Cooperative Optimization And Regulation Strategies Of Dynamic Services In Agricultural Internet Of Things

Internet of Things(IoT)service systems should be capable of intelligent learning,processing,decision-making and control.However,there are still many challenges to achieve this goal.At present,the intelligent service of IoT is still relatively weak,and it is urgent to improve the coordination ability of information resources which intelligently adapts to service requirements,and the ability of intelligent decision control.It is the key to design a cooperative regulation mode in the autonomous services to realize the dynamic adaptation of IoT and the global optimal services.In this dissertation,the regulation mechanism of biological network and heuristic intelligence idea are used,such as nervous system dominance,endocrine regulation and immune system defense.Focusing on the collaborative regulation and optimization of IoT service resources,a series of in-depth studies have been carried out on such issues as intelligent IoT service modeling,services provision of static requests sequences,dynamic services provision of dynamic requests sequences,and coevolution.This dissertation aims to dig into the deep hidden relationship between requests and services in IoT service system.This dissertation studies the pyramidal evolution model from the candidate solutions of the foundation layer to the elite solutions of the top layer,and gives consideration to and maintains the balance between convergence and diversity,so as to construct a reasonable and efficient collaborative regulation and optimization algorithm.The main work of this dissertation is as follows:(1)The basic characteristics,research background and research goal of agricultural IoT and IoT services are summarized.This dissertation also introduces the neuro-endocrine-immune regulation system and the relationship between these systems,and analyzes the feasibility of applying the regulation mechanism of biological systems to solve the problem of IoT services provision.They provide a basic idea for the studies of resources autonomous allocation,intelligent collaboration,regulation and optimization in IoT services.(2)Aiming at the characteristics of multi-source,multi-type and uneven tasks of IoT service data,this dissertation considers the multiple requests at random time and constructs the agricultural IoT service scenario and single-objective optimization model.Drawing on the regulation rule of hormones in the endocrine system,an adaptive immune algorithm is designed and applied to the calculation of service equipment selection in agricultural IoT.Simulation results show that the algorithm is effective and feasible in the application scenarios of agricultural IoT service.(3)The intelligent devices in IoT not only provide services,but also consider how to allocate heterogeneous resources and reduce resources consumption and service time as far as possible.This dissertation constructs a multiobjective optimization model of IoT services.Drawing on the coordination mechanism between immune system and endocrine system,we propose a hierarchical coevolutionary multiobjective optimization algorithm(IE-HCMOA).In IE-HCMOA,a multiobjective immune algorithm based on global ranking with vaccine is designed to choose superior antibodies.Meanwhile,we adopt clustering in top population to make the operations more directional and purposeful and realize self-adaptive searching.And we use human forgetting memory mechanism to design two levels memory storage for the choice problem of solutions to achieve promising performance.The simulation results demonstrate that the proposed algorithm has stronger exploration ability and better robustness,and can obtain globally optimal aggregation services.(4)To solve the multiobjective optimization problem in IoT service system aiming at the dynamic requests at random time,this dissertation proposes a bio-inspired self-learning coevolutionary dynamic multiobjective optimization algorithm(BSCA).Inspired from the cooperative mechanisms among nervous,endocrine and immune systems in the human being,a three-layer progressive architecture is devised.The first layer is composed of multiple subpopulations evolving cooperatively to obtain diverse Pareto fronts.Based on the solutions obtained by the first layer,the second layer aims to further increase the diversity of solutions.The third layer refines the solutions found in the second layer by adopting an adaptive gradient refinement search strategy and a dynamic optimization method to cope with changing concurrent multiple service requests,thereby effectively improving the accuracy of solutions.Experiments on agricultural IoT services in the presence of dynamic requests under different distributions are performed based on two service-providing strategies,i.e.,single service and collaborative service.The results show that the performance of the proposed algorithm is better,and it can obtain the global optimal regulation schemes,realizing the environmental adaptability of the dynamic services in IoT.Finally,the dissertation summarizes the collaborative regulation and optimization problem of IoT service resources.Then it points out the problems to be solved urgently in this field,and prospects the future research direction and methods.