Power Grid Load Forecasting Method and System Based on Neural Network and Dynamic Mode Decomposition
A technology of power grid load and dynamic mode, applied in the direction of biological neural network model, neural learning method, prediction, etc., can solve the problems of poor prediction accuracy of power grid load data, failure to consider the fluctuation of power grid load, low robustness, etc., and achieve cost Low, predictive accuracy and robustness, strong implementation effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0056] The application environment of the embodiment of the present invention: different weather, seasonal conditions, holidays, work cycles, and economic fluctuations, etc., the data generated by these factors have subtle time patterns, which will lead to deviations in grid load forecasting, making the forecasting results accurate The degree is not high and the stability is low.
[0057] Such as figure 1 As shown, this embodiment provides a grid load forecasting method based on neural network and dynamic mode decomposition, which specifically includes the following steps:
[0058] Step 1: Obtain the original power grid load data through observation data.
[0059] Step 2: According to the Cover theorem, the original grid load data is expressed in a linear model in the delayed coordinate space, and the sliding window sampling is performed on the original grid load data in the delayed coordinate space to construct a sliding window matrix. The specific construction method is as...
Embodiment 2
[0120] Such as figure 2 As shown, this embodiment provides a grid load forecasting system based on neural network and dynamic mode decomposition, the system includes a raw data acquisition module, a raw data processing module, a sliding window matrix building block, a grid load linear model building block, and a singular index Time series and multifractal spectrum time series building block, time series forecasting neural network training module, power grid load forecasting model building block and power grid load data forecasting module.
[0121] The original data acquisition module is used to obtain the original grid load data.
[0122] The original data processing module is used to transform the original grid load data into high-dimensional space to obtain high-dimensional grid load data.
[0123] The sliding window matrix construction module is used to perform sliding window sampling on high-dimensional power grid load data to construct a sliding window matrix;
[0124]...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More