Search

• The MPC strategy is proposed to optimize the energy performance of an ASHP heating system. • The ASHP heating system operates under the variable water temperature difference. • A high-fidelity physical model is built in Dymola to provide the raw data source. • A data-driven control-oriented model is derived by dynamical system identification. • The simulation is conducted to validate the effectiveness of the proposed strategy. The heat pump heating system has been widely accepted to provide space and water heating in commercial and residential buildings. However, the operating energy performance in practice conditions is limited due to the traditional control strategy. An efficient-cost advanced control strategy is desired to improve the overall operating performance. This paper presents a model predictive control (MPC) strategy for an air source heat pump (ASHP) heating system to optimize the overall operating performance. Based on the control-oriented model and objective function, MPC can provide the optimized compressor frequency and water mass flow rate in real-time, which can lead to the optimized real-time variable water temperature difference. Thus, the total power consumption can be minimized for the heating system by coordinating the compressor power consumption and water pump power consumption in the real-time operating condition. A high-fidelity physical model is first built to produce the required data source, and then the control-oriented model is derived based on the machine learning technique. MPC is designed to minimize the total power consumption providing that the heating capacity is well maintained. The MPC strategy is evaluated and compared with the traditional PI control method under three operating cases. [ABSTRACT FROM AUTHOR]

The use of model predictive control (MPC) to optimally control district heating (DH) networks can support the transition to a carbon-neutral heating sector. DH systems are inherently subject to non-linear physics and integer controls, which results in a mixed-integer non-linear program (MINLP). In this work, an MINLP-based MPC strategy is developed for the optimal control of a DH network, building upon an existing decomposition approach (combinatorial integral approximation). The main novelty of this work is the application of an integrated MINLP-based MPC to a DH network and its comparison to a non-linear program (NLP)-based MPC. To successfully develop this MINLP-based approach, the pycombina tool is efficiently integrated in the existing NLP-based MPC toolchain (TACO) and the concept of an augmented time horizon is introduced to manage dwell time constraints. The MINLP-based MPC is applied to two use cases: a relatively simple nine-zone terraced house and a more complex fourth generation DH network. The simulation study shows that the MINLP-based MPC yields a comparable control performance to that of a previously developed NLP-based MPC, but the CPU time is approximately eight times higher. However, the absence of a post-processing step (which requires ample engineering practice and time) and the improved match between the MPC's controller model and the actual system show promise for the MINLP-based MPC to control complex DH networks. • An MINLP-based MPC methodology for district heating systems is developed. • The MINLP-based MPC methodology can cope with dwell time constraints. • NLP- and MINLP-based MPC reach similar performance for the envisaged use cases. • NLP-based MPC has a computational advantage. • MINLP-based MPC avoids parameter tuning and better matches the actual system. [ABSTRACT FROM AUTHOR]

The problem addressed in this work is the enhancement of energy efficiency in buildings, primarily focused on heating control. The paper highlights the importance of addressing this challenge, given that HVAC systems are responsible for over 60% of total energy consumption in buildings, making them the largest energy consumers in both the residential and nonresidential sectors. This paper presents a methodology for developing a model predictive control (MPC) system based on artificial neural network (ANN) models, with the aim of improving the energy efficiency of buildings. This approach involves the creation of a training dataset using dynamic thermal simulation tools, enabling the learning of the ANN model that is responsible for predicting future states within an MPC optimization loop. Additionally, the implementation leverages advanced technologies such as Field-Programmable Gate Arrays (FPGAs), specifically the PYNQ board, Wi-Fi modules, and MQTT communication to enhance scalability, deployment, and adaptability. The study evaluates the proposed Neural Network Model Predictive Control (NNMPC) strategy implemented on the PYNQ, reporting a substantial reduction of 38.53% in heating energy consumption in buildings compared to a basic On/Off control approach for 24-hour simulation. This outcome underscores the solutions significant potential for energy savings in building heating systems. [ABSTRACT FROM AUTHOR]

District energy systems through fourth and fifth generations have shown great promises to help integrate renewable energy sources at large scale and to decarbonize the built environment. However, older generations using boilers represent a large proportion of systems currently in operation, and significantly contribute to greenhouse gas (GHG) emissions. Moreover, operational data has become increasingly available for these older generation systems and represents a suitable opportunity for Artificial Intelligence (AI)-based modelling and advanced controls. In this context, model predictive control (MPC) has appeared as a powerful control solution; however, field implementation remains relatively scarce. This paper aims to develop an AI-based MPC strategy for multi-boiler district heating systems. It relies on heating load forecasting machine learning models and data-driven boiler performance curves to optimize boiler thermal outputs. This strategy was implemented in two Canadian demonstration sites and showed GHG emissions reductions of 1.3 % and 2.8 %. Although relatively modest, statistical analysis confirmed the realization of these savings. In absolute terms, the strategy helped avoid 45 t and 77 t CO2eq emissions and save $10,268 and $19,975 CAD during the testing period of 2–3 months. The model and strategy developed in this work could be easily scalable to similar district heating systems. • AI predictive control was developed for multi-boiler district heating systems. • The strategy relies on data-driven load forecasting and boiler performance curves. • The strategy was implemented in two Canadian demonstration sites. • Natural gas boiler consumption was reduced by 1.3 and 2.8 %. • GHG emissions, energy costs were reduced by 45–77 t CO2 eq and $ 10,268–19,975 CAD. [ABSTRACT FROM AUTHOR]

Organic Rankine Cycle (ORC) has received its wide application in low-grade waste heat recovery (WHR) technology for its significant performance and easy access to its components. Given the highly fluctuating nature of the waste heat source, Model Predictive Control (MPC) is usually utilized to realize the reasonable adjustment of ORC based WHR systems and has performed satisfactorily. In order to apply MPC, a relatively precise model should be built up for the ORC system which ensures acceptable control performance. However, popular modeling methods based on the mechanism of the ORC establish the high-dimensional nonlinear model and suffer from computational costs when utilizing complicated nonlinear MPC. Even if linear MPC is employed for the purpose of reducing the calculation amount, the model obtained by linearization near the operating points usually makes it valid locally, thus bringing about suboptimal or unstable control performance. To address this problem, the Koopman operator is introduced for the data-driven identification and MPC of the ORC system. Koopman identification constructs a linear model in the lifted space in which the ORC system possesses nearly global linear evolution and enables the prediction of the nonlinear ORC dynamics from measurement data. In view of the possible online calculation burden increment caused by the rise of variable dimension through lifting action, the fast Koopman MPC (FKMPC) algorithm is thus proposed based on the invariance of the Hessian matrix of the optimization problem to shorten the computing time. Simulations on setpoint tracking and disturbance rejection are performed to verify the established model accuracy and the control effectiveness of the proposed strategy in comparison with other approaches. [ABSTRACT FROM AUTHOR]

This article presents the results of a 182 day implementation of a model predictive controller (MPC) in a test house in Ottawa, Canada. The MPC controls the timing of hydronic floor heating and cooling within the highly glazed house (13.5% south-facing window-to-floor ratio). The objectives of the MPC are to optimally control the on/off times of active heating and cooling to minimize energy use while also minimizing air temperature violations outside of a prescribed range. It was found that 71% of the days in the experimental period saw successful operation of the MPC towards these goals. Days that experienced overheating were found to be concentrated in January–February and days that experienced overcooling were found to be concentrated in March–May. Problematic decision making by the MPC was found to be rooted in 4 main issues: (1) global horizontal irradiance (GHI) is a bad indicator of passive solar on its own; (2) weather forecasting (specifically GHI forecasting) is critical to MPC success in highly glazed buildings; (3) the GHI history regressor used in the prediction model regularly influenced predictions illogically; and (4) the MPC lacked decision-making logic on the timing of hydronic floor cooling. • A model predictive controller was implemented in a highly glazed, full-scale, test house. • Performance of the model predictive controller was evaluated over 7 months. • The design of the MPC is simple and practical for implementation in a real building. • A recommendation is made on what a model predictive controller could look like for a highly glazed house. [ABSTRACT FROM AUTHOR]

• A grey box model of a district thermal station was established. • The feasibility of the method was verified using an existing district heating substation. • The proposed method was flexible and performed well in terms of anti-interference. • The proposed method can reduce thermal energy consumption by 7.4%. In China, the regulation of a district heating (DH) station is mainly based on weather compensation control. This control method leads to large room temperature fluctuations and high energy consumption as the regulation of supply water temperature is solely based on outdoor meteorological parameters, which the influence of pipe delay, thermal inertia of buildings, and indoor temperature were not considered. In this study, a model predictive control (MPC) method was adopted to achieve high room temperature stability and high flexibility in DH station regulation. A dynamic model of the entire DH station, which considers the thermal inertia of buildings and delay of pipe, was established and verified using actual operation data. The performance of the MPC method was then evaluated and compared with that of the conventional control method (CCM) using the dynamic system model. When thermal comfort was considered as the single objective of the MPC, the optimal control step of the DH station was 8 h. The room temperature stability improved significantly as the room temperature fluctuation amplitude reduced from 2.64 °C to 0.35 °C. When thermal comfort and energy consumption were combined by the objective function, the MPC method could reduce the system energy consumption by 7.4 %. The MPC method could help improve the stability of the pipe network and had a strong anti-interference ability. In practical applications, during the two years after renovation, energy consumption was reduced by 5.9 % and 7.9 %, respectively. Further, the thermal comfort rate of households was improved by 0.27 % and 0.03 %, with relative incomes of 0.72 yuan/m2 and 1.04 yuan/m2, respectively. [ABSTRACT FROM AUTHOR]

Organic Rankine Cycle (ORC) stands out in low-grade waste heat recovery (WHR) technology for its significant performance. Considering the system's coupling dynamics, model predictive control (MPC) has adopted its wide application in realizing reasonable adjustment of the ORC based WHR system and has been proved to be effective. MPC is usually applied under the premise of having established a relatively accurate model, thus achieving a satisfactory control performance. However, the popular first principles model of the ORC system turns out to be high-dimensional and will result in computationally costly during typical nonlinear MPC adoption. While model linearization around operating points would enable the employment of linear MPC and reduce online calculation amount to some extent, the obtained local model loses global validity, leading to possible unstable control performance. To address these problems, a practical input–output data-driven quasi-linear parameter varying (QLPV) model is constructed for the ORC based WHR system by introducing the Koopman operator to ensure the global control effect. The corresponding MPC algorithm is thus presented via QLPV representation which solves the constructed constrained optimization problem iteratively in the form of a series of quadratic programming (QP) problems at each time step. Moreover, considering the possible lack of adequate training data covering the important dynamics of the ORC based WHR systems in practical application, an online updating mechanism that involves recursive equations is proposed to realize prediction accuracy improvement. Simulations on prediction, setpoint tracking and disturbance rejection are performed to verify the established model accuracy and the control effectiveness of the proposed strategy. • The practical data-driven QLPV model and MPC algorithm is proposed for ORC system. • QLPV model outdoes those models built up through local linearization. • QLPV model update mechanism is presented considering lack of training data. • QLPV-IMPC strategy has shown its advantages compared with NMPC algorithm. [ABSTRACT FROM AUTHOR]

The capability to dynamically plan, predict, and control indoor conditioning allows to adapt to individual preferences of inhabitants and enables demand side management. While former mainly improves thermal comfort of inhabitants so does latter unlock ecological and financial opportunities mostly for energy utilities. Commonly, dynamic indoor conditioning is based on piece-wise constant indoor temperature constraints. This paper's contribution is the presentation of additional constraints: in particular ones expressed relative to the nominal behavior of a hydronic heating system. This allows to simultaneously harness the relevant process variables in particular during the pre-loading and post-loading phases of a load reduction. The findings are based on data sets acquired on 10 inhabited, residential buildings in Stockholm over a whole year. One of the findings is that building models need to be adaptable if predictive control is applied in practice. This adaptability is assured by a novel concept, i.e. a so called model manager on which the control is relying for the selection of the most accurate model. Centralized optimal control of buildings connected to a district heating network is challenging in practice due to a high computational load. In order to reduce it, the herein presented method elaborates plans only every hour instead of at every control step for optimal control. Since these plans cannot be optimal due to the lack of regular update a hitherto unknown cascaded control logic has been developed that corrects planning errors and other disturbances. Capabilities are demonstrated and compared to conventional controllers in dynamic simulations of a multi-zoned building. The herein presented method is to our knowledge the first to provide all flexibility desired by energy utilities and inhabitants alike through harnessing the consequences of transitions. [ABSTRACT FROM AUTHOR]

This paper aims to quantify the performance of Model Predictive Control (MPC) for a typical commercial building heating, ventilation and air conditioning (HVAC) system across a wide range of climate and weather conditions. The motivation of the study comes from the fact that although there is a large body of work on MPC for HVAC systems, there is a lack of studies that examine the range of possible performance of MPC, in terms of both energy savings and maintaining indoor climate (temperature and humidity) as a function of outdoor weather. A challenge in conducting such a study is developing an MPC controller that can be used in a wide range of weather. The root cause of this challenge is the need for a tractable cooling and dehumidification coil model that can be used by the MPC controller, since the coil may operate in quite distinct modes depending on weather. We present such an MPC controller, and then leverage it to conduct an extensive simulation campaign for fourteen climate zones in the United States and four weather conditions (winter, spring, summer, and fall) in each climate zone. The performance of the proposed controller is compared with not only a rule-based baseline controller but also with a simpler MPC controller that ignores humidity and latent heat considerations. There are several results the arise from this comparative study. One such result is that energy savings from MPC over baseline can vary dramatically based on climate and season. Another is that the effect of ignoring humidity in the MPC formulation can lead to poor indoor humidity control more in milder weather rather than in hot weather. The results from this study can help practitioners and researchers assess costs and benefits of proposed MPC formulations for HVAC control. • Performance of two MPC controllers for HVAC systems is studied. • The study spans a wide range of climate and weather conditions, a first of its kind. • MPC controller can perform poorly unless latent heat is considered in design. • Poor performance occurs typically in moist and marine climate zones. [ABSTRACT FROM AUTHOR]

District heating systems may support an increased penetration of stochastic renewable energy technologies and a reduction in centralized combined heat and power plants to reduce carbon dioxide emissions. Ultra low temperature district heating minimizes transport heat losses while enabling the utilization of low-grade surplus heat. Local heat booster substations can heat water to useable temperatures using a heat pump and a hot water tank for storage and flexible operation. This paper proposes a hybrid model predictive control strategy in which an existing heat booster substation is modelled and its charging schedule optimized in real-time over a 24-h forecasted prediction horizon. This enables load shifting whereby scheduling of the heat pump minimizes operation costs. The realisation of energy flexibility can support greater utilization of renewable energy sources and surplus heat in energy supply systems to reduce primary energy consumption. The linear hybrid model predictive controller was successfully implemented in a real 22-flat multifamily building in Copenhagen to verify the control strategy. A comparison of the proposed model predictive control scheduling to the standard rule-based control showed average savings of 23 % on the electricity costs. • Modelling of a heat booster substation (HBS) for a cost-effective optimal operation. • Field test of a hybrid Economic MPC (EMPC) in the HBS with real-time feed-back. • Average saving of 23 % by comparing the proposed hybrid EMPC to a rule-based control. • Load shifting of EMPC for HBS enabling renewables integration in ULTDH. [ABSTRACT FROM AUTHOR]