Study of Building Energy Consumption Prediction Model Based on Improved CGWO-SVM Algorithm
The prediction of building energy consumption can respond to whether there is any space for reduction of consumption in each sub-energy consumption system within a building,to provide guidance for energy saving and emission reduction work.In order to improve the accuracy of building energy consumption prediction,a building energy consumption prediction model based on improved chaotic gray wolf optimization-support vector machine(CGWO-SVM)algorithm is proposed.To address the problems of premature convergence and easy to fall into local optimal solutions of traditional gray wolf optimization(GWO)algorithm,Tent chaotic sequence initialization of the population,control parameter stochastic dynamic adjustment strategy is used to balance the search ability.Cauchy mutation operation reduces the probability of the algorithm to fall into the local optimal solution,and the improved CGWO is proposed.Through the simulation tests,it is proved that the CGWO algorithm is better than the GWO algorithm in terms of solution accuracy and stability.The algorithm is used to optimize the support vector machine(SVM)and establish the CGWO-SVM building energy consumption prediction model.The actual building energy consumption data are used for testing.The test results show that the prediction performance of the CGWO-SVM algorithm is better than other models.The CGWO-SVM algorithm can be used in the prediction work of building energy consumption,which has strong practical application value.
Building energy consumption predictionImproved chaotic gray wolf optimization(CGWO)Tent chaotic sequenceVariationSupport vector machine(SVM)
NETL
NSTL
万方数据
