面向智能配电的无线融合网络关键问题研究

发布时间：2018-03-08 22:17

本文选题：智能配电网　切入点：无线融合网络　出处：《华北电力大学》2014年硕士论文　论文类型：学位论文

【摘要】：智能配电通信网络建设的薄弱之处在接入网部分,目前不存在任何一类网络可以完全为该部分通信业务提供满足工业化需求的服务。无线融合网络通过多种无线接入技术为用户提供可靠、安全、高效的服务,极其适合应用在智能配电通信网络接入部分,因此对无线融合网络关键问题的研究显得尤为重要。其中异构融合网络中负载均衡及无线传感器网络中路由选择是无线融合网络研究的两个关键问题,且是影响系统资源利用率及能量消耗的首要因素。因此,设计适于智能配电通信接入网的异构融合网络负载均衡策略及无线传感器网络路由选择策略,在无线融合网络的研究中扮演着至关重要的角色。在对智能配电通信网络及业务模型分析的基础上,构建了面向智能配电的无线融合网络模型。该模型不仅在配电网接入层支持多种无线接入技术重叠覆盖,而且以无线传感器网络作为感知延伸层,满足未来智能电网无缝覆盖的通信需求。针对异构融合网络中负载均衡问题,首先结合智能配电通信业务需求,提出一种基于物元分析的负载均衡算法,算法从接入控制的角度考虑负载均衡问题,将信号强度、网络可用资源及网络阻塞率作为评价指标,通过物元分析和熵权法得到各个网络质量等级排序,根据排序结果对不同类型新到配电通信业务按照优先级进行有差别的服务。其次,综合考虑重载网络的情况和业务的准入控制,结合资源借用和流量转移技术,提出一种混合动态负载均衡算法。首先利用三角模融合算子选择可借用资源的网络,通过同构网络间借用资源减轻重载网络的负载或根据转移非实时、占用资源多业务优先原则转移部分业务到轻载网络以缓解网络的重载程度；其次对新产生的配电通信业务,将其接入到可用资源率最大的网络,以均衡各个网络间的负载。对无线传感器网络路由选择问题,在研究了现有链式路由算法的基础上,提出一种基于D-S证据理论的链式路由算法。该算法在建链阶段通过考虑已经加入链的节点,使距离最近的两个节点直接相连而形成一条有分支的链,以此保证形成的链中不存在长链；在决策链首节点时,利用D-S证据理论,对节点到基站距离和节点剩余能量的评价结果进行融合,从而决策出符合要求的链首节点。仿真结果表明,该算法在均衡网络能耗的同时,大大延长了网络寿命。
[Abstract]:The weakness of the construction of intelligent distribution communication network is in the access network. At present, there is not any kind of network that can completely provide services to meet the industrial requirements for this part of the communication services. Wireless convergence networks provide reliable, secure and efficient services to users through various wireless access technologies. It is very suitable for intelligent distribution and communication network access, Therefore, it is very important to study the key problems of wireless fusion network, in which load balancing in heterogeneous convergence network and routing selection in wireless sensor network are two key issues in the research of wireless fusion network. Therefore, a load balancing strategy for heterogeneous fusion networks and a routing strategy for wireless sensor networks are designed for intelligent distribution communication access networks. It plays an important role in the research of wireless fusion network. Based on the analysis of intelligent distribution communication network and service model, a wireless fusion network model for intelligent distribution is constructed, which not only supports overlapping coverage of multiple wireless access technologies in the access layer of distribution network. Furthermore, wireless sensor network (WSN) is used as the sensing extension layer to meet the communication requirement of seamless coverage of smart grid in the future. In order to solve the load balancing problem in heterogeneous fusion network, a load balancing algorithm based on matter-element analysis is proposed, which considers the load balancing problem from the point of view of access control, and combines the demand of intelligent distribution communication service. The available resources and blocking rate of the network are used as evaluation indexes, and the ranking of each network quality grade is obtained by matter element analysis and entropy weight method. According to the results of sorting, different types of new distribution communication services are provided with different services according to their priorities. Secondly, considering the situation of heavy load network and the access control of service, combining the technology of resource borrowing and flow transfer, In this paper, a hybrid dynamic load balancing algorithm is proposed. Firstly, the triangular modular fusion operator is used to select the network with borrowed resources, which can reduce the load of the heavy-load network or reduce the non-real-time load according to the transfer between isomorphic networks. The principle of multi-service priority transfers some services to the light load network to alleviate the heavy load degree of the network. Secondly, the newly generated distribution communication services are connected to the network with the highest available resources to balance the load between the various networks. In this paper, a chain routing algorithm based on D-S evidence theory is proposed to solve the routing problem in wireless sensor networks. The nearest two nodes are connected directly to form a branched chain, so as to ensure that there is no long chain in the formed chain, and the D-S evidence theory is used to make use of the D-S evidence theory when the first node of the chain is decided. The evaluation results of the distance from the node to the base station and the residual energy of the node are fused to determine the chain head node which meets the requirements. The simulation results show that the proposed algorithm not only equalizes the network energy consumption but also greatly prolongs the network lifetime.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM76

【参考文献】