Heating Control System Research Articles

A trustworthy reinforcement learning framework for autonomous control of a large-scale complex heating system: Simulation and field implementation

Traditional control approaches heavily rely on hard-coded expert knowledge, complicating the development of optimal control solutions as system complexity increases. Deep Reinforcement Learning (DRL) offers a self-learning control solution, proving advantageous in scenarios where crafting expert-based solutions becomes intricate. This study investigates the potential of DRL for supervisory control in a unique and complex heating system within a large-scale university building. The DRL framework aims to minimize energy costs while ensuring occupant comfort. However, the trial-and-error learning approach of DRL raises concerns about the trustworthiness of executed actions, hindering practical implementation. To address this, the study incorporates action masking, enabling the integration of hard constraints into DRL to enhance user trust. Maskable Proximal Policy Optimization (MPPO) is evaluated alongside standard Proximal Policy Optimization (PPO) and Soft Actor–Critic (SAC). Simulation results reveal that MPPO achieves comparable energy savings (8% relative to the baseline control) with fewer comfort violations than other methods. Therefore, it is selected among the candidate algorithms and experimentally implemented in the university building over one week. Experimental findings demonstrate that MPPO reduces energy costs while maintaining occupant comfort, resulting in a 36% saving compared to a historical day with similar weather conditions. These results underscore the proactive decision-making capability of DRL, establishing its viability for autonomous control in complex energy systems.

Fixed-time active fault-tolerant control for a class of nonlinear systems with intermittent faults and input saturation

In this paper, the problem of fixed-time active fault-tolerant control is studied for systems with sector-bounded nonlinearities, intermittent faults, and input saturation. Since intermittent faults appear and disappear randomly, a fixed-time scheme is considered in the active fault-tolerant control algorithm, composed of detection, isolation, estimation, and the control unit. Utilizing homogeneity-based observers, the fixed-time state estimation is available in the presence of unknown but bounded disturbances, and a fault diagnosis unit is proposed. An input saturation compensator is introduced to analyze the effect of input saturation, and its auxiliary variables are used in the reconfigurable control law. The fault-tolerant controller, which is constructed via the information provided by the fault diagnosis unit and saturation compensator, has two switching modes. As a consequence, intermittent faults are compensated via the designed active fault-tolerant control method and the system reaches practical stability with the entire convergence time bounded in a fixed time. Finally, the example of a heat control system is exploited to demonstrate the effectiveness of the developed active fault-tolerant control scheme.

Exploring the use phase social impacts of smart heating, ventilation and air conditioning control systems: A scoping review

Exploring the use phase social impacts of smart heating, ventilation and air conditioning control systems: A scoping review

Dyna-PINN: Physics-informed deep dyna-q reinforcement learning for intelligent control of building heating system in low-diversity training data regimes

This paper introduces Dyna-PINN, a novel physics-informed Deep Dyna-Q (DDQ) reinforcement learning (RL) approach, designed to address the data-intensive training requirements and model-agnostic nature of the conventional model-free RL methods. The DDQ approach blends model-based and model-free elements to enhance both learning and decision-making processes. By utilizing a physics-informed neural network (PINN) based model, our method enriches the learning process with physical information, enhancing the agent's planning capabilities and leading to faster learning compared to conventional model-free RL methods like Deep Q-Network (DQN) in scenarios with low-diversity training data availability. Our results demonstrate that Dyna-PINN has 50% greater sample efficiency than DQN and outperforms rule-based control in terms of thermal discomfort. Due to physics incorporation, the Dyna-PINN implements a more logical and interpretable control policy. It shows consistently good performance compared to all control variants across low-diversity data scenarios, i.e., 6 weeks of building data, and in higher-diversity data regimes, i.e., 6 months of building energy data, demonstrating the value of physics incorporation into the RL training. Additionally, we present two other DDQ-based techniques, RC-DDQ and NN-DDQ, exploring the synergy between neural networks and physical data in intelligent control designs for building energy systems. Rigorous controller testing is performed using the Building Optimization and Testing Framework (BOPTEST), a high-fidelity simulator that closely represents a real building's operation. Through comprehensive comparisons and realistic simulations, our study underscores the effectiveness of incorporating physics-informed approaches into RL-based control strategies, paving the way for more efficient and robust building energy management systems.

Enhanced temperature regulation in compound heating systems: Leveraging guided policy search and model predictive control

With the widespread application of solar-energy-air-source heat pump composite heating, optimizing the regulation of air temperature in heating zones, enhancing system thermal comfort, and reducing energy consumption have become crucial. To ensure residential comfort while minimizing HVAC system energy costs, a control method combining guided policy search (GPS) with model predictive control (MPC) is proposed for composite heating systems. This method replaces state estimation in MPC with policy search and substitutes the offline trajectory optimization used in guided policy search with MPC, thus integrating the strengths of both approaches. This strategy not only improves control precision but also reduces energy consumption and enhances system robustness. The GPS-MPC control algorithm was validated against reinforcement learning and MPC. A simulation model was developed and validated on a real physical platform. Simulations compared the control effects of on-off control and MPC on pump frequency, flow rate, stratified thermal storage tank temperature, component, and system energy consumption. The data results indicate that the GPS-MPC algorithm offers superior predictive accuracy, efficiency, and robustness in composite heating control systems compared to conventional methods. Under the GPS-MPC strategy, indoor temperature fluctuations, pump frequency response speed, and energy consumption were significantly improved, with temperature fluctuation limited to only 0.8 °C, and the system achieving energy savings of over 12 %.

Thermal comfort driven space heating control via hierarchical state machine strategy interacting with multiphysics simulation

Improving energy efficiency of space heating systems is as important as adopting emerging clean energy technologies for decarbonising heating. While exploring strategies for reducing energy consumption, the space occupants’ expectations to thermal comfort should not be sacrificed. Conventional space heating control systems are often developed under the hypothesis of uniform temperature distribution in a bounded space and the heating system controller is primarily operated based on the information from the temperature sensor at a fixed position. That is, the measured temperature does not represent the temperature surrounding the occupant’s location. This paper explores a new control method; the heating system control is led by the occupant’s thermal experience, expressed by the temperature surrounding the occupant. A new hierarchical state machine control strategy interacting with a Multiphysics simulation model to dynamically estimate the temperature at the point of the occupant is developed. The accuracy of the Multiphysics simulation model is verified based on the corresponding experimental test room data. A co-simulation platform is built to allow COMSOL and SIMULINK to communicate in real-time for control strategy realisation. The experimental results show that the maximum and average mean square error between the measured and simulated temperature under different operating modes is less than 0.21℃ and 0.06℃, respectively. The simulation results indicate that energy consumption could be potentially reduced around 20% with an improved thermal comfort of the occupant by 30%, compared with the conventional fixed sensor control method.

System Design, Modelling, Energy Analysis, and Industrial Applications of Ohmic Heating Technology

AbstractThe food and drink industry is gradually gaining sustainability pressure due to the high carbon emissions rate and limited availability of traditional energy supplies. Thus, it is essential to consider energy efficiency and decarbonisation of the ever-expanding food and drink industry. Ohmic heating technology is a highly energy-efficient and quick method of electrically heating food products. It uses the resistance of the food material to convert electrical energy into heat. However, the heating rate of ohmic heating depends on several process variables, such as electrical conductivity, voltage gradient, and voltage frequency. Therefore, it is important to study and understand the effects of these process variables to optimise the modelling and control of ohmic heating systems to obtain desirable output from food products. Hence, this study presents a review of the effects of process parameters on nutritional and organoleptic qualities, product yields, and energy efficiency of ohmic heating, along with its various industrial applications. The technology shows promising advancements in colour and nutritional quality, increasing energy efficiency, shorter processing times, and higher product yields compared to conventional heating processes.

Towards various occupants with different thermal comfort requirements: A deep reinforcement learning approach combined with a dynamic PMV model for HVAC control in buildings

Reinforcement learning (RL) has great potential in achieving energy-efficient, comfortable and intelligent control of heating, ventilation and air conditioning (HVAC) systems. Although research on RL-based HVAC control has attracted increasing interest, current studies generally use simple building simulation as the environment to train agents, and the definition of thermal comfort is limited to a wide temperature range, which cannot meet the different thermal comfort requirements of various occupants. This study proposes a deep reinforcement learning (DRL) control framework based on the Dueling Deep Q-network (DQN) algorithm, combined with a self-designed environmental model and reward function, for HVAC control meeting different thermal comfort requirements. Specifically, based on the theory of building thermal dynamics, a nonlinear equation modified by experimental data is used for the environmental model that reflects the actual thermal change of building. Different thermal comfort requirements are considered and analysed through a dynamic predicted mean vote (PMV) model that focuses on the metabolic rate and clothing level of occupants. By systematically exploring different heating modes for occupants and control time intervals, the proposed framework demonstrates that heating energy consumption can be reduced by 4.8%-39.58% under various conditions compared to rule-based control. In addition, the study found that the HVAC control based on DRL has greater potential in saving energy when the heating demand of building is higher. Our study is helpful for researchers to make HVAC control more energy-efficient and user-friendly with the help of artificial intelligence.

Influence of Molasses and Caesalpinia spinosa Meal Inoculums on Biogas Production from Cattle Manure

The management of organic waste through anaerobic digestion is an alternative to energy recovery. This research focused on evaluating the production of biogas with different inoculums. For this purpose, two types of systems were implemented—one used a heating system controlled by an STC-1000 thermostat, while the other used a solar heating system under a polycarbonate parabolic trough. The experiment was carried out at laboratory level with 3 L PET bottle biodigesters and the biogas produced was collected with the water displacement technique in 3 L bottles, calibrated every 50 mL, over 43 days. Inoculums of the following manure concentrations were used: water (1:5, 1:2, 1:3) mixed with Caesalpinia spinosa meal and molasses. The results determined that the thermostat-controlled heating system generated 69.6 mL/day of biogas while the other system produced 610.9 mL/day. On the other hand, the T1 treatment with a manure:water ratio of 1:5 and molasses and Caesalpinia spinosa meal inoculums in both systems had a higher average biogas volume. In terms of methane (CH₄), the highest value of 76.9% was obtained through the T1 treatment under the controlled heating system. This allows the production of biogas with a high concentration of methane, which in future applications can be utilized for residential or industrial purposes, promoting economic, social and environmental development. Since the main challenge in the production of biogas is to reduce the digestion time, which is influenced by the temperature of the site, two types of inoculums with a low cost and easy access were used.

Experimental investigation on an innovative serpentine channel‐based nanofluid cooling technology for modular lithium‐ion battery thermal management

AbstractMany nations have committed to becoming carbon neutral by 2050 as a means of addressing the global warming challenge. To achieve carbon neutrality, transportation is one of the most essential and important tasks. Energy‐efficient pure electric vehicles (EVs) and hybrid electric vehicles (HEVs) with green energy power are being developed in response to the worldwide energy and environmental crises, as the potential replacements for the current generation of combustion‐engine automobiles. EVs require batteries more than ever before. In this perspective, lithium‐ion batteries (LIBs) stand out as remarkable energy storage technologies and have been widely used due to their numerous impressive benefits. Owing to LIBs sensitivity to temperature, EVs typically use the battery thermal management system (BTMS). The working temperature span of a lithium‐ion battery in an electric car is 15°C–35°C, which is achieved by the use of a BTMS. The production of internal heat during charging and discharging also affects how well lithium‐ion batteries work. A battery heat control system is therefore required. The temperature of the LIB pack might be efficiently controlled by liquid‐cooled systems in discharge and charge scenarios. Based on Al2O3 nanofluid (NF), the current experimental study suggests a novel active cooling technology for regulating the heat produced by the 18650‐format lithium‐ion batteries. A thorough analysis is conducted on the impact of charge/discharge C‐rates, Al2O3 nanoparticle (NP) volume fractions, inflow coolant velocity, and intake liquid temperature on the thermal efficiency of the LIB pack. By incorporating aluminum oxide NPs into the water at varying volume fractions of 0.3%, 0.5%, and 1%, the LIB pack's maximum temperature was significantly reduced by 7.9%, 18.09%, and 19.56%, respectively. With increase in mass flow rate of coolant from 0.0290 to 0.5810 kg/s, the maximum temperature has been substantially reduced by 3.7%–8.6%. Results show that using higher fluid inflow temperature significantly increased both the highest experienced temperature and temperature diversity throughout the discharge operation by about, 6°C and 5°C, respectively. The outcomes of the study indicate that NFs exhibit superior cooling performance compared to conventional coolants such as water and ethylene glycol.

Optimal control for district heating systems based on imbalance analysis of pipe networks

In district heating (DH) systems, the transmission and distribution process occupy large amount of energy consumption. Hydraulic and thermal imbalance in pipe networks often lead to uneven heating, affecting user comfort and resulting in energy losses. To solve this problem, this study investigates the root causes of hydraulic and thermal imbalances in DH systems. Key factors have been identified such as initial flow rates, initial loads, and positions in the network of end-users. And the relationship between hydraulic and thermal imbalances have been analyzed. Building on this analysis, we further propose a novel strategy for differential pressure set-point adjustment. This strategy is developed based on a quantitative understanding of the flow rate changes and positions of end-users. It aims to rectify imbalances efficiently, thus enhancing the overall performance of the network. The effectiveness of the proposed strategy is validated through comparative analysis with the conventional constant differential pressure (CCDP) control strategy and an exist variable differential pressure (VDP) control strategy. Results demonstrate significant improvements in network performance, including enhanced stability and reduced energy losses. The proposed strategy reduces the average pressure difference in the main pipe by up to 10.9 %, improves the supply and return temperature difference by 10.9 %, and maintains valve openings above 70 % during peak periods. Additionally, it achieves 23.5 % energy savings in transmission and distribution, highlighting its potential to enhance system performance and efficiency. In addition, this strategy is more in line with the actual situation of end-user demand change for valve adjustment.

Design of Small-Size Lithium-Battery-Based Electromagnetic Induction Heating Control System

This paper presents the design and optimization of a small-size electromagnetic induction heating control system powered by a 3.7 V–900 mAh lithium battery and featuring an LC series resonant full-bridge inverter circuit, which can be used for small metal material heating applications, such as micro medical devices. The effects of the resonant capacitance, inductor wire diameter, heating tube material, and wall thickness were studied to maximize the heating rate of the workpiece and simultaneously reduce the temperature rise of the NMOS transistor. The optimal circuit configuration meeting the design requirements was finally identified by comparing the operational parameters and NMOS transistor loss under different circuit conditions. Validation experiments were conducted on designed electromagnetic induction smoking devices. The results indicate that under an output current of 4.6 A, the heating tube can reach the temperature target of 250 °C within 11 s, and all NMOS transistors stay below 50 °C in a 5 min heating process.

A simple load model based on hybrid mechanism and data-driven approach for district heating in building complex

Accurate prediction of heating loads in district heating systems is essential for the implementation of demand-driven heating. This work presents a novel heating load prediction model that is particularly suitable for complex multi-user buildings. The input characteristics of the model are established through the heat transfer mechanism, considering factors such as the cumulative impact of outdoor temperature and user demand (indoor temperature). The specific form of the heating load function is determined using the MLR-PSO (Multiple Linear Regression-Particle Swarm Optimization) method. Only the indoor and outdoor temperatures need to be provided for the model to calculate future heating loads. Practical engineering tests demonstrated that the model achieved normalized mean bias errors of daily loads between 4.98 % and 5.54 % across different heating seasons, with a minimum annual relative deviation of 0.75 % for annual loads. Additionally, the model helps guide the operation of heating systems. For example, during the 2021–2022 heating season, setting the target indoor temperature at 18 °C reduced weekly energy consumption by 15.3 % compared to the previous season. This approach may be employed to construct a simple load model for existing heating systems to accurately predict both short-term and long-term loads, providing valuable insights into the management and control of heating systems.

MGMI: A novel deep learning model based on short-term thermal load prediction

Accurate heat load prediction is the key to stable operation and control of district heating system. However, current heat load forecasting methods tend to treat individual buildings as isolated entities, ignoring the temporal and spatial correlation between buildings. In this paper, a heat load forecasting model is proposed which fully considers spatiotemporal information. Firstly, the spatial and temporal relationship between buildings is analyzed by statistical method. Secondly, synchronous wavelet transform (SWT) is used to denoise the heat load data to reduce the difficulty of prediction. Then, a hybrid model of multi-modal graph attention network (MG) and multi-scale Informer (MI) is constructed to capture spatiotemporal information in the data. Finally, an example of DHS in a campus in Liaoyang is analyzed. The results showed that compared with LSTM, CNN-Informer, GCN-LSTM and GAT-Informer, the MAE of MGMI on 168 h test set was reduced by 71.4%, 61.5%, 58% and 41.6%, respectively. And has the highest computational efficiency. The validity of the model is verified, which provides an important basis for the operation control of the thermal system.

A novel improved hybrid neural network for predicting heating load in airport building

In order to guarantee the efficient and energy-saving operation and control of airport heating systems, accurate heating load prediction is extremely critical. In view of the relatively little research on heating load prediction of airport buildings, besides, out of these, the fact that the actual characteristics of the airport itself and its production operations should be considered, resulting in inferior performance. Framed in this goal, Initially, the heating area is partitioned. Subsequently, a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm synthesizes it with fuzzy entropy (FE) to decompose the heating load, and the maximum information coefficient integrated with the distance correlation coefficient is utilized for the feature selection. After that, a multi-strategy improved artificial bee colony algorithm (MIABC) algorithm is used to optimize the radial basis function (RBF)neural network and the gated recurrent unit (GRU) neural network, respectively, and the two are combined to construct the hybrid prediction model of the heating load. Finally, actual heating load data is used to verify the prediction performance of the proposed model. Besides, the results show that the root mean square error (RMSE), mean absolute percentage error (MAPE), range of relative error (RRE), and mean bias error (MBE) of the proposed model are 0.465, 1.08 %, 11.42 % and 0.082, respectively, the coefficient of determination (R2) is 0.997, and the prediction accuracy reaches 98.92 %. Additionally, it has more robustness and generalization power, effectively overcoming the heating load's uncertainties and laying a foundation for controlling, monitoring, and managing the buildings' operation.

A framework for assessing the energy performance of Personalized Environmental Control Systems (PECS) for heating, cooling and ventilation

This study presents a methodological framework for assessing the energy performance of Personal Environmental Control Systems (PECS), aiming to address the existing gaps in standardized evaluation approaches. Initially, a systematic approach to delineate the distinct spatial control volumes associated with PECS is presented, facilitating the classification and comparison of various system types. Utilizing this framework, the study introduces definitions for energy efficiency and effectiveness specific to PECS, providing a procedure for the experimental assessment of the quantities related to these definitions. To validate the proposed framework, the method was applied to an illustrative case study involving two PECS under varying operational conditions – heating and cooling. The experimental results provide insights into the energy performance of the evaluated PECS, highlighting the framework's applicability and effectiveness in assessing their energy efficiency. The study underscores the importance of systematically evaluating PECS energy performance, offering a robust methodology that can guide future research and development efforts in this domain.

RANCANG BANGUN MODUL TRAINER KIT SISTEM KENDALI BERBASIS DIGITAL

A control system is defined as a combination of several components that aim to produce a system response in accordance with the desired process variables. Therefore, control systems are now widely used in various fields such as industry, automotive, public facilities, and household appliances. Especially in the industrial field, control systems play a role in improving production quality and reducing production costs. To create qualified human resources who have a good understanding in this field, universities have an important role in realizing this. One of them is Udayana University through the Electrical Engineering Study Program in the Independent Learning Curriculum in the Control System + Lab Course. This study aims to design a Digital-Based Control System + Lab Trainer Kit Module using Arduio Uno to support the process of implementing practicum in the laboratory. In the trainer kit module, there will be three systems designed, which are DC motor direction and speed control system using L298N motor driver, water heater control system using DS18B20 temperature sensor, and also water level control system using HC-SR04 ultrasonic sensor.

Predicting individual thermal preferences in an office: Assessing the performance of mixed-effects models

Multiple studies suggested that existing thermal comfort models inadequately predict occupants' individual thermal preferences. Personalised comfort models offer an alternative to conventional comfort models aiming to forecast individual's thermal preference. Implementation of these personalised models in occupant-centric control of heating, ventilation, and air-conditioning (HVAC) systems can enhance their performance.A promising technique for personalised comfort modelling is mixed-effects (ME) modelling. A ME model accounts for fixed effects, representing the trends in the general sample, and for random effects, representing variations of specific clusters in the data. In contrast to fixed-effects (FE) models, ME models can capture individual differences. However, its effectiveness in predicting occupants' thermal preferences based on field measurement data, as well as the influence of variations in ME models on prediction accuracy, remains to be thoroughly investigated.This study aims to assess the prediction accuracy of ME models in contrast to FE models using field measurement data, including thermal preference votes from 30 unique occupants. The prediction performance was evaluated across three testing scenarios, each representing a different application of the models. Furthermore, two random effect structures were tested for the ME model: an intercept-only model and an intercept and slope model. The results show that ME models, in comparison to FE models, achieve an improved prediction performance of 8.0 % on average and up to 28.4 % for individual occupants. Moreover, the addition of a random slope to the ME resulted in deteriorated predictions. Finally, occupants’ individual variations were determined with an uncertainty of 6 % after 20 observations.

Experimental and numerical analyses of water droplet condensation on photovoltaic surfaces under clean and dusty conditions

Several studies have investigated the condensation effect on dust agglomerate located on PV surfaces. The phenomena of dust condensation remain minimally addressed by researchers and not understanding well in the field of PV systems. Using rigorous experimentation, this research study explored the accumulation of dust on photovoltaic units. Furthermore, we thoroughly examined the effect of condensation on the glass surface of the unit. The formation of condensed droplets and their consequences were examined through meticulous experiments and mechanical analyses. The experiment was made possible by an ingeniously designed experimental platform, which enabled monitoring the influences of condensation on dust aggregation. The condensation process was carefully observed and analysed using microscopic techniques, while its direct impact on the dust accumulated on the PV module's surface was also examined. The experimental and numerical investigations of droplets condensation, condensing water droplets speed, condensation film thickness, and mud formation layer on the glass cover of PV module were highlighted. The droplets condensation on a glass plane (similar surface to covering glass for PV module) was recorded at different air conditions. The values of air velocity were chosen 0, 2.4, 4.1, 6.3 m/s, to provide a constant air temperature and humidity of the inflow air, the channel was supplied with heating control system. Furthermore, two dust seeding quantities 150 and 300 g were tested to provide a range for different cases for thickness of dust layers with different air velocities. According to the study, condensate time directly correlates with drop equivalent diameter, demonstrating that drop diameter increases proportionally to condensation time. It is indicated that the droplet formation and condensation period accelerated with increasing dust seeding quantity. It is noteworthy that the period of droplet formation and condensation accelerates with the increase in the amount of dust seeding. The results showed that by increasing the air speed from 2.4 m/s to 6.3 m/s, the weight of accumulated dust was decreased from 0.166 g to 0.151 g. Also, the average mud layer thickness was reduced from 27.6 μm to 25.1 μm.

Improved performance of a fiber-optic hydrogen sensor based on a controllable optical heating technology.

A novel, to the best of our knowledge, and compact fiber-optic hydrogen sensor based on light intensity demodulation and controllable optical heating technology is proposed and experimentally investigated. This system employs three photodetectors for optic signal transformation. The first PD is used to receive a little fraction of the amplified spontaneous emission (ASE) for calibration, and the second PD is utilized to detect optic signal reflected by a single mode fiber deposited with WO3-Pd2Pt-Pt composite film. The last PD is utilized to receive the optical power reflected by the short fiber Bragg grating (SFBG) with a central wavelength located in a steep wavelength range (the intensity decreases approximately linearly with the increase of the wavelength) of the ASE light source. A 980 nm laser and proportion integration differentiation (PID) controller were employed to ensure the hydrogen sensitive film working at an operating temperature of 60°C. This sensing system can display a quick response time of 0.4 s toward 10,000 ppm hydrogen in air. In addition, the detection limit of 5 ppm in air can be achieved with this sensing system. The stability of this sensor can be greatly enhanced with a controllable optical heating system, which can greatly promote its potential application in various fields.