In order to improve the utilization of CO2 fertilizer, the photosynthetic rate of tomato plants in the flowering stage was studied. A wireless sensor network system was used to real time monitor greenhouse environmental parameters. A LI-6400XT portable photosynthesis analyzer was used to measure the photosynthetic rate of tomato plants, and the environmental parameters of leaves were controlled according to the presetting rule. The photosynthetic rate prediction models of single leaves were established based on the back-propagation (BP) neural network. The environmental parameters were used as input neurons after processed by principal component analysis (PCA), and the photosynthetic rate was taken as the output neuron. The performance of the prediction model was evaluated. The prediction results of the models showed that the correlation coefficient between the simulated and observed data sets was 0.99, RMSE was 0.288. Furthermore, when different CO2 concentrations were selected as the input to predict the photosynthetic rate, the simulated and observed data showed the same trend. According to the above analysis, it was concluded that the model could be used as the basis of the quantitative regulation of CO2 fertilization to tomato plants in greenhouse.