Voltage Overrun Evaluation and Prediction Research Based on Combinatorial Weighting and CNN-GRU

Aiming at the problem of assessing the severity of voltage overrun in distribution networks, a voltage overrun assessment and prediction and warning method based on combinatorial assignment and the CNN-GRU model is proposed. Firstly, a new type of voltage overrun assessment index considering the influence of user density is selected, the subjective and objective weights are optimized and coordinated based on the theory of combinatorial assignment, and the regional division of monitoring points is used to give appropriate weight correction from the perspective of voltage propagation to establish the voltage deviation severity assessment level and determine the voltage deviation severity degree of assessment samples; the data features are extracted based on the assessment results after the overrun assessment of the data; then the data features are extracted through the gating network (CNN), and the data features are extracted through the gating network (CNN). (CNN) for the extraction of data features, and then through the gated recirculation unit (GRU) for the prediction and judgment of the voltage overrun problem in the next time period and timely warning; finally, some of the measured voltage data in a certain area are analyzed as an example to verify the effectiveness of the assessment level system and prediction and warning model