基于改进型鱼群算法的番茄光环境调控目标值模型

Light Environment Regulation Target Model of Tomato Based on Improved Fish Swarm Algorithm

针对光温耦合条件下番茄光环境调控目标值难以快速、精确获取的问题,在光温嵌套光合速率试验结果基础上,为提高人工鱼群算法寻优速度,基于视野和步长动态调整思想,提出了改进型鱼群算法的光温耦合寻优方法,对不同温度下光饱和点进行快速精准寻优,建立了番茄光环境调控目标值模型,模型决定系数为0.999 9。验证试验结果表明,不同温度下光饱和点的模型计算值与实测值高度线性相关,相关系数为0.988,最大相对误差在±2%内,明显优于遗传算法构建模型的相对误差(±6%)。快速、动态获取不同温度下光饱和点,对设施光环境精准调控效率具有重要意义。

To struggle with the problems of hard to acquire the optimum light value for tomato planting rapidly and precisely,a model was developed to control the light staying around the optimum value in the environment. In order to evaluate the optimum light saturation points under different temperatures,a novel light and temperature coupling optimizing method based on improved fish swarm algorithm was proposed. This new method effectively improved the optimum speed of traditional fish swarm algorithm through adjusting the vision and step dynamically. In addition,the method could avoid trapping into local optimum,and get more accurate optimal solution than genetic algorithm. Based on the light saturation points by optimizing this method,the light environment regulation target model was established with nonlinear regression. For verifying the accuracy of the method,a set of light and temperature coupling photosynthetic rate test was performed. The results showed that the model determination coefficient can reach 0. 999 9,the squared error term was 1. 543,and the root mean square error was 0. 712. A comparison between simulation results and testing results was made,which showed the highly linear correlate relation with a value of 0. 988 between them. In addition,the maximum relative error was less than ± 2%,which is obviously better than the results of genetic algorithm. At last,a positive conclusion was obtained that the proposed light and temperature coupling optimizing method in this study can acquire the optimum light saturation points rapidly and dynamically,and has great significance to the precise control of light environment in facility.