Greenhouse Multi Environmental Factors Control Decision Method Driven By Crop Response

The modernization of facility agriculture has become an important part of the country’s agricultural development,and facility environment regulation,as a key technology in facility agriculture,has also achieved rapid development.Greenhouse regulation technology can improve crop quality,increase resource utilization,improve crop production conditions,and achieve high-efficiency and high-quality production of greenhouse crops.At present,the research of greenhouse regulation and control with tomato as the research object has made great progress,but there are still low regulation accuracy,less multi-factor coupling regulation,and often the regulation and optimization of the fixed growth period of the crop,which is difficult to achieve.Follow-up regulation during growth period.Aiming at the existing problems,this paper combines intelligent optimization algorithms to achieve precise control decisions for the growth environment suitable for greenhouse crops.The main research content and work of this paper are as follows:（1）A crop response prediction model based on an improved ant colony optimization algorithm optimized by a regression support vector machine algorithm（IACO-SVR）is proposed.In view of the slow convergence speed of the standard ant colony optimization algorithm and the tendency to fall into local optimality,the adaptive update method is used to improve,and the gray correlation analysis method is used to establish a relationship table with a high correlation with photosynthetic response,and realize IACO-SVR The algorithm accurately predicts the photosynthetic rate of crops in different growth periods.The model verification results show that the prediction performance is better than the BP neural network,SVR and ACO-SVR models,and can be used to make environmental decisions based on tomato responses.（2）An environmental factor optimization model based on improved particle swarm optimization algorithm（IPSO）is proposed.In view of the low optimization accuracy of the standard PSO algorithm and the problem of convergence to local extremes,by improving the dynamic optimization parameters,the photosynthetic response prediction model is used as the optimization fitness function to achieve the combination of different environmental factors during the whole tomato growth period.Search for the saturation point of CO₂ and light intensity when the photosynthesis rate is the highest.The test function experiment shows that the IPSO algorithm has a faster convergence speed and a stronger global search ability than the standard PSO algorithm,and provides reliable sample data for environmental factor control and decision-making.（3）The IACO-SVR algorithm is used to realize the control and decision-making of greenhouse environmental factors.Using the IACO-SVR modeling method,training the obtained sample data of the optimized environmental factor target value,respectively constructing the CO₂ concentration and light intensity control decision model for obtaining the full growth period,and achieving the CO₂ and light intensity of the tomatoes under different environmental conditions Optimal regulatory decision.Through experimental analysis,the actual measured value and the predicted value are compared and analyzed,which shows that the regulatory decision model has excellent generalization ability and predictive performance,and provides multi-factor regulatory decision information for the creation of a reliable and suitable growth environment for greenhouse tomatoes.