Substation line load prediction method based on deep BiLSTM

Abstract

The invention discloses a substation line load prediction method based on deep BiLSTM, and the method comprises the steps: collecting substation line load data, carrying out data preprocessing, and forming a time series data set; taking the time sequence data set as a model training sample, and constructing a deep BiLSTM model; optimizing the deep BiLSTM model by adopting an improved Adam algorithm, outputting optimal parameters of the model, and carrying out iterative updating training on the deep BiLSTM model; and inputting substation line load data acquired in real time into the trained deep BiLSTM model, and predicting and outputting the load condition of the substation line by the model. According to the invention, the load condition of the substation line can be rapidly and accurately predicted, enough time is reserved for maintenance of the power supply line, and long-term stable and reliable operation of the power supply line can be ensured.

Description

technical field [0001] The invention relates to the technical field of power supply capability prediction, in particular to a substation line load prediction method based on deep BiLSTM. Background technique [0002] In recent years, the increasing demand for residential electricity has put forward higher requirements on the security, stability and reliability of power supply. In order to ensure the long-term stable operation of power supply lines, power supply companies hope to have a set of prediction methods that can quickly and accurately judge whether the substation lines may be overloaded in the future, so that maintenance personnel can intervene in time or early to maintain or repair the lines. . [0003] However, at present, power supply companies mainly rely on the relevant evaluation indicators to judge whether the substation lines are overloaded. By collecting the operation data of the substation lines, the index values ​​of the relevant evaluation indicators are...

AI Technical Summary

Problems solved by technology

However, this evaluation method has a certain lag in the statistics and calculation of data, and it is difficult to evaluate whether the line is overloaded in real time, let alone predict in advance whether the line will be overloaded in the future

In addition, at present, artificial methods are mostly used to evaluate whether the line is overloaded. The artificial evaluation method depends on the evaluation level of the evaluators themselves, which is quite subjective, and the evaluation results are not scientific, objective and accurate enough.

Therefore, most of the time, power supply companies can only arrange maintenance after a line fault actually occurs, which is difficult to meet people's general requirements for long-term power supply stability and reliability

Images