Online dynamic security assessment method based on mutual information and iterated random forest

Abstract

The invention provides an online dynamic security assessment method based on mutual information and iterated random forest, which comprises the steps of creating a data set containing a large number of operation variables and transient stability margins (TSMs) based on historical operation data of the power system and simulation of a series of faults; carrying out feature selection on the data setby using partial mutual information (PMI) to screen out key variables; sending the key variables to an iterated random forest (IRF) regressor for training; and immediately providing a real-time assessment result by a trained dynamic security assessment (DSA) model by receiving the selected variables in real time from a wide area measurement system (WAMS) server. The online DSA model of the powersystem provided by the invention can provide rapid and efficient assessment for the power grid, thereby facilitating the system maintenance and security measure prevention work of the power personnel,and having important significance for the operation security and quality of the power grid.

Description

technical field [0001] The invention belongs to the field of power system safety analysis and evaluation, and in particular relates to a mutual information tool and a dynamic safety evaluation method of a random forest algorithm. Background technique [0002] With the continuous expansion of the scale of the power system and the continuous access of new equipment, the safe operation of the power system is facing unprecedented challenges. Quickly and accurately assessing the security status of power systems is one of the key concerns of researchers. Dynamic Security Assessment (Dynamic Security Assessment, DSA) is of great significance for maintaining the safe and stable operation of the system. However, due to the large scale of the actual power system and various operating conditions, online DSA is still a challenging task. For example, the Chinese patent application publication number CN104657915A discloses a dynamic adaptive power system terminal security threat assessme...

AI Technical Summary

Problems solved by technology

[0005] However, there are still some shortcomings in the existing methods, as follows: ①Due to the continuous expansion of the scale of modern power systems, traditional tools are often difficult to train large-scale samples, and are relatively insufficient in terms of computational efficiency and accuracy

In addition, it is difficult to deal with the problem of data loss and easy to overfitting, which cannot be ignored in these commonly used tools; ② Some commonly used mutual information tools often have more or less overestimation and underestimation problems when performing correlation detection. The selected features are not accurate enough; ③ Due to some unexpected situations often appear in the actual power system operation, the traditional DSA model is difficult to adapt to the new operating state of the system and it is difficult to evaluate the new state

Images