A novel building cooling load prediction based on SVR and SAPSO

Abstract

Accurate building cooling load forecasting is the precondition for the optimal control and energy saving operation of HVAC systems. Hourly cooling load forecasting is a difficult work as the load at a given point is dependent not only on the load at the previous hour but also on the load at the same hour on the previous day. So the accuracy of forecasting is influenced by many unpredicted factors. Support vector machine (SVM) is a novel type of learning machine, which has been successfully employed to solve nonlinear regression and time series problems. The key problem of SVM is the selection of SVM free parameters. In this paper, it is proposed a new optimal model, which is based on Stimulated Annealing Particle Swarm Optimization Algorithm (SAPSO) that combines the advantages of PSO algorithm and SA algorithm. The strong searching ability of SA was employed to PSO algorithm to avoid the premature convergence with better stability and astringency. The SA based PSO algorithm was used to optimize the free parameters of SVM model. The study used the proposed model to forecast load of cooling load system. The numerical simulation results demonstrate that the accuracy of SA-PSO based SVM model outperforms that of the traditional SVM load forecasting model.

The Research on the Load Prediction for Ice Storage System Based on Rough Set

“…MacArthur, et al, [19] and Spethmann [20] developed a prediction method based on the autoregressive integrated moving average (ARIMA) model and applied it to an optimal cold storage controller. Li Xuemei, et al, [21] proposed an optimal model which is based on stimulated annealing particle swarm optimization algorithm that combines the advantages of PSO algorithm and SA algorithm. In Ref [21], it illuminated that because of its strong non-linear mapping ability, artificial neural networks (ANN) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems, which have been popularly applied to predict the building cooling load and building energy consumption.…”

“…Li Xuemei, et al, [21] proposed an optimal model which is based on stimulated annealing particle swarm optimization algorithm that combines the advantages of PSO algorithm and SA algorithm. In Ref [21], it illuminated that because of its strong non-linear mapping ability, artificial neural networks (ANN) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems, which have been popularly applied to predict the building cooling load and building energy consumption. ANNs-based models seem to obtain improved and acceptable performance in cooling load forecasting issue, however, the conventional ANNs still suffer from several weaknesses such as the need for a large number of controlling parameters, the difficulty in obtaining stable solutions, the danger of over fitting and thus the lack of generalization capability.…”

A Data-Driven Approach to Forecasting Heating and Cooling Energy Demand in an Office Building as an Alternative to Multi-Zone Dynamic Simulation

“…The selection of the three non-negative parameters (epsilon, C, and gamma) is determinant to forecasting accuracy. The use of a combinatorial method is often suggested to optimize the parameter selection procedure [18,[25][26][27][28]. The authors selected simulated annealing (SA) to approach the parameter selection stage.…”

“…Nevertheless, as the building analyzed in this article has its heating and cooling needs provided by district heating (further addressed in Section 2.1), electricity consumptions are quite more stable and easier to predict. For similar reasons, a considerable number of researchers are tackling the challenge of estimating building heating and cooling needs [3,5,[15][16][17][18][19].…”