Submitted by: Michael
Firm Name: AKF Group
ASHRAE Climate Zone: 4A
Who performed the simulation analysis?
What tools were used for the simulation analysis?
What phase of the project was analysis conducted?
What are the primary inputs of the analysis? Annual Chilled Water Load Profile from eQUEST model, efficiency matrices for several chiller configurations.
What are the primary outputs of the analysis? Annual energy cost to operate each chilled water plant, taking into consideration both consumption and significant, highly-variable demand charges.
List the investigations questions that drove your analysis process.
The primary question that drove the study was how chiller sizing would impact annual energy consumption and cost. The chillers being evaluated were very efficient at low part-loads and low entering condenser water temperatures; the fact that the manufacturer was able to provide us with a wide range of efficiency points enabled a more targeted and granular analysis than otherwise would have been possible.
How was simulation integrated into the overall design process?
How did you set up the simulation analysis and workflow?
The initial chiller plant alternates were defined by the MEP Design team and owner: - (3) 1,000-Ton Conventional Centrifugal Chillers - (3) 1,000-Ton Magnetic Bearing Centrifugal Chillers - (1) 750-Ton and (2) 1,250-Ton Conventional Centrifugal Chillers - (1) 300-Ton and (2) 1,250-Ton Conventional Centrifugal Chillers - (3) 1,000-Ton Conventional Centrifugal Chillers plus an Ice Storage plant We used an eQUEST model developed for the client's existing office space, re-calibrated the chilled water load based on actual condenser water trending data, and then made adjustments for anticipated changes in square footage and critical loads in order to develop an hourly chilled water profile for the new space. The hourly annual chilled water load profile was then entered into eQUEST models for each plant alternate. Load management strategies were developed within eQUEST to maximize each chiller's efficiency based on part-load and condenser water temperature. Hourly results for chiller loads, condenser water temperature, pump energy and cooling tower fan energy were then exported from each simulation using Python. Once the results were in Python, we used SciPy to develop a multiple regression for the manufacturer-provided efficiencies, and we post-processed chiller compressor energy for each alternate accordingly. Hourly results for total plant kw were re-grouped and transformed using Pandas to conduct a time-of-day, day-of-week, and seasonal demand charge analysis.
How did you visualize the results to the design team? What was successful about the graphics that you used to communicate the data?
Because the entire project team was invested in the simulation results and chiller optimization strategies, we developed an easily-shareable interactive visualization platform that provided several different visualization types, including scatter plots, heatmaps, histograms and line charts. This enabled us to investigate the study in granular detail in real-time in project meetings, as well as share the platform to team members for their own targeted exploration. We found that being able to explore the dataset in real time enabled the project team to identify potential issues with their own assumptions, as well as help them understand trends and results in a deeper way than a simple annual cost table would have allowed.
Most importantly, what did you learn from the investigation? How did simulation and its outputs influence the design of the project?