Rooftop greenhouse system based on photovoltaic and Internet of Things research and design

In the context of the “dual-carbon” strategy, the integration of renewable energy with agriculture has emerged as a novel approach, offering a new model to address land tension during urbanization, enhance the efficiency of utilizing urban idle spaces, and foster sustainable green urban development. This paper introduces the design of a rooftop greenhouse system based on renewable energy and the Internet of Things (IoT), built upon the study of photovoltaic power generation and rooftop greenhouses. Initially, we propose a feasible solution for rooftop greenhouse space agriculture and construct an architectural model to optimize the layout of photovoltaic arrays. Subsequently, we design a greenhouse remote monitoring system with STM32 as the control core, employing LoRa modules for information transmission from the collection terminal and enabling remote interaction with the upper computer through the NB-IoT module. Next, we introduce an improved APSO-optimized fuzzy PID algorithm (APSO-FUZZY-PID) and compare its control effects with two other temperature control models. The results demonstrate that the APSO-FUZZY-PID control model achieves higher efficiency and accuracy. Finally, the optimized GWO-BP neural network is employed to forecast the photovoltaic greenhouse temperature. Our MATLAB simulation analysis reveals that the GWO-BP prediction method exhibits higher accuracy compared to traditional prediction methods.