Photovoltaic (PV) systems need measurements of incident solar irradiance and PV surface temperature for performance analysis and monitoring purposes. Ground-based network sensor measurement is preferred in many near real-time operations such as forecasting and photovoltaic (PV) performance evaluation on the ground. Hence, this study proposed a Fuzzy compensation scheme for temperature and solar irradiance wireless sensor network (WSN) measurement on stand-alone solar photovoltaic (PV) system to improve the sensor measurement. The WSN installation through an Internet of Things (IoT) platform for solar irradiance and PV surface temperature measurement was fabricated. The simulation for the solar irradiance Fuzzy Logic compensation (SIFLC) scheme and Temperature Fuzzy Logic compensation (TFLC) scheme was conducted using Matlab/Simulink. The simulation result identified that the scheme was used to compensate for the error temperature and solar irradiance sensor measurements over a variation temperature and solar irradiance range from 20 to 60 °C and from zero up to 2000 W/m2. The experimental results show that the Fuzzy Logic compensation scheme can reduce the sensor measurement error up to 17% and 20% for solar irradiance and PV temperature measurement.
The increased demand for solar renewable energy sources has created recent interest in the economic and technical issues related to the integration of Photovoltaic (PV) into the grid. Solar photovoltaic power generation forecasting is a crucial aspect of ensuring optimum grid control and power solar plant design. Accurate forecasting provides significant information to grid operators and power system designers in generating an optimal solar photovoltaic plant and to manage the power of demand and supply. This paper presents an extensive review on the implementation of Artificial Neural Networks (ANN) on solar power generation forecasting. The instrument used to measure the solar irradiance is analysed and discussed, specifically on studies that were published from February 1st, 2014 to February 1st, 2019. The selected papers were obtained from five major databases, namely, Direct Science, IEEE Xplore, Google Scholar, MDPI, and Scopus. The results of the review demonstrate the increased application of ANN on solar power generation forecasting. The hybrid system of ANN produces accurate results compared to individual models. The review also revealed that improvement forecasting accuracy can be achieved through proper handling and calibration of the solar irradiance instrument. This finding indicates that improvements in solar forecasting accuracy can be increased by reducing instrument errors that measure the weather parameter.