Planning And Optimization Methods Of Smart Grid Commnication Networks

With the development of Smart Grid, the convergence of advanced information communication technology and power network forms the smart grid communication networks to support power grid operating. Compared with the common communication networks, the smart grid communication networks put forward high requirement in economy, reliability, security, etc. To comply with these requirements, planning optimization of the smart grid communication networks is needed, make the network construction basis fully considering various index to better support the smart grid. Therefore, it is very important to study the planning optimization methods of smart grid communication networks.The research on smart grid communication networks includes power backbone communication network, power distribution communication network and power usage communication network. For the power backbone communication network, the index and constraints of existing optical cable planning method are brief and simple, that fail to reflect the power communication networks’demands, therefore, it is necessary to study the new optical cable planning method with consider of the various constraints. For the power distribution communication network, the (n-1)-resilient is the important character and target, but the exsiting gateway location planning method didn’t consider the (n-1)-resilient, therefore, it is necessary to study the new gateway location planning method which includes the (n-1)-resilient to improve the network reliability under the failure condition. Furthermore, in the power distribution communication network, the requirements of network delay of different services can be quite different; however, the existing gateway location planning method didn’t consider the delay constraint, it can’t adjust the end-to-end delay according to the service, so it is necessary to study the new gateway location planning method which considers the delay factor. For the power usage communication network, redundancy optimization for communication nodes is the effective way to reduce damage caused by node failures, the existing redundancy optimization method select the redundancy target node just by the reliability factor and ignores the users’number, that has one-sidedness, therefore, it is necessary to study the new redundancy optimization method which can balance the reliability and users’number to improve the network economy and reliability.In allusion to the problems mentioned above, this paper studies the planning optimization methods of smart grid communication networks based on three scenarios:power backbone communication network, power distribution communication network and power usage communication network, solves the problem of optical cable lines planning, power line communication gateway location planning optimization and data concentrator redundancy optimization, separately. The innovation of research on the following four aspects:(1) In the power backbone communication network, for the shortage of the existing optical cable line planning methods does not consider the station voltage, which leads to large differences between planning model and real networks. This paper proposes a multi-objective based cable path planning method. Firstly, a graph-based power backbone communication network model is developed, which identifies the voltage class by the weight of the nodes and the network reliability function is developed based on station ring content. Then, an optimization model of cable construction with minimum network cost as objective function, the reliability and connectively as constrain conditions is established, and an improved immune algorithm based planning method is proposed to solve the problem. The simulation result shows that the proposed planning method can be combined actual network and provide effective optical cable planning solution.(2) In the power distribution communication network, for the shortage of the existing power line communication (PLC) gateway location planning methods does not consider the (n-1)-resilient, which leads to low reliability under the link failures condition. This paper proposes a new reliability based PLC gateway location planning method. Firstly, the network cost function is developed, which as objective function. Then, to solve the problem of (n-1)-resilient, the proposed method not only consider the reliability under the normal condition, but also consider the one link failures condition,so the (n-1)-resilient can be added into the optimization model. After that, a genetic algorithm based planning method is proposed to solve the problem. The simulation result shows that the proposed PLC gateway location method can balance the economy and reliability.(3) In the PLC network, for the shortage of existing PLC gateway location methods lead to high end-to-end delay and load imbalance. This paper proposes a delay controllable based PLC gateway location optimization method. Firstly, the network cost and reliability function are established. Then the end-to-end delay model is developed using queuing theory, and that is added to the optimization model as a new constraint. The proposed method also optimization and reallocation the routing, and make the end nodes re-election the destination PLC gateway, thus the load balance is realized. The simulation result shows that the proposed optimization method can significant reduce the end-to-end delay, meanwhile, meet the economy and reliability requirement.(4) In the power usage communication network, the data concentrator (DC) failures would cause imbalance between power supply and demand, which leads to huge economic loss. Thus, this paper proposes a redundancy optimization method of DC in Advanced Metering Infrastructure (AMI). Firstly, analyses the characteristics of AMI and Demand Response (DR), and a DC failures damage model is developed using probability theory to quantify the loss caused by DC failures. Then, considering the network reliability, economy and users’number, a DC redundancy optimization model is established. The simulation result shows that the proposed optimization method can provide better reliability and economic than other traditionally redundancy optimization methods.