Development of IoT Data Logger for Photovoltaic Monitoring System

Abstract

Generally, the degradation in performance due to environmental factors are some significant flaws of solar photovoltaic (PV) systems, rendering them unreliable for isolated or remote facilities. Thus, remote PV systems require extensive Real-Time monitoring systems to gather all parameters when estimating system performance and functionality. Consequently, data loggers and monitoring systems are essential for an effective, dependable, and trouble-free operation of PV solar energy systems. Also, data recorder and monitoring systems aids in the identification of system malfunctions prior to adverse failure. However, the Internet of Things (IoT) enables wireless collection and transmission of data without human intervention. Therefore, this system integrates a data logger so that application users can view data from an inaccessible location. Also, using two sensors, the LM-35 and the LDR sensor, this system displays the incident light intensity and operating temperature. Overall, this research suggests a data logger system that is capable of measuring solar PV voltage and temperature then transmits information via mobile networks to the internet for data recording and user review. Additionally, this ESP 32-based data recorder stores all monitoring parameters on a micro-SD card and displays them on a Blynk application. Hence, data can be downloaded directly from the webpage for system analysis and verification. However, to monitor the associated parameters, a PV panel with electrical parameters that is fed into the data logger's input channel is utilized. Moreover, the system converts the acquired unprocessed data into digital input for data acquisition and stores the data. Thus, when data is sent via internet media, a WIFI module (NodeMCU) is used to send Arduino readings to the Blynk platform. Finally, a comparison of IoT and SM206 solar irradiance meter readings provided an average voltage reading error of lesser than 15 % and the temperature reading showed an error of lesser than 1 %.