Enterprise monthly maximum demand prediction method

Abstract

The invention discloses an enterprise monthly maximum demand prediction method which is used for reducing the maximum demand prediction error problem. The method comprises the following steps: step 1,obtaining enterprise related historical data, and obtaining a monthly maximum load value and an average temperature value of a reference month; step 2, adopting a maximum load characteristic intervalanalysis method, calculating the maximum load value of each part of the monthly peak, average and valley and the probability value of the maximum load day number appearing in the monthly peak, average and valley in the total days, obtaining the monthly maximum load characteristic value according to the maximum load value of the monthly peak, average and valley and the corresponding probability value, and setting a weighting sequence according to the near-large-far-small setting to obtain a historical data prediction value; step 3, classifying factors influencing enterprise load fluctuation byadopting a load characteristic coefficient classification method, obtaining a predicted monthly average temperature and a maximum load estimated value, and obtaining different characteristic functions according to different working conditions; and step 4, multiplying the characteristic function by the historical data predicted value to obtain a demand predicted value, and further giving a declaration suggestion.

Description

technical field [0001] The invention relates to the technical field of electric power forecasting, in particular to a method for forecasting the maximum monthly demand of an enterprise. Background technique [0002] According to the "Interim Measures for the Administration of Sales Electricity Prices", in the two-part electricity price adopted by large electricity users, there are two charging methods for the basic electricity fee, charging according to the transformer capacity and charging according to the maximum demand. According to the requirements of the maximum demand policy, double penalties will be imposed after exceeding 105% of the set value of the contract demand. When the approved maximum demand value of the application is lower than 40% of the sum of transformer capacity and high-voltage motor capacity, it will be 40% of the total capacity Approved contract maximum demand. The transformer capacity of electricity users is generally fixed, and its capacity electr...

AI Technical Summary

Problems solved by technology

[0003] After searching the existing technical literature, it is found that there is a method for predicting the maximum monthly demand of electricity consumption of enterprises, patent number 201410074421.9. Meteorological forecast average temperature value, monthly meteorological forecast sunshine intensity value, and reference monthly average temperature value, monthly reference value, etc., but this method does not consider the difference and variability between actual demand and historical demand value, and the load characteristic function There is no classification processing, so there is a certain error in the prediction of this method

A method and device for determining the maximum monthly demand of an enterprise, patent number 201610796775.3, which uses the weight of the maximum monthly demand change to calculate the forecasted demand value of each month in the next n months, and determines a consecutive The maximum and minimum values ​​of the monthly forecasted maximum demand, and set the value of the step size, calculate the a electricity charges corresponding to each value, and obtain the minimum demand value as the predicted value, but this method requires at least two years of work Historical data, and the weight of historical data is treated as the same, without considering the factor of "nearly large and far small" of different monthly data weights, there is a certain error

Images