Power system fast state estimation method based on deep learning

Abstract

The invention provides a power system rapid state estimation method based on deep learning, and belongs to the technical field of power system monitoring, analysis and control. According to the technical scheme, the power system fast state estimation method based on deep learning comprises the steps that a DNN network is selected as a deep learning model, feature input is selected through a correlation analysis method, and the robust capacity of the model to bad data is improved through a noise network. The technical problems that a traditional state estimation method based on a physical modelis low in calculation efficiency, low in operation speed and too complex in grid structure and even possibly causes poor state estimation convergence and stability are solved. The method has the advantages that the calculation speed is obviously improved compared with a traditional estimation method, and the estimation precision and robustness for measuring bad data of the method are also greatlyimproved compared with the traditional estimation method.

Description

technical field [0001] The invention relates to the field of power system monitoring, analysis and control and technology, in particular to a method for fast state estimation of power systems based on deep learning. Background technique [0002] State estimation is the foundation and core part of the energy management system. It obtains the optimal estimated value of the operating state of the power system by processing the measured data. The state estimation method actually used in the current power system is still weighted least squares (Weighted Least Square, WLS) state estimation. Due to the high measurement redundancy of the power system, WLS estimation uses the weighted least squares estimation criterion to estimate the state of the power system in real time through linearized measurement equations and Newton iteration method. WLS estimation is a uniform unbiased minimum variance estimation under the ideal condition that the system measurement noise only contains Gaus...

AI Technical Summary

Problems solved by technology

Due to the introduction of non-quadratic estimation criteria, the solution model of WLAV estimation is complex, the calculation time is long, and it is difficult to meet the needs of actual engineering due to the limitation of node scale and computer performance.

Although there are improved algorithms, such as the bilinear method, the problem of robust state estimation in practical applications has not been fundamentally resolved

[0004] With the rapid expansion of the scale of the power grid, the amount of data to be processed has greatly increased. The traditional state estimation method based on the physical model will have the problems of low calculation efficiency and slow operation speed. An overly complex grid structure may even lead to the convergence of state estimation. and poor stability, which poses a severe challenge to the power system state estimation problem

Images