Abstract:Effective feature extraction of the heat exchange tube fouling signal is the essential step for fouling thickness detection. In view of the echo energy decentralizing and model aliasing, a signal processing method based on CEEMD wavelet adaptive threshold is proposed. Firstly, the similarity between intrinsic mode function (IMF) and original signal is calculated by the angle cosine method. The signal and noise mode segmentation point is determined and evaluated combining with the energy spectrum. Besides, the wavelet adaptive threshold is used to collect detail information in noise modes. Finally, all of the remained IMFs are reconstructed to obtain a noise suppressed signal. The results show the accuracy of segmentation point is high. Improved CEEMD has better denoising performance than wavelet threshold. The numerical simulation matches the test results, proved that the proposed method is significant to extracting the feature of the thin fouling signal.

Abstract:In order to reliably diagnose the open circuit fault of IGBT in active power filter, a fault feature extraction method of IGBT in active power filter is proposed based on multi  feature fusion. The method acquires the voltage across the clamped diode bridge arm in the threelevel APF main circuit as the measurement signal, to which wavelet decomposition is conducted. The energy coefficient, power spectrum entropy and singular spectrum entropy of each frequency band are extracted to compose the multifeature parameter matrix. Then, the feature dimension reduction is conducted to compose the eigenvector matrix. On the basis of theoretical analysis, corresponding experimental analysis was performed. Firstly, the measurement waveforms under different working states were obtained based on the above measurement signals, which were compared with those of other measurement signals; then, the kernel fuzzy Cmeans clustering method was used to analyze the distinguishing performance of the extracted features for the fault type, and the adaptive experiment of feature extraction on a threephase rectifier bridge harmonic source under load mutation and triggering angle change was conducted. Finally, the practical experimental platform of APF is built for further testing.The experiment results show that the measurement method based on the voltage across the diode can effectively distinguish different working states, and the adopted multifeature fusion extraction method overcomes the onesidedness of single feature extraction method, and has good distinguishing performance under various working conditions.

Abstract:With the development of high speed and high precision NC machining technology, high cutting stability of NC machine tool is required. The uncertainty processing is insufficient in the traditional cutting state monitoring. In this paper, an uncertainty processing method for cutting state monitoring is proposed based on modal interval theory. The uncertainty in traditional monitoring methods is described by using the width of modal interval to solve the monitoring uncertainty problem. In order to verify effectiveness of the proposed method, a cutting experimental platform is built. The cutting information of the NC machining is obtained by acceleration sensor. The cutting states are divided into three processing stages: stable, transition and chatter state by using timefrequency analysis. The interval feature of the different stages is extracted by using the wavelet packet energy percentage based on modal interval. The interval feature is encoded by Lloyd algorithm, and regarded as the input vectors of generalized hidden Markov model. Finally, the cutting stats of NC machine are identified by generalized hidden Markov model state recognition method. The experimental results show that the proposed generalized hidden Markov model recognition method based on modal interval is superior to the traditional hidden Markov model recognition method.

Abstract:The phase difference estimation methods for Coriolis Mass Flowmeter are unsuitable for many high precision flow measurement areas, due to high computational load, high calculation complexity, low realtime performance, and poor accuracy. To solve these issues, a new phase difference estimation method for Coriolis mass flowmeter is proposed based on correlation and Hilbert Transform. Firstly, the signal frequency is estimated by FFT, which is used to determine the integrer period of the sampling signal, and the noninteger period sampling signals need to be extended. Then, the same frequency reference signal can be generated by using autocorrelation of original signal. Then, the Hilbert Transform is conducted to these three signals, and the correlation functions of these three signals with the transformed three signals can be computed. Finally, the formula of phase difference can be obtained by utilizing the sine functions. Compared with traditional methods, the proposed method is suitable for both integerperiod and nonintegerperiod sampling signals, and its accuracy, realtime and dynamic performance is superior. Simulations and experiment results verify the effectiveness and superiority of the proposed method.

Abstract:Using the maximum of twodimensional wavelet transform coefficient modulus as wavelet ridge will produce large error for the fringe image with noise interference. In view of this problem, wavelet ridge extraction algorithm utilizing a cost function in twodimensional wavelet transform is proposed. Firstly, the maximum point is extracted from twodimensional wavelet transform coefficient modulus, and the local maximum points exceeded 90% of maximum point are also obtained, these points are selected as wavelet ridge candidates. Then, the gradient of scale factor is introduced into the modulus, the cost function is established to evaluate the value of all candidate points. The logarithmic Logistic model is used to adjust the weights to improve the estimator. Finally, the dynamic programming is applied to accurately identify the optimal wavelet ridge, and the wrapped phase can be obtained by extracting the phase at the ridge. Consequently, the fringe pattern with low signaltonoise ratio can be demodulated accurately, and its noise immunity is better than wavelet ridge extraction from direct maximum modulus. At the same time, only one fringe pattern can be projected to reconstruct the shape of object, which can be used for dynamic 3D measurement in harsh environment. Simulation and experimental results show that, for the fringe pattern with noise, the accuracy of 3D surface recovery by the proposed algorithm is increased, compared with the maximum modulus of the wavelet ridge extraction algorithm. And the computation time is shortened by 46.9% compared with the extraction of the whole local extreme points. In addition, , simulation results show that the 2D Cauchy wavelet has better directivity and higher accuracy by applying different mother wavelets to the proposed method.

Abstract:The accelerometer’s dynamic model plays an important role in describing and analyzing its dynamic characteristics. For the influence of unknown delay time on the accuracy of accelerometer’s dynamic model parameters identification using the calibration upon shock excitation, a dynamic model parameter identification method of accelerometer using delay time correction (DTC) and least square (LS) is presented. In this method, the delay time is firstly introduced into the dynamic model of accelerometer, and the frequency response function is estimated by using Fourier transform of the accelerometer’s shock excitation and response data. Then, the LS method and minimizing the error criterion function of phase are utilized to correct the delay time through continuous iteration. Finally, the dynamic model parameters of accelerometer are estimated, based on the delay time corrected data which consists of shock excitation and response of accelerometer. The experimental results of simulation and the calibration upon the shock excitation show that the presented method can reduce the influence of the delay time effectively on the accuracy of accelerometer’s dynamic parameters identification, and the dynamic model parameters of accelerometer can be obtained with higher accuracy.

Abstract:In order to establish the relationship between the microstructure of surface topography and its functional properties, the percolation model of surface topography is established based on percolation theory. The percolation characterization of three dimensional (3D) surface topography is achieved with the percolation probability, the mean size of void clusters and the distribution coefficient of void clusters. The digital rough surfaces with given autocorrelation function (ACF) and orientation parameters are generated based on digital filter method. The percolation characteristics of 3D rough surfaces with the same root mean square (RMS) roughness and different structures are analyzed. The effects of surface texture parameter and autocorrelation length on the percolation characteristics of surface topography are obtained. The quantitative relationship between surface topography and percolation parameters is established by 3D characterization parameters (ISO25178). The results show that the surface height, the percolation threshold and percolation volume decrease with the increase of the surface texture parameter while searching the spanning void cluster of the anisotropic surfaces along the transverse direction. The distribution coefficient of void clusters increases with the increase of surface texture parameter. The changing law is opposite to the transverse while searching along the longitudinal direction. For the isotropic surfaces, the surface height and the percolation threshold decrease firstly and then increase slightly. Besides, the percolation volume and the distribution coefficient of void clusters decrease gradually with increase of the autocorrelation length. This study provides a theoretical basis for the design of functionoriented surface topography.

Abstract:The vertical wind array commonly used in threedimensional wind measurement technology has defects in eliminating the shadow effect. This paper designs a nonorthogonal wind array based on special tetrahedron edges according to the characteristics of flow field. The array is designed to reduce the influence of wake region turbulence on the mean wind velocity of the wind path, and to compensate the wind path which the round flow interferes severely. The wind array is modeled with GAMBIT software, and the flow field of gas velocity and Reynolds number of different environmental parameters are changed in FLUENT software to simulate the different performances of the two arrays in the low velocity laminar flow region and the high velocity turbulence region. The cloud chart of velocity distribution and the velocity line graph based on the path of three groups of transducers are obtained to prove that the nonorthogonal array can improve the accuracy of the threedimensional wind measurement.

Abstract:Aiming at the problem that inevitable outliers occur in GNSS position and velocity for land vehicular navigation under complex urban environment, which would deteriorate the estimation accuracy of GNSS/MEMSINS navigation state parameters and even lead to the filtering divergence, in this paper an adaptively outlierrestrained GNSS/MEMSINS integrated navigation algorithm is proposed to improve the accuracy and reliability of integrated navigation based on the faulttolerant adaptive Kalman filtering. This algorithm establishes accurate noise model for MEMSbased inertial sensors with Allan variance analysis technique, which reduces the influence of kinematic model mismatch and state disturbances effectively. The innovation sequences are used to construct the test statistic for detecting observation outliers. The adaptive innovation weighting factor is constructed according to the statistic to adjust the filter gain matrix, and weaken the adverse influence of observation outliers on state estimation. The field test result indicates that the proposed algorithm can effectively control the influences of GNSS position and velocity outliers, and has strong realtime and faulttolerant ability for GNSS/MEMSINS integration navigation. The estimation accuracies of position, velocity and attitude determination are improved by 35.78%, 60.19% and 82.41%, respectively compared with those of traditional algorithm, which verifies the effectiveness and practicability of the proposed algorithm.

Abstract:Industrial measurement can not only obtain product quality status information, but also can be an important information source for cyber physical system (CPS) in Industrial 4.0. However, industrial measurement is a complex technical system, including the definition of error/tolerance, measurement and evaluation method, comparison between measured results from different measurement methods, measurement system development and its performance evaluation, determination and acceptance of product quality, and etc. Besides, there is a complete set of ISO/GB (GPS&V) standards, as well as strict measurement process management system and product acceptance process. Industrial measurement operator must follow these management system and process. Focusing on product geometry quality and measurement, this paper analyzes the metering/measuring principles and methods, theory and methods of advanced quality design and control, standard system and its related theory and methods in industry measurement, and etc. Some development directions of industrial measurement technology are also discussed.

Abstract:Strokes per minute (SPM) is one of the important parameters in beam pumping unit operation condition. Realtime accurate measurement of SPM is the basic principle to enhance the pump efficiency substantially. In view of the traditional methods by the formula calculating and sensor measuring, many problems of the additional measurement components, the high malfunction ratio, the difficulties of repairing and the accuracy influenced in the field operating conditions need to be considered. A softsensing approach for beam pumping unit SPM based on autocorrelation algorithm is proposed. The periodic correlated model is formulated among the SPM, load function and the driving motor input power function. The realtime input power is derived according to the driving motor input current and voltage. There are two kinds of features on the autocorrelation function. One is the periodic signal autocorrelation function period which is the same as input power function. The other is the noise signal autocorrelation function concentrated on the origin. On account of these two features, the input power function period can be identified accurately after interfering removed by the autocorrelation processing. Furthermore, the SPM is calculated according to the relationship between the SPM and the period of the electric power function The beam pumping unit SPM softsensing is realized accurately without additional peripherals. Simulation and an oilfield test results show that the proposed approach is practical and effective, high antiinterference, and the measurement error is less than 1%.

Abstract:This paper mainly proposes an online interference measurement approach of the film thickness in sliderondisc conformal contact. With a monochromatic laser, the physical characteristic of the parallel straight interference fringes translation induces by variation of the film thickness. Based on optical flow calculation and dynamic time wrapping technique, a hybrid algorithm is introduced to calculate the displacement of an onedimensional intensity curve across the parallel fringes between two adjacent frames, therefore, the corresponding film thickness change can be obtained. From the start of the test rig, the film thickness variation of each frame is recorded and the online film thickness measurement is achieved. The film thickness by the presented approach is validated by its good correlation with that by the old offline method. With the new measurement system, the transient film thickness is measured in real time under conditions of step loads, uniform acceleration and deceleration, and some characteristics of film building are demonstrated.

Abstract:Laser scanning differential confocal microscopy measurement method is characterized with its nanoscale axial measurement precision. However, the method suffers potential low lateral resolution because its signal is acquired at defocusing positions. Furthermore, its measurement speed is limited by the sequential scanning of a laser beam. The structured light illumination microscopy imaging method based on spatial light phase modulator can achieve superhigh lateral resolution imaging, but does not have a matching high axial measurement precision. Therefore, neither of these two methods meet the requirement of online, onsite measurement of the microscopy surface topography of complex object in micronano machining manufacturing process. In this paper, combining the spatial light phase modulated structured light illumination microscopy imaging technique with differential axial measurement method, a parallel objectside differential rapid measurement method based on structured light illumination is proposed. The method uses only an area array camera as the detector to acquire multiple phase images of a sample at opposite but equal distance away from the focal plane under the structured light illumination phase modulated mode. The acquired images are then synthesized to obtain corresponding high resolution images, IA and IB at opposite defocused positions respectively. The differential signal ID is calculated as the difference between the two images. The surface heights of the sample at different positions can be obtained according to the precalibrated differential curve. The proposed method was applied to measure a standard block with the step height of 500 nm at the pitch of 10 μm. The experiment results show that the proposed method achieves the standard deviation of 2.8 nm and the relative error of about 0.6%; it takes only 65 ms to complete the surface topographical measurement with the image resolution of 2 048×2 048. The measurement results show that the method is suitable for fast online nanoscale high precision axial measurement, and can achieve fast 3D topographical nanoscale precision measurement with a speed of 15/s.

Abstract:Regular infrared dualband scene simulators are mostly designed with single DMD, which cannot modulate the two wave bands severally. The single DMD based design can only meet the requirements of working wave band, but is not that practical. To solve this problem, a dualDMD based dualchannel coaperture switchzoom compactstructure MWIR (Medium Wave Infrared)/LWIR(Long Wave Infrared) scene simulator is designed. The opticalmechanical structure of its main parts, including projection system, illumination system, and dichroic beamcombiner, are designed in detail. The optical engine adopts telecentric system to illuminate DMD target surface directly, which uses vertical space overall arrangement to avoid interference from different light path. Moreover, the engine uses two smallsize metal plane mirrors to compress illumination light path, so that system integration degree is increased. The beamcombiner is fixed in half kinematics elastic way, we calculate pinchcock’s deflection for the least preload, and its bending stress formed in material. Working temperature of black bodies in two wave bands can be calculated with Matlab, which also fits the proportional relation ψ(T) curve of afteropticalsystem radiation remittance of two wave bands and black body radiation remittance of overall infrared wave band. The simulation testing shows that when black body temperature reaches 850 K, it meets the highest apparent temperature requirement. The contrast ratio of midwave image is 250∶1, and the contrast ratio of longwave image is 14∶1, which satisfies the operating requirements of infrared mid/longwave scene simulator at the present stage.

Abstract:The comparison and validation of ozone Standard Reference Photometer (SRP 48) by Institute for Environmental Reference Materials of MEP(IERM of MEP) and SRP 2 by National Institute of Standards and Technology(NIST) are performed in order to establish the metrological traceability. The evaluating the measurement uncertainty of SRP is focused and the main sources of measurement uncertainty are identified. Take SRP 48 for example, the measurement uncertainty components such as the optical path length (Lopt), the pressure (P), the temperature (T), the product of transmittances ratio of the two cells (D) and the crosssection (σ) are calculated, respectively. A measurement mathematical model is established and the standard uncertainty is combined. The results of the comparison show that SRP 48 and SRP 2 are comparable over an ozone mole fraction range of 0~500 nmol/mol. The relationship between SRP 48 and SRP 2 is xSRP48= ［(0.998 96(xSRP2)+0.025］ nmol/mol. The expanded uncertainty of SRP 48 ozone concentration measurement is 2×((0.28)2+(1.10×10－2×x)2) nmol/mol at 95% confidence interval using a coverage factor k=2.

Abstract:This paper optimally designs a hollowcore inductive sensor with bandwidth range from 1 Hz to 10 kHz. Firstly, we utilize a magnetic core with hollow structure, and analyze demagnetizing factor and efficient magnetic permeability of hollow core. The magnetic flux of hollow core sensor coil is simulated and measured, and is found equal to the one of rod core sensor coil with the same ratio of length and diameter. Then, sensitivity and noise of inductive magnetic sensor with hollow core are analyzed and the performances of low noise and low weight are achieved by optimizing the noise formula using mathematic algorithm. To verify the theoretical analysis, this paper designs an inductive magnetic sensor with hollow core and tests the performance of the inductive magnetic sensor with hollow core in shielded environment yielding sensitivity of 0.73 V/nT corresponding the frequency higher than 400 Hz and noise of 0.06 pT/Hz1/2 at frequency of 100 Hz. Under this condition, the whole configuration weighs 80 g. The process is consistent with the theoretical analysis. Compared with THEMIS, the inductive magnetic sensor with hollow core has merits of low noise and light weight, and can satisfy the needs of Space Electromagnetic Science.

Abstract:To detect the temperature and alcohol concentration during the liquor distillation, a dualparameter optical sensor based on MachZehnder interferometer (MZI) and fiber Bragg grating (FBG) cascade structure is fabricated. The sensor is formed by a 4order FBG which is fabricated by the femtosecond laser with linebyline method in single mode fiber (SMF). The transmitted MZI is based on coreoffset and large overlap splicing method between thin core fiber and SMF. The period of the proposed FBG is 2.2 μm. The Bragg wavelength is 1 591.21 nm. The depth of the transmission spectrum is 23 dB. For MZI, the length of thin core fiber is 8.7 mm and the interference contrast fringe is 28.5 dB. The temperature and alcohol solution concentration features of the sensor are analyzed by beam interference theory. FBG and MZI have different sensitivities to temperature and alcohol solution concentration. The sensitivity matrix is constructed and the temperature and alcohol solution concentration can be measured simultaneously. In the experiments, the sensitivities of concentration of alcohol solution and temperature can reach up to -41.37 pm/% and 58.96 pm/℃. Therefore, the proposed sensing structure has the potential application prospect in the liquor manufacturing industry.

Abstract:In regard to the technical problems of ordinary temperature sensors with the inadequacy of long response time to rapidly measure transient temperature and the difficulty to lead wires for thin film thermocouple, a thin film transient temperature sensor is developed with quick response time, high measurement accuracy and convenient to lead wires. The DC pulsed magnetron sputtering technology is applied to deposit NiSi functional thin film and SiO2 insulating thin film on the surface of ceramic substrate embedded with NiCrNiSi parallel electrode wires. The static performance of developed sensor is studied by utilizing a selfdeveloped static calibration system. Experimental results show that developed sensor has good linearity and thermal stability in the range of 50 to 400℃. The seebeck coefficient is 41.2 μV/℃ with a nonlinear error less than 0.05%, which remains stable with the change of NiSi film thickness. The dynamic performance of developed sensor is studied by using ANSYS finite element simulation and dynamic calibration experiments, which has a microsecond response time increased with the increasing of NiSi film thickness. In addition, the response time of developed sensor has little effect with the change of energy of laser pulse. The temperature test of thin film transient temperature is conducted through utilizing the temperature calibration furnace. The results show that the sensor can quickly respond to temperature changes and the developed sensor can provide effective methods and technical approaches for transient temperature test.

Abstract:This paper presents a current sensor for DC superimposed with impulse current waveform. The current sensor consists of two magnetic cores: one core works in fluxgate principle, and the other is based on Rogowski coil principle. A feedback winding is introduced into the sensor in order to integrate fluxgate principle with Rogowski coil technique. Introducing the feedback winding effectively overcomes the influence of the impulse magnetic field and DC magnetic field on the magnetic core working state. A prototype is developed based on the theoretical analysis. Test results show that the proposed current sensor can effectively measure DC superimposed with impulse current waveform with a wide bandwidth up to 340 kHz. The measurement error for DC is limited to 0.6%, and the maximum error is 1.5%, which is at cutoff point between the low frequency band and high frequency band.

Abstract:In order to automatically classify weld surface defect in header pipe joint, Computer Vision based defect classification is studied. The texture features of different weld defects are analyzed, Grey level cooccurrence matrix (GLCM) is applied to extract features from digital images, and 15 types of statistical indexes are obtained to characterize the weld surface defects. Backpropagation artificial neural network method is used for defect classification. The influence of GLCM parameters, the neural network structure and the number and variety of input parameters on the defect classification performance is analyzed, and optimal neural network structure and input parameters are selected. In further, the optimized network is utilized for training and classifying the images of different weld defects acquired by industrial endoscope. The results show that weld defects detection rate of overall classification can be up to 91%. The proposed method can be used for automatic classification of weld surface defect in header pipe joint.

Abstract:In order to solve the problem of fracture caused by the failure of the native defects of wind turbine blades, the mechanism of crack initiation is defined by analyzing the quantitative relation of energy release during the transition from mesoscopic defects to macroscopic cracks under external loading. Firstly, a new stress function is constructed according to the load characteristics of wind turbine blade. Based on the basic formula of the orthotropic composite, the stress intensity factor, the stress strain and the displacement of the native defect are calculated, and the plastic strain energy of the mesoscopic defects can be obtained. Using the infrared thermal imager, the temperature field of transition process of native defects is collected and the thermal energy dissipation is calculated. Then, based on the irreversible thermodynamics, the variation law of the internal storage energy with the fatigue cycles is obtained. Finally, the blade specimen with bubble and fiber fracture is selected for the fatigue test. The results show that the error between the calculated displacement using the new stress function and the test value is smaller. This indicates that the proposed stress function can be applied to calculate the plastic strain energy for the mesoscopic defects. When the native defects are transitioned into a small crack, the change of the internal energy storage can be used as to determine the type and extent of the defects. The fatigue energy theory of multilayer composites is explored in this study, which helps to realize the life cycle monitoring of key components of wind turbine.

Abstract:The equipment condition monitoring and fault prediction technology plays an increasingly important role in industrial equipment health management with the development of science and technology on instrument measurement and analysis, Internet of things, cloud computing, data mining, artificial intelligence. This paper studies a condition monitoring and fault prediction technology based on stress wave analysis. The electronic signal of friction, mechanical shock and dynamic load on equipment moving parts are detected and processed by stress wave sensor. The stress wave analysis is fulfilled by using the time domain and frequency domain feature extraction software and sensor data fusion is conducted based on neural network. The equipment states are quantitatively analyzed and the equipment fault is accurately predicted, so as to provide the equipment health diagnosis reports. The test shows that, compared with the traditional vibration analysis, the proposed system can monitor the equipment operation condition better in real time, predict the fault earlier. The production safety can be guaranteed, the equipment maintenance cost can be reduced, and the production efficiency can be improved.

Abstract:Natural gas pipeline leak monitoring is entering the age of big data. Aiming at the problems of traditional methods, such as redundant data, subjective feature extraction and identification, an intelligent pipeline leak aperture identification method is proposed combined compressed sensing (CS) and deep learning theory, which can achieve compressed sampling, adaptive feature extraction and recognition. The random Gaussian matrix is used to acquire the compressed acquisition data, and the aperture information contained in measured samples in CS domain is analyzed by deep learning. The sparse filtering is applied to realize the automatic feature selection. Finally, the high precision classification and recognition of the aperture is obtained by softmax regression. Experimental results show that this method realizes the compression of the monitoring data, and the identification performance for data of compressed sensing domain is better than traditional methods.

Abstract:Current vehicle detection method for fake plate vehicles has a high computational complexity, low detection accuracy, lack of robustness. This paper presents of fake plate vehicles detection method based on multitask Faster RCNN (regionbased convolutional neural network). Firstly, spatiotemporal constraint is used to obtain the suspected fake plate vehicle. Then, front part of the vehicle is located in the image using Faster RCNN. Next, the public face (basic characteristics of a vehicle) of suspicious fake plate vehicles is contrasted. In further, the subtle features of a private face (Annual inspection certificate for vehicles) is contrasted. This hierarchical visual inspection method, detected from macroscopic features of vehicles to microscopic features, has the advantages of fast detection speed, high robustness, strong generalization ability, convenient deployment and high detection precision. Experimental results show that detection accuracy are 99.39% and 99.22% on the Vehicle ID data set and the Hangzhou bayonet data set, respectively.

Abstract:Lake surface water temperature (LSWT) is an important factor in aquatic environment, which directly affects watershed ecosystem and biodiversity. Precise LSWT measurement and prediction is essential to control and improve the aquatic ecological environment of the river basin, and is also the key to prevent and control the outbreak of cyanobacteria bloom. Focusing on Dianchi Lake, 54 water quality parameters (LSWT, chlorophyll a, pH, permanganate index, dissolved oxygen, etc.) of 10 water quality monitoring sites from 2005 to 2016 are used as the data set. A hybrid forecasting model is presented composed of εsupport vector regression (εSVR), principal component analysis (PCA) and back propagation artificial neural network (BPANN). Moreover, Kriging method is combined with geographic information system (GIS) to realize the scene reproduction of the historical changes of the Dianchi lake LSWT and water quality in the past 12 years and the trend simulation of the next 5 years. Results show that the average relative error of the model is 0.5%, the mean square error is 1.452 3, R2 is 0.904 9. Spatial visualization results indicate that the region with LSWT over 20 ℃ diffuses obviously from north to south. The outbreak of cyanobacteria bloom changes from locally to globally, which is related to the expansion of Kunming urbanization and meteorological environment.

Abstract:Highway tunnel maintenance technical specifications clearly pointed out that the investigation of tunnel cracks is one of the special inspection items. At present, manual detection is often used and the missing detection is inevitable. In order to overcome this shortcoming, automatic detection with machine vision has become the main method in this field at home and abroad. In this paper, the background in machine vision methods, the present research on crack of tunnel concrete lining detection algorithms at home and abroad are reviewed. Lining image preprocessing, crack detection, interference elimination, crack width measurement and error analysis are included, while the advantages and disadvantages of different algorithms are compared. Finally, the conclusion and future vision are introduced.

Abstract:In order to register targets between two wave band infrared images, the overall structure feature is studied in this paper. A novel target registration algorithm for two wave band infrared images based on improved multiscale edge structural similarity is proposed by means of improving structural similarity algorithm, multiscale structural similarity algorithm and edge structural similarity algorithm in Gauss scale space. Improved multiscale edge structural similarity is calculated through using images of different scales and their edge images in Gauss scale space. Then the performance of target registration can be improved due to that target’s complete edge structure information in the infrared images is utilized fully. The experimental results show that the accuracy of target registration based on the proposed method is better than other algorithms, and the target similarity between two wave band infrared images is higher than 0.98.

Abstract:The bearing model and rotor model are firstly built in this paper. The connection relationships between the housing, bearing, shaft and tool of highspeed motorized spindle are analyzed. A dynamical model of highspeed motorized spindle is established, and the transfer functions of different parts of the motorized spindle to the tool are deduced. Then, based on the D62D24A type motorized spindle, the weakening effect of speed on bearing dynamical stiffness is analyzed. The first order inherent modal shape and inherent frequency are investigated numerically and experimentally for both situations of with and without the tool. The results show that both tool and speed have a “softening” effect on the inherent characteristics of system.

Abstract:In MMCHVDC system, the AC transient faults will bring overcurrent and overvoltage occurred on arms of rectification and inverter station, as well as abnormal operating state. To evaluate the operating state of MMCHVDC system is a new issue by using electrical information of MMC when AC transient faults or disturbance occur. To address this issue, this work firstly analyzes the influence of AC transient faults on the electrical signals in MMC in detail, and investigates the relationship of electrical information in MMC. Then, considering system damping, irregular change and complex frequency of the currents and voltages and disturbance feature with low energy in MMC, Renyi wavelet packet energy entropy (RWPEE) is utilized to extract feature of AC transient faults. Based on the feature extraction, the evaluation index set is obtained. In combination of Renyi entropy weight and grey relational analysis, a novel evaluation method is proposed to evaluate operating state of MMCHVDC system when AC transient faults occur. Finally, 2terminal 31level MMCHVDC is used to generate experimental data, and operating state of MMCHVDC system is evaluated. The experimental results prove the effectiveness of the proposed method.

Abstract:The energysaving issue attracts more and more attentions recently with the increasing installations of the elevators in the buildings. And super capacitors (SUPCAPs) are the preferred energyconserving devices equipped in the elevators to store the energy regenerated by the tracking motor. A new back propagation neural network based (BPNNbased) predictive control strategy is employed to predict the energy required by the elevator at every trip. According to the information of the current stop floor, the destination floor, and the weight of the passengers, the BPNN model is setup and trained by the samples. Then the trained BPNN is applied to predict the energy required by the elevator at the beginning of each trip. In this way, the energy provided by the SUPCAPs is determined and the balance voltage (BV) can be regulated to fully compensate the peak power. Moreover, the power capacity of the gridside pulsewidth modulation (PWM) rectifier can be reduced because the SUPCAPs provide as much energy as possible in every trip. Not only the peak power which emerges when the elevator moves in heavy load or full load eliminated, but also the rated power can be partly counteracted by the SUPCAPs. Finally, a simulation based on MATLAB/Simulink as well as the corresponding experimental prototype is setup to verify the proposed method, and results from the simulation and experiments prove the effectiveness of the proposed method.