IES incomplete data load prediction method and system based on C-GAN transfer learning

Abstract

The invention provides an IES incomplete data load prediction method and system based on C-GAN transfer learning. The method comprises the following steps: firstly, collecting original sample data andnormalizing the data; secondly, extracting sample features of the normalized sample data by adopting a depth variation self-encoding network; inputting the extracted sample features into a constructed first C-GAN generator; when a game of the generator and a discriminator reaches Nash equilibrium, expanding the incomplete sample data; inputting the expanded sample data set into a constructed generator of a second condition C-GAN; when the game of the generator and the discriminator reaches Nash equilibrium, predicting electricity, gas and heat loads in parallel; judging the prediction precision based on the C-GAN discriminator, continuously correcting and improving the prediction precision of comprehensive energy load prediction in the process that the generator and the discriminator playa game to achieve Nash equilibrium. The prediction system provided by the invention is used for load prediction, parameters required by network training are reduced, and meanwhile, the prediction time is shortened.

Description

technical field [0001] The invention relates to the technical fields of comprehensive energy load forecasting and artificial intelligence, in particular to a method and system for IES incomplete data load forecasting based on C-GAN migration learning. Background technique [0002] At present, the scale of my country's integrated energy system (referred to as IES) continues to expand, and the real-time scheduling of integrated energy systems has become a top priority. Improving the speed and accuracy of load forecasting in the integrated energy system plays a vital role in realizing the real-time scheduling and optimal operation of the integrated energy system. However, in practical applications, sample collection plays a very important role in comprehensive energy load forecasting. Due to power outages, sample collectors not working, etc., the data collection of comprehensive energy systems is incomplete. have a non-negligible impact. [0003] Among the existing load forec...

AI Technical Summary

Problems solved by technology

However, in practical applications, sample collection plays a very important role in comprehensive energy load forecasting. Due to power outages, sample collectors not working, etc., the comprehensive energy system data collection is incomplete. have a non-negligible impact

[0003] Among the existing integrated energy system load forecasting methods, the algorithm with the advantage of forecasting accuracy has insufficient speed, and the direct method based on energy function, which has better calculation speed and accuracy, cannot be applied to complex systems

Moreover, the existing comprehensive energy load forecasting does not take into account the situation of incomplete data

In recent years, machine learning methods have been applied to load forecasting problems, such as artificial neural networks, support vector machines and other methods, and have made great progress. However, due to their weak feature learning ability, the prediction accuracy is difficult to guarantee.

Deep learning methods have also been introduced into this field, such as deep belief networks and long-short-term memory, etc., but there are still deficiencies in the processing of samples and the precise application of deep learning networks.

The current integrated energy system load does not prevent local minimum points to ensure the uniqueness of the network solution

Images