Photosynthetic Rate Prediction of Tomato Plant Population Based on PSO and GA

Abstract Photosynthesis is the basis of plant growth and it has a great significance to study the photosynthetic rate of plant population. In this paper, the LI-6400XT photosynthetic rate instrument and assimilation chamber were used to establish the plant photosynthetic rate prediction models. The wireless sensor network (WSN) system and assimilation chamber were used to monitor environmental information in real time, including time, air temperature, air humidity, temperature of assimilation chamber, temperature of gases in closed system, CO 2 concentration in assimilation chamber, light intensity, pressure of the atmosphere and leaf area. The Grid and pixel conversion method were used to measure the whole plant leaf area of tomato. As a semi-closed measurement system, the assimilation chamber was used to calculate the plant population photosynthetic rate together with LI-6400XT. To establish the plant population photosynthetic rate prediction models based on support vector machine (SVM), the greenhouse environmental parameters were used as input parameters and the photosynthetic rate was taken as the output parameters. In order to improve the prediction accuracy of the model, the input neurons were standardized using Z score method and then processed by principal component analysis. The principal components were selected according to the principal components’ cumulative contribution rate. The particle swarm optimization (PSO) and grid search method (GA)were used to optimize the parameter of SVM The results indicated that the correlation coefficient of the photosynthesis prediction model based on PSO and GA parameter optimization were 0.9883 and 0.9878 respectively. Experimental results show that this model has a high accuracy.