Universality of an improved photosynthesis prediction model based on PSO-SVM at all growth stages of tomato

CO2 concentration is an environmental factor affecting photosynthesis and consequently the yield and quality of tomatoes. In this study, a photosynthesis prediction model for the entire growth stage of tomatoes was constructed to elevate CO2 level on the basis of crop requirements and to evaluate the effect of CO2 elevation on leaf photosynthesis. The effect of CO2 enrichment on tomato photosynthesis was investigated using two CO2 enrichment treatments at the entire growth stage. A wireless sensor network-based environmental monitoring system was used for the real-time monitoring of environmental factors, and the LI-6400XT portable photosynthesis system was used to measure the net photosynthetic rate of tomato leaf. As input variables for the model, environmental factors were uniformly preprocessed using independent component analysis. Moreover, the photosynthesis prediction model for the entire growth stage was established on the basis of the support vector machine (SVM) model. Improved particle swarm optimization (PSO) was also used to search for the best parameters c and g of SVM. Furthermore, the relationship between CO2 concentration and photosynthetic rate under varying light intensities was predicted using the established model, which can determine CO2 saturation points at the various growth stages. The determination coefficients between the simulated and observed data sets for the three growth stages were 0.96, 0.96, and 0.94 with the improved PSO-SVM and 0.89, 0.87, and 0.86 with the original PSO-SVM. The results indicate that the improved PSO-SVM exhibits a high prediction accuracy. The study provides a basis for the precise regulation of CO2 enrichment in greenhouses.