Research On Adaptive Regulation Of Multiagent Manufacturing System Based On Immune Mechanism

As an aggregation of enabling technology,modern manufacturing systems suffer various disturbances(e.g.rush orders,machinery failures,stock changes,etc.)from the industrial field,which are under constant changes.When referring to build an intelligent manufacturing system with selfadaption ability to improve the dynamic response capabilities,which can quickly perceive the cause of disturbance and eliminate the influence of unpredictable factors,is an important research topic.Combining the theory of multi-agent systems,this paper proposes a scheduling and control model of intelligent manufacturing system embedded iterative and heterarchical structure,in which apply the biological characteristics and immune mechanism.Aiming at intelligent disturbance perception and dynamic scheduling in a complex manufacturing environment,referring to self-organization,selflearning,and self-adapting features of the biological immune systems are introduced into the manufacturing system.In this paper,a control model of an intelligent manufacturing system with recursive control structure is built to further improve the adaptiveness of the manufacturing system.The main contents of this paper are as follows:(1)The adaptive control model of a multi-agent intelligent manufacturing system based on the immune mechanism is constructed.Firstly,by comparing the manufacturing system with the biological immune system,it shows that the manufacturing system and the immune system have similar organizational structures,and the manufacturing system can be controlled by referring to the biological system’s related immune algorithm.Secondly,based on the characteristics and related mechanisms of the biological immune system,combined with the structural characteristics of an intelligent manufacturing unit,the immune control unit model of an intelligent manufacturing system based on the immune mechanism is constructed.The model consists of the sensor,the controller,and the actuator.Finally,the multi-agent intelligent manufacturing system framework structure and adaptive immune regulation network are designed,and the immune regulation operation process of the intelligent manufacturing systems and the communication mode between immune control units are introduced.(2)Research on health status immune evaluation of manufacturing system based on Grey correlation is built.Firstly,the disturbance and state of the intelligent manufacturing system are perceived by the perceptron,and the data are processed by immune enhancement.Secondly,the analytic hierarchy process is used to construct a health assessment model with a hierarchical intelligent manufacturing system,which can assess the health status of the manufacturing system and its changing trend.Finally,the accuracy of the immunoassay method is verified by an example.(3)Research on multi-layer immune response mechanism of a manufacturing system based on the distributed agent.Aiming at the problem that manufacturing system is difficult to perceive disturbance,the immune response mechanism in the biological immune system is used in this paper,and negative selection algorithm is cited,and an antibody training system for disturbance problem is established in the workshop of manufacturing system,and a multi-layer immune response mechanism of a manufacturing system based on multi-agent is formed.At the same time,considering the incompleteness and timeliness of sample collection,an adaptive immune regulatory layer is established to realize the adaptive regulation of antibodies,eliminate the disturbance and maintain the efficient and stable operation of the system.Taking a discrete machine shop as the research object,the validity of this method is verified by experiments.(4)Research on artificial immune adaptive scheduling method based on reinforcement learning.Aiming at the problems of poor dynamic response-ability and lack of intelligence of traditional production scheduling methods,and immune scheduling agent was designed by referring to the immune system of the organism in this paper.Based on the self-learning theory and training mechanism,an immune scheduling model for the manufacturing system was constructed in this paper.The scheduling knowledge of the model is stored in the form of parameters,which are modifiable and can be further studied and optimized.At the same time,the disturbance processing mechanism is constructed,and the feasibility of the proposed scheduling model and algorithm is verified by algorithm analysis and simulation.The model has an excellent performance in disturbance processing.(5)An experimental verification platform for a multi-agent manufacturing system based on the immune mechanism was constructed.To verify the applicability of the adaptation mechanism of the multi-agent intelligent manufacturing system based on immune mechanism proposed in this paper,a hardware platform of multi-agent manufacturing system based on immune mechanism and a hormonelike communication protocol between intelligent manufacturing units were constructed.Finally,the feasibility and effectiveness of the adaptive control method of a manufacturing system based on the immune mechanism were verified by comparative experiments and data analysis.