Abstract. Existing carbon dioxide concentration (CO2) control systems commonly use a quantitative replenishment of CO2, without considering the effects of multiple environmental factors on a plant’s photosynthetic rate and its characteristic impact on CO2 demand, resulting in improper control of CO2 concentration. Accordingly, in this article, a nested combination experiment for the photosynthetic rate of cucumbers is presented. In order to obtain a continuous carbon dioxide response curve, the photosynthetic rate prediction model is established using the cucumber experimental data based on the support vector machine (SVM) algorithm. The network of the support vector machine photosynthetic rate prediction model is used as an optimal objective function and the improved artificial fish swarm algorithm (IAFSA) is employed to search for the saturation of CO2 in a multidimensional nesting condition. Further, the optimal CO2 regulation model based on multi-factor coupling is established using the results of the above-mentioned experiments. Moreover, XOR verification of the proposed model showed that the maximal relative error of the proposed optimal CO2 regulation model is 3.898%. Consequently, using a wireless sensor network platform, a multi-sensor fusion-based CO2 control system is realized and verified. The verification showed that the average relative error between the target CO2 value and the actual CO2 value is 2.88%. At the same time, the average photosynthetic rate (Pn) of the crop increased by 26.94% compared to the contrasting region, which proves that the proposed system can achieve a stable and reliable operation, greatly improving the efficiency of the environment of the facility. Keywords: CO2 control system, Optimal CO2 regulation model, Photosynthetic rate, Verification.
Robust real-time heart rate prediction for multiple subjects from facial video using compressive tracking and support vector machine