Recent Advances in Electrical & Electronic Engineering - Volume 9, Issue 3, 2016

Background: The global registration algorithm is adopted to optimize the ICP algorithm, each match point is matched simultaneously for the entire depth image registered with ICP, in order to obtain the optimal solution of translation and rotation matrix for images of all perspectives. Methods: In the process of dynamic reconstruction of a three-dimensional (3D) non-rigid image, some image regions overlap, in the traditional 3D non-rigid dynamic image reconstruction method based on FDK, all the overlapping regions are calculated with ICP, which greatly increases computational complexity. A dynamic reconstruction optimization method for the 3D non-rigid image is raised based on the global registration, the disadvantages of ICP algorithm are analyzed, and the global registration algorithm is adopted to optimize the ICP algorithm, each match point is matched simultaneously for the entire depth image registered with ICP, in order to obtain the optimal solution of translation and rotation matrix for images of all perspectives. Result: The feature vectors in the process of 3D image reconstruction are classified. The empirical knowledge between the feature vectors and information are utilized to train 3D reconstruction of learning samples. The simulation results show that the method shortens the time of 3D image reconstruction and improves the accuracy of 3D image reconstruction.

Background: The need for random and pseudorandom numbers arises in many cryptographic applications. Many cryptographic protocols require random or pseudorandom numbers at various points, e.g., for auxiliary data used in digital signatures, or challenges in authentication protocols. In NIST SP800-22 and Chinese randomness test specification, the focus is on those applications where randomness is required for cryptographic purposes and a set of statistical tests for randomness is described. In practice, these tests can be used to evaluate data generated by cryptographic algorithms. It is studied that the fast software implementations of Chinese randomness test specification in this paper. Methods: In this paper, the experiments show that the detection time of linear complexity test accounts for about two-third of all the testing time. We propose a type of fast implementation of the linear complexity test. The method is to follow the basic steps according to the test, replacing bit operations with word operations, where a word can be 32/64/128/256 bits, according to CPU processing power. Results: When the word width is set to 32, the improvement of the linear complexity test is much more significant compared with the basic implementation. Through the method, the efficiency of Chinese randomness test suit is increased by 2.41 times, compared with the basic implementation. Conclusion: we study the fast software implementation of Chinese randomness test specification. The hotspot item is detected and a type of fast implementation method of this test item is proposed in this paper. Through this method, the efficiency has been improved relative to the basic implementation. The method also can be used to NIST randomness test suit which has the same bottleneck item.

Background: With the rapid development of information technologies, digging out useful information from mass data has become a hot issue. We should cluster the data before the analysis. Human clustering of mass data cannot meet the requirement of data mining, therefore, various auto clustering algorithms come out successively. Spectral Clustering is a commonly-used cluster algorithm and the effect of spectral clustering highly depends on similarity matrix. Gaussian kernel method has the problem with selecting the good parameter. In real world data set, there is always noise. It is hard to select a good parameter to construct an ideal similarity matrix by Gaussian kernel function. Method: This paper proposes a similarity matrix constructing method based on locally linear embedding. This kind of graph is sparser than Gaussian method and has little noise. This method is not sensitive to noise compared with Gaussian kernel function. The experiments on real world data sets prove the effect of this method. Result: This paper starts from the locally linear expression relationship, uses the non-negative linear value constructing similarity matrix and gets a better experiment result.

Background: Multi-spectral imaging detection technology is a new generation of optical remote sensing detection technology, it can obtain exact color information, and according to features of acquired data it can get more spectral information, at the same time, the multi-spectral imaging technology can overcome the phenomenon of “same color different spectrum”, it can make imaging detection and identification of target object accurately and effectively, according to the structure of the image sensor, imaging principle, data read mode, and on the basis of a variety of spectral imaging system, the principle and structure of multi-spectral imaging system were proposed based on single camera and FPGA chip, the designed system can receive the four parallel outputs of CMOS sensor data, and the system takes them as four separate signals respectively in image processing. Methods: Multi-spectral imaging system overall design system overall design scheme of structure, it includes both hardware and software systems, hardware system includes the construct of multi-spectral imaging system, system calibration, spectral imaging experiment and imaging difference comparative study, etc. Software system includes the design of the spectrum reconstruction model and algorithm. Multispectral imaging detection technology can be in more than one through pre-selected characteristic spectrum band imaging detection of target scenario, and can well realize the authenticity of a color close target recognition, clutter background suppression and other traditional image detectors can't complete the task of wide band. Although, multi-spectral imaging detection technology compared with hyperspectral imaging detection technology in spectral imaging can greatly reduce the number of wavelengths, it cannot reach identification precision of the target species, and be adapt to more class scene, the pixels, target recognition, and other functions. But in specific scenarios, through study in advance good target and the background signal intensity differences of spectral bands, the background noise ratio target can be improved, the detection performance can also be improved. Most importantly multi-spectral imaging detector has simple structure, small volume and can realize miniaturization of tactical observation equipment, and thus it can meet the needs of variety of occasions. For a long time, multi-spectral imaging sensor does not get more attention not like a hyperspectral image sensor, and related theoretical innovation and technological innovation are lacked. Aimed at the problem according to the image plane segmentation imaging patent, we design a novel visible wavelengths integral four band spectral imaging optical path structure. The four bands imaging optical path structure which is similar to color multi-spectral imaging of target identification, in the experiment narrow-band filters imaging in different regions of the spectrum, clustering techniques to target the multispectral image data are analyzed, and the multispectral imaging detection technology in target recognition application are verified. Conclusion: In the paper, it presents overall structure of multi-spectral remote sensing imaging system and the imaging unit, optical path system, the control unit and the design of the mechanical structure, and repeated in different time, different illumination conditions, different imaging angle repeated multispectral imaging test, through the preliminary analysis of the experimental data, the multi-spectral imaging system can satisfy the specific spectrum, multi-spectral imaging of different bandwidth requirement, but in different frequency band its imaging ability can be further improved.

Background: The remaining satellite lifetime for scientific and commercial space mission can depend on the available electrical power from the solar generator. The prediction requires a detailed understanding of the degradation mechanism of the applied solar cells, their electrical parameters and the space weather conditions such as ultraviolet light, micrometeoroids or charged particles. Methods: A network simulation is performed for a high power solar generator containing solar cells, wires and protection diodes. The daily in orbit performance has been studied for operational cells, short circuited cells and cells in open circuit mode. The simulation is reinforced by comparing it to daily flight data of single cells, sections and the whole solar generator. Results: The daily in-orbit power has been successfully predicted and satellite lifetime has been assessed. For this goal the available telemetry only needs to contain view cell/section parameters but no current voltage curve measurements. Operational versus short circuit current coefficients offer an alternative approach for maximum power point voltage knowledge. Conclusion: The total generated in-orbit bus current at an operational point and the current/voltage characteristics can be simulated for various flight mission cases, so that the satellite lifetime can be assessed.

Background: This paper presents an indirect matrix converter (IMC) fed permanent magnet synchronous motor (PMSM) drive system using only one current sensor in dc link. Method: On the input side (rectifier) of IMC, the space vector modulation (SVM) algorithm based on the dc current is proposed to realize unity power factor. On the output side (inverter) of IMC, the reconstructed algorithm of three-phase output currents is proposed. SVPWM (Space vector pulse width modulation) for the inverter stage of IMC and vector control of PMSM are presented. Conclusion: The prototype of IMC-fed PMSM drive system is implemented and the experimental results validate the performance of the proposed method.

Background: Modern radars have adopted CFAR type of adaptive processing techniques to alleviate the damaging effects of the unwanted returned signals. Owing to the presence of spurious targets in practical operational environments, it is important to search a CFAR algorithm that performs well in this situation. An interference-saturated environment is commonly encountered as one of such background environments. Additionally, the characteristics of the radar targets can be regarded as the actuating tool for the radar systems to be practically designed and operationally assessed. If there is a relative motion of the target with respect to the radar, the aspect angle will be varied resulting in fluctuations in the radar cross section (RCS). The rate of these variations changes from independently in the case of consecutive pulses to significantly in the case of successive scans. Methods: The SWI model is characterized by full-correlation between the two consecutive echoes while they are de-correlated for the SWII models. Among these two boundaries, there is an interesting class which is termed as moderately fluctuating χ targets. If a coherent pulse train illuminates this class of targets, a train of partially-correlated pulses will be returned. Results: The CFAR detection of this class of radar targets in an interference-saturated environment is therefore of practical interest. The scope of this paper is to evaluate the performance of the doublethreshold (DT)-CFAR processor for the situation where the radar receiver based its detection on a collection of M-pulses. An exponentially correlated return from targets exhibiting two degrees of freedom χ statistics is the nature of these pulses. Conclusion: The detection performance analysis, in its exact form, is presented in the case where the operating environment is free of or contaminated with outliers. For specific values of the trimming threshold, the processor detectability loss becomes very low and the performance impairment, owing to interferers, is extremely small even for large number of interfering samples. Additionally, there is an enhancement in the detection performance when the size of the post-detection integrated pulses augments and/or the dependence of the target returns decreases, as was predicted.

Background: This paper designs an inverter conversion device to solve the problems in the conventional control technologies including the larger occupation, the worse transplantability and the inconvenient maintenance, etc. Methods: The sine pulse width modulation based on the equivalent area method is reviewed. Results: This design emphasizes on the study of the equivalent area method, and simulates the generation algorithm of the sine pulse width modulation through MATLAB software. Comparing with the traditional method based on the analog circuit modulation, the equivalent area method performs the good effect in the waveforms of the voltage output and the current output. After completing the design of the main circuit and the control circuit, the paper compiles the algorithm program based on the equivalent area method and realizes the control of the frequency conversion output. The test results are almost the same as the expected results so as to validate the feasibility of the designed inverter conversion device based on the equivalent area method. Conclusion: The device exhibits not only the good sine degree of the output waveforms on the alternating current side, but also the better transplantability, portability and practicability.

Background: The use of battery energy storage in the renewable energy applications presents a problem related to the limit of their lifetime. This limit of battery lifetime is generally related to the intermittent operation conditions or the low quality of the manufacturing process. Among the processes involved in the manufacturing of lead acid battery, the formation process is a key stage in which the cured plate is converted into active mass such as lead dioxide (PbO2) in the positive electrode and spongy lead (Pb) in the negative electrode. During this process, the formation current profile should be taken into accounts the specific chemical processes that occur in the battery and reduces the gassing amount to increase the formation efficiency with minimum time. Methods: The goal of this work is to analyze the impact of the formation current profiles on the lead acid battery quality by adopting an approach based on the dependability analysis tools. This approach presents a functional study of the soaking and formation processes of lead acid battery with Structured Analysis and Design Technique (SADT). The detailed analyze of the lead acid battery degradation during the formation process is described by the Causal Tree Analysis (CTA). The limit of battery performances is generated by the low electrical penetration into the pores or the corrosion of the electrodes. Then, qualitative and quantitative analysis of the battery performance under the different formation profiles is given by analyze the gassing amount. Results: The results indicate that the multistep current profile ensures the high efficiency of formation with the minimum gassing energy compared to the other profiles. Conclusion: This profile allows avoiding the corrosion of electrode by overcharge in order to increase the battery performance and lifetime during the operation conditions.

Background: Because the permanent magnet servo motor (PMSM) has the advantages of high efficiency, big torque, energy saving and environment protection, more and more researchers pay attention to it in recent years. This paper did research on the application of the equivalent thermal network method in the motor temperature field calculation. Methods: The basic theory and the calculation of traditional thermal network method were analyzed. Considering the relatively high deviation of traditional thermal network method, the paper proposed a new method of decreasing the deviation. Taking a 10kW PMSM as an example, the theories of energy conservation and heat conservation are considered, and a new thermal network node model was built. In this model, the key node division was optimized in the position where the heat transfer difference is great. The difference of heat transfer capacity at the edge of the thermally conductive materials and the influence of external heat flow on the calculation of equivalent thermal resistance had been considered properly. Results: The temperature field of the motor was calculated by traditional and optimized thermal network methods respectively. In order to verify the optimized method could improve the accuracy of the traditional method, the temperature field of the motor was also calculated by the finite element method. Comparing the results of traditional and optimized methods with the finite element method, it proved that the optimized method could decrease the difference between the results of equivalent thermal network method and finite element method, and the practicability of the optimized method is also proven. Conclusion: The optimized method provides a new method to increase the accuracy of equivalent thermal network method.

Background: To make up for the deficiency existing in single method for transformer fault diagnosis, a model of multi-algorithm fusion based on improved Dempster-Shafer (D-S) evidence theory was proposed through analyzing the implementation process of quantum particle swarm optimized BP neural network (QPSO-BP). Methods: According to the failure modes of transformer, the primary fault diagnosis was achieved using a model group formed by several single methods, such as QPSO-BP, the inertia weight PSO optimized BP network (IWPSO-BP) and the constriction factor PSO optimized BP network (CFPSOBP), then the fusion decision was implemented by D-S theory. In view of the defect of standard D-S which can not synthesize the highly conflicting evidences, the credibility factor was used to improve the capability of information fusion. Results: Diagnostic results show that, compared with the single models and standard D-S, the proposed method has stronger fault tolerance, and improves the accuracy of transformer fault diagnosis. Conclusion: The method based on the multi-algorithm fusion can enhance effectively the diagnostic efficacy, and suitable for the pattern recognition of transformer fault.