Design of supervisory controllers and ultra-low power data loggers for hybrid power systems

Abstract

Standalone photovoltaic (PV) systems are crucial in the transition towards sustainable energy, offering reductions in fossil fuel dependence and lower electricity costs. Optimizing these systems' performance requires effective data loggers to monitor and record operational data. Furthermore, proprietary and non-configurable data logger available in the current market impede their massive adoption. In this thesis four different designs of the data logging system are proposed to solve the existing data logging system problems. The first design emphasizes low power consumption and comprehensive data management. Key low-power strategies are presented, including reducing supply voltage and CPU frequency, using a data buffering mechanism, and optimizing the Wi-Fi connection interval. In the second design the novel Human-Machine Interface (HMI) and data storage solution are proposed. A mobile application as the HMI receives data via Bluetooth Low Energy, and then carries out historical data analysis and trend identification. Data is also intermittently transferred to a website via Wi-Fi for substantial remote storage free of charge. Following is the third design, which put emphasis on supervisory control of the load by using a relay, the deployment of Master Terminal Unit (MTU), and the use of the middleware Node-RED. Data is transmitted from an ESP32-E microcontroller to a Banana Pi M4 Berry (MTU) via the Message Queuing Telemetry Transport (MQTT) protocol. The Node-RED platform on the MTU provides customizable dashboards. In the last design, two Internet of Things (IoT) platforms are utilized to build a scalable IoT-Supervisory Control and Data Acquisition (SCADA) system, which facilitates the implementation of a complete five-layer IoT architecture. The load images are available on a web camera server for inspection. Verified under the laboratory setup, the ultra-low power, open-source, IoT-based data loggers have showcased their great potential in monitoring, controlling, and logging PV system operational data and process.