Applied Mechanics and Materials Vols. 530-531

Abstract: Detection technology of thermal infrared targets in scene image is to automatically detect target in busy background and intensive environmental noises by image processing algorithm and then to obtain the gray-scale targets image. Intensity of thermal radiation have business with temperature, so when the temperature is high then the radiation intensity is high and the image tone is shallow, but when temperature is low then the intensity is low and the tone is deep. This article researched the firelight in scene image, and combined the second prediction difference method and Top-Hat target detection based on gray-scale morphology to obtain the thermal infrared target image. Further more, it did simulation experiments and research deeply image processing technology according to the actual scene image. Finally it summarizes the effective algorithms of detection and identification which have instructional meaning to optical remote sensing, targets detection and fire control engineering and so on.

Abstract: Multi-domain perimeter intrusion detection system by one machine is based on leaky cable. To solving the problem that the multi-frequency electromagnetic disturbance signal detection circuit is complex and repeated, a method is raised, which is without processing by different frequency, sending the intrusion disturbance signal from multisport to processing circuit to perform FFT, contrasting it with signal spectra without disturbance by frequency analysis, and determining if intrusion happened, the number of intrusion case, detailed place and domain. Simulation study is performing by Matlab, which demonstrated the feasibility and advantage in the project.

Abstract: Brain computer aided diagnostic system based on CT image has been widely applied for medical clinical field, which studies image preprocessing, feature extraction and image classification diagnosis based on digital image processing technology. This paper presents system design and realization of aided diagnostic technology for brain CT image. The dynamic grey level range of CT image is extended by adopting segmental linear stretching method at first. Then textural features of CT image are extracted based on GLCM (grey level concurrence matrix). BP neural network algorithm is used to design a classifier for textural features vector of CT image at last, which identifies normal and abnormal brain CT image. Test result shows that the method presented has good accuracy, quick speed and strong robustness for realtime application.

Abstract: The frame structure of monitoring system is introduced and the network structure of monitoring system is analyzed in this paper. Fault diagnosis expert system is put forward and the fault inference rules are elaborated according to the fault of natural gas compressor of a company. The significance of the system is summarized in the end.

Abstract: Power equipment condition monitoring diagnosis (CMD) system includes 4 parts, master stations in control center master station system, integrated processing unit in subsation, monitoring device, and communication network. To achieve a seamless communication with the condition monitoring device and master station, IEC 61850 shall be used. Based on IEC 61850, the configure file, information model, information exchange model are standardized.

Abstract: With the extensive use of electronic current transformer, check the original method cannot meet the requirements of electronic current transformer performance testing technology. In 500 kV Jincheng a variable electronic current transformer as the test object, the principle and design characteristics of electronic current transformer, proposes an effective test method, ratio and through field test of low power coils and Rogowski coils 2 coil error, phase error and polarity, verify the validity and accuracy of the method.

Abstract: The partial discharge (PD) phenomenon is very harmful for electrical appliances and its early detection could be a cost effective approach for the industry. Although many techniques are used for PD detection yet no technique has presented widely acceptable solution. Still the subject needs parallel study of the detection techniques. In this study, partial discharge signal has been captured by the three techniques using fiber optic sensor (FOS), Piezoelectric Sensor (PZT), leakage current (LC) techniques. In these experiments, FOS shows good sensitivity in the range of applied high voltage > 5 kV. The sensitivity and noise level of PD signal was different in these two experiments. The location of a PD was demonstrated in a 500 gallon water tank using convolution method.

Abstract: Damage detection is the most difficult part of structural health monitoring Systems. In order to acquire response information as much as possible, sensors should be carefully placed. Taking Tai Zhou bridge as an example and using the genetic algorithms (GAS), this paper discussed the optimal placement of sensors on suspension bridges for detection of accumulated damage for structural health monitoring systems. After a brief discussion on the classical GAs, and used an example to compare them with the classical GAs. The results show that the generalized GAs are clearly more effective than the classical GAs. On the basis of displacement and curvature modes, three expressions of the fitness were defined for different sensors and tasks. The generalized GAs were used to search the optimal placement of accelerometers and strain gages on the stiffening decks and the towers of Tai Zhou bridge. The results show that the generalized GAs give reliable and stable results with rapid convergence.

Abstract: In this work optical chromatic approch has been use to analyze PD signals in insulation oil. The study of chromatic approach of acoustic emission for PD phenomenon addressing PD signals with three detectors red, green and blue (R,G,B) having non-orthogonal (overlapping) responses. The outputs of the three detectors are transformed into a number of chromatic parameters designated x, y, z as well as hue, lightness and saturation (H,L,S). The x, y, z represents the proportion of the signal from each of the three processors while H, L, S represents the dominant phase, effective strength and notional spread of the PD signals respectively. This technique provides a high degree of traceability in relating the chromatic parameters to the raw PD signals. Changes in the values of the chromatic parameters are shown to reflect various PD trends and to provide early indications of incipient insulation failure and PD source discrimination.

Abstract: In order to study the sensor distribution program in health monitoring of side forklift steering, we visited many ports and warehouses for inspection of side forklift steering safety in Huangpu district of Guangzhou, and summarized the most dangerous part of steering failure. Then we built a model of the steering mechanism which was simplified, and did motion simulation analysis .Then the dynamic simulation analysis was carried out on the ball pin. Finally, through the simulation calculation of maximum stress under various working conditions, we can determine the position of sensors stationing, and it provided with the train of thought of sensors distribution in health monitoring of side forklift steering mechanism.