An ANFIS-Based indirect control strategy for solar heating system: Exploring PMV approach

Abstract

A new path in solar heating system development is toward the control aim of thermal comfort. Considering the impact of dynamic behaviors on comfort, a Predicted Mean Vote (PMV) indirect control strategy based on an Adaptive Neuro-Fuzzy Inference System (ANFIS) improvement is proposed. Employing a three-speed fan coil unit as the terminal heat dissipation device, flow adjustment is achieved using a variable-frequency water pump, and humidity is maintained through a simple humidifier. By controlling the temperature and adjusting the fan coil unit speed to control the indoor airflow velocity, indirect control of PMV in the room is achieved. The working characteristics of the fan coil unit and the PMV control characteristics of different speed levels are analyzed, and three different PMV control modes are proposed based on the working characteristic curves of each speed level. The corresponding temperature and indoor airflow velocity are obtained by reverse-solving the PMV comfort zone. Simulation results show that after improving the fuzzy controller with ANFIS, the accuracy is increased by 17.65% compared to traditional control, and it has good control effectiveness. The proposed segmented PMV adjustment strategy combines the advantages of the three-speed levels, resulting in smaller overshoot, faster heating speed, and more stable PMV control, providing an effective solution for similar indoor environmental control problems.