Metodologia de estimação de parâmetros via filtragem robusta aplicada ao motor de indução trifásico

Abstract

The study and development of a parametric estimator applied to the three-phase induction motor is presented in this work. Using an adaptive estimation methodology, based on the motor’s states and outputs, a filtering technique obtained via LMIs — with guaranteed H∞ norm to assure the robustness — is applied to obtain the required informations, being the method robust to parametric uncertainties and disturbances. In this way, a short circuit failure estimator is proposed, using a new model in a state space representation with a pseudo-parameter proportional to the failure severity, which is intended to be estimated. A feasibility analysis is exposed, addressing the influence of the parameterization conditions on the performance and the synthesis of the filters. Its also presented the development of the proposed method, computational simulations and the validation with experimental data, which were obtained in laboratory tests, considering different failure, load torque and voltage supply conditions. The method demonstrated feasible to state estimation and short-circuit fault identification, proving to be viable for qualitative fault estimation.