Simple Linear Regression

IntroductionHVAC systems typically consume around 70% of a base building’s energy usage, with 25-35% of energy consumed by chillers producing chilled water for air-conditioning. Therefore, the efficiency of chillers and the optimization of their performance within the HVAC systems they operate are important in achieving high performance.Purpose of the AssignmentIn this assignment a one-year historical data (refer to Daily Chilled water-1 excel file attached) for a chiller within a building has been measured. The main aims of this assignment are that students· Taste the flavor of forecasting methods within buildings.· Understand basic concepts of short-term/midterm forecasting.· Learn how to· capture the salient features using different regression models· build regression models using Excel· justify how well the models have fitted the data· develop a model for one day ahead load forecastingQuestionsStudents must answer the following question in details.1. Plot the data and apply the simple linear regression (SLR) model and second order regression (SOR) model for the chilled water and HDD. In both cases, find the mathematical models for producing daily chilled water as well.2. Use the obtained models above to forecast the chilled water that you would expect for the next January (2014) if the HDD varies as shown in Table 1 (attached).3. Apply multiple linear regression (MLR) model for the chilled water and all parameters impacting the amount of chilled water. Find the mathematical model for producing daily chilled water as well.4. Use the obtained MLR model above to forecast the chilled water that you would expect for the next January (2014) if the RH, Tdew-C and HDD vary as shown in Table 1.5. Calculate MAPE and RMSE for each model (SLR, SOR, and MLR) and discuss which model(s) is more fitted with the historical data?