Design and implementation of a solar array diagnostic system

Abstract

Solar modules are the most important part of a Photo-voltaic (PV) solar system as light energy from the sun is converted into electrical energy. The basic component of a solar module is a solar cell that has the capacity of producing about 0.5V of electricity. A solar panel or module is a combination of several solar cells connected in series to generate usable voltage. A solar array is a collection of solar modules connected to form a single unit that feeds power to the solar system. Solar modules are often installed as an array to generate the required power. However, in Uganda energy losses in PV arrays due to faults such as shading, dusting, and open-circuit within the PV array represent 49.92 % of the total energy production. The purpose of this project is to design and implement a solar array diagnostic system that detects and identifies faults on the DC side of the PV array in particular shading, soiling or dusting, and open-circuit in the PV array. This was realized by first collecting the PV array time-series data on an SD module for each operating condition using a data acquisition system which was then visualized and pre-processed respectively using MATLAB. The pre-processed data for each operating condition was analyzed using statistical values in MATLAB to obtain trends of power. A diagnostic algorithm was modelled from the statistics functions of the plotted graphs for each operating condition. The solar array diagnostic algorithm was tested and verified by burning it on the PV array system while manually creating shading, dusting, open-circuit, and normal operation and inspecting whether the serial monitor displays the created operating condition. In conclusion, a solar array diagnostic algorithm that displays the operating condition of the PV array was successfully designed and implemented. For further work, a plan to deploy the solar array diagnostic algorithm in solar charge controllers since they contain current and voltage sensors is being considered.