15 December 2023 in IEEE Transactions on Industrial Informatics
This article deals with the neural networks (NN)-based classification of current sensor (CS) faults in an induction motor drive system designed according to the idea of fault-tolerant control (FTC). The CS-FTC system ensures continued system functionality, even after a failure has occurred, not only through its ability to detect and isolate faults but also through its ability to compensate for them. The article focuses on the development of the technical implementation of a CS faults classifier based on NN in considering the problem. The proposed method uses the principle structure of a multilayer perceptron (MLP) based on single samples of the stator phase current signal. The tests demonstrate techniques that allow the development of NN input vectors while eliminating the constraints resulting from the machine's operating conditions (load change, rotational speed change). In addition, the possibility of optimizing the input vector of the network in terms of the analyzed damages is demonstrated by using harmonic analysis. The article shows the possibility of full use of the mathematical model in the practical implementation of diagnostic systems implemented on a real object. MLP aims to determine the CS fault category (no fault, gain change, offset, saturation, open circuit) and fault localization (phase A or B). The experimental verification on a laboratory stand with an induction motor of 1.1 kW confirmed the efficiency of the proposed method in 97.2% of the cases.
1 January 2023 in Power Electronics and Drives
In modern areas of knowledge related to electric drive automation, there is often a need to predict the state variables of the drive system state variables, such as phase current and voltage, electromagnetic torque, stator and rotor flux, and others. This need arises mainly from the use of predictive control algorithms but also from the need to monitor the state of the drive to diagnose possible faults that have not yet occurred but may occur in the future. This paper presents a method for predicting stator phase current signals using a network composed of long-short-term memory units, allowing the simultaneous prediction of two signals. The developed network was trained on a set of current signals generated by software. Its operation was verified by simulation tests in a direct rotor flux-oriented control (DRFOC) structure for an induction motor drive in the Matlab/Simulink environment. An important property of this method is the possibility of obtaining a filtering action on the output of the network, whose intensity can be controlled by varying the sampling frequency of the training signals.
20 May 2024 in Bulletin Of The Polish Academy Of Sciences-Technical Sciences
In modern drive systems, the aim is to ensure their operational safety. Damage can occur not only to the components of the motor itself but also to the power electronic devices included in the frequency converter and sensors in the measurement circuit. Critical damage to the electric drive that makes its further exploitation impossible can be prevented by using fault-tolerant control (FTC) algorithms. These algorithms are very often combined with diagnostic methods that assess the degree and type of damage. In this paper, a fault classification algorithm using an artificial neural network (ANN) is analyzed for stator phase current sensors in AC motor drives. The authors confirm that the investigated classification algorithm works equally well on two different AC motors without the need for significant modifications, such as retraining the neural network when transferring the algorithm to another object. The method uses a stator current estimator to replace faulty sensor measurements in a vector control structure. The measured and estimated currents are then subjected to a classification process using a multilayer perceptron (MLP), which has the advantage of small structure size as compared to deep learning structures. The uniqueness of the method lies in the use of data in the training set that are not dependent on the parameters of a specific motor. Four types of current sensor faults were studied, namely total signal loss, gain error, offset and signal saturation. Simulations were performed in a MATLAB/SIMULINK environment for drive systems with an induction motor (IM) and a permanent magnet synchronous motor (PMSM). The results show that the algorithm correctly evaluates the type of damage in more than 99.6% of cases regardless of the type of motor. Therefore, the results presented here may help to develop universal diagnostic methods that will work on a wide variety of motors.
1 January 2026 in Power Electronics and Drives
The growing demand for fault-tolerant systems requires the use of algorithms that will maintain continuity of operation in the event of faults, e.g., in measuring sensors. Sensorless control methods, originally developed for electric motor drives without speed measurements, can be applied within fault-tolerant control (FTC) strategies when speed sensors fail. Recent research also increasingly addresses failures of current sensors (CS). This study proposes a permanent magnet synchronous motor (PMSM) drive control method based on a current-sensorless vector control structure, eliminating the need to measure phase currents and enabling compensation for damaged CS. Three open-loop estimator types derived from a PMSM mathematical model are introduced, all operating without feedback from state variables. The first estimator uses fixed design parameters optimized via particle swarm optimisation (PSO). The second relies on the relationship between stator flux and dq-axis currents. The third extends this relationship by incorporating rotor position to account for spatial harmonics. The relevant flux functions are stored in look-up tables (LUTs). This study discusses challenges in selecting estimator parameters and issues related to voltage and speed measurement. The proposed solutions were tested on a 2.5 kW PMSM experimental setup across a wide range of operating conditions.