It is very important for effective acquisition of temperature and humidity change trends in solar greenhouse to realize the accurate control of greenhouse environment.In order to effectively improve the prediction accuracy and reliability of temperature and humidity in sunlight greenhouses,a greenhouse temperature and humidity environmental prediction model of optimized Elman neural network based on the Sparrow Search Algorithm(SSA)is proposed.The study utilizes the main environmental influencing factors selected through Spearman correlation analysis as input variables,with future temperature and humidity inside the sunlight greenhouse as output variables.The Sparrow Search Algorithm is adopted to optimize and adjust the parameters of the Elman neural network model,thus completing the prediction of temperature and humidity trends in the sunlight greenhouse.Experimental validation is conducted by using monitoring data from a winter facility tomato sunlight greenhouse in Shandong Province from October 1,2022,to January 1,2023.The results show that the root mean square error,mean absolute error and coefficient of determination of the SSA-Elman model are 0.592,0.320 and 0.963 for temperature prediction,and 0.120,2.530 and 0.972,respectively,for humidity prediction,indicating that the proposed model can effectively make an accurate prediction of the temperature and humidity of the solar greenhouse.The proposed model can effectively predict the temperature and humidity of solar greenhouses,which can provide reliable data support and a decision-making basis for the future accurate regulation of the greenhouse environment.
关键词
Elman神经网络/温室温湿度/农业物联网/麻雀搜索算法
Key words
Elman neural network/greenhouse temperature and humidity/agriculture Internet of Things/Sparrow Search Algorithm
维普
万方数据