Big data based power load prediction method

Abstract

The invention discloses a big data based power load prediction method. The method comprises the steps of step one, providing data information of N periods, obtaining a first power load predictive value of the (N+1) periods through a reinforcement learning load prediction data model directed at same data information and obtaining a second power load predictive value of the (N+1) periods in a data driving mode; step two, performing information fusion on the first power load predictive value and the second power load predictive value through a D-S evidence theory to obtain a final predictive result of the (N+1) periods. By the aid of the method, directed at a power load prediction system containing multiple dimensions and multiple stages of space, time, attributes and the like, a data driving theory based non-model load prediction controller and wavelet neural network based accumulative learning prediction are combined, information fusion is performed on the predictive values through the information fusion technology to obtain an optimal predictive value, and accordingly, the accuracy and the timeliness of load prediction are improved greatly.

Description

technical field [0001] The invention relates to a large data-based power load forecasting method. Background technique [0002] As far as the current situation in our country is concerned, the current situation of load characteristics, the main factors affecting load characteristics and the changing trend of future load characteristics have been systematically analyzed and studied, and some conclusions that have guiding significance for power planning and power grid operation and The conclusions of reference value, but these studies are more at the theoretical level, not enough to do enough data deduction and verification, and its main shortcomings are as follows: [0003] 1. Dimensional limitations of factors affecting load. In the existing load forecasting system, the forecasting factors are mainly limited to common data such as historical load and meteorological data. [0004] 2. Insufficient analysis of the internal law of meteorological elements and loads. The impact...

AI Technical Summary

Problems solved by technology

[0002] As far as my country's current situation is concerned, the current situation of load characteristics, the main factors affecting load characteristics and the changing trend of future load characteristics have been systematically analyzed and studied, and some conclusions that have guiding significance for power planning and grid operation have been obtained. The conclusions of reference value, but these studies are more at the theoretical level, and there is not enough data deduction and verification. The main shortcomings are:

[0003] 1. Dimensional limitations of factors affecting load

In the existing load forecasting system, the forecasting factors are mainly limited to common data such as historical load and meteorological data.

[0004] 2. Insufficient analysis of the internal laws of meteorological elements and loads

The influence model of meteorological factors established by the existing system can not fully reflect the real change of load, and the work done in aspects such as the cumulative effect of temperature and delay effect is limited, and the depth of research needs to be further deepened

[0005] 3. The load characteristic index is a time-point index, and the load characteristics of different regions and different times cannot be directly superimposed, which makes it more difficult to analyze the load characteristics of a large area; and the typical load characteristic curves of various industries and the non-grid unified load characteristic curves It is difficult to obtain and process

[0006] 4. Lack of research on the similarity between ground load regulation and provincial load regulation

[0007] 5. Lack of analysis of relevant factors affecting load

Limited to past conditions, in most systems, the influence of weather and other factors on load is not considered, or only limited meteorological information (maximum, minimum and average temperature) is used, and the prediction accuracy is not high

Images