Research And Application Of Greenhouse Tomato Fertilization Regulation Based On Generalized Regression Neural Network

The greenhouse tomato is a commonly seen greenhouse solanaceous vegetable.To raise production and quality through precise fertilization remains a problem to be solved.At present,the fertilization of greenhouse tomatoes is determined by the experience of growers,which fails to raise production and results in waste of resources.To this regard,this paper puts forward a fertilization control model that is able to realize the precise control of fertilization of greenhouse tomatoes.From the perspectives of yield prediction and fertilization control,this paper proposes a predictive model based on Generalized Regression Neural Network(GRNN)with the help of salp swarm algorithm to predict the yield of tomatoes.In view of the problem of Particle Swarm Optimization(PSO)in the stage of fertilization control,this paper proposes an improved version of Particle Swarm Optimization.With the yield predictive model as the target function,this paper realizes the iterative refinement of yield to get the target value of fertilization at different temperature,humidity,light intensity,and carbon dioxide concentration.Experimental simulation can be used to verify the validity of the proposed algorithm in this paper.In the process of building the yield predictive model of tomatoes,the algorithm in this paper is higher in precision and faster in convergence compared with the Generalized Regression Neural Network and the Generalized Regression Neural Network with the help of salp swarm algorithm.In the construction of the model of fertilization control,the Particle Swarm Optimization(PSO)in this paper is also higher in precision and faster in convergence compared with the current Particle Swarm Optimization.It is able to work out the optimal amount of fertilization in a quick and stable way.Through practical verification,the application rate regulation model proposed in this paper increased the yield in 80.5% of the experimental plots,while the application rate decreased in 69.4% of the experimental plots,which proved the superiority of the fertilization control algorithm based on the yield of tomatoes in this paper.It can be applied in real life.