Torque estimation of electric vehicle motor using adaptive-network based fuzzy inference systems

Abstract

This paper presents to estimating studies of the torque data of the Electric Vehicle (EV) motor using Adaptive-Network Based Fuzzy Inference Systems (ANFIS). The real-time data set of the Outer-Rotor Permanent Magnet Brushless DC (ORPMBLDC) motor which was designed and manufactured for using in ultralight EV, was used in these estimation process. The current, the power and the motor speed parameters are defined as input variables, and the torque parameter defined as output variable. Five distinct ANFIS models were designed for torque estimation process and the performances of each model were compared. The most effective model for testing data set among the ANFIS models was anfis: 2 with 98 nodes and 36 fuzzy rules, and the worst model was anfis: 5 with 286 nodes and 125 fuzzy rules. Performance results of all designed models were presented in tables and graphs.

Load Torque Observer for BLDC Motors Based on a HOSM Differentiator

“…The proposed method estimates the residual torque, defined as the sum of all the friction and viscous torques, from data acquired during the functioning of the EMA; simulations are used to validate this scheme. Using another approach, in the work [10] the torque is estimated from real-time data using five distinct Adaptive-Network Based Fuzzy Inference Systems. Finally, the article [11] presents a comparison between Kalman and Extended Kalman Filters algorithms for torque estimation; simulations are also used to validate the proposed approach.…”