A Cucumber Photosynthetic Rate Prediction Model in Whole Growth Period with Time Parameters

Photosynthetic rate prediction models can provide guidance for crop photosynthetic process optimization, which has been widely used in the precise regulation of protected environment. The capacity crops continuously changes during their whole growth process. Previous studies on mainly consider interaction between a crop’s and its outer environmental conditions have able to predict certain period. However, periods not proposed yet. To solve this question, paper introduces growing time into variable set proposes method building cucumber model periods. First, leaves under different (light, temperature, CO2 concentration) period was obtained through multi-gradient nested test. With data cultivation as inputs, built using Support Vector Regression algorithm. In order obtain better modeling results, multiple kernel functions were pretraining, parameters algorithm optimized based population genetic algorithms. Compared with Back Propagation neural network Non-linear method, had highest accuracy, coefficient determination test 0.998, average absolute error 0.280 μmol·m−2·s−1, provides theoretical solution