Line dynamic capacity increasing method based on multi-source data fusion

Abstract

The invention discloses a line dynamic capacity increasing method based on multi-source data fusion, which has the effects that in a neural network structure model based on a probability sparse self-attention mechanism, multi-source information such as feature sequence information, hierarchical time sequence information and the like is fused, a neural network embedded layer is established, and the variables are utilized for modeling, so that the probability sparse self-attention mechanism is established. The method is used for extracting a period rule of power transmission line capacity change, realizing long-time sequence prediction based on a historical sequence, and performing reasonable pre-judgment on a future working condition change trend. The method has the advantages that prediction and judgment in a period of time in the future can be achieved, the comprehensive change condition of a certain line or multiple lines in the period of time in the future can be known through judgment in a period of time, and the rated load of the power transmission line can be better and more reasonably allocated in advance.

Description

technical field [0001] The invention belongs to the technical field of power systems, and relates to a dynamic capacity increase method for high-voltage lines, in particular to a deep learning modeling method for line dynamic capacity increase based on multi-source data fusion. Background technique [0002] At present, the dynamic capacity expansion technology of transmission lines is relatively mature. It mainly monitors the transmission lines and environmental conditions. On the premise of not breaking the current technical regulations, the maximum allowable current carrying capacity of the wire is calculated according to the Morgan Equal-Ampacity Model. The actual transmission capacity of the transmission line is improved by utilizing the hidden capacity that exists objectively in the line. The parameters monitored in real time include the temperature, tension, sag, temperature, sunshine, customs and other parameters of the conductor, and the maximum allowable current car...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, single working condition prediction is mostly used in the process of dynamic capacity increase of transmission lines. For example, the patent with publication number CN110321601A discloses a method and system for advanced prediction of dynamic current-carrying capacity of overhead lines. It introduces attention mechanism The cyclic neural network is used to make predictions. Such a result can only predict one point or several points in the future, such as predicting a line point in the next 15 minutes, and the capacity required for this line at that time node. This kind of prediction iterative learning

For the era of multi-source power data fusion, it is obvious that such a prediction cannot achieve the purpose of prediction.

Images