Data‐driven online temperature compensation for robust field‐oriented torque‐controlled induction machines

Abstract

Squirrel-cage induction machines (IMs) with indirect field-oriented control are widely used in industry and are frequently chosen for their accurate and dynamic torque control. During operation, however, temperature rises leading to changes in machine parameters. The rotor resistance, in particular, alters, affecting the accuracy of the torque control. The authors investigated the effect of a rotor resistance parameter mismatch in the control algorithm on the angular rotor flux misalignment and the subsequent deviation of stator currents and motor torque from their setpoints. Hence, an online, datadriven torque compensation to eliminate the temperature effect is proposed to enable robust torque-controlled IMs. A modelbased analysis and experimental mapping of the temperature effect on motor torque is presented. A temperature-torque lookuptable is subsequently implemented within the control algorithm demonstrating the ability to reduce the detrimental effect of temperature on torque control. Experimental results on a 5.5 kW squirrel-cage induction motor show that the proposed datadriven online temperature compensation method is able to reduce torque mismatch when compared to having no temperature compensation. Up to 17% torque mismatch is reduced at nominal torque and even up to 23% at torque setpoints that are lower than 20% of the nominal torque. A limited torque error of <1% remains in a broad operating range.

Preventing Catastrophic Forgetting using Prior Transfer in Physics Informed Bayesian Neural Networks

“…Minimizing the difference between the variational distribution and the exact posterior is thus translated into a cost function comprised of two terms that will be minimized simultaneously. The first term in (10) minimizes the KL-divergence to the prior function, regularizing the solution to obey this prior to some extent. Secondly there is a likelihood cost, which encourages to learn a parameter distribution that maximally explains the training data.…”

“…Friction is a good example as this phenomenon depends on various environmental factors, is difficult to directly measure and to account for in the physical models [5]- [9]. Other uncertainties are e.g effects of temperature dynamics in electrical actuators that give rise to different torque or force levels [10]. Secondly, the fixed structure of physical modelling makes it hard to C. Van Heck, A. Coene and G. Crevecoeur are with the Department of Electromechanical, Systems and Metal Engineering, Ghent University, B-9052 Ghent, Belgium e-mail: {cevheck.vanheck, annelies.coene, guillaume.crevecoeur}@ugent.be.…”

Robust iterative learning model predictive control for repetitive motion of maglev planar motor

“…Note that the planar motor always executes a repetitive motion task in practical industry application, for instance, laser cutting and micromachining process [17]. Among the various repetitive control methods [18][19][20][21], iterative learning control (ILC) stands out for its excellent ability to capture the repeated information hidden in the previous iteration to improve the closed-loop tracking performance gradually and therefore, it is widely applied in modern precision industry [22]. In Ref.…”

Inverse Thermal Identification of a Thermally Instrumented Induction Machine Using a Lumped-Parameter Thermal Model

“…Thermal prediction of electric motors is becoming increasingly important in the drive for higher power density, energy efficiency and cost reduction [1]. Accurate temperature estimation of motor hotspots is key for condition monitoring and fault detection [2], as well as motor control [3,4]. As the stator windings are surrounded by thermally vulnerable insulation material, real-time prediction of the temperature of the windings allows enhancement of the motor end-of-life duration.…”

“…A large portion of the industrial induction motors use Field-Oriented Control (FOC) schemes. They rely on having accurate knowledge of the rotor resistance parameter value for their torque control [4]. Having accurate real-time temperature prediction capabilities thus allows improvement to the torque performance of DTC (i.e., knowledge of the stator windings temperature) and FOC (rotor temperature) as well as enhance end-of-life duration.…”