فصلنامه روشهای هوشمند در صنعت برق
پیاپی 42 (تابستان 1399)

The widespread utilization of induction motors as a driving force of electric vehicles has recognized the necessity for upgrading control system of these motors even more than ever before, in order to improve efficiency and reduce the torque ripple. This matter can lead to increase in the distance traveled by the electric vehicle at each charge and ultimately Increase the battery life. To this end, a predictive direct torque control method, as well as an optimal direct torque control method, was proposed. In the predictive direct torque control method, the reference voltage vector based on the predictive control is determined so that both the torque value and the charge value are equal to the reference values as quickly as possible. The optimal direct torque control method is also based on calculating the optimal stator reference flux according to the load torque. For comparison and evaluating the performance of controllers optimal direct torque control method and predictive direct torque control method along with the conventional direct torque control method, are simulated on an induction motor. Simulation results demonstrate that optimal direct torque control method in no-load mode and predictive direct torque control method when applying load have the highest efficiency, lowest current amplitude and torque ripple. Therefore in this paper, direct torque compound control method is presented. This method it uses optimal direct torque control in no-load and predictive direct torque control when applying load. This method has the best performance to increase battery life in electric vehicles.

Using the adaptive radial basis function (RBF) neural network dynamic surface control design method, a controller design approach is presented in order to the stabilization of strict-feedback nonlinear stochastic systems subjected to Prandtl-Ishlinskii nonlinearity in the actuator. This method is capable to be applied to nonlinear stochastic systems with any unknown dynamics. According to the universal approximation capability the RBF neural networks make it possible to approximate the unknown dynamics of the nonlinear stochastic systems. Using the minimal-learning-parameters algorithm the approximation procedure is done with a minimum complexity and required calculations. The stability of the proposed control system is proven analytically and its results are demonstrated using a simulation example. It is shown that the proposed design approach guarantees the boundedness in probability for adaptive control system, and in turn the uniformly ultimately boundedness of all closed-loop signals. It is also shown, that using this method the tracking error can be made arbitrarily small.

Optimal energy management in multi area smart grids will increase social welfare, reduce economic costs and environmental pollution. Power management solutions for smart grids include issues such as economical distribution of load, suitable load management, optimized charging and discharging of energy storages, and the availability of renewable resources considering limitation of power exchange in different area, all of which are issues in an intelligent grid, that in this paper has been considered. This paper presents a bi-level mixed integer quadratic programming (MIQP) model for energy management in multi-are smart grids with the aim of reducing economic costs and environmental pollution and increasing social welfare by considering energy storage systems, load management and Renewable resources are presented. In this paper presents a bi-level approach that the upper level is formulated to minimization economic cost and pollution of resource and lower level is presented to maximization social welfare in the form of Karush–Kuhn–Tucker (KKT) conditions. The simulation is implemented in MATLAB with Gurobi solver that the results show that the proposed bi-level model is also an efficient way to optimize energy in multi-area smart grids compared to Pareto front and Weight methods.

The reference trajectory tracking is one of the most important issues in the field of tractor-trailer wheeled mobile robots control. In this paper, thetrajectory tracking control issues of a tractor-trailer wheeled mob ile robot has been significantly solved in the presence of structural uncertainties,non-hol o n o mic constraints and external disturbance. The proposed scheme is based on a design that the tractor-trailer’s state space representation is extracted from its dynamic and ki n e matic models and presented ina companion format first. In the following,by considering the state space representation of system, the control algorithm is presented includingtwo external and internal control loops. Toward this end, the control law has been developed in the inner loop via input-output feedback linearization in a nonlinear feedback formwh i ch is continuously eliminating the nonlinear dynamics of the system. Then,by using a comb ination of the output that is pr o duced in linearization steps with a terminal sliding mode control algorithm and sketching a neural robust ad aptive finite time controller in the outer loop, the accurate and fast performance of the closed loop system has been guar a nteed in the presence of uncertainties. The proposed control algorithmfinally guarantees the boundedness of closed-loop signals and accurate finite time convergence of tracking errors. At the end, the effectiveness of the proposed sch eme has been demo nstrated and shown through the extended Lyapunov theorem and simulated by MATLAB application.