Cold load prediction method based on genetic algorithm-support vector machine

Abstract

The invention discloses a genetic algorithm-based method. The cold load prediction method of the support vector machine comprises the following steps: firstly, simulating to obtain a feature vector and a to-be-predicted tag within a period of time; Dividing the simulated data according to time; Recycling z Carrying out normalization processing on the feature vectors of the two data sets and a to-be-predicted tag by using a scroe algorithm; setting the optimization ranges of the penalty functions c and g parameters of the initial support vector machine SVM; setting each parameter of the adopted genetic algorithm GA according to use requirements; Generating a plurality of initial individuals in a random form based on the set parameters of the genetic algorithm GA; calculating fitness valuesof the plurality of initial individuals; and training a training sample set formed by a plurality of initial individuals through a genetic algorithm GA to obtain an optimal support vector network, and predicting the cold load. The method can overcome the defect that traditional simulation calculation by using simulation software is time-consuming, and can quickly and effectively predict the cooling load.

Description

technical field [0001] The invention relates to the technical field of operation control of building energy supply systems, in particular to a cooling load prediction method based on a genetic algorithm-support vector machine. Background technique [0002] At present, my country's energy demand is increasing day by day, and the construction industry ranks among the three "big energy-consuming households". Building energy consumption accounts for about 20% of the world's energy consumption, and my country's building energy consumption accounts for about 28% of the total social energy consumption. Building energy-saving work such as optimizing the building envelope, intelligent control of the building environment, and system operating equipment has been in full swing. However, due to the complexity of the building structure, the influence of human factors, and thermal delays, there is energy waste in the actual use of buildings. , Insufficient energy supply and unbalanced ene...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a cooling load prediction method based on genetic algorithm-support vector machine, which can overcome the time-consuming shortcomings of the traditional simulation software simulation calculation, and can quickly and effectively predict the cooling load

Images