Voltage sag risk early warning method based on fusion of simulation and multi-source measured data

Abstract

The present invention proposes a voltage sag risk early warning method based on the fusion of simulation and multi-source measured data, which includes the following steps: Step S1, select simulation data and measured data with complementary information correlation, and form a support for voltage sag risk early warning after preprocessing data; step S2, on the basis of the supporting data, introduce sub-item indicators, and obtain comprehensive quantitative indicators of the influence domain after co-chemotropy processing, feature extraction, dimension reduction and fusion; step S3, design and train multiple regression based on the improved gradient descent method Model, the model update strategy includes the knowledge transfer between simulation and measured data, and the step size update strategy is improved to an adaptive step size update based on the Armijo-Goldstein criterion; step S4, mining the support data with a multiple regression model, predicting the residual voltage amplitude and according to it. The voltage sag risk early warning can be implemented by combining different user needs and sag tolerance characteristics into early warning levels; the invention can integrate simulation and measured data information from the power grid and the user side to realize sag risk early warning.

Description

technical field [0001] The invention relates to the technical field of power quality and power grid operation and maintenance, in particular to a voltage sag risk early warning method based on the fusion of simulation and multi-source measured data. Background technique [0002] With the improvement of the industrial level and the continuous expansion of the power grid, the equipment tends to be integrated and sophisticated, and people's requirements for power quality are getting higher and higher. Voltage sag is one of the most serious power quality problems. Every year, complaints caused by voltage sag events exceed half of power quality problems. The economic losses caused by voltage sag problems are also huge. [0003] In order to reduce the impact of the voltage sag problem, accurate early warning of the voltage sag risk is crucial. Accurate voltage sag risk warning can send early warning information to the power grid and users in advance. The power grid and users can ...

AI Technical Summary

Problems solved by technology

[0004] At present, there are few studies on the early warning of voltage sag risk. In the past, most of them were carried out through simulation analysis. The method based on simulation analysis mainly obtains the sag and level of the entire network through fault calculation. In today's complex large power grid environment, new energy , Distributed power generation is developing rapidly, there are many system components, complex parameters, large amount of fault simulation, cumbersome calculation process, long simulation time, low calculation efficiency, and it is difficult to reflect the influence of external factors (such as weather) in actual scenarios, and most of them only consider The voltage sag caused by the fault is ignored, the influence of other sag causes is ignored, and the consideration of the sag effect is not comprehensive enough

However, limited by the cost of monitoring devices, the number of monitoring devices is small, the monitoring data samples are insufficient, and the data fidelity is not high, which affects the convergence speed, convergence and accuracy of the model using measured data.

Images