Trial results from a model predictive control and optimisation system for commercial building HVAC

doi:10.1016/j.enbuild.2013.12.037

Energy and Buildings

Volume 72, April 2014, Pages 271-279

https://doi.org/10.1016/j.enbuild.2013.12.037 Get rights and content

Highlights

•
A novel predictive control system for commercial building heating ventilation and air-conditioning (HVAC).
•
Multi-objective optimisation framework simultaneously considers energy, operating cost and occupant comfort.
•
Online occupant comfort feedback helps to fine-tune occupant comfort satisfaction.
•
Results presented from real-world trials have demonstrated energy improvement between 19 and 32% over existing control strategies.

Abstract

This paper presents the results from two real-world trials of an optimised supervisory model predictive control (MPC) system for heating, ventilation and air conditioning (HVAC) in commercial buildings. The system learns a model from historical data and uses weather forecasts and a given temperature set-point profile to predict building zone conditions and thermal comfort with the aim of optimising building controls for a number of HVAC zones throughout a day. The multi-objective optimisation minimises running cost and CO₂ emissions, subject to operator preferences, while constraining occupant thermal discomfort to an acceptable range to find the best zone temperature set-point schedule for the building. This schedule is then applied to the building by a feedback control loop, which balances the power supplied to each zone for heating, cooling and ventilation.

A complementary online occupant comfort feedback tool was deployed to all occupant computers in the trial office buildings. This tool allows occupants to submit feedback on their thermal comfort and satisfaction at any point in time via electronic surveys, as well as allowing these surveys to be issued to occupants at scheduled times. This feedback fine-tuned the thermal comfort model used to constrain the optimisation, allowing for errors in the comfort model to be compensated. Thermal comfort feedback was also used to measure and compare relative occupant comfort levels with a baseline.

This control system was trialled on two office buildings in Australia, over two winter months and results compared with the performance of the incumbent building management and control system (BMCS). An average energy reduction of 19% and 32% was achieved in the two buildings over 51 and 10 days of operation respectively without substantially affecting measured or modelled occupant thermal satisfaction levels.

Introduction

This paper presents a supervisory control and optimisation system for commercial HVAC systems, aimed at minimising energy consumption and occupant thermal discomfort, and an accompanying online occupant comfort feedback tool which can be installed on occupant computers.

Occupant feedback from the comfort feedback tool is used to assess the impact of the optimised HVAC control strategy, to fine-tune the standard comfort model used in the optimisation process and to inform users of the state of the HVAC system, such as when natural ventilation or economy modes are active.

The supervisory HVAC control and optimisation system (referred to hereafter by the authors’ internal designation ‘OptiCOOL’ interfaces with a commercial building management & control system (BMCS) to read required data from HVAC zones and air handling units (AHUs) and to perform required control actions. It learns a model that can be used to predict zone conditions and comfort levels using information about HVAC power consumption and weather, which is used to discover the optimal power consumption schedule throughout the day-balancing cost, energy and emissions while maintaining comfort. The control module then overrides zone or supply air set-points to balance the heating/cooling power supplied to each zone and to track the best power profile from the optimisation.

Both systems were installed in two office buildings on the east coast of Australia and trialled over two months during winter to evaluate their energy savings potential. Thermal comfort levels were monitored throughout the trials to ensure consistency with pre-trial levels.

Section snippets

Background

Model predictive control (MPC) has been widely used for some form of building HVAC optimisation. The majority of work to date concentrates on examining control strategies for specific subs-systems [1], applying offline optimisation of particular buildings or building configurations using simulation models [2], or considers only specific approaches such as set-point resetting, peak demand limiting or load shifting [3], [4], [5], or examines techniques using small-scale test beds [6], [7]. Whole

Thermal comfort

Human perception of thermal comfort is highly subjective, and modelling it accurately for the purposes of predicting occupant comfort from measured environmental conditions is an active area of research, with de facto standard models having been shown to bias predictions such that excess cooling energy is often consumed unnecessarily [22]. Given that the primary aim of building HVAC systems is to maintain a level of occupant thermal comfort, it is critical to have an accurate comfort metric by

Trial details

The system was deployed and experiments run at two sites during winter 2011 (July–August) in two buildings on the east coast Australia.

The first building was a 3-floor 3322 m² office building in Newcastle, New South Wales (NSW) with an under-floor air distribution system driven by 15 AHUs. The water loops to each AHU were supplied by two centralised chillers and a Combined Heat and Power (CHP) system with two gas micro-turbines and two gas boilers. The second building was a 3-floor 1808 m² office

Conclusions

We have presented a novel method of optimising the operation of commercial HVAC systems using model predictive control, to minimise a weighted combination of operating cost, greenhouse gas emissions and occupant thermal comfort. The system was trialled in two buildings in Newcastle, and Melbourne Australia during winter 2011 to compare performance to standard building management and control systems.

Performance results obtained from the Newcastle Office Wing over winter 2011 show an average

Acknowledgement

This research was funded in full by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) Energy Transformed Flagship, Australia.

References (35)

S. Yuan et al.
Multiple-zone ventilation and temperature control of a single-duct VAV system using model predictive strategy
Energy and Buildings
(Oct. 2006)
M. Kintner-Meyer
Optimal control of an HVAC system using cold storage and building thermal capacitance
Energy and Buildings
(1995)
K. Ling et al.
Expert control of air-conditioning plant
Automatica
(1994)
M.A. Humphreys et al.
The validity of ISO-PMV for predicting comfort votes in every-day thermal environments
Energy and Buildings
(Jul. 2002)
M. Morari et al.
Model predictive control: past, present and future
Computers & Chemical Engineering
(May 1999)
P.T. Hartman
Designing efficient systems with the equal marginal performance principle
ASHRAE Journal
(2005)
M. Gwerder et al.
Predictive control for integrated room automation
K. Fong et al.
HVAC system optimization for energy management by evolutionary programming
Energy and Buildings
(2006)
A. Aswani et al.
Reducing transient and steady state electricity consumption in HVAC using learning-based model-predictive control
Proceedings of the IEEE
(Jan. 2012)
V.M. Zavala
Real-time optimization strategies for building systems
Ind. Eng. Chem. Res.
(Sep. 2012)

V. Zavala et al.

Proactive energy management for next-generation building systems

J. House et al.

A system approach to optimal control for HVAC and building systems

ASHRAE Transactions

(1995)

M. Cascia

Implementation of a near-optimal global set point control method in a DDC controller

ASHRAE Transactions

(2000)

G. Henze et al.

Experimental analysis of model-based predictive optimal control for active and passive building thermal storage inventory

HVAC&R Research

(Apr. 2005)

S. Wang

Supervisory and optimal control of building HVAC systems: a review

HVAC&R Research

(Jan. 2008)

N. Nassif et al.

Optimization of HVAC control system strategy using two-objective genetic algorithm

HVAC&R Research

(Jul. 2005)

Y. Sawaragi et al.

Theory of multiobjective optimization

(1985)

Cited by (129)

Efficient management of HVAC systems through coordinated operation of parallel chiller units: An economic predictive control approach
2024, Energy and Buildings
In developed countries, air conditioning systems have become major contributors to energy consumption in buildings. Cooling installations made up of independent chiller units connected in parallel pose a challenge in finding the most energy-efficient management strategy. This article proposes a novel economic model predictive control strategy to optimize the operation of multiple-chiller HVAC systems according to an economic cost comprising energy consumption as well as a thermal comfort index. Provided that the gradient of the economic cost function can be calculated or estimated, the proposed controller only entails the solution of a single quadratic programming (QP) problem at each sampling period, reducing computational requirements and thus facilitating the deployment on commercial embedded HVAC management platforms. The proposed controller improves economic performance. As our numerical analysis shows, optimal sequencing is achieved for the case of dual-chiller plants. Moreover, the controller adapts to possible variations in the economic criteria, enabling the system to respond to changes in electricity price and user preferences. A realistic case study, using a high fidelity model of a building simulated in TRNSYS, demonstrates the effectiveness of the proposed methodology. In particular, the proposed controller demonstrates to be more efficient than state-of-the-art QP-based economic predictive controllers in achieving a reduction of the energy consumption up to 5.19%, which is in line with the targeted reduction of 7.9% by 2030 in the EU Green Deal's ‘Fit for 55’ package.
Key district heating technologies for building energy flexibility: A review
2024, Renewable and Sustainable Energy Reviews
In the background of the continued integration of renewable energy sources (RES) and the increasing flexibility on the demand side, the diversity and complexity of new technologies for heating present increased challenges for design and operation of district heating systems (DHS). This work first reviews the progress of the new generation of DHS, followed by providing an overview of investigations on building energy flexibility in the field of heating, with a focus on the characterization and quantification of energy flexibility, the realization of thermal flexibility, and the use of building thermal mass in demand side management (DSM). Different technologies were categorized and summarized according to the composition of the new generation of DHS. Control strategies such as model predictive control were also examined. In particular, the concept of building thermal battery is used to analyze buildings or prosumers thermal energy flexibility. Finally, new elements of DHS development and potential challenges were discussed.
A day-ahead operation regulation method for solar water heating based on model predictive control
2023, Energy and Buildings
The operation performance of a solar heating system is largely influenced by the coupling of outdoor climate conditions and building energy operation patterns, and is characterized by strong randomness and large heating fluctuations, which makes the optimal control rules for solar heating systems in dynamic state. Although the traditional RBC(rule-based control), whose rules is preset, has a fast response speed, the fixed-rules are hardly to make the solar heating system in optimal state. Furthermore, solar heating systems is generally characterized by a large time delay in the heat transfer process. The control point lag occurs in the control signal of the dynamic control strategy. To address these issues, this paper proposes a model predictive control-based (MPC) control strategy for solar heating systems. The method uses seq2seq-LSTM to predict major operating parameters including ambient temperature, solar radiation, the load inside the building for the next 24 h and combines the heating system operation model to simulate and obtain the control signal at the next moment. The results showed that, compared with traditional control strategies including RBC and PID control strategies, the efficiency of solar collector regulated by MPC increased by 9%. Taken the tank temperature at the simulation end as the energy calculation reference point, the solar heating system’s total energy consumption employing MPC is 34.2% lower than that employing traditional control strategy.
Artificial intelligence enabled energy-efficient heating, ventilation and air conditioning system: Design, analysis and necessary hardware upgrades
2023, Applied Thermal Engineering
Literature search across different databases showed that the application of artificial intelligence (AI) in heating, ventilation and air conditioning (HVAC) equipment has been extensively studied. On the commercial front, Internet search suggested that numerous AI-equipped HVAC products have been launched. These products apply AI in very different ways, and their energy-saving effects are also different. Such divergence and uncertain energy-saving effects may hinder AI application. To overcome this difference and accelerate the development of AI applications, the present study proposed a double diamond preferred reporting items for systematic reviews and meta-analysis (PRISMA) method—an analysis method that combined literature review with design thinking. Through a process of divergence-convergence-re-divergence, this study described how to design AI functions for energy-efficient HVAC systems, taking into account more than 1,700 research papers it had reviewed. However, there was a limitation on the part re-divergence. Because the vast majority of research papers only published results of successful AI applications, no cases of failed applications were available for review, making it impossible to re-think profoundly. Instead, this study collected raw data from 88 research papers and used these data to analyze the effectiveness and ineffectiveness of AI in depth. It was concluded that AI application must be accompanied by necessary hardware improvements to achieve effective energy savings. AI-enabled energy-saving effects for chillers, air-handing units, heating systems, and air conditioners, as well as corresponding hardware upgrades, were discussed.
Real-world implementation and evaluation of a Model Predictive Control framework in an office space
2023, Journal of Building Engineering
In this work, Model Predictive Control (MPC) is experimentally implemented on a Heating, Ventilation and Air Conditioning (HVAC) system of a large-scale office space. As controller model, a physics-based Modelica building model is calibrated based on historic data pursuing an iterative, nonlinear optimization approach. For the calibration period with a horizon of seven weeks, the calibrated model exhibits a high model accuracy with a Root Mean Square Error (RMSE) of 0.49 $° C$ between the measured and estimated room temperature. The MPC toolchain includes modules for state estimation and forecasts of disturbances quantities (such as outdoor temperature, solar radiation including calculation of the direct and diffuse fraction, supply temperatures and occupancy). The MPC execution comprises the operation of heating based on radiators and floor heating (via regulation of valve openings) as well as shading of three Venetian blind systems (via regulation of vertical position and slat inclination angle). The experiment is conducted during a heating period with a duration of three weeks from October 21 to November 11, 2022. The heating actuators are controlled considering their typical dynamics and take into account the night setback during unoccupied office periods. User acceptance of the automated shading control is included through additional cost function terms for the shading operation. The field test reveals the predictive control capabilities of the proposed MPC toolchain in a real-life scenario, demonstrating energy-efficient building operation and total average discomfort of 0.53 Kh/d. The MPC formulation provides flexibility regarding adjustability of the control towards energy efficiency, thermal comfort, daylight transmission and non-oscillating shading control. Finally, the disturbance forecast accuracies for outdoor temperature and the solar radiation quantities are evaluated and the MPC control performance is compared against a conventional control approach.
MF^2: Model-free reinforcement learning for modeling-free building HVAC control with data-driven environment construction in a residential building
2023, Building and Environment
Reinforcement Learning (RL) has advanced energy-efficient control of building Heating, Ventilation and Air Conditioning (HVAC) systems. Constructing a suitable RL environment for buildings is a crucial challenge. Compared to widely-used simulation-based environments, data-driven approaches offer higher training efficiency but face convergence difficulties due to influential factors, limiting their current application.
To explore data-driven construction of RL environments for building HVAC systems, this study proposes two strategies for controlling room temperature setpoints in a residential building. XGBoost and Long Short-Term Memory Network (LSTM) are trained for energy consumption and room temperature prediction. One strategy predicts parameters for on-off states, while the other for power-on states. The XGBoost models are integrated into an OpenAI Gym environment. The first strategy achieves 0.8634 R² and 0.2423 Root Mean Squared Error (RMSE) for energy consumption prediction. The R² of room air temperature models are approximately 0.99 and the RMSE are lower than 0.31. The second strategy achieves 0.9181 R² and 0.1042 RMSE for energy consumption prediction and similar performance for room temperature prediction. Deep Q-learning (DQN) and Deep Deterministic Policy Gradient (DDPG) algorithms are separately trained using these environments. Results show that the first strategy fails to induce the correct training of RL models, while the second strategy successfully induces a useable DDPG model for controlling building HVAC systems but fails to induce a useable DQN model. We analyze the reasons behind these observations. Compared to the original room temperature setpoint method, the DDPG-based HVAC control logic achieves a 10.06% energy-saving effect while ensuring comfort.

View all citing articles on Scopus

View full text

Trial results from a model predictive control and optimisation system for commercial building HVAC

Highlights

Abstract

Introduction

Section snippets

Background

Thermal comfort

Trial details

Conclusions

Acknowledgement

Energy and Buildings

Energy and Buildings

Automatica

Energy and Buildings

Computers & Chemical Engineering

Designing efficient systems with the equal marginal performance principle

ASHRAE Journal

Predictive control for integrated room automation

HVAC system optimization for energy management by evolutionary programming

Energy and Buildings

Reducing transient and steady state electricity consumption in HVAC using learning-based model-predictive control

Proceedings of the IEEE

Real-time optimization strategies for building systems

Ind. Eng. Chem. Res.

Proactive energy management for next-generation building systems

A system approach to optimal control for HVAC and building systems

ASHRAE Transactions

Implementation of a near-optimal global set point control method in a DDC controller

ASHRAE Transactions

Experimental analysis of model-based predictive optimal control for active and passive building thermal storage inventory

HVAC&R Research

Supervisory and optimal control of building HVAC systems: a review

HVAC&R Research

Optimization of HVAC control system strategy using two-objective genetic algorithm

HVAC&R Research

Theory of multiobjective optimization