Research And Implementation Of Partial Discharge Device For Power Equipment Based On Acousto-optical Joint Detection

Partial discharge refers to the uneven distribution of the electric field in the insulation system,which causes the insulation material to generate a discharge that does not penetrate the insulation material in the local area under the action of the electric field.Partial discharge of power equipment is a sign of deterioration of insulation materials.If measures are not taken in time,large-scale power accidents may be caused by it.In order to prevent such accidents,aiming at the problem of partial discharge detection in power equipment,a traditional ultrasonic detection method and ultraviolet detection method are combined in this paper to propose a combined acoustic-optical detection method.Based on this method,a set of devices that can be used normally is researched and implemented.According to actual measurement,the device can complete the detection of partial discharge of electric equipment within 10 m,and perform preliminary classification on the intensity of partial discharge.The specific work of this article is as follows:1.Analyzing the research status of partial discharge detection in power equipment.Aiming at the shortcomings of the traditional single detection method,such as weak anti-interference ability,short detection distance,and dangerous detection process,this paper combines the traditional ultrasonic detection method and ultraviolet detection method,and proposes a combined acoustic and optical detection method.This method can not only combine the advantages of the two methods,and simultaneously improve the detection accuracy and the long-range detection sensitivity of the local close-up,but also has a strong anti-interference ability and a simple and safe detection way.2.Based on the acousto-optical joint detection method,a set of high-sensitivity,long detection distance,and safe detection mode power equipment partial discharge detection device is designed.A sensing probe and a demodulation host are mainly included in the device.A software named SolidWorks is used to complete the design of the sensor probe housing,3D printing technology is used to make the sensor probe housing in this paper;The demodulation host's various modules(including the main control module,signal conditioning and acquisition module,signal storage module,display Screen module and power module)is completed(including schematic and PCB layout),and the device is finally successfully implemented.TheSTM32F407ZGT6 chip is used as the MCU in this device,which can realize the acquisition of ultrasonic signals at a sampling rate of 400 kHz,the acquisition of ultraviolet signals in the form of pulse capture,and signal processing and partial discharge intensity judgment.3.Based on the acousto-optical joint detection method,a method of acousto-optical signal processing is proposed.Firstly,the ultrasonic signal and ultraviolet signals are collected in real time to determine whether a partial discharge has occurred.When it is judged that there is a partial discharge of the acousto-optical signal,the ultrasonic signal is first extracted and wavelet packet denoised,and then the ultrasonic signal and the ultraviolet signal are separately extracted and statistically related to the partial discharge related feature information,and the extracted acousto-optical signal features are fused.Finally,the partial discharge intensity is judged based on the KNN method.After verification,the denoising effect of this method is obvious.The average accuracy rate of partial discharge intensity classification is 98.78%,the average recall rate is 98.80%,and the F1 score is 98.79%.Under the condition of similar accuracy,the calculation time is much lower than the traditional SVM algorithm.It's suitable for deployment on embedded devices with limited computing performance.4.In order to verify the effectiveness and practicability of the partial discharge detection device,an indoor test and a field test were performed.The test results show that the device can sensitively identify partial discharges within 10 m and successfully display,verifying the effectiveness of the device and the feasibility of application to partial discharge detection and fault diagnosis of power equipment.