MPPT Of PV Systems Under Partial Shading Conditions Using Optimization Algorithms

In recent years,Photovoltaics(PV)electrical energy generation has become an important source of energy.PV systems provides a technique for the conversion of solar energy into electrical energy.PV systems use photoelectric effects to convert light into direct current(DC).To increase the output of PV systems,we can either work on the quality of PV panels or operate them at an optimum level.The most effective way to obtain maximum power out of a PV system is by implementing an effective control system.we need to maximize the PV systems efficiency under various weather.The changing conditions of weather and the nonlinear behavior of current-voltage(I-V)I-V curves make the task further challenging.Thus,maximum power point tracking(MPPT)has become a vital part of the PV system.which is essential to acquire maximum output with improved efficiency and cost savings.The basic purpose of the control system in a PV system is to locate the maximum power point(MPP)and enforce the system to operate near the point where maximum power can be extracted.Under non-uniform irradiance this problem is further complicated by partial shading(PS)and complex partial shading(CPS).Under non-uniform irradiance the current-voltage(I-V)curve and the power-voltage(P-V)is further complicated by the existence of multiple maxima on the curves.Since I-V and P-V curves can have only one global maxima(GM),hence to locate the GM among multiple Local Maxima(LM)is challenging task for the PV system.The main objective of this dissertation is to develop control techniques for MPPT of PV systems under various weather conditions.Utilizing evolutionary computational techniques to operate PV systems to yield maximum power under all conditions.In this dissertation,five novel techniques are proposed.The first technique is proposed by using the conventional incremental conductance(IC)technique.Genetic algorithm(GA)based incremental conductance GA-IC technique is developed.This technique utilizes the evolutionary behavior of GA to optimize the controller gain of the PID controller which further controls the step size of the IC.The parameters of the PID controller are fine-tuned for optimized adaptive step size of duty cycle for IC technique.GA-IC technique is able to track the MPP under uniform irradiance condition.The second technique is bio-inspired heuristic technique named adaptive cuckoo search optimization algorithm(ACOA).ACOA is used to get the maximum output under PS and CPS conditions.Cuckoo search optimization(CS)technique makes use of levy flights in their search mechanism.Although levy flight and fixed switching of CS helps to break free of local maxima traps in partial shading conditions.However,it lowers efficiency.The proposed system successfully lowers the drawbacks of CS.Highly effective MPPT control is implemented and tested against existing particle swarm optimization(PSO),CS and artificial bee colony(ABC)to validate its practicality.This technique attains high robustness,can be implemented on a low-cost controller with high efficiency.In the third technique,PS and CPS conditions are tackled using novel bio-inspired dragonfly optimization(DFO).The proposed technique is simple and derivative-free nature.High efficiency and accuracy are prominent features of the DFO technique.These features are able to resolve the drawbacks of existing swarm-based techniques.The limitations of particle swarm optimization(PSO)are successfully countered in the proposed technique.The efficiency of the proposed technique is 99.7%.Ripple reduction is achieved up to 90%as compared to perturb and observed(P&O).Overshoot reduction is kept well under 60%.Fourthly a novel salp swarm optimization(SSO)based MPPT control is developed.SSO is able to solve the single-objective and multi-objective function problems.This technique utilizes leader-follower notation and long chains to improvise local and global exploration and exploitation highly.This ability helps to obtain highly coherent behavior suitable for nonlinear optimization like MPPT.Efficiency up to 99.75%is achieved using SSO.Moreover 100%GM tracking is obtained.Statistical analysis confirms the proposed technique is highly successful in achieving the objectives of MPPT control.The fifth technique is categorized in Intelligent Control of PV systems.General Regression Neural Networks(GRNN)is used to perform the MPPT control.Fruit Fly Optimization Algorithm(FFOA)is used to train the networks extensively.Many sets of current and voltage inputs and optimized output of duty cycle are used to train GRNN-FFOA based MPPT controller for the PV system.FFOA is used for training because of coherence properties,and the comprehensive mathematical model of this optimization algorithm gives least ripples.Therefore,the smoothing of the sigma function of GRNN is naturally coherent to the task being performed.This proposed technique has the least oscillations around GM.The proposed techniques are verified by a low-cost experimental setup and comprehensive simulation setup.High efficiency and robustness are prominent features of these techniques.Under different irradiance and temperature pattern,100%MPPT is successfully achieved.