Generative adversarial network-based power load data generation method

Abstract

The invention discloses a power load data generation method based on a generative adversarial network, and the method comprises the steps: building a generative adversarial network model, taking power load data as an input sample of the model, obtaining generated data which is difficult to distinguish from the authenticity of the input sample after the model is sufficiently trained, and enabling the input data of the generated model to be noise data, the output is generated data, and the output is a judgment result of the two kinds of data; and finally, judging that the model cannot accurately distinguish the authenticity of data generated by the model. A generative adversarial network model method is used for generating load data, the capability of the method in the aspects of retaining original data sample space-time law characteristics and the like is verified through experiments, the feasibility of the generative adversarial network method in the aspect of generating the load data is proved, and a large amount of load data meeting requirements can be generated; meanwhile, compared with a traditional scene modeling method, the generative adversarial network can better reflect the superiority of an unsupervised neural network, the influence of human factors can be reduced, the training time is shortened, and the cost is reduced.

Description

technical field [0001] The present invention specifically relates to a method for generating data, in particular to a method for generating power load data based on a generative adversarial network. Background technique [0002] Energy is an important material basis for the survival and development of human society. Facing the pressure of energy supply, we should adhere to the principle of giving equal attention to development and conservation, and giving priority to conservation. The party and the government attach great importance to energy conservation. Power load monitoring is the basis for energy-saving work. It can use energy units to fully grasp the energy consumption and usage of each equipment, and use this to formulate a reasonable energy-saving plan and purchase energy-saving equipment in a targeted manner. At the same time, it can also use power load monitoring to test the energy-saving effect, thereby reducing energy consumption and energy expenditure. Theref...

AI Technical Summary

Problems solved by technology

The data collection method is that the experimenter enters the load user and collects the load data using the electric meter. However, data collection is a waste of financial, material, manpower and time. In addition, when the load types are complete enough In this situation, it is difficult to list the multi-load data in different scenarios, and a large amount of data of different lengths is also required, which brings great difficulty to the data collection work.

Images