Prediction of Tomato Yield in Chinese-Style Solar Greenhouses Based on Wavelet Neural Networks and Genetic Algorithms

Yield prediction for tomatoes in greenhouses is an important basis making production plans, and yield accuracy directly affects economic benefits. To improve the of tomato Chinese-style solar (CSGs), a wavelet neural network (WNN) model optimized by genetic algorithm (GA-WNN) applied. Eight variables are selected as input parameters output. The GA adopted to optimize initial weights, thresholds, translation factors WNN. experiment results show that mean relative errors (MREs) GA-WNN model, WNN backpropagation (BP) 0.0067, 0.0104, 0.0242, respectively. root square (RMSEs) 1.725, 2.520, 5.548, EC values 0.9960, 0.9935, 0.9868, Therefore, has higher precision better fitting ability compared with BP models. research this paper useful from both theoretical technical perspectives quantitative CSGs.