An Improved Stator Resistance Adaptation Mechanism in MRAS Estimator for Sensorless Induction Motor Drives
A comparative study of the conventional fixed gain PI and Fuzzy Logic based adaptation mechanisms for estimating the stator resistance in a Model Reference Adaptive System (MRAS) based sensorless induction motor drive is investigated here. The rotor speed
- PDF / 759,330 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 101 Downloads / 237 Views
Abstract A comparative study of the conventional fixed gain PI and Fuzzy Logic based adaptation mechanisms for estimating the stator resistance in a Model Reference Adaptive System (MRAS) based sensorless induction motor drive is investigated here. The rotor speed is estimated parallely by means of a PI control based adaptive mechanism and the electromagnetic torque is also estimated to add more resilience. By considering the external Load torque perturbation as a model perturbation on the estimated stator resistance, the effects of the same on the estimated parameters are observed. The superiority of the Fuzzy based stator resistance adaptation mechanism is observed through detailed simulation performed offline using Matlab/Simulink blocksets. Furthermore, a sensitivity analysis of the stator resistance estimate with respect to load torque is also done to verify the effectiveness of the above concept.
⋅
Keywords Speed estimation Adaptive control model Computational intelligence
⋅
⋅
Model reference
⋅
Machine
Notation iSds , iSqs ψ̂SqrV , ψ̂SdrV
d and q axis stator currents in stationary reference frame d and q axis Voltage model rotor flux linkages in stationary reference frame d and q axis Current model rotor flux linkages in stationary ψ̂SqrI , ψ̂SdrI reference frame Lr, Lm, Ls, σ Rotor Magnetising and Stator inductance, Reactance Actual and Estimated Stator resistances RS , R̂ S ωr , ω̂r , Tr Actual and Estimated Rotor speeds, Rotor Time constant K p, K I Proportional and Integral gains
S. Mohan Krishna (✉) ⋅ J.L. Febin Daya School of Electrical Engineering, VIT University, Chennai, India e-mail: [email protected] © Springer Science+Business Media Singapore 2017 J.K. Mandal et al. (eds.), Proceedings of the First International Conference on Intelligent Computing and Communication, Advances in Intelligent Systems and Computing 458, DOI 10.1007/978-981-10-2035-3_38
371
372
S. Mohan Krishna and J.L. Febin Daya
1 Introduction The indirect rotor flux oriented control implementation for a speed encoderless induction motor was significant in many ways. The presence of the speed encoder meant additional electronics, space constraints and cost. Besides, it would negate the inherent robustness of the induction motor. Consequently, many research efforts focused on the concept of sensorless speed estimation [1–4]. The speed estimation can be classified as one exploiting the concept of rotor spatial harmonics and the other depending on the machine model. Though the former is independent of machine parameters and considered as an accurate speed measurement, it introduces considerable measurement delays and cannot be used as a feedback signal for high performance drives. This led the researchers to focus more on machine model based speed estimation schemes, as they are relatively easy to implement, but the disadvantage was that, they were parameter dependent. Furthermore, the recent research focuses on machine model fed speed estimation mainly based on adaptive control. The classification of adaptive control base
Data Loading...
