Power system transient stability situation assessment method based on random matrix theory

Abstract

The invention discloses a power system transient stability situation assessment method based on random matrix theory, comprising the following steps: 1. Selecting power network state quantity data andconstructing system state matrix according to splicing matrix method; 2, according to the state matrix constructed in the step 1, combining the real-time sliding time window to construct a random matrix; 3, calculating that average spectral radius kVS of the random matrix obtain in the step 2 according to the random matrix theory; 4, if that kVS (t) is less than beta; kVS, N, the system is in anabnormal state; Otherwise in a normal state; If the system is in an abnormal state, judging whether the kVS value decreases continuously with time, and if the kVS value decreases continuously and reaches 80% of Beta- kVS, the transient instability of the system is judged. The invention considers the power network history and real-time state quantity data, can judge whether the system has transientinstability, and the analysis process is definite and applicable, which is convenient for practical engineering use.

Description

technical field [0001] The invention relates to a method for assessing a transient stability situation of a power system, in particular to a method for evaluating a transient stability situation of a power system based on a random matrix theory. Background technique [0002] At present, the traditional transient stability assessment relies on model parameter selection and modeling process, which cannot make full use of high-dimensional operational data with spatio-temporal characteristics, and large-scale new energy grid connection increases the randomness of the power side; leading to the possibility of traditional research assumptions No, there is currently no data-driven transient stability assessment method for power systems. [0003] For example, a static stability situation assessment method based on random matrix theory is proposed in "Static Stability Assessment Method of Power Grid Based on Random Matrix Theory"; a method based on random matrix theory is proposed in...

AI Technical Summary

Problems solved by technology

[0002] At present, the traditional transient stability assessment relies on model parameter selection and modeling process, which cannot make full use of high-dimensional operational data with spatio-temporal characteristics, and large-scale new energy grid connection increases the randomness of the power side; leading to the possibility of traditional research assumptions No, there is currently no data-driven transient stability assessment method for power systems

[0003] For example, a static stability situation assessment method based on random matrix theory is proposed in "Static Stability Assessment Method of Power Grid Based on Random Matrix Theory"; a method based on random matrix theory is proposed in "Transient Stability Analysis of Power System Based on Random Matrix Theory". Transient stability index, but it can only analyze the transient stability macroscopically, and fails to further quantitatively evaluate the stability situation in combination with the actual data situation; "Distribution Network Situation Awareness Method Based on Historical Data Mining" proposes a method based on historical data modeling Distribution network situation awareness method for real-time state estimation, but only applicable to low-voltage distribution network; "Transient stability prediction model based on characteristic ellipsoid and support vector machine compound mapping" proposes a method based on characteristic ellipsoid and support vector A Transient Stability Prediction Model Based on Machine Composite Mapping

Traditional transient stability analysis usually adopts a model-based analysis method, and decouples the actual system according to some assumptions and simplified conditions. The results obtained depend on the model parameter selection and modeling process, unreasonable or inaccurate formula expressions, and biased parameters All choices will lead to poor results and low applicability; the existing transient stability analysis methods mainly focus on the judgment of transient stability in the current operating state of the system, without considering the operating trend of the system to the future state, the analysis level Single, unable to identify system security and stability based on a global perspective; existing data-driven analysis methods are only for distribution network situation awareness or static stability situation assessment

Images