Research On Optimization Of IoT Agricultural Service Based On Multi-objective Evolutionary Algorithm

Internet of things(IoT)intelligent service system has the ability of intelligent decision-making,its main purpose is to provide convenient services for people in various fields.In order to achieve this goal,there are still many difficulties and challenges,especially in the large-scale agricultural IoT services.At present,the ability of intelligent IoT to reasonably allocate agricultural resources is still relatively weak,and its ability to adapt to different scale scenarios needs to be improved.There are many factors affecting the service capability of IoT system,among which there are many multi-objective problems,such as service cost and service time.How to solve the multiobjective problem in the service has become the key to improve the regulation ability of agricultural IoT service.Intelligent evolutionary algorithm(EA)is an algorithm based on the idea of natural evolution,which is one of the key technologies to solve multi-objective problems.The decomposition based multi-objective evolutionary algorithm is more effective for the search and solution of practical multi-objective optimization problems.It decomposes the complex high-dimensional multi-objective problem into several standard single objective problems,and optimizes them at the same time to improve the optimization speed of the algorithm;and add appropriate calculation operators to improve the diversity of candidate solutions and enhance the ability of the algorithm to optimize problem.In this paper,based on the decomposition of multi-objective optimization algorithm,the optimization operator is improved reasonably.Around the agricultural IoT service optimization problem,the static service and dynamic service are studied.For the static IoT service problem,a multi-objective optimization model is constructed to minimize the service time and total service cost,and an improved multiobjective evolutionary algorithm based on decomposition for IoT service is proposed.According to the characteristics of the problem,the appropriate real coding method,population initialization method and solution modification method are designed,and the simulated binary crossover and Gaussian mutation operator are used to increase the diversity of candidate solutions.In order to evaluate the effectiveness of the algorithm,it is applied to three scenarios of agricultural IoT services.The simulation results show that the algorithm can better achieve the trade-off of IoT service solutions,reduce the total cost of services,and also shorten the service time.In addition,due to the static IoT service problem does not take into account the dynamic changes of service requests in real life,this paper also studies and analyzes the dynamic IoT service problem.and establishes a multi-objective optimization model of dynamic IoT service under the dynamic changing environment;According to the characteristics of the service problem,the decomposition based evolutionary algorithm is improved,and a dynamic multi-objective evolutionary algorithm is proposed to optimize the problem.In order to deal with the changing environment,dynamic monitoring operator is added to the algorithm.Due to the dynamic nature of the dynamic IoT services,the algorithm adopts two service strategies,namely single target service and collaborative service,and five distributions are set for experimental comparison.Experimental results show that the proposed algorithm can effectively find the optimal solution in the dynamic agricultural IoT environment,and its performance is better than other comparative algorithms.