective, and the compensation accuracy is higher than the traditional gravity field calibration method, which provides a certain theoretical basis for engineering application.

Abstract:Particle concentration is an important parameter describing the flow state of gas-solid two-phase flow. In this paper, the relationship between the particle concentration and the output capacitance signals of the ECT sensor is derived, and the on-line nondestructive particle concentration measurement approach is studied to detect the two-phase flow. The simulation experiment of particle concentration measurement is performed using COMSOL and MATLAB, by which the accuracies of the image-based and capacitancebased particle concentration measurement approaches are quantitatively evaluated. A preliminary experimental study on particle concentration measurement of gas-solid two-phase flow is carried out in a circulating turbulent fluidized bed, and the experimental results show that the image-based particle volume concentration measured at the bed height of 200, 300 and 400 mm fluctuates in the range of 0. 26~ 0. 75, 0. 09~ 0. 33, 0. 04~ 0. 31, respectively, while the particle volume concentration measured by capacitance-based method is in the range of 0. 27~ 0. 68, 0. 09~ 0. 47, 0. 04~ 0. 26, respectively.

Abstract:Aiming at the problem that when the Sage-Husa adaptive Kalman filter ( SHAKF ) algorithm is processing inertial measurement units ( IMU), random errors are likely to accumulate over time and cause filter divergence, an improved Sage-Husa adaptive robustness Kalman filter (MSHARKF) algorithm is proposed. First, build a suitable model for IMU, then combine SHAKF with adaptive robust Kalman filter (ARKF) and incorporate it into the improved time-varying noise estimator, and then introduce the optimal adaptive scale factor (αk ) to the measurement equation Iteratively update, and finally get a new predicted covariance matrix to be substituted into the original equation. The experimental results show that through the Allan variance and root mean square error (RMSE), the static / dynamic data before and after the MEMS-IMU filtering is calculated and the random error noise is reduced to one ten thousandth and one percent of the original data, respectively. Compared with other algorithms in this paper, this method effectively suppresses the filtering divergence of the algorithm, thereby improving the measurement accuracy and long-term stability of the IMU.

Abstract:Aiming at the problem that it is difficult to extract the early characteristics of pipeline micro-leakage under complex environmental noise, this paper proposed a method based on variational mode decomposition-singular value decomposition(VMD-SVD) self-optimizing pipeline microleakage signal enhancement method. Firstly, the genetic iterative algorithm was used to optimize the VMD parameters [α,k] , and the singular value kurtosis difference spectrum was used to optimize the reconstruction order of SVD. Then, the leakage signals were decomposed by VMD with optimized parameters, and the decomposed modal components were screened and reconstructed by kurtosis analysis. Finally, order optimized SVD is used to nonlinear filter the reconstructed signals, so as to improve the signal-to-noise ratio (SNR) of micro-leakage signals. Simulation and experimental results show that: the signal enhancement method proposed in this paper improves the signal-to-noise ratio of simulation signals by 9. 32 dB, the correlation of pipeline micro-leakage signals has been increased by 5. 92 times, and the relative positioning error of cross-correlation leakage is reduced by 14. 34%.

Abstract:The cogging force of permanent magnet linear synchronous motor (PMLSM) affects the control performance of the motor. In this paper, a novel composite magnetic slot wedge ( CMSW) based on soft magnetic and hard magnetic materials is proposed to effectively suppress the thrust fluctuation of PMLSM. Firstly, the influence of material, size and spatial position distribution of magnetic slot wedge on the output thrust and thrust fluctuation of PMLSM is analyzed. the effects of the material proportion and position distribution of soft and hard magnetic materials on the electromagnetic properties of output thrust, thrust fluctuation and cogging force are studied. Taking maximum thrust and minimum thrust fluctuation as optimization objective, the size of composite magnetic slot wedge was optimized by orthogonal optimization method. It is shown that the new composite magnetic slot wedge can effectively reduce loss and thrust fluctuations by 79. 4%, while the average thrust is basically unchanged, which provides a new technology way for thrust ripple suppression of PMLSM.

Abstract:Aiming at the issues such as measurement ability limitation caused by the mismatch between the test probe and the radome under test, as well as the traditional calibration methods apply only for a particular single operating frequency, one kind of radome electrical thickness reflection test probe is proposed innovatively by use of magic-T and half-star cross section prism. By which, the mismatch suppression can be up to 30 dB within 8 ~ 12 GHz full frequency range of X band, and the linearity of the phase mapping between the test signal and insertion phase difference of radome is optimized, and with no calibration required. Using this probe, the reflection test signal can be easily converted into transmission signal, the thickness information of the radome can be obtained by simple phase comparison, and the probe can be used at different operating frequencies at same time.

Abstract:In order to meet the needs of online, automatic and intelligent measurement of the articulated arm coordinate measuring machine (AACMM), a self-driven AACMM is proposed and its measurement and control system is studied in this paper. The hardware system uses the computer as the controller, and is built with six driving joints and trigger measurement circuit. The software is designed with the structural framework of state machine and event structure. The motion control and the automatic measurement of sampling points in the planning path of the self-driven AACMM are realized. The test results show that the measuring machine runs smoothly with the support of the measurement and control system, and the self-driven AACMM can achieve a repeatability of 0. 038 mm for the small size ball and 0. 192 mm ( k = 2) for the larger size gauge block. The study on the measurement and control system in this paper lays a foundation for the calibration technology, error analysis and application research of the self-actuated articular arm coordinate measuring machine. .

Abstract:Aiming at the problem of attitude resolution distortion caused by angle of inclination measurement errors in measurement while drilling, a joint estimation method based on Kalman filter and cross-correlation extraction was proposed. Firstly, a limiting filter is used to filter the impact noise, and then Kalman filter is used to remove most of the whitening noise caused by vibration. Finally, the crosscorrelation detection method is used to extract the accurate radial tangential gravity acceleration, and the axial acceleration is smoothed, so as to complete the estimation and compensation of angle of inclination measurement error. Simulation experiments show that angle of inclination measurement error is less than 0. 1°. Compared with traditional methods, this method can greatly improve the precision of deviation angle measurement.

Abstract:A fault diagnosis method based on multi-task deep learning is proposed, which classifies fault diagnosis tasks into fault classification and defect severity recognition. The shared layer uses convolution neural network to extract fault characteristic information contained in monitoring vibration signal, and the two subtask modules use gated recurrent unit to classify fault and recognize defect severity respectively. In multi-task deep learning method, the two subtask modules can interact with each other through the shared layers to promote the feature extraction ability of the model and make the whole model have better fault diagnosis performance. Fault diagnosis experiments are carried out on bearing dataset and compared with fault classification single-task model and defect severity identification single-task model to verify the fault diagnosis performance of multi-task deep learning method. The experimental results show that the accuracy of the multi-task deep learning model is 99. 79% for both tasks on the test set. In order to further verify the feature extraction capability of the multi-task deep learning method, different degrees of Gaussian noise were added to the test set for fault diagnosis experiment. Under the condition of strong noise, the accuracy of the multi-task deep learning model was significantly higher than that of the single-task deep learning model. The research results show that the multi-task deep learning model can diagnose fault more accurately and has better denoising than the single-task deep learning model, which has certain practical value.

Abstract:In view of the nonlinear dynamic characteristics of rolling bearing vibration signal and the low accuracy of reliability evaluation, a rolling bearing health condition assessment method based on improved cross fuzzy entropy (ICFE) and Weibull proportional hazards model (WPHM) was proposed. Firstly, the original vibration signal is decomposed by improved DLMD (Crt- DLMD), and the effective component with the most fault information is selected for reconstruction. Then, the ICFE of the reconstructed signal is calculated by using the sliding mean instead of the original coarse-grained process. Finally, the ICFE is used as the covariate of WPHM for health status assessment. The life cycle data and experiments of rolling bearing from national aeronautics and space administration (NASA) and Xi′an Jiaotong University Changxing Shengyang technology (XJTU-SY) show that the proposed method can accurately and effectively evaluate the health status of rolling bearings.

Abstract:Indicator diagram is a curve that describes the pressure of the gas in the cylinder of a diesel engine as the crankshaft Angle changes. The combustion and running state of diesel engine can be obtained by analyzing the indicator diagram curve. Due to the characteristics of high noise and vibration in the working process of diesel engine, there are a lot of interference signals in the measured indicator diagram curve, which seriously affects the accuracy of analysis. In order to solve the above problems, a processing algorithm to eliminate the interference signal of the indicator graph curve is proposed. The algorithm effectively integrates mean filtering, median filtering, anomaly data replacement processing algorithm, improved anomaly data local filtering algorithm and five-point cubic smoothing algorithm, and it can effectively eliminate the interference signal in the indicator diagram curve. A 4190ZLC diesel engine was used to set the engine speed at 945 r/ min, the load rate at 30% and the speed at 1 000 r/ min, and the load rate at 50% for experimental verification. The results show that the algorithm can effectively filter out the high-frequency interference signals in the indicator diagram data, and obtain smooth and continuous indicator diagram curve while maintaining a small distortion of the indicator diagram curve.

Abstract:An infrared dim small target detection algorithm based on local product weighted contrast is proposed for the low detection rate and high false alarm rate of infrared dim small targets in complex backgrounds caused by pixel noise and high-bright edge interference. First, the mean value of the target area and the background area is calculated respectively, and the difference between target and local background is obtained. A local product weighting method is proposed, which greatly improves the salience of small targets and the suppression ability of background clutter. Second, multi-scale algorithm is used to enhance the adaptive ability of the algorithm. Finally, adaptive threshold segmentation is performed on the saliency image to obtain the real target to be detected. Simulation results show that compared with the existing algorithms, SCRg and BSF of the proposed algorithm are improved to a certain extent, and still have good accuracy and robustness under the complex background and strong noise interference, achieving the purpose of improving the detection rate and reducing the false alarm rate.

Abstract:Aiming at the problem of low accuracy of transformer fault diagnosis, a random forest-recursive feature elimination (RFRFE) algorithm and an improved sparrow algorithm ( ISSA) optimization of the extreme gradient boosting tree (XGBoost) transformer fault diagnosis method are proposed. First, based on the diagnostic accuracy, the RFRFE algorithm is used to select important feature variables to remove redundant features; then the traditional sparrow algorithm ( SSA) is improved by the uniform distribution random adjustment strategy and the Levi flight strategy, and the ISSA and SSA and particle swarm optimization ( PSO) performs algorithm performance testing, which proves that its classification accuracy and network optimization capabilities have been improved; finally, the improved sparrow algorithm is used to optimize XGBoost related hyperparameters to obtain the synthesis of the combination of RFRFE and ISSA-XGBoost. The fault diagnosis model is compared with the PSO-XGBoost and SSA-XGBoost fault diagnosis models. The results show that the fault diagnosis rate of ISSA-XGBoost is 91. 08%, which is 9. 9% and 6. 93% higher than that of PSO-XGBoost and SSAXGBoost, respectively. The proposed method can effectively improve the performance of transformer fault diagnosis.

Abstract:In order to obtain a more accurate three-dimensional distribution of the magnetic field under the transmission line, an intelligent optimization algorithm is used to optimize the number and position parameters of the simulated current in simulation current method, which solves the problem that the position and number of the simulated current in the normal simulation current method can only be determined by experience, so as to improve the calculation accuracy. A three-dimensional calculation model based on the actual physical shape of transmission lines was established to calculate the three-dimensional magnetic field distribution of Dongjing-Sijing 500 kV line in Songjiang district, Shanghai by different methods. By comparing the calculation results with the actual measurement results, the results show that the calculation error of the optimized simulation current method is 4. 54%, which is 7. 67% lower than the 12. 21% of the traditional calculation method. It provides a theoretical basis for optimized simulation current method to calculate the magnetic field distribution of transmission lines, and has a theoretical guiding significance for the construction of high-voltage transmission lines.

Abstract:Aiming at the problem of low accuracy of straw coverage automatic recognition, it is proposed that a more accurate and adaptive method is used to detect straw coverage rate. Firstly, based on the color component spatial distance graying algorithm, the object and background of straw image collected by camera are separated; Secondly, the color image is grayed; Lastly, the straw image is binarized by improved Bernsen algorithm, and the straw coverage rate is calculated. In the experiment, 200 pictures of straw coverage are selected with coverage range of 20% ~ 30%, 30% ~ 40%, 40% ~ 50%, 50% ~ 60%, 60% ~ 70%, 70% ~ 80% and 80% ~ 90% respectively. The straw coverage rate is calculated respectively by the improved Bernsen algorithm and unimproved Bernsen algorithm. The result shows that the improved Bernsen algorithm is more accurate when the straw coverage is between 30% and 80%, and the error is less than 5%. In other cases, the calculation error of straw coverage rate is between 5% and 10%.

Abstract:It is of great significance to realize the accurate prediction of the temperature of power equipment to ensure the safety of the power system and to improve maintenance efficiency. Traditional forecasting methods cannot meet the requirements of high-precision forecasting. A temperature prediction method for power equipment based on an improved long short-term memory ( LSTM) neural network is proposed, which uses de-pooling convolutional neural networks (CNN) to extract local features of time series, and then use the recursive layer designed by LSTM to extract the long-term features of the time series to realize the temperature prediction of electrical equipment. Experimental results on the Power monitoring temperature data set of Capital International Airport show that the prediction accuracy of the temperature prediction value is better than 1 ℃ within 20 to 60 minutes, and the root mean square error (RMSE) 0. 12 is smaller than other temperature prediction models.

Abstract:In order to accurately obtain the global characteristics of motor imaging EEG signals and the common characteristics between individuals, and then improve its classification accuracy and model robustness, a fusion convolutional neural network algorithm with parameter sharing transfer learning is proposed. First, the trained model on the source domain is migrated layer by layer to the target network to obtain the best migration layers. Secondly, after the migration layers, different numbers of convolution-pooling blocks are connected to form four convolutional networks with different depths, and they are merged in parallel and finally the classification results are obtained through the classifier. Use the BCI competition IV Datasets 2a to conduct experimental analysis on the proposed method. The results show that the average recognition rate of all subjects when using 100% and 50% samples is 80. 85% and 78. 9%, respectively, which verifies the effectiveness of the proposed method on global feature extraction and the advantages of small sample problems.

Abstract:The vehicle honking detection system is an innovative application of traffic management department in the field of intelligent transportation in recent years. The sound source location unit can capture the illegal honking signal of vehicle and locate the vehicle honking source by microphone array technology. In this paper, the feasibility of microphone array technology applied to the vehicle honking detection system is verified by simulation of sound localization algorithm, where we clearly point out the importance of calibration of vehicle honking detection system. Then, two calibration methods of vehicle honking detection system are studied, which include anechoic chamber testing and field testing. The validation in anechoic chamber mainly measures the location accuracy of microphone array to the sound source, while, the field testing is designed to measure the accuracy rate, the capture rate and the false alarm rate under different working conditions. According to the validation method, an experiment has been done and the results show the validity of this calibration technology in the application of vehicle honking detection system.

Abstract:The current verification of DC energy meters ignores the impact of dynamic changes in power on the measurement results, and lacks DC energy metering standards under non-ideal conditions. For this reason, proposed a method of using pulse virtual power source to calibrate DC electric energy meters. This method applies pulsed virtual power to the tested meter, and obtains the error by comparing the output electric energy with the electric energy measured by the tested meter. It realizes the traceability of the value of the output electric energy. Based on this method, developed a calibration device, and calculated the error of the instrument’s output energy through verification and experiments. Using this device to test the existing DC electric energy meters on the market, and the influence of the dynamic change of power on the DC electric energy measurement is analyzed by combining the test results and the algorithm simulation results. The results show that the pulse virtual power source method can be used to establish DC electric energy metering standards and realize the verification of DC electric energy meters under dynamic power changes.

Abstract:To improve the measurement accuracy in wind tunnel experiments for short blunt model which has larger axial force and shorter model length such as landing patrol device and lunar return cabin. A six-component large axial force strain gauge balance based on complex structure was developed. The complex structure balance was compact design and has larger axial force capacity compare with the conventional series structure balance, meanwhile, the complex structure balance could be installed in the model lumen completely to reduce the influence of temperature. The results of numerical simulation indicate that the mean strain, sensitivity and strength of complex structure balance meet the measurement requirements of the wind tunnel experiments. Calibration data show that the accuracy of the complex structure balance is less than 2. 1%, which satisfied the technical index of National Military Standard.

Abstract:The detection of high pressure steam leakage in power plant is related to the long-term stable operation of power plant equipment. In order to improve the accuracy of high-pressure steam leakage detection in power plants and solve the problem of wrong segmentation and leakage segmentation of leakage areas, this paper proposes a high-pressure steam leakage detection algorithm based on CBAM-Res_UNet image segmentation network. The residual_block of ResNet is added to the UNet structure to obtain more semantic information of leakage images, and CBAM is integrated to strengthen the learning of regional characteristics of high-pressure steam leakage images. According to the influence of different loss functions and evaluation criteria on image segmentation results, the loss function Focal Loss+Dice Loss and performance index F1_score are selected. Through the experiment on the image data set of highpressure steam leakage in power plant, the F1_score obtained by CBAM-Res_UNet network is 0. 985. The experimental results show that the network can more completely segment the steam leakage area, and has a strong generalization ability for the variety of high pressure steam leakage images.