Design of multi-sensor data fusion monitoring system for plant factories

To address the problems of sensing blind spots and insufficient multi-parameter coordinated control in plant factory environmental monitoring, this study designs and implements an intelligent monitoring system based on multi-sensor data fusion. The system constructs a Zigbee-based wireless sensor network to collect real-time environmental data from multiple sources, including temperature, humidity, light intensity, and CO 2 concentration. A hierarchical fusion algorithm combining Kalman filtering and adaptive weighted averaging is proposed to achieve optimal estimation and dynamic correction of environmental states. Experimental results show that the root mean square error of temperature and humidity measurements is reduced by 49.3% and 50.0%, respectively. Even when sensors fail, the system can still maintain a state recognition accuracy of over 94%. Compared with traditional proportional-integral-derivative control, the system narrows the temperature fluctuation range from ± 2.5 °C to ± 0.8 °C, shortens the humidity recovery time by more than 50%, and reduces overall energy consumption by 15.7%. The results demonstrate that the system achieves high monitoring accuracy, stable environmental regulation, and strong resistance to interference, thereby providing technical support for refined environmental monitoring and intelligent control in plant factories.