Current and Speed Sensor Fault Diagnosis Method Applied to Induction Motor Drive

Cuong Dinh Tran,Martin Kuchar,Pavel Brandstetter,Bach Hoang Dinh,Petr Palacky

doi:10.1109/access.2021.3064016

Abstract

The paper proposes a novel approach based on a current space vector derived from measured stator currents to diagnose speed and current sensor failures in the field-oriented control of induction motor drives. A comparison algorithm between the reference and measured rotor speed is used to detect the speed sensor faults. A counter is added to eliminate the influence of the encoder noise in the diagnosis method. In this approach, estimated quantities are not used in the proposed speed sensor fault diagnosis strategy, which increases the independence between the diagnosis stages in the fault-tolerant control (FTC) method. Moreover, in order to discriminate between the speed sensor faults and the current sensor faults, a new approach combining the current space vector and a delay function is proposed to reliably determine the current sensor failures. The MATLAB-Simulink software was used to verify the idea of the proposed method. Practical experiments with an induction motor drive controlled by DSP TMS320F28335 were performed to demonstrate the feasibility of this method in practice. The simulation and experimental results prove the effectiveness of the proposed diagnosis method for induction motor drives.

Highlights

In recent years, induction motor drive (IMD) systems overcame the challenges of complex nonlinear control structures to become the most common machine type for applications in industry and electric cars, HVAC, home appliances, etc
The simulations have been implemented in the low-speed region, where the operating motor speed is set to 10% of the rated value under three sensor fault types:
The experimental structure includes an induction motor, a controllable load, an inverter controlled by the TMS320F28335-DSP, and a three-phase power supply

Summary

Introduction

Induction motor drive (IMD) systems overcame the challenges of complex nonlinear control structures to become the most common machine type for applications in industry and electric cars, HVAC, home appliances, etc. A typical IMD system consists of four essential parts: an induction motor (IM), a power converter, measurement sensors, and a controller [1]. To control the induction motor operating at a specific speed, the controller uses feedback measurement signals from sensors and compares them to desired values to generate a control command for a power converter supplying an IM. This process performance relates to the control algorithm ability and the quality of hardware devices in which the sensor system keeps a significant position. In PFTC, the control system of IMDs can only deal with simple sensor failures and work offline. This research aims to propose an improving AFTC method against the speed sensor and current sensor faults

Objectives

Results

Conclusion