An active disturbance rejection control (ADRC) of induction motor based on adaptive particle swarm optimization (APSO) algorithm is proposed in this paper, in order to realize the precise decoupling of induction motor and the disturbance compensation. The novel control method employs APSO as the automatic tune mechanism for ADRC controller. According to the feedback information of induction motor, an optimal solution can be achieved via the optimization mechanism and self-learning ability of APSO, so the reliance of ADRC controller on parameters can be reduced. In order to make the optimal solution more reasonable and reliable, the aggregation degree and evolution speed are introduced into the APSO to dynamically modify the inertia weight based on the practical optimization process. Simulation and experimental results indicate that the robustness of the proposed optimal design method for ADRC is better than the conventional ADRC when the disturbances occur, and the method is feasible and effective.
31 December 2020 in IEEE Transactions on Power Electronics
Internal model control (IMC) has been used widely for its less computational burden, and simple implementation. But the time-varying disturbance cannot be eliminated effectively by IMC, while the disturbance caused by model mismatch, parameter variations and other unstructured dynamic uncertainties in the induction motor (IM). To optimize the current-control performance of induction motor drives, a nonlinear current-control algorithm for the IM systems using internal model control strategy based on a Luenberger disturbance observer (IMC-LDO) is proposed in this paper. The time-varying disturbance has been considered in the IM system. Based on this system, the proposed method is derived, where a Luenberger disturbance observer is introduced to estimate the disturbance. To tune the proposed scheme properly, a stability analysis based on Jury criterion has been used to determine the selection of the observer gain. Comparative experimental results indicate that the proposed method improves the dynamic property and the robustness in case of parameter variations and step load change. The feasibility and effectiveness of the proposed method are verified by the experimental results.
7 January 2020 in IEEE Transactions on Power Electronics
A novel speed estimation method of induction motors based on active disturbance rejection observer (ADRO) is proposed in this paper, in order to realize the sensor-less control of induction motors. Such an approach permits to cope with the unmodelling dynamics as well as the uncertainties in the model parameters and exogenous disturbance, so that ADRO is robust to the external disturbances and model errors. Meanwhile, the system uncertainties and disturbance can be estimated instantaneously, so that ADRO observer realizes the disturbance estimation accurately without the knowledge of the exact system parameters. Therefore, the disturbance including rotor speed information is estimated in real-time by ADRO, and the rotor speed is estimated from the observed disturbance variation. The experimental results indicate that the novel speed estimation is feasible and effective.
To address the issue of inaccurate load forecasting affecting the effectiveness of minimum demand scheduling in railway traction stations, this study introduces a multi-objective grey wolf optimizer (MOGWO) to jointly optimize the parameters of variational mode decomposition (VMD) and long short-term memory network (LSTM) within the forecasting framework. The proposed MOGWO-VMD-LSTM model enhances the data decomposition capability of VMD and improves LSTM training, prediction accuracy, and inverse normalization reconstruction. Using a 10-day load dataset from a traction station, the model’s performance is compared against LSTM and VMD-LSTM baselines. Simulation results demonstrate superior performance in terms of mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE) metrics. Application of the forecasting results to traction station scheduling reduces the single-peak power purchase from 22.279 MW to 20.052 MW, achieving a 9.995% reduction, indicating strong practical potential.