Electric quantity prediction method based on industrial chain product output

Abstract

The present invention relates to a method of electricity forecasting based on the output of industrial chain products, comprising the following steps: 1) Screening industrial chain indicators, and extracting index data, said industrial chain indicators include social overall economic development level indicators and major industry economic development levels 2) Construct the industrial chain index database; 3) Extrapolate the power forecasting model through the above index data and the industrial chain index database. The beneficial effects are as follows: firstly, the present invention comprehensively considers the relevance of various related industries and their industrial chains, screens key indicators that have an impact on the electricity market, identifies deterministic factors such as product output of the industry, and finally reflects the output of related products in the industrial chain. The trend of future power consumption is forecasted.

Description

technical field [0001] The invention belongs to the field of electric power, and in particular relates to a power forecasting method based on product output of an industrial chain. Background technique [0002] At present, there are many methods for forecasting power system electricity at home and abroad, and there are mainly two types: traditional methods, including time series analysis, regression analysis, gray forecasting methods, etc.; emerging methods, including fuzzy forecasting, expert systems, etc. artificial neural network and other artificial intelligence methods, as well as emerging wavelet analysis methods and combined forecasting methods. [0003] Time series methods can model loads based on historical data and use the model to predict future electricity. Its advantages are: less historical data is required, and the workload is small; the disadvantage is: it does not consider the factors of power changes, only focuses on data fitting, and does not deal with re...

AI Technical Summary

Problems solved by technology

Its advantages are: less historical data is required, and the workload is small; the disadvantage is: it does not consider the factors of power changes, only focuses on data fitting, and does not deal with regularity enough, and is only suitable for short-term predictions with relatively uniform power changes Happening

The advantages are: less load data required for modeling, no consideration of distribution laws, no consideration of changing trends, convenient calculation, high short-term prediction accuracy, and easy inspection; The worse the accuracy

Its disadvantages are: limited by the total amount of knowledge accumulated in the database, poor adaptability to emergencies and changing conditions, it is difficult to obtain definite long-term forecast results

The artificial neural network method has inherent defects such as difficulty in scientifically determining the network structure, slow learning speed, local minimum points, and unstable memory.

Images