Research On Security Situation Awareness Of Large Power Grid Based On Deep Neural Network

With the rapid development of Chinese power industry,China has entered an era of "large power grid",the scale of the power grid is continuously expanding,and the national interconnected power grid has also been bulit.However,blackouts often happen due to the weak power grid structure,the high failure rate of electrical equipment and lines and other factors;and with the development of new technologies such as wind power,photovoltaics,and DC,the power supply and load characteristics have undergone tremendous changes,which has increased the uncertainty and complexity of the grid’s operating characteristics and increased the difficulty of power system characteristics recognition,operation scheduling and accident prevention,so security situation awareness of large power grid becomes more and more important timely and effectively;deep neural networks have been widely used in various fields in recent years,and it show huge advantages in data processing,data prediction and other aspects.Based on this,this paper introduces the deep neural network into the research of security situation awareness of large power grid,which is used to evaluate and predict the safe operation state of large power grid.The situation awareness of the large power grid mainly includes the extraction of situational elements,situational understanding and situational prediction.Based on this,the main work of this paper is as follows:Firstly,by extracting the factors that affect the safety situation of the large power grid,the internal factors that affect the safety situation of the large power grid are collected from the aspects of lines,loads,transformers and the system as a whole,which are used to reflect the static safety,transient safety,voltage safety and Structural safety;and by analyzing blackouts in the national power grid from 1981 to 2015,the indicators of the external factor of affecting the security situation of the large power grid were analyzed,and the degree of influence of the indicators of the external factor on the security situation of the large power grid was analyzed,and then a relatively complete assessment system of safety situation for large power grid was constructed.Then,in the stage of situational understanding,through the analysis and understanding of elements of the safety situation of the large power grid,the comprehensive evaluation of the safety situation of the large power grid is completed,which mainly includes the determination of the index weight and the calculation of the evaluation value.In terms of the determination of the weight,the analytic hierarchy process and the improved entropy weight method are used to obtain the comprehensive weight of each index through weighting method of the game theory,and the weighted sum method is used to obtain the safety situation assessment value of the large power grid;and the IEEE39 node system is used as an example to illustrate the specific process of calculating the assessment value of the safety situation of the power grid to lay the foundation for the subsequent assessment and prediction.Finally,in order to verify the feasibility and accuracy of the deep neural network’s assessment and predition of the safety situation of the large power grid,three models of BP neural network,RBF neural network and deep neural network were constructed to completed the assessment and predition of the security situation of the large power grid.