基于DFNN的智能配电异构无线网络准入控制算法

发布时间：2018-01-11 15:06

本文关键词：基于DFNN的智能配电异构无线网络准入控制算法　出处：《现代电力》2014年06期 　论文类型：期刊论文

【摘要】：为提高智能配电通信业务的服务质量,根据智能配电网对通信技术的要求,提出一种基于动态模糊神经网络(DFNN)的智能配电异构无线网络准入控制算法。在智能配电网络的异构准入控制模型中构建神经网络系统,以网络的接入阻塞率差作为系统参数强化学习的目标,对网络的负载均衡具有较好的动态适应性。神经网络系统在输入层较多时容易产生太多规则而影响决策结果,而DFNN通过计算当前系统规则的完备性,动态添加规则,并通过计算所有规则的重要性,动态删除规则,使得系统的规则有效而不冗余。仿真结果表明,该方法较多接入选择算法(MLB)明显降低了网络的接入阻塞率,相对于模糊神经网络算法(FNN)而言简化了系统结构,突出了规则的重要性,具有较低的接入阻塞率和更好的均衡效果。
[Abstract]:In order to improve the intelligent distribution communication quality of service, according to the requirements of intelligent distribution network communication technology, this paper proposed a dynamic fuzzy neural network (DFNN) based on the intelligent distribution of heterogeneous wireless network access control algorithm. In the heterogeneous access intelligent distribution network control based on neural network system model, the network access blocking rate is the difference the system parameters of reinforcement learning, has better dynamic adaptability to the network load balancing system based on neural network. The input layer is more easy to produce too many rules and the impact of the decision-making results, and through the calculation of the current DFNN completeness of system rules, add dynamic rules, and by calculating the importance of all the rules, dynamic delete rules, which makes the system the rules without redundancy. Simulation results show that the method of access selection algorithm (MLB) significantly reduces the network access blocking rate, phase The fuzzy neural network algorithm (FNN) simplifies the system structure, highlights the importance of the rules, has a lower access blocking rate and better balance effect.

【作者单位】：新疆电力公司;华北电力大学;
【分类号】：TM76;TN92
【正文快照】： 0引言在飞速发展的社会经济环境中,智能电网是未来提高电能质量、加强企业服务水平等的重要工具[1-2]。作为智能配电网的重要组成部分,智能配电通信网是实现智能配电网的基础[3]。其骨干网和部分接入网络首选光纤通信网络,并辅以宽带无线通信网络方式[4-5]。但是在一些比较偏

【参考文献】