Key Technology Research Of Power Communication Network Risk Management

As infrastructure of smart grid, power communication is to support service between power sub-systems, such as relay protection, stability control and so on. In order to improve reliability and support safe and stable operation of smart grid, one of important security technologies of smart grid is to study how to reduce operation risk of power communication network. It is importance to research on power communication network risk management technologies, because it can reduce the number and influence of power communication failures and support power communications service channel to be continuously available.The main research object of power communication network risk management is uncertainty of power communication network operation and impact of it. Power communication network risk management typically includes risk assessment and risk control two kinds of technology, which are used to to achieve the qualitative and quantitative analysis of electric power communication network risk, confirm the possibility and severity of all kinds of electric power communication network operation risk, control and solve the potential risk, prevent the occurrence of traffic accidents. Domestic and foreign power grid corp and research institutions in the power communication network risk assessment and risk control field has carried out a lot of research, and achieved some results, but there are still some deficiencies that have not been solved. When we evaluate the risk of power communication networks in the period of year, month, week and day, traditional methods are complex and lack of adaptive. There are few control mechanisms in the existing research results in Network equipment random failure, repair time correlation, unbalanced business distribution and so on, the risk of which has not been effectively controlled. For the above problems, this paper studies risk assessment method and risk control mechanism of power communication network. Specifically, contributions of this paper are shown followed:(1) To meet requirements of adaptive capacity of power communication network risk assessment and risk early warning mechanism in year, month, week and day, a risk assessment method for power communication network based on the fuzzy radial basis function neural network is proposed. In this paper, an adaptive learning rule of hidden layer nodes of neural network is defined on the basis of the hidden layer nodes of the Chebyshev distance to dynamically adjust number of hidden nodes, which makes the fuzzy radial basis function neural network can track the changes of business parameters and risk factors in the learning process and dynamically adjust its structure, so that the network risk can be more accurately quantified.(2) For the risk of service channel protect-ring faults caused by randomness of resource failures and repair time, based on network equipment and line fault arrival rate and repair time distribution function, a mathematical model for measuring violation risk of power communication service channel is established. The violation risk of power communication service channel is used as constraint condition for power communication service channel routing algorithm to control risk of service channel protect-ring faults caused by random failure. According to probability generating function and moment generating function, the mathematical model of violation risk of power communication channel is simplified to reduce its computational complexity and truncation error and make it suitable for actual routing problem. Simulation experiments were carried out using actual data of power communication network and results show that the proposed algorithm can effectively reduce risk caused by service channel protect-ring faults.(3) To solve the problem of risk caused by service channel protect-ring faults in power communication network, a routing algorithm for working channel and protect channel of power communication service is proposed. Firstly, fault correlation model for power communication service channel is established by correlation distance between repair time of equipment and cable. Subsequently, violation risk of power communication channel is used as decision-making parameters and I_Dijkstra algorithm is adopted to search k uncorrelated and minimizing correlation degree routes as candidate routes. Finally, service channel routes of minimal fault incidence are treated as the decision-making target to select working path and protect path for service channel. Simulation results show that this algorithm can simultaneously reduce the probability of working channel and protection channel, thus reduce risk of power communication.(4) For the problem of increasing operation risk of power network caused by power communication services centralized on a few paths, a routing algorithm based on risk balance is proposed. This algorithm based on approximate ideal solution sort method, according to availability, switching delay, error rate and channel protection index of power communication service, a model for evaluating importance degree of power communication service is established to calculate service importance. A mathematical model for measuring risk of power communication channel sections is built, which based on importance of power communication service and violation probability of power communication channel. Taking the risk of power communication channel path as the constraint, the service channel route is chosen according to the principle of MIN-MAX. During selection process, risk of channel path is updated according to risk update function after each operation. Ssimulation results show that this algorithm can make power communication communication service evenly distributed in the network, and avoid important service to convergence to a few routes.