PMU Optimization Configuration Based On The 0-1 Integer Programming

With the expansion of the scale of power system, any kind of power accident may bring great impact on all aspects of society as a whole. Therefore, online real-time monitoring of power system is in great need. At the same time, Synchronous phasor measurement(PMU) mature technology developments meet the demands of monitoring the power system. However, the price of PMU is way too expensive. It can achieve the goal of real-time monitoring power system when all nodes in the entire network deploy PMU while it could have greatly increased the one-time investment of power grid. So we need to consider the optimization of PMU device configuration. The configuration of PMU requires the least PMUs under the premise of complete network observability. The 0-1 mixed integer programming formulation has a great advantage of solving the multi-objective PMU placement problem. However, there are plenty of zero power injection nodes in power system. It's advantage is that it can reduce the configuration number of PMUs, but at the same time, when it is introduced to the integer programming calculation, it will model nonlinear and difficult to solve. The traditional methods have different levels of problems when solving the nonlinear constraints.The paper revise the constraints which contains zero injection node, and a new condition function inequality are given, it can well solve the nonlinear problem. At the same time, we can easily get the new function constraint on the basis of the original adjacency matrix, it has a strong operability. When considering the case of N-1 PMU optimal configuration, the proposed method has strong ability of inheritance, it can effectively solve the problem of N-1 optimized configuration when the traditional method can't. Then, in the multi-stages of global observability configuration, the article deploy the PMU configuration with the target of reducing the unobservability, put forward to increase the virtual connection scheme to modify topology and condition function. It can effectively apply to the 0-1 linear integer programming. In the multi-stages of N-1 configuration process, the article deploy the PMU configuration with the target of key line observability. Different from the case of traditional literature's demanding key line' observability under condition that one line is lost, the article achieve that key line is observed when one PMU is lost.Finally, the article do the PMU optimal configuration in both the IEEE 14 nodes system and the IEEE 39 nodes system, the obtained solution is better compared with the traditional algorithm. It can reduced the number of the PMU when guarantee right configuration results quantity.