Fast-Accurate Dual-Axis Solar Tracker Controlled by P&O Technique with Neural Network Optimization

This paper proposes a fast-accurate method to locate the sun through dual-axis sun tracker rapidly and accurately. The tracking time is reduced to decrease the power consumption and improve the dynamic performance. The controlling procedure is closed-loop combination of Multi-Layer Perceptron (MLP) neural network and Perturbation & Observation (P&O). If the sun deviates from solar panels significantly, the sun tracker will operate quickly, but inaccurately by using the neural network. But if the sun gets away slightly, the sun tracker will operate accurately, but slowly by using P&O. So, by combining these two given cases will cause to track the sun fast and accurately. The error back propagation training algorithm is firstly used to train the neural network and in the next step the Modified Particular Swarm Optimization (MPSO) method used to optimize the neural network. The results indicate that the tracking time of proposed method is 42%–49% less than other which is used for comparison. So, the sun tracker has better dynamic performance, lower power consumption and increased efficiency.