CAJ | 학술논문

建筑周围气象参数的不确定性和持续波动性，为建筑系统动态负荷预测及实时优化控制带来困难。文章以最小二乘支持向量机（LSSVM）作为预测算法，运用粒子群优化算法（ PSO）优化 LSSVM 模型参数，建立基于历史信息的多输入多输出（MIMO）建筑周围气象参数预测模型，对影响建筑负荷的室外温度、湿度及风速进行超短期预测。结果表明：PSO 算法可对模型参数进行优化，基于 PSO-LSSVM 算法构建的建筑周围气象参数超短期预测模型能够实现未来140 min 气象参数的预测，为建筑供能和用能系统动态优化运行提供数据。
건축주위기상삼수적불학정성화지속파동성，위건축계통동태부하예측급실시우화공제대래곤난。문장이최소이승지지향량궤（LSSVM）작위예측산법，운용입자군우화산법（ PSO）우화 LSSVM 모형삼수，건립기우역사신식적다수입다수출（MIMO）건축주위기상삼수예측모형，대영향건축부하적실외온도、습도급풍속진행초단기예측。결과표명：PSO 산법가대모형삼수진행우화，기우 PSO-LSSVM 산법구건적건축주위기상삼수초단기예측모형능구실현미래140 min 기상삼수적예측，위건축공능화용능계통동태우화운행제공수거。
The uncertainty and volatility of weather around buildings bring difficulty for dynamic load forecasting and building system's real-time optimal control. A multi-input/ multi-output( MIMO) weather forecasting model is built based on historical information,using the least squares support vector machine(LSSVM)as the model prediction algorithm and particle swarm optimization(PSO)to optimize the parameters of the LSSVM,this article predicts the temperature,humidity and wind speed which influences the building load in ultra-short term by the proposed model. The results show that, PSO algorithm can optimize the model parameters,and this ultra-short term forecasting model based on PSO-LSSVM( least squares support vector machine based on the particle swarm optimization) predicts weather parameters in 140 minutes ahead,and provides data for dynamic optimal operation of building supply and use system.