ANFIS Controller-Based Cascaded Nonisolated Bidirectional DC-DC Converter

摘要

The development of bidirectional DC-DC converters has become important because of their requirement in energy-storage systems. The simple structure of nonisolated bidirectional DC-DC converter types includes multilevel, switched-capacitor, buck-boost, and coupled inductor type. In multilevel and switched-capacitor types, if large voltage gain must be provided, more switches and capacitors are required. Since the leakage inductor energy cannot be recycled, voltage stresses on the switches are present. Therefore, the control strategy is easily implemented in the system operation. This paper presents a cascaded nonisolated dc-dc switched coupled converter for enhancement of the switching operation. For the optimal switching performances, an Artificial Intelligence (AI) technique is utilized. The AI technique is the Adaptive Neuro-Fuzzy Inference System (ANFIS) for generating the optimal control pulses to enhance the performance of boost and buck switch. In addition, the proposed technique is utilized in cascaded nonisolated DC-DC switched coupled converter to reduce the losses. In the ANFIS technique, the error voltage and change in error voltage are given as inputs. At the same time, the ANFIS controller is employed to reduce the error value and produce the optimized gain pulses. In the buck and boost switch mode of operation, it is enhanced with the help of the proposed technique. Moreover, the operating principle and voltage conversion ratio are discussed. It is seen that the implementation of the proposed controller improves the efficiency of the system and also reduces the voltage drop across the switching operation. Then the proposed ANFIS technique with bidirectional converter topology was implemented in MATLAB/Simulink working platform and the output performance is analyzed. Then the proposed circuit performance is compared to the existing circuit such as proportional integral derivative (PID), artificial neural network (ANN) and Fuzzy, respectively.