基于BP神经网络的灌溉决策系统设计开发

Design and Development of Irrigation Decision System Based on BP Neural Network

适时适量的灌溉对提高农作物的产量至关重要。针对中国传统粗放型农业生产中的盲目灌溉,开发了基于BP神经网络的灌溉决策系统。根据气象信息,运用彭曼公式或BP神经网络预测计算出ET0值(参考作物蒸发蒸腾量),再结合农作物的生长周期、生长环境得出精确的作物需水量。其中,BP神经网络预测算法写成m文件的形式,系统检测现有气象数据,通过动态调用MATLAB执行对应的m文件来完成灌溉决策。实验表明,该系统各部分工作良好,具有一定的实用价值。

Timely and appropriate rrigation is crucial for improving crop yield.A irrigation decision-making system based on BP neural network was developed to address blind irrigation in traditional extensive agricultural production in China,Based on meteorological information,use the Penman formula or BP neural network to predict and calculate the ETO value(referring to crop evapotranspiration),and then combine it with the growth cycle and environment of crops to obtain accurate crop water demand.Among them,the BP neural network prediction algorithm is written in the form of m files,the system detects the existing meteorological data,and completes the irrigation decision by dynamically calling MATLAB to execute the corresponding m files.The experiment shows that all parts of this system work well and have certain practical value.