Electric heating load prediction method

Abstract

The invention provides an electric heating load prediction method. The electric heating load prediction method comprises steps that main influence factors of an electric heating load are analyzed; theBP neural network calculation method is constructed; a genetic algorithm is constructed to optimize the BP neural network calculation method; integral data is selected as a training sample and a prediction sample, the BP neural network calculation method and the genetic algorithm are utilized to optimize the BP neural network calculation method to respectively predict an electric heating load, and effectiveness and scientificity of the genetic algorithm for optimizing the BP neural network calculation method are verified. The method is advantaged in that the electric heating load can be precisely predicted, a type of heating equipment of a heating system and a new load size can be accurately determined, heating parameters, idle capacity and a system operating status, etc., can be accurately determined, so operation, maintenance and overhaul of the heating equipment are made to be more scientific and reasonable.

Description

technical field [0001] The invention belongs to the technical field of load forecasting, in particular to an electric heating load forecasting method. Background technique [0002] The heating industry is closely related to people's daily life and plays a decisive role in the development of the entire national economy. In order to reduce energy consumption and reduce urban environmental pollution, most cities in my country currently use centralized heating for heating. Cities and towns in northern my country generally use central heating for heating, and heating energy consumption has become the largest component of building energy consumption. Therefore, heating energy saving is one of the most potential and most effective ways of building energy conservation in my country. [0003] Relying solely on the voluntary behavior of users and enterprises in the central heating system, as well as the mandatory implementation of building energy-saving standards can only play a cert...

AI Technical Summary

Problems solved by technology

However, the neural network also faces many problems: there is no theoretical guidance for the structural design of the neural network, and it is difficult to optimize the structure of the forward network; the learning and training speed of the neural network is slow, especially for large-scale neural networks. During training, the time required is too long, and the initial value has a great influence on the convergence speed; the traditional gradient descent algorithm is easy to fall into the local minimum point and cannot reach the global optimum

Images