Guo Yajuan

Research on information security testing technology for smart Substations

As a critical infrastructure, smart substation is one of typical industrial control systems, whose information security evaluation has been one of the most pressing issues. As much as a system is critical, chances of attack from outer or from inner are more. The attack could be on hardware device, software solution and on integrated database with in substation. This paper introduces some experiences from the practices of smart substations information security testing in Jiangsu Electric Power Company Research Institute (JSEPRI). We first give information security threats faced by ICS in the filed of smart grid, analyze the requirements of the information security testing system construction, proposes a method to build a testbed for substations, and describes several key technologies involved in the testbed.

Research and application of load balance approach for distribution based on big data

For solving the common existence of feeder overload problem, a load balance approach based on big data is proposed in this paper. Because of the dynamic characteristics of distribution load, a feeder overload quantitative model is raised, while multi data sources are employed, such as total contents of feeder/distribution transformer load, grid fault information, business expanding plan, work/operation tickets etc. The approach changes the open/closed status of sectionalizing and tie switches of the overload feeder and its adjacent counterparts, calculates the comprehensive overload index of all the possible operation modes and chooses the most appropriate strategy. Example shows the effectiveness of this approach. This approach has been actually applied for the decision-making of Jiangsu distribution dispatching.

An implementation of complex alert rules for online monitored equipment in the power grid

Smart Grid is the future direction of the power grid development. As one of the Smart Grid key issues, online monitoring of power equipment is of increasing importance because it can effectively collect, integrate and analyze real-time data from the power grid and provide accurate operating status of equipment to help operators making precise decisions. The alert generation scheme is crucial for the online system to ensure the power grids steady operation. However, most of existing alert logics are hard coded into the front-end monitoring devices and the threshold of alert is difficult to be customizes based on individual variability of equipment. A high extensible implementation method of alert logic is proposed in this paper, by which professional equipment status evaluators can flexibly customize the thresholds. Time and space efficiency is obviously enhanced comparing to the method that directly taking individual equipment as unit to calculate and store alert thresholds. The proposed method took only 16.22% of calculation time and 27.16% of storage space for the alert information and extensively reduced the false alert rate of monitored equipment in the power grid from 29.67% to 12.82%.