基于ANFIS的温室小气候环境因子预测模型辨识

Identification of Greenhouse Microclimate Environmental factors Prediction Model Based on ANFIS

为了提高日光温室小气候环境因子的预测精度,达到对温室内环境因子综合控制的目的,提出一种基于Takagi-Sugeno模糊模型的自适应神经模糊推理系统(ANFIS)对温室小气候环境因子进行辨识,建立温室小气候环境因子预测模型,并用所建立的模型对温室内的温度、湿度、光照等小气候环境因子进行预测。结果表明:预测值和实测值的拟合关系较好,尤其在光照环境因子的模拟上,相关度甚至达到0.9976,说明基于ANFIS网络进行温室小气候非线性系统辨识是有效的,且该成果对温室小气候智能调控的发展具有一定的参考价值。

In order to improve the prediction accuracy and achieve comprehensive control of the environmental factors in the solar greenhouse climate, we proposed adaptive neural fuzzy inference system(ANFIS) based on the Takagi-Sugeno fuzzy model to identify the greenhouse microclimate environmental factors, thus the prediction model was established. The greenhouse temperature, humidity, illumination were predicted with the model. The simulation results showed that the curve fit the predicted data and the measured data was validated, and the correlation coefficient even reached 0.9976 in the simulation of radiation,which demonstrated that the identification of nonlinear system such as greenhouse microclimate based on ANFIS was effective.The achievement has certain reference value for the development in intelligent control of the greenhouse microclimate.