实现负载均衡的电力通信网络优化策略研究

发布时间：2019-03-12 16:11

【摘要】：国内外经济、能源形式的深刻变化，给电网的发展带来新的机遇和挑战。近年来，智能电网的建设不断深入，现代电网信息化、自动化、互动化的特征愈发明显，为电网起着强力支撑作用的电力通信网络也随之发生着深刻的变革。随着信息通信技术与电力生产、经营的深度渗透，电网对电力通信网络的依赖性越来越强，通信网络的可靠性和传输性能也成为影响电网安全稳定运行的关键因素。随着网络拓扑的日益复杂、业务需求的逐步丰富，电力通信网络中的负载不均衡问题也愈发严重，部分核心节点或重要链路承载着大量的业务数据，一方面增加了网络的脆弱性；另一方面降低了网络资源的利用效率。传统的优化策略由于未考虑电力通信网络拓扑和业务流量的特殊属性，难以直接应用。本文依据电力通信网络的拓扑特性和业务流量属性，针对电力通信网络拓扑优化、业务流量分析和预测及路由优化问题进行了研究，主要内容包括以下几个方面：（1）在鲁棒性和脆弱性分析的基础上，提出了基于节点介数加矩阵的拓扑优化算法，主要针对网络中的薄弱环节进行拓扑优化。优化算法首先基于节点介数加矩阵生成优化链路集，然后通过增加有限数量的链路来实现优化目标，同时采用重复节点回避机制避免网络业务的过度集中。仿真实验表明，该算法能够有效增强网络抵御选择性攻击的能力，延缓网络的一次效能突降现象，同时可以提高关键节点的带宽利用效率，降低关键节点发生拥塞的几率，对电力通信网络拓扑的优化具有更强的针对性和更明显的效果。（2）基于ARIMA（Autoregressive Integrated Moving Average Model）模型提出了适用于电力通信网络流量特点的流量分析预测方法，对电力通信业务流量进行了分析和预测。按照ARIMA建模流程和方法，重点对生产数据业务和视频业务进行了流量分析与预测，通过数据预处理、确定模型参数、模型建立等几个步骤构建电力通信业务流量分析预测模型，并采用残差检验的方法对模型的适用性进行验证。仿真实验表明，本文提出的流量分析和预测方法充分考虑了流量自身变化趋势、周期性等因素，能够较好的适应电力通信业务流量特点，有效提高了模型拟合的精确度和预测的准确度。（3）提出了一种基于链路权重控制的可变等价多路径模型VECMP（VariableEqual-Cost Multi-Path），，并基于改进遗传算法设计IGA-VECMP（ImprovedGenetic Algorithm for Variable Equal-Cost Multi-Path）求解算法，对VECMP模型进行求解。其中，VECMP模型基于链路权重可变原则，在传统ECMP模型的基础上引入链路状态控制因子，依据节点配置比例，通过动态选择策略按照配置优先级为相应节点开通ECMP功能，从而实现网络负载的有效分担；IGA-VECMP算法通过对选择、交叉、变异三种遗传算子的优化改进，有效提升了VECMP模型的求解效率。仿真实验表明，本文提出的IGA-VECMP求解算法能够快速得到网络的最佳权重配置方案，实现网络负载的有效均衡，降低最大链路利用率，提升网络资源的利用效率，在电力通信网络路由优化领域具有广泛的应用价值。
[Abstract]:The profound changes in the form of economy and energy at home and abroad bring new opportunities and challenges to the development of the power grid. In recent years, the construction of the intelligent power grid has been continuously deepened, the characteristics of the information, the automation and the interaction of the modern power grid become more obvious, and the electric power communication network which plays a strong supporting role for the power grid also has a profound change. With the deep penetration of the information communication technology and the power production and operation, the dependence of the power grid on the power communication network is becoming more and more strong, and the reliability and transmission performance of the communication network are also the key factors that affect the safe and stable operation of the power grid. With the increasing complexity of network topology and the gradual increase of service demand, the problem of load imbalance in the power communication network is becoming more and more serious. Some core nodes or important links carry a large amount of service data, and on the one hand, the fragility of the network is increased. and on the other hand, the utilization effect of the network resources is reduced, The traditional optimization strategy is difficult to direct due to the special attributes of the network topology and traffic flow of the power communication. In this paper, according to the topology and traffic flow properties of the power communication network, the paper studies the network topology optimization, traffic flow analysis and prediction and route optimization of the power communication network. The main contents of this paper are as follows: Face: (1) Analysis of Robustness and Vulnerability On the base of the base, a topology optimization algorithm based on the node-number-plus-matrix is proposed, which is mainly aimed at the weak links in the network. the optimization algorithm firstly generates an optimized link set based on the node interface number and the matrix, and then realizes the optimization target by adding a limited number of links, and meanwhile, the repeated node avoidance mechanism is adopted to avoid the over-interference of the network service, The simulation experiment shows that the algorithm can effectively enhance the ability of the network to resist the selective attack, delay the one-time performance of the network, improve the bandwidth utilization efficiency of the key nodes, and reduce the congestion of the key nodes. The probability of the power communication network topology is more specific and more obvious (2) Based on the ARIMA (Autoregressive Integrated Moving Average Model) model, the traffic flow analysis and prediction method, which is suitable for the characteristics of the network traffic of the electric power communication, is put forward, and the traffic flow of the power communication is carried out. Analysis and prediction. According to the ARIMA modeling process and method, the traffic analysis and prediction of the production data service and the video service are mainly carried out, and the traffic flow rate of the power communication service is constructed by the steps of data pre-processing, determining the model parameters, establishing the model and the like. Analysis of the prediction model and the application of the residual test method to the model The simulation experiments show that the flow analysis and prediction method proposed in this paper takes fully into account the change tendency and periodicity of the flow, can meet the traffic characteristics of the power communication service well, and effectively improves the accuracy and the pre-set of the model fitting. (3) A variable-equivalent multi-path model VECMP (VariableEqual-Cost Multi-Path) based on link weight control is proposed, and an improved genetic algorithm is designed to solve the VECM. based on the variable principle of the link weight, the VECMP model introduces a link state control factor on the basis of the traditional ECMP model, The effective sharing of the load is achieved. The IGA-VECMP algorithm effectively improves the VECMP by improving the selection, crossover and mutation of the three genetic operators. The simulation results show that the proposed IGA-VECMP algorithm can quickly get the optimal weight configuration of the network, realize the effective balance of the network load, reduce the maximum link utilization rate, and improve the network. The utilization efficiency of the resources has the advantages that the routing optimization field of the electric power communication network has the advantages of
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN915.853;TM73

【参考文献】