Emergency Control Of Transient Stability Of Power System Based On LSTM

As an important guarantee of modern economic development and social progress,the safety and reliability analysis of electric energy has been highly valued by scientific research workers.In recent years,with the expansion of the interconnection scope of the power grid,the operating environment of the power grid is becoming more complex and changeable,and many power outages have exposed the shortcomings of the traditional power system security and stability control system.The rapid and accurate emergency control of power system transient stability is of great significance to stabilize the second line of defense of power system,prevent the stability of power system from being destroyed and even cause the breakdown of power grid.Most of the traditional power system transient stability control methods quantitatively evaluate the stability margin of the measured curves of phasor measuring devices based on the extended isometric criterion.The effectiveness of this kind of method depends largely on the accuracy of fast prediction of disturbed feature trajectory,and it is difficult to match the data of power grid with multi-source heterogeneous characteristics.In recent years,the growing artificial intelligence method is active again in every field of power grid construction with its powerful self-learning ability.Artificial intelligence method can effectively get rid of the shackles of power grid model,take multi-source heterogeneous power grid dynamic data as input,establish the accurate mapping relationship between many influencing factors and system state in transient problems,and provide new ideas for improving economic level and ensuring local electricity safety.In the research of power system transient stability emergency control based on response information,how to quickly and accurately identify the disturbed system homo-modulator group and the ultra-real-time prediction of the iso system trajectory information is of great significance to the optimization of transient stability emergency control strategy.(1)Based on the concept of motion similarity of voltage phase trajectory,a method for rapid identification of generator synchronicity is proposed,and a characteristic plane can be constructed to characterize the differential degree of motion of voltage phase trajectory,so as to improve the timeliness of identity identification of homodone group under the premise of ensuring accuracy.(2)Given the anticipated faults,a long-short term memory network is constructed to quickly predict the disturbed trajectory of the isoline system,and the input data is processed in 3D to ensure that the prediction model can fully excavate the evolution law of the characteristic disturbed trajectory in each time section,and effectively improve the prediction accuracy of the model.(3)In order to ensure the reliable return stability of the system and improve the effectiveness of the emergency control system action,based on the concept of transient deviation energy,the cycle monitoring function is proposed to continuously monitor the stability of the system with the difference between the relative dynamic energy of the target system and the remaining deceleration area after the cutting machine action,so as to ensure the stable operation of the system.A IEEE-39 system example is used to verify the effectiveness of the proposed method.It is very important to strengthen the level of safety prevention and control of power system only through the information of disturbed trajectory for a short time after failure,and the quantity of cutting machine control measures is given.