Abstract:Vacuum AC contactor is a kind of important power control switch, which is widely used in power transmission system. The mechanical life of the contactor shall be much longer than electrical life, which ensures the reliable operation of the product. The fracture cycle of parts determines the mechanical life, and affects more than other mechanical failure. On the premise that material and other essential factors remains unchanged, the fracture cycle of parts in contactor can be predicted according to manufacturing parameters,which can effectively evaluate product quality. Recently, researchers pay attention to the influence of manufacturing parameters on performance, but pay less attention to the mechanical life prediction of batch contactors considering manufacturing parameters. In order to solve the time-consuming in mechanical life prediction for batch products, it is necessary to use the neural network with non-linear mapping to reduce time cost.This paper builds a virtual prototype considering manufacturing parameters, which based on multi-body dynamics and electromagnetic theory. Based on the stress of parts in virtual prototype, the fracture cycle is calculated by Gerber model. Manufacturing parameters of batch products are measured as input, and fracture cycle calculated by virtual prototype is used as output, these data are used to train the neural network. The forecasting accuracy of standard BP neural network is limited by training samples. The weights and thresholds of standard BP neural network can be optimized by whale optimization algorithm（WOA） to improve the forecasting accuracy. WOA builds a mathematical model to capture optimal parameters by simulating the whale search,enclosure and predation.According to the distribution characteristics of manufacturing parameters, the virtual samples of contactor is extracted by Monte-Carlo method, and the fracture cycle of parts in virtual samples is predicted by the WOA-BP neural network. A contactor is selected as object to verify the effectiveness of this method, clearance in the joint of rotating lever is used as manufacturing parameters to predict the fracture cycle distribution of rotating lever in a batch of products. The accuracy of prediction results is verified by mechanical life experiment. The following conclusions can be drawn from the simulation analysis:（1） The electromagnetic and dynamic characteristics of vacuum AC contactors are affected by manufacturing parameters at the same time. By changing the electromagnetic system and transmission position, the manufacturing parameters make the load and load position on parts change at the same time, resulting in the fluctuation of fracture cycle.（2） The mechanical life of 10 000virtual contactors is predicted by using the trained neural network. The prediction time consumed is only the same as that of a single calculation by virtual prototype. The weight and threshold of BP neural network are optimized by whale optimization algorithm, which can effectively improve the prediction accuracy of rotating lever fracture cycle and reduce the prediction error from 233.3% to 12.1%.（3） The calculated results of virtual prototype are compared with the experimental results by a high-speed camera. The maximum error between calculated value and measured value is 7.3%. The central value of fracture cycle predicted by WOA-BP is in good agreement with the experimental value, which indicates that the prediction accuracy of this method can meet the needs of batch production.

Mechanical Remaining Life Prediction of Circuit Breaker Based on Long-Term and Short-Term Memory Neural Network

Remaining Life Prediction of Conventional Low-Voltage Circuit Breaker Contact System Based on Effective Vibration Signal Segment Detection and MCCAE-LSTM

Prediction of Residual Electrical Life in Railway Relays Based on Convolutional Neural Network Bidirectional Long Short-Term Memory

Mechanical Condition Identification and Prediction of Spring Operating Mechanism of High Voltage Circuit Breaker

Remaining useful life prediction techniques for electric valves based on convolution auto encoder and long short term memory

Mechanical Life Prognosis of High Voltage Circuit Breakers Based on Support Vector Machine

A Remaining Useful Life Prediction Method of Mechanical Equipment Based on Particle Swarm Optimization-Convolutional Neural Network-Bidirectional Long Short-Term Memory

A life prediction method of mechanical structures based on the phase field method and neural network

Mechanical Life Prediction of Batch Electromagnetic Switches Considering Manufacturing Parameters

Remaining Useful Life Prediction of High-power DC Contactors Based on BiLSTM

Remaining useful life prediction for equipment based on RF-BiLSTM

Remaining Useful Life Prediction For Mechanical Equipment Based On Temporal Convolutional Network

Research On Mechanical Fault Prediction Algorithm For Circuit Breaker Based On Sliding Time Window And Ann

A Lifetime Prediction Model Based on Two-Path Convolution with Attention Mechanism and Bidirectional Long Short-Term Memory Network

Journal bearing seizure degradation assessment and remaining useful life prediction based on long short-term memory neural network

Prediction of Remaining Service Life of Rolling Bearings Based on Convolutional and Bidirectional Long- and Short-Term Memory Neural Networks

Remaining Useful Life Prediction of Rolling Bearings Based on Multi-scale Permutation Entropy and ISSA-LSTM

A novel approach based on spatio-temporal attention and multi-scale modeling for mechanical failure prediction

A Hybrid Transformer-GRU Network for Remain Useful Life Prediction of Aero-Engine Systems

A Fuzzy Reinforcement LSTM-based Long-term Prediction Model for Fault Conditions in Nuclear Power Plants

Life Prediction of Bearings in Rotating Machinery Based on Grey Model and LSTM Network