Research On Evaluation Method Of Performance State Of Circuit Breaker Energy Storage Spring

As a power component of circuit breaker,the reliability of energy storage spring plays an important role in the drive and control the operation of circuit breaker motion process.Normally,the energy storage spring needs to be checked during regular inspections,but the spring is inconvenient to be disassembled and tested on site,and the fatigue state and internal defects of the spring cannot be accurately determined by manual inspection.Therefore,there is currently no suitable technical means to test the performance of the energy storage spring in the field.After receiving the switching signals,the energy storage spring releases energy instantly,and the intense movement is easy to deepen the crack of the spring with fatigue defects.The cracks gradually expand over time and finally cause fracture.Therefore,the key to evaluate the performance of the energy storage spring is to understand the deformation and output of the energy storage spring during the release period.In this paper,a non-contact testing method of spring deformation feature based on image matching tracking technology was proposed: the high-speed image sequence of energy storage spring releasing process was captured by high-speed camera.Normalized cross correlation(NCC)algorithm was utilized to track the position of the first turn of the active end reflecting the change of spring length,and obtain the coordinates of the target of each key frame and the characteristic parameters related to spring deformation;The Golden Tower search strategy is used to improve the efficiency of the algorithm.According to the shape and deformation characteristics of the spring,a normalized cross correlation-Pyramid-estimation(NCC-P-E)algorithm was proposed to eliminate the interference of similar structure on matching results,in order to ensure the accuracy and fast tracking of reversing moving targets.In addition,while experimentally obtaining the deformation characteristics of the spring,the pressure sensor was used to collect the energy storage spring output changes during the switching process,extracted the characteristics of the energy storage spring output,and established the performance evaluation index of the circuit breaker energy storage spring.Reliable samples were accumulated in the experiments,and the deformation characteristic parameters of normal springs and old springs with reduced stiffness were compared to screen the deformation characteristics that mutated when the spring performance decreased.Finally,a BP neural network optimized based on the firework(FWA-BP)algorithm was proposed to transform the deformation characteristic parameters of the breaker energy storage spring which was changed by deterioration to the evaluation performance index,so as to realize the performance evaluation of the breaker energy storage spring.After verification,this method can not only detect the circuit breaker energy storage spring under "non-contact" conditions,but also provide a basis for the status maintenance and replacement of the spring,and has broad practical application value.