A Study On The Identification Method Of Pointer Instrument Reading In Substation

In a substation,there are a large number of meters that are used to show the current operation of the equipment.Due to the poor resistance of the digital instrument to electromagnetic interference,and the substation itself is a powerful source of electromagnetic interference,there is still a pointer meter in the substation under the complex electromagnetic environment.The accuracy of the pointer meter reading recognition directly affects the safety of the substation.operation.At present,most of the pointer meters are manually copied by people.The accuracy of manual operation readings is low,the labor intensity of the personnel is large,and it is easily affected by external factors.These effects will bring hidden dangers to the safe operation of the substation.In order to avoid the error caused by manual reading,improve the accuracy of substation pointer meter reading and realize the development of substation automation,this paper studies the identification of substation pointer meter reading.In this paper,we study image acquisition,image preprocessing,image segmentation,pointer feature extraction and angle recognition,display interpretation,MATLAB experimental simulation and other technologies.First use the camera to photograph the pointer instrument,and then transfer the captured instrument image to the computer through the data transmission device.In the computer,the digital image processing technology is used to preprocess the collected pointer instrument image in MATLAB software.In the process of preprocessing,by analyzing the existing noise and the performance of the de-noising algorithm,the application of the mixed filter method combined with median filter and wavelet transform in the identification process of the pointer instrument of substation is proposed.The experiment shows that the hybrid filter method has better noise removal effect than the single filter method for the noise in the process of the substation pointer meter recognition.The image segmentation method based on genetic algorithm is put forward to identify the image of the pre-processed image in the process of substation pointer instrument recognition.This method determines the optimal threshold by genetic algorithm,and then uses the threshold segmentation method to separate the target(pointer)from the background(dial)of the image.Using the Hough transformation method to extract the straight line where the instrument pointer is located,through the determination of the rotary center,the direction of the pointer fingertip and the deflection angle,the improved angle method is used to interpret the instrument,and finally the substation pointer meter reading recognition system is experimented and analyzed.The experimental results show that the algorithm is simple,the recognition speed is fast,and the recognition precision is better than the artificial reading,Therefore,it has a certain value of application.