IMPROVING ENERGY SIMULATION ACCURACY OF NEW AND RENOVATED HEALTHCARE BUILDINGS

Open Access
Author:
Alanqar, Ibrahim W
Graduate Program:
Architectural Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
November 06, 2013
Committee Members:
  • Jelena Srebric, Thesis Advisor
  • Chinemelu Jidenka Anumba, Thesis Advisor
  • Stephen James Treado, Thesis Advisor
Keywords:
  • Healthcare
  • energy
  • energy simulation
  • sensitivity analysis
  • simulation accuracy
  • retrofit
  • laboratory
  • hospital
Abstract:
The purpose of this research is to improve the accuracy of building energy simulation for healthcare buildings by identifying and modifying key input parameters. The thesis will use the modified parameters in simulating case-study hospital buildings to correctly predict the buildings’ energy behavior. Hospital buildings have been major energy consumers with a significant contribution to the total annual energy consumption in commercial buildings. Based on Commercial Building Energy Consumption Survey (CBECS), the total energy used by a typical hospital building per square-foot is more than two times the use of a typical office building. Therefore, there is a need to enable accurate and reliable energy simulation models to accurately predict the energy-use of new buildings and study the feasibility of different renovation and retrofit strategies that could lead to energy savings for retrofit projects. There are three case studies for this research. The first case-study is the five-story Hershey Cancer Institute, part of the Penn State Milton S. Hershey Medical Center, located in Hershey, Pennsylvania. The cancer institute is an example of a new LEED® certified hospital Building. The second case-study is the twelve-story Bluemle Life Sciences building, part of the Thomas Jefferson University Hospital, located in Philadelphia, Pennsylvania. This building is an example of a simple-renovation project. The facility is mainly biomedical laboratory spaces. The renovation of this building was mainly the conversion of one of the air-handling units from constant-air-volume (CAV) system to variable-air-volume (VAV) and the installation of new controls on that unit and the supported air-distribution boxes. The third and final case-study is the nine-story Clinical Sciences building, also part of the Hershey medical campus. This is an example of a building that has gone through a major renovation where the entire air distribution system of the building was converted from CAV system to VAV system in three phases. The energy managers of both facilities provided building plans as well as metered energy consumption data. In addition, energy predictions provided by the consulting firm overseeing the first case-study, the Cancer Institute, were provided for comparison with the research simulation results. Recent studies examine impacts of influencing input parameters that affect the energy simulation results for hospital buildings. The influencing parameters that many research publications recommend in their sensitivity analyses are: load and lighting densities, schedules of equipment, lighting and occupants, ventilation, and temperature and humidity set-points. Each input parameter is then modified to values obtained from applicable building codes, previous research, and technical reports. The research simulates the energy consumption of the case-studies using EnergyPlus engine. Comparison between simulation models using the program defaults and the modified values are then compared to actual metered data to measure the accuracy and the influence of the parameters on the simulation. In Hershey Cancer institute, the simulation accuracy of the electric energy of the modified model shows a considerable improvement over the default model and a good agreement with the metered data (8.70 % average monthly and 16% annual). In addition, the pattern of predicted energy, which represents the electricity demand, also followed the actual energy use pattern. In Bluemle Life Sciences building, the facility manager provided both metered electrical (including cooling) and metered gas (representing heating) data. The comparisons between electric metered data and modified simulation show that the monthly and annual simulated energy consumptions are very close having an average monthly accuracy of 4.6% and 4.8% annually. Thermal consumption, on the other hand, was more challenging in meeting uncertainty indices but the overall accuracy of the total thermal energy remained at an average monthly of 10.3% and 9.3% annual . In the Clinical Science building, the comparisons between metered and modified simulation show that the monthly and annual simulated electricity consumptions remain within a good agreement, 12.63% average monthly and 7.2% annual, with the metered electric data. Although no metered thermal data were provided, analysis of simulation results was significant. Changes of CAV to VAV systems had noticeable savings on the heating and cooling energy consumptions in both default and modified models; however, modified model predicts 26.7 % and 15.1% heating and cooling energy savings, respectively, compared to the default model that predicts 42% reduction in heating and 18.8% reduction in cooling energy savings. The results from the accuracy measures obtained from each case-study after the parameter modification provide a great example of the effectiveness and influence the changes of parameters have on energy modeling. For future research and to improve the validation of this current research, there is a need for more case-studies from different climate zones where these parameters can be tested. Another potential research that can stem from this thesis is using the same process of accuracy improvement by identifying and modifying key parameters for different types of commercial buildings.