Advances in Civil Engineering

Research Article

Development of Regression Models considering Time-Lag and Aerosols for Predicting Heating Loads in Buildings

Table 26

Multiple regression equations using new variables based on building size.


	Building Types	Linear Regression Model

Original variables only	DOE small office (time-lag0)	Y = −179888432.974 + (−685748.939)X₁ + (86,402.645)X₂ + (−47343.593)X₃ + (1811.848)X₄ + (24,711.629)X₅ + (−49815.842) X₆ + (34,307.818)X₇ + (61,552.875)X₈ + (−371456.480)X₉ + (−1367478.673)X₁₀ + (−544550.045) X₁₁
	DOE medium office (time-lag1)	Y = −242862773.124 + (−57775.571)X₂ + (2360.095)X₃ + (76,361.042)X₄ + (−82461.104)X₅ + (56,837.692)X₆ + (101106.744)X₇ + (−4510822.624)X₈ + (−1387805.281)X₉ + (−677958.805)X₁₀ + (−799488.056)X₁₁
	DOE large office (time-lag2)	Y = −242862773.124 + (2360.095) X₂ + (76,361.042)X₃ + (56,837.692)X₄ + (−4510822.624)X₅ + (−1387805.281)X₆ + (−57775.571)X₇ + (−82461.104)X₈+ (101106.744)X₉ + (−677958.805)X₁₀ + (−799488.056)X₁₁

With new variables	DOE small office (new model)	Y = −84468362.750 + (−153900.636)X₂ + (820.654)X₄ + (7302.493)X₅ + (−28329.843)X₆ + (22,573.966)X₇ + (31,803.161)X₈ + (−380750.599)X₁₀ + (−215562.688)X₁₂
	DOE medium office (new model)	Y = −137086417.229 + (−581774.230)X₁ + (−112436.116)X₂ + (1264.005)X₄ + (47,644.241)X₅ + (16,462.981)X₇ + (−2787320.969)X₁₀ + (−566944.126)X₁₂ + (1076401.110)X₁₃
	DOE large office (new model)	Y = −137086417.229 + (−581774.230)X₁ + (−112436.116)X₂ + (1264.005)X₄ + (47,644.241)X₅ + (16,462.981)X₇ + (−2787320.969)X₁₀ + (−566944.126)X₁₂ + (1076401.110)X₁₃

X₁ = dry bulb temperatures, X₂ = dew point temperatures, X₃ = relative humidity, X₄ = atmospheric station pressure, X₅ = horizontal infrared radiation, X₆ = global horizontal radiation, X₇ = direct normal radiation, X₈ = diffuse radiation, X₉ = wind speed, X₁₀ = total sky cover, X₁₁ = visibility, X₁₂ = AOD, X₁₃ = precipitable water, and Y = heating load.