Greenhouse Microclimate Control Optimization Based on Improved NSGA-H Algorithm

Optimization of systems can increase the efficiency of a control system. The greenhouse microclimate includes temperature, light, water, gas, fertilizer and other environmental parameters within a controlled environment. Usually, the temperature or humidity is controlled through a single factor, but several control objectives may be contradictory. To solve this problem, an improved non-dominated sorting genetic algorithm-II (NSGA-II) is presented. The adaptive mutation method is used to replace the traditional polynomial mutation method, which can increase the rate of obtaining the Pareto optimal solution while maintaining the diversity of the solution set. The arithmetic crossover method is used to replace the original crossover method, which can enlarge its search domain, make the algorithm have better global searching ability and keep the diversity of the knowledge set better. The simulation results show that when the improved multi -objective optimization algorithm is applied to the geenhouse environment model, the Pareto optimal frontier solution set has a more reasonable distribution and has good efficiency and performance in greenhouse environment control.