CAJ | 학술논문

以精确的空调负荷预测为前提,方能使得冰蓄冷空调在融冰供冷过程中采取最为合理的运行策略.提出一种改进的增量型极限学习机(PSO-IELM)的建筑物空调负荷预测模型.通过粒子群优化算法,克服传统极限学习机(ELM)在预测中存在的不稳定性.并结合对西安地区某购物中心夏季不同月份的空调负荷进行训练和预测.实例分析结果表明,粒子群优化增量型极限学习机(PSO-IELM)具有更好泛化能力和更高预测精度.是对建筑空调负荷预测的有效手段.
이정학적공조부하예측위전제,방능사득빙축랭공조재융빙공랭과정중채취최위합리적운행책략.제출일충개진적증량형겁한학습궤(PSO-IELM)적건축물공조부하예측모형.통과입자군우화산법,극복전통겁한학습궤(ELM)재예측중존재적불은정성.병결합대서안지구모구물중심하계불동월빈적공조부하진행훈련화예측.실례분석결과표명,입자군우화증량형겁한학습궤(PSO-IELM)구유경호범화능력화경고예측정도.시대건축공조부하예측적유효수단.
On the premise of accurate prediction of air conditioning load,the ice storage air conditioning can operate in the most reasonable way in the process of melting ice and supplying cooling,and a prediction model can also be estab-lished for the building of particle swarm optimization incremental type extreme learning machine (PSO-IELM) air conditioning load.By PSO algorithm,instability existing in the prediction of traditional extreme learning machine(ELM) can be overcome.