Enhanced control of Photovoltaic Power Converters under Mismatching Conditions

Abstract

Exhausting fossil fuel, a huge increase in oil prices, global warming, damage to environment, increasing energy demand are major problems being faced. In order to avoid these problems, power generation is being done using renewable energy sources. Among the renewable energy sources, solar photovoltaic (PV) is dominant because of long operational life, lesser emission, decreasing cost of solar photovoltaic panels. Photovoltaic sources exhibit unique maximum power point under uniform conditions. Under mismatching conditions, there will be multiple peak points because of the presence of bypass diodes. Maximum power point tracking algorithm is used to track the maximum power from the PV source. This thesis presents a literature review of maximum power point tracking (MPPT) algorithms for tracking the global peak. The methodology employed for tracking maximum power point is classified as empirical methods, perturbation methods, model-based methods, artificial intelligence methods, evolutionary computing methods, scanning-based methods, and modified perturbation methods. Based on the literature survey, research gaps are identified and are presented as objectives for this thesis. Four maximum power point tracking algorithms capable of tracking global peak under mismatching conditions are proposed. The first algorithm is based on searching technique and bisection method in which zone wise division of characteristics is performed based on open circuit voltage and panel characteristics. It is a duty ratio based control method and the value of duty ratio is calculated based on bisection method until the global peak is detected. Once the global peak is detected, conventional perturb and observe method is used to retain the operating point at GP. The second algorithm is based on current control in which reference current is moved in the forward and backward direction by multiplying or dividing PV current with 0.9. The movement of PV current is continued in the backward direction until the operating voltage is less than minimum voltage below which there is no chance of occurrence of global peak. After that, the perturbation of PV current is continued in the forward direction until the operating current is less than minimum current below iiiwhich there is no chance of occurrence of global peak. During the process of perturbation, the maximum power point is identified and a conventional algorithm is used to retain the operating point at that point. The third algorithm uses reference voltage control and reference current control to track the global peak. The choice to use voltage or current control is made using a decision variable. The algorithm operates in the current control mode to find the nearest peak and operates in voltage control mode to identify the inflection point. Initially, the voltage below which there is no chance of occurrence of the global peak is identified and it is initialized as the reference voltage. Then the succeeding peak is identified using reference current control. Once the peak is determined, reference voltage control is used to identify the inflection point. This process is continued until the operating PV current is less than the minimum possible current. The fourth algorithm tracks the global peak by sampling variations in the transient period during charging of the input capacitor. The algorithm operates in three stages viz., scanning, correcting and retaining the operating point at MPP. In the scanning stage, the maximum power and voltage at maximum power are identified by changing the value of duty ratio from maximum to minimum value. The correcting stages bring the operating point close to the voltage at maximum power point by varying the duty ratio and retaining stage maintains the operating point at MPP. The simulation studies of all the four MPPT algorithms are performed in MATLAB. All the methods are compared with recent existing MPPT methods in the literature. Hardware implementation is performed using solar array simulator, the boost converter, and resistive load.