Neural Adaptive Kalman Filter for Sensorless Vector Control of Induction Motor

Abstract

This paper presents a novel neural adaptive Kalman filter for speed sensorless field oriented vector control of induction motor. The adaptive observer proposed here is based on MRAS (model reference adaptive system) technique, where the linear Kalman filter calculate the stationary components of stator current and the rotor flux and the rotor speed is calculated with an adaptive mechanism. Moreover, to improve the performance of the PI classical controller under different conditions, a novel adaptation scheme based on ADALINE (ADAptive LInear NEuron) neural network is used. It offers a solution to the PI parameters to stabilize automatically about their optimum values and speed estimation to converge quicker to the real. The proposed adaptive Kalman filter represents a good comprise between estimation accuracy and computationally intensive. The simulation results showed the robustness, efficiency, and superiority of the proposed scheme compared to the classical method even in low speed region.

MRAS Speed Sensorless Vector Control of Induction Motor Drives using Predictive Adaptation Mechanism

“…After use the vector control, the induction machine can be represented as a two-phase motor in a stationary reference frame ) , (   and then convert in synchronously dynamic reference frame ) , ( q d by applying Park transformation. The mathematical models of an IM can be described by the following state equations [4], [7], [15], [16].…”

“…The purpose of the flux rotor orientation is to decouple the stator current into flux and torque producing components, regulated separately, to obtain a good performance IM drive [7], [17], [18]. For vector control, the rotor flux is oriented in the d-axis…”

“…For many years, lots of efforts have been made in ac drives to eliminate the speed sensor mounted on the machine shaft this technology is referred as sensorless control [3]. In sensorless IM drives, several techniques have been proposed for rotor speed estimation such as extended kalman filter [4], sliding mode control [5], MRAS [6], artificial intelligence-based estimators [7] and direct calculation method.This models suffer from many problems which become dominant at low speed range including sensitivity to machine parameter variation, pure integration effects and inverter nonlinearity. Among these techniques, model reference adaptive system MRAS schemes are the most common strategies employed due to their relative simplicity and low computational effort [8], [9].…”

Comparative study between the NLPI controller and the CPI controller

“…Obviously, the CPI controller makes it possible to correct within a suitable time the difference that exists between a setpoint (speed, temperature, pressure, position) and its associated measurement return, by increasing or decreasing the output of the MAS and this, by the intermediate of two main parameters called gains (Proportional gain and Integral gain), which can accelerate, delay and stabilize this correction. The regulation function can be embodied in the form of a black box that can be represented by a comparator with 2 inputs (setpoint and return), the result of this comparator controls the output of the MAS [9,[11][12][13][14][15].…”