基于k-临近算法构建大棚番茄灰霉病预警及其生物农药干预系统

Construction of Early-warning System and Biological Pesticide Intervention System of Tomato Gray Mold Based on k-approaching Algorithm

采用大棚气象条件监测系统记录了大棚番茄两种种植季的温度、湿度和光照的变化,结合大棚番茄灰霉病发病率的规律,使用机器学习中的k-临近算法构建了大棚番茄种植过程中的灰霉病预警系统。监测结果表明,生物农药小檗碱或枯草芽孢杆菌均可针对灰霉病的发生起到良好的预防作用,且小檗碱的预防作用更佳,可将发病率控制在5%以下。运用该种预警系统与生物农药干预的结合,为大棚番茄灰霉病的防治提供了良好的理论基础。

An early-warning system on tomato gray mold was established based on the greenhouse temperature,humidity,light intensity and incidence rate during two planting seasons.The k-approaching algorithm in machine learning was used for calculation of the possibility of gray mold.At the same time,the results indicated that biological pesticide berberine and Bacillus subtilis are useful for prevention of gray mold,and berberine displayed better result of less than 5%of total gray mold rate.These early-warning system and biological pesticide intervention system are promising for the control of gray mold in greenhouse planting.