Machine Learning-based Pest Count Forecasting for Solar Insecticidal Lamps

Agricultural pests severely threaten food security, resulting in significant economic, social, and environmental impacts. This paper uses weather data and machine learning models to forecast pest counts. Data from solar insecticidal lamps, including pest counts and meteorological conditions, were collected. Using this data, we developed a multidimensional forecasting model. The model integrates crucial variables, including pest counts, temperature, humidity, and geographical coordinates. The genetic algorithm back propagation neural network model has demonstrated excellent performance in forecasting pest numbers, with an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> value of 0.73, confirming its effectiveness in the early detection of crop diseases and pest infestations. Moreover, this high-accuracy forecasting model provides a scientific basis for intelligently adjusting the insecticidal working time of solar insecticidal lamps, thereby enhancing pest control efficiency and energy utilization. This research not only reinforces the application of machine learning in agricultural pest management but also provides significant support for precision agriculture.