The major problem in renewable energy system is that the variation in power generation from time to time because of the intermittent nature of the renewable sources. In this paper presents the Artificial Neural Network (ANN) based intelligent control strategy for hybrid standalone microgrid system for varying wind, varying solar irradiation, varying load and symmetrical and asymmetrical fault conditions is presented. A dynamic model of hybrid standalone microgrid consisting wind and solar PV renewable sources with maximum power point tracking control algorithm is developed in MATLAB/Simulink software. A solar PV panel, wind energy conversion system and dynamic model of hybrid energy storage consisting Lithium-ion batteries (Li-Ion) and supercapacitor is connected with the DC Bus. The DC link is connected through the power electronic interfacing circuit and converters connected to a source for a diverse electricity generation. The main objective of this thesis is to improve the power quality of the hybrid microgrid system by solving the voltage sag and swell problem and reduce total harmonics distortion occurring due to various symmetrical and asymmetrical condition. The simulation studies have been carried out to determine system performance with different scenarios of the sources such as typical solar radiation, temperature, wind, battery and supercapacitor charge or discharge conditions The proposed strategy gives better performance characteristics and reduces the harmonics of the system compared to the conventional solution.Keywords: PV, WECS, Battery, ANN, Standalone Hybrid System.
This paper focuses on the development of ANN based MPPT interfaced Permanent magnet synchronous generator (PMSG) for wind energy conversion system (WECS). There are many drawbacks to conventional energy sources, such as higher fossil fuel prices, damaging to the environment, a shortage of resources, and increased pollution. Due to the unpredictable and arbitrary nature of wind energy, it is crucial to introduce several control strategies in order to maximize its efficiency. In this work, Levenberg algorithms are developed for Artificial Intelligence Technique using MPPT algorithms for PMSG-based wind energy conversion systems. The proposed system consists of 3 kW PMSG integrated WECS, DC to DC boost converter, multilevel inverter and different loads. The DC to DC converter is connected to the multilevel inverter through a DC link capacitor. The inverter power is fed to the grid through a step-down transformer. MPPT controllers are used in the standalone wind energy conversion system. The goal of this research is to develop an MPPT controller using Levenberg-based Artificial Intelligence Technique for wind energy conversion systems. With the proposed Levenberg-based Artificial Intelligence Technique, MPPT uses voltage and current as input variables for DC to DC boost converter duty ratios. The Artificial Intelligence Technique of MPPT controller for standalone wind energy conversion systems with various different input/output conditions is modeled and simulated in MATLAB/ Simulink environment. The performance of the Artificial Intelligence Technique-based MPPT controller versus the conventional incremental conductance MPPT controller. Simulation results show that for a wide range of input wind speed, Artificial Intelligence Technique based MPPT controller shows improved performance than the conventional MPPT at various operating conditions. The response of the developed Levenberg based Artificial Intelligence Technique based MPPT algorithms for PMSG based wind energy conversion system in MATLAB/ Simulink environment to investigate the robustness of the developed control strategy and validate the reliability and stability under different input/output conditions for grid connected wind generation system.Keywords: ANN, MPPT, P&O, MLI, DC to DC boost converter.