中文摘要：

论文以温室内外的气象数据为输入量,以温室内温湿度等气象因子为输出量,使用改进PSO算法优化的RBF神经网络构建温室内环境温湿度的预测模型。通过实验对预测模型进行仿真测试与性能评估,验证该方法的可行性和有效性。该模型数据获取方便、所需参数少、模拟精度高,为温室内极端温度的预测、调控和管理优化提供了科学依据。

英文摘要：

Based on the meteorological data and outside greenhouse as input,the greenhouse temperature humidity and other meteorological factors as the output,the prediction model of greenhouse environment temperature and humidity with improved RBF neural network based on improved PSO algorithm.The simulation test and performance evaluation are carried out to verify the feasibility and effectiveness of the proposed method through the experiment.The model is convenient for data acquisition,few parameters and high accuracy,which provides scientific basis for the prediction,regulation and management of extreme temperature in greenhouse.