A low-voltage direct-current contactor state evaluation method and device

The invention relates to a low-voltage direct-current contactor state evaluation method and device, and the method comprises the steps of collecting data representing the state of a low-voltage direct-current contactor, then carrying out the preprocessing of the collected data through employing an innovation adaptive Kalman filtering algorithm, obtaining data after interference elimination, and carrying out the dimension reduction of the data through employing a principal component analysis method; calculating the importance of the data by adopting an XGBoost algorithm based on machine learning so as to extract feature information; and finally, performing state evaluation on the direct current contactor by adopting the optimized extreme learning machine. According to the invention, the importance of the data is calculated by adopting the XGBoost algorithm based on machine learning, and the state of the direct current contactor is evaluated by adopting the optimized extreme learning machine, so that the state