A feedback control method for plant factory environment based on photosynthetic rate prediction model

In facility agriculture, environmental parameters are usually associated, uncertain and fluctuating. In the existing environmental control system, environmental parameters are regulated according to human experience, and the feedback control of environment and equipment energy consumption are not considered enough. It is of great significance to study an environmental optimization control system that can not only realize the feedback control of greenhouse climate but also reduce energy consumption, which can improve yield and reduce regulation cost. To this end, a feedback control system for plant factory environment based on photosynthetic rate prediction model is proposed in this paper. The system is composed of a master node and multiple sub-nodes. Firstly, low-power wide area network (LPWAN) technology is used to build a multi-network structure to achieve monitoring area network coverage. The master node is compatible with ZigBee, LoRa and Cat.1 communication technologies. Depending on the communication distance, ZigBee or LoRa communication technology is used for data transmission of the sub-nodes. Secondly, the photosynthetic rate prediction model was established by the least squares support vector machine (LSSVM) algorithm. Finally, the optimal control strategy for the plant factory environment was developed by integrating the multi-objective compatible control (MOCC) algorithm with a photosynthetic rate prediction model. The experimental results show that the coefficient of determination (R2) of the prediction model is 0.992. In the optimal control condition, the average leaf temperature was 27.55℃, the average CO2 concentration was 1214.20µmolmol−1, and the average PPFD was 1709.51µmolm−2s−1.