Research On Key Technologies Of Quality Of Service Optimization For Industrial Wireless Networks In Intelligent Manufacturing

Intelligent manufacturing system is characterized by small batch and flexible demands.The distributed manufacturing cells complete the tasks in a collaborative and self-organized form,so the interconnections between all units are great of significance.With various advantages,wireless networks have become the key communication technology for maintaining connectivity within intelligent components.However,emerging features of intelligent manufacturing system,such as intelligentization,flexibility,mobility,reconfiguration and other aspects,are bringing great challenges to the quality of services(QoS)of conventional industrial wireless networks(IWN).Specifically,real-time capacity and energy efficiency of IWN will directly affect the performance of intelligent manufacturing system.Knowing the importance of QoS in IWN,we focus on real-time capacity and energy efficiency.Considering the clustered network topology and special requirements of QoS in manufacturing IWN,in this dissertation,we provide an extensive research on the optimization of QoS in IWN from the perspective of real time capacity and energy efficiency.Concretely,we conduct studies from node deployment,mobile handover,data transmission and load balance,in each part,relevant optimization methods are provided.The major contributions of our work are as follows:From aspect of nodes,in view of the limitations of state-of-art wireless network deployment strategies and the characteristics of IWN,the deployment strategy of critical IWN nodes considering real time constraints is proposed.Firstly,the wireless coverage problem on stationary and mobile nodes is transformed to one-dimensional target coverage problem,and the optimal deployment algorithm of industrial wireless network cluster considering real-time constraint is then provided.Secondly,in order to reduce the inter-cluster interference,we design the strategy of re-deploying of cluster heads based on improved virtual force algorithm.The experimental results show that the proposed node deployment strategies can guarantee the real-time capacity and reduce energy consumption.In light of the shortcomings of traditional mobile handover and access control strategies in IWN,a cloud-assisted fast handover and wireless access strategy is presented.Firstly,a cloud-assisted algorithm is designed to reduce the network scanning and decision time of mobile nodes in the handover process.Moreover,in order to improve the real-time performance of the mobile nodes accessing new subnet,the optimal access strategy of mobile nodes is generated based on ant colony algorithm.The experimental results verify that the proposed handover and access control strategy has advantages in terms of energy consumption,real time capacity and handover requests.In the Transmission level,we propose real-time transmission strategies based on time-constraint classification and distributed data caching.For data communication between inter-cluster members,the transmission time limit is classified.In regard of the lower time limited transmission constraint,an optimal-path data transmission method based on Path Difference Degree(PDD)is proposed,meanwhile optimal transmission power control strategy is adopted for higher time limited scenario.For swift data distribution,an optimal distributed data caching algorithm based on effective path coverage along with a rapid data acquisition method for mobile nodes are designed.Finally,simulation results prove the advantages of the approaches.From the network layer,in regard of the periodical and spatio-temporal variation of loads in industrial wireless network,a load balancing strategy based on real time capacity and energy efficiency is proposed.First of all,temporally,we divide the duty cycle of the system into several segments,and then spatially,industrial wireless network is divided into multiple subnets according to existing clusters,thereby we vectorize the load pattern of industrial wireless network.Then,an industrial network load balancing strategy based on real-time constraint is proposed to balance the load distribution of the whole network.Moreover,a wireless cluster head dynamic dormancy strategy to reduce network energy consumption is designed.Finally,simulation results show that our strategy can alleviate the load aggregation problem,while maintaining reasonable energy consumption and real time performance.Finally,based on our own prototype system,the intelligent manufacturing wireless network verification platform is implemented.Strategies of this research are implemented and verified.Specially,the real world performances of QoS in IWN are discussed from the perspective of transmission delays,successful delivery ratio of data and energy consumption.The promising results convince us that our work is practically valuable.