Control Sequence Simulates Multizone HVAC Systems in Direct Digital Control Systems
From eSociety, August 2019
Engineers have been struggling with using demand control ventilation as required by ANSI/ASHRAE/IES Standard 90.1, Energy Standard for Buildings Except Low-Rise Residential Buildings, and also still complying with the ventilation requirements in ANSI/ASHRAE Standard 62.1, Ventilation for Acceptable Indoor Air Quality.
A recently completed ASHRAE research project developed a practical, programmable control sequence to simulate the energy and ventilation performance in several buildings with multiple zone HVAC systems. The investigators used RP-1547’s logic. Researchers also tested the sensitivity of CO2 and airflow sensors.
The final report for RP-1747, Implementation of RP-1547, CO2-based Demand Controlled Ventilation for Multiple Zone HVAC Systems in Direct Digital Control Systems, details findings including a 7%-44% HVAC energy savings compared to non-DCV baselines. TC 4.3, Ventilation Requirements and Infiltration, sponsored the research. Zheng O’Neill, Ph.D., P.E., Member ASHRAE, was the principal investigator.
C. Hwakong Cheng, P.E., Member ASHRAE, who was a key contributor, discusses the research’s significance and challenges.
1. What is the significance of this research?
The main issue that this work addresses is the fact that ASHRAE Standard 90.1 has required the use of demand-controlled ventilation (DCV) for years, but that there is no direction on how to do that while still complying with the ventilation requirements in Standard 62.1. The goal was to develop a practical solution that designers can directly apply to actual projects. This work goes even further though.
The Ventilation Rate Procedure (VRP) in ASHRAE Standard 62.1 is notoriously complex and poorly understood, and there is literally no one right answer as to how much to ventilate a building according to the VRP. The control sequence developed from this applied research solves these problems by simplifying the design process and utilizing the building automation system (BAS) to evaluate the ventilation requirements in real time.
2. Why is it important to explore this topic now?
With the ever increasing focus on finding ways to reduce building energy use, there is more and more reliance on advanced control systems. DCV allows us to ensure that we provide the right amount of ventilation in each moment based on real-time CO2 and occupancy presence sensing in buildings. From an energy standpoint, it simply does not make sense to over-ventilate buildings at fixed rates based on their peak design population anymore.
This research provides a way to minimize energy use while ensuring consistent compliance with the ventilation requirements in Standard 62.1, which is important for occupant well-being in buildings.
3. How did you use RP-1547 during the project?
The previous research project RP-1547 successfully developed the fundamental approach to apply DCV in the context of Standard 62.1. Unfortunately, that mathematical approach was not something that could be practically implemented in a typical building automation system. Our project adapted that mathematical approach into a practical control sequence and included simulations and lab-scale testing to evaluate and validate its performance in realistic building systems.
4. What lessons, facts, and/or guidance can an engineer working in the field take away from this research?
DCV can provide significant energy savings by adjusting ventilation during periods of partial occupancy. In addition to CO2-based DCV, this approach also accounts for occupancy presence sensing and occupied-standby mode in Standard 62.1, which allows zone ventilation to be reduced to zero when a zone is detected to be empty. Though the savings from applying this approach to any particular building will depend on its occupancy profile, our simulations of an office building with realistic occupancy schedules showed 10%-30% HVAC energy savings for various climates.
The recent publication of ASHRAE Guideline 36, High-Performance Sequences of Operation for HVAC Systems, provides a platform onto which this approach can be overlaid, and a convenient mechanism for broad distribution.
Though we conducted lab-scale testing of this DCV approach, we are hoping to be able to do further field-scale testing as part of a separate project to consider system stability in real operating buildings. Once that stability has been validated, we intend to offer this as an optional ventilation control approach in Guideline 36.
Having the control logic pre-programmed by the control manufacturers and shifting the responsibility of evaluating the VRP from the designer to the BAS will improve efficiency, ventilation compliance and make things a lot easier for designers. Stay tuned!
5. How can this research further the industry’s knowledge on this topic?
This research project was perhaps unique in requiring not just a theoretical answer to a problem, but a practical solution that could be applied to real buildings. A key part of our project’s success was our team’s complementary backgrounds and skill sets, which brought together academic rigor and hands-on field experience from practicing engineers.
The formula from this research project, of translating primary research into a practical solution, is being repeated on the related project RP-1819, which will address the DCV issue for multiple path systems like fan powered boxes, and the proposed RP-1865, which will investigate supply air temperature control for dedicated outdoor air systems.
6. Were there any surprises or unforeseen challenges for you when preparing this research?
One challenge was during our lab-scale testing phase where we were iteratively improving the methodology, though this was mostly by design, so not a surprise. We were running the lab facility each day according to a carefully scripted schedule to mimic realistic building conditions. At the end of each day, we downloaded all of the operating data, and then analyzed and adjusted the control sequences at night so that we could pass those on to the team to reprogram, and then retest the next day. These iterations were compressed into a short but busy period because of the limited availability of the test facility. Ultimately though, the process worked exactly as planned.
The initial equations that looked good on paper didn’t work so well when applied to a real system—issues like sensor error, transients and network delays caused the equations to blow up in unexpected ways. Testing the algorithms in a controlled setting allowed us to improve them for these real-life considerations to develop a complex but robust and implementable control methodology.