An improved Wang-Mendel algorithm in Engine Test Bench control system

The Engine Test Bench system needs to tune the PID parameters every time the engine or dynamometer is replaced. In order to realize the self-tuning of PID parameters, a kind of Wang-Mendel algorithm (WM) which is improved by the Fuzzy C-Means Clustering (FCM) and Genetic Algorithm (GA) is applied to the system. According to the characteristics of the Engine Test Bench system, Data of different characteristics is selected to train the prediction model, so as to predict the PID parameters. This paper first introduces the control structure of the Engine Test Bench system, next introduces the fuzzy prediction model, and then uses the FCM algorithm for data mining and GA to optimize the weight in the WM algorithm. Finally, the prediction model is obtained and the prediction results are used in the actual engineering for testing. The results show that the improved algorithm has higher accuracy and practicability.