Intelligent Health Diagnosis of Lithium Battery Pole Double Rolling Equipment Driven by Hybrid BP Neural Network and Expert System

Abstract

At present, the fault diagnosis methods of lithium battery pole rolling mill mostly rely on manual experience and the self-test function of mature control devices such as frequency converters and lack the ability of intelligent fault diagnosis for the whole equipment and the ability to evaluate the health state of the equipment during operation. To improve the intellectual health diagnosis ability of lithium battery pole double rolling mill equipment, starting from the structure and technology of lithium battery pole double rolling equipment, this paper analyzes its common fault types. It summarizes the shortcomings and common fault types of existing equipment. Then, we introduce fuzzy reasoning into the fault diagnosis method based on Expert Systems and establish the FEFDM of lithium battery pole double rolling equipment. Finally, we introduce the concept of health degree, effectively connect BP neural network and health degree through the fuzzy set, and establish an equipment operation health state evaluation method based on an improved BP Neural Network, which realizes the evaluation ability of the health state of double roller equipment. In addition, we use Extended Kalman Filtering (EKF) to clean the “dirty data” and filter out the Gaussian white noise from the signal. The health diagnosis method proposed in this paper can meet the ability to accurately locate and diagnose the fault of lithium battery pole double roller equipment and evaluate the health state of equipment operation and maintain the equipment in advance.

Keywords

• Compound fault diagnosis
• health status assessment
• BP neural network
• FEFDM