Thermal Load Prediction Research Articles

Experimental study of performance of building heating system based on HPS and equipped with thermal potential recovery system and heat load prediction

The results of experimental studies of the heating system performance of a building located in the area of temperate continental climate with temperatures in the heating period down to −30 °C are discussed. The studied heating system comprises an air–liquid heat pump, a liquid–liquid heat pump with a ground circuit, and an electric boiler for peak demands.The paper presents the results of the influence of the ground thermal potential recovery system on the liquid–liquid heat pump performance. Also the influence of the used method of heat load prediction on the system efficiency and indoor air temperature stability is shown.Implementation of the thermal potential recovery system allows to increase the average value of the transformation coefficient for the liquid–liquid heat pump from 2.7 to 2.9 during three years of operation.Heat load prediction leads to stable indoor air temperature in the range of 22 ± 1 °C. Also it allows to level out the dependence of indoor temperature fluctuations on outdoor air temperature fluctuations.

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.

Multi-objective optimization of buildings in urban scale for early stage planning and parametric design

Passive design of buildings in early design stage is essential for energy efficiency and sustainability. In this study, a streamlined tool for parametric study and passive design of buildings in urban scale considering thermal loads and natural lighting performance is presented, which markedly enhances the efficiency of early-stage urban planning. The tool is comprised of three sub-models, including (1) degree-month based annual building thermal load prediction model, (2) simplified climate modification model that takes into account the effects of building layout on solar heat gain and urban heat island, (3) multi-objective optimization algorithm NSGA-II, which optimizes both energy-saving and natural lighting performance. Model validation results show that this tool has an average error of 11.44% and 15.82% in simulating building thermal load compared with Dragonfly and CitySim, and has an average error of 9.9% in predicting average daylight factor compared with Radiance. Multiple optimization scenarios and sensitivity analysis in 10 typical climatic cities were conducted. It is found that in cold and hot climatic cities, it is more beneficial to set energy target as the optimization objective, meanwhile, establishing natural lighting as a constraint, rather than solely minimizing thermal loads. In contrast, the inherently lower thermal loads of buildings in milder climatic cities help to achieve both energy-saving and natural lighting goals, and it is also found that the increased thermal loss due to a higher window-to-wall ratio can be offset by employing thicker wall and roof insulation or by improving building layouts.

Thermal Load Prediction in Residential Buildings Using Interpretable Classification

Energy efficiency is a critical aspect of engineering due to the associated monetary and environmental benefits it can bring. One aspect in particular, namely, the prediction of heating and cooling loads, plays a significant role in reducing energy use costs and in minimising the risks associated with climate change. Recently, data-driven approaches, such as artificial intelligence (AI) and machine learning (ML), have provided cost-effective and high-quality solutions for the prediction of heating and cooling loads. However, few studies have focused on interpretable classifiers that can generate not only reliable predictive systems but are also easy to understand for the stakeholders. This research investigates the applicability of ML techniques (classification) in the prediction of the heating and cooling loads of residential buildings using a dataset consisting of various variables such as roof area, building height, orientation, surface area, wall area, and glassing area distribution. Specifically, we sought to determine whether models that derive rules are competitive in terms of performance when compared with other classification techniques for assessing the energy efficiency of buildings, in particular the associated heating and cooling loads. To achieve this aim, several ML techniques including k-nearest neighbor (kNN), Decision Tree (DT)-C4.5, naive Bayes (NB), Neural Network (Nnet), Support Vector Machine (SVM), and Rule Induction (RI)- Repeated Incremental Pruning to Produce Error (RIPPER) were modelled and then evaluated based on residential data using a range of model evaluation parameters such as recall, precision, and accuracy. The results show that most classification techniques generate models with good predictive power with respect to the heating or cooling loads, with better results achieved with interpretable classifiers such as Rule Induction (RI), and Decision Trees (DT).

Computational Fluid Dynamics Prediction of External Thermal Loads on Film-Cooled Gas Turbine Vanes: A Validation of Reynolds-Averaged Navier–Stokes Transition Models and Scale-Resolving Simulations for the VKI LS-94 Test Case

Given the increasing role of computational fluid dynamics (CFD) simulations in the aerothermal design of gas turbine vanes and blades, their rigorous validation is becoming more and more important. This article exploits an experimental database obtained by the von Karman Institute (VKI) for Fluid Dynamics for the LS-94 test case. This represents a film-cooled transonic turbine vane, investigated in a five-vane linear cascade configuration under engine-like conditions in terms of the Reynolds number and Mach number. The experimental characterization included inlet freestream turbulence measured with hot-wire anemometry, aerodynamic performance assessed with a three-hole pressure probe in the downstream section, and vane convective heat transfer coefficient distribution determined with thin-film thermometers. The test matrix included cases without any film-cooling injection, pressure-side injection, and suction-side injection. The CFD simulations were carried out in Ansys Fluent, considering the impact of mesh sizing and steady-state Reynolds-Averaged Navier-Stokes (RANS) transition modelling, as well as more accurate transient scale-resolving simulations. This work provides insight into the advantages and drawbacks of such approaches for gas turbine hot-gas path designers.

Ensemble learning for predicting average thermal extraction load of a hydrothermal geothermal field: A case study in Guanzhong Basin, China

Accurate prediction of the average thermal extraction load (ATEL) in hydrothermal heating systems optimizes energy recovery, though numerical models are constrained by modeling accuracy and computational time costs. Data-driven modeling can efficiently screen alternative production plans and enhance daily operational decisions. However, ensemble learning applications for forecasting ATEL lack comprehensive comparisons and appropriate algorithm selection. This study evaluates six ensemble learning algorithms: random forest (RF), adaptive boosting (AdaBoost), gradient boosting decision trees (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightBGM) and categorical boosting (CatBoost) on 1159 data points from a homogeneous thermal-hydraulic coupling numerical model. The superior performance of the CatBoost is characterized by the highest accuracy (with the lowest root mean square error (RMSE) of 150.6 kW–170.7 kW, minimal mean absolute percentage error (MAPE) of 0.66%–0.73%, and exceptional R2 values of 0.999), commendable stability, reasonable computational cost and low uncertainty. LightGBM, XGBoost, and GBDT are identified as suitable substitutes, whereas RF is a possible substitute. Moreover, the production rate is identified as the most significant factor contributing to the ATEL, followed by the well spacing and injection temperature. While perforation depth has the least influence on heat recovery.

A Review of Cooling and Heating Loads Predictions of Residential Buildings Using Data-Driven Techniques

Energy efficiency is currently a hot topic in engineering due to the monetary and environmental benefits it brings. One aspect of energy efficiency in particular, the prediction of thermal loads (specifically heating and cooling), plays a significant role in reducing the costs associated with energy use and in minimising the risks associated with climate change. Recently, data-driven approaches, such as artificial intelligence (AI) and machine learning (ML) techniques, have provided cost-effective and high-quality solutions for solving energy efficiency problems. This research investigates various ML methods for predicting energy efficiency in buildings, with a particular emphasis on heating and cooling loads. The review includes many ML techniques, including ensemble learning, support vector machines (SVM), artificial neural networks (ANN), statistical models, and probabilistic models. Existing studies are analysed and compared in terms of new criteria, including the datasets used, the associated platforms, and, more importantly, the interpretability of the models generated. The results show that, despite the problem under investigation being studied using a range of ML techniques, few have focused on developing interpretable classifiers that can be exploited by stakeholders to support the design of energy-efficient residential buildings for climate impact minimisation. Further research in this area is required.

Seasonal performance of an energy pile heat pump system and prediction of building thermal load

Ground source heat pumps (GSHPs) are widely used for cooling and heating buildings, which can reduce the burning of fossil fuels and contribute to the reduction of carbon emissions. Energy piles, as underground heat exchangers, have advantages over conventional buried borehole pipes, saving drilling costs and accelerating the construction period. However, few studies have focused on the actual operation of an energy pile heat pump system, and the prediction of the building thermal load can reduce the instrument arrangement and test recording time. Hence, a heat pump system was built in this study to serve a single lab using energy piles with a cap structure as the energy supply, and summer cooling and winter heating tests were conducted. Under the test conditions in this study, the GSHP had a faster start-up speed and less power consumption than the air source heat pump (ASHP), and the average coefficient of performance (COP) of the energy pile heat pump system was 3.78, which is 40.0 ∼ 51.2% higher than that of the corresponding ASHP (the normative reference COP is 2.50 ∼ 2.70). The methods of higher-order surface fitting and an artificial neural network model were both suitable for predicting building thermal load.

Surrogate modeling for long-term and high-resolution prediction of building thermal load with a metric-optimized KNN algorithm

• Surrogate modeling, to predict all-year hourly building loads in pre-design stage. • Load database contains 16384 models with crucial features and thermal load data. • Manhattan distance is the optimal metric when performing KNN in load prediction. • Propose thermal load prediction process at district level and apply it to a case. During the pre-design stage of buildings, reliable long-term prediction of thermal loads is significant for cooling/heating system configuration and efficient operation. This paper proposes a surrogate modeling method to predict all-year hourly cooling/heating loads in high resolution for retail, hotel, and office buildings. 16 384 surrogate models are simulated in EnergyPlus to generate the load database, which contains 7 crucial building features as inputs and hourly loads as outputs. K-nearest-neighbors (KNN) is chosen as the data-driven algorithm to approximate the surrogates for load prediction. With test samples from the database, performances of five different spatial metrics for KNN are evaluated and optimized. Results show that the Manhattan distance is the optimal metric with the highest efficient hour rates of 93.57% and 97.14% for cooling and heating loads in office buildings. The method is verified by predicting the thermal loads of a given district in Shanghai, China. The mean absolute percentage errors (MAPE) are 5.26% and 6.88% for cooling/heating loads, respectively, and 5.63% for the annual thermal loads. The proposed surrogate modeling method meets the precision requirement of engineering in the building pre-design stage and achieves the fast prediction of all-year hourly thermal loads at the district level. As a data-driven approximation, it does not require as much detailed building information as the commonly used physics-based methods. And by pre-simulation of sufficient prototypical models, the method overcomes the gaps of data missing in current data-driven methods.

Thermal Load Model of a Proportional Solenoid Valve Based on Random Load Conditions

Drastic changes in the random load of an electromechanical system bring about a reliability problem for the proportional solenoid valve based on a thermal effect. In order to accurately and effectively express the thermal load of a proportional solenoid valve under random load conditions and to meet the requirements of online acquisition, adaptive anomaly detection, and the missing substitution of thermal load data, a thermal load prediction model based on the Kalman filter algorithm is proposed. Taking the compound operation process of an excavator as the object and based on the field testing of an excavator and the independent testing experiment of a proportional solenoid valve in a non-installed state, a method of obtaining historical samples of the proportional solenoid valve’s power and thermal load is given. The k-means clustering algorithm is used to cluster the historical samples of the power and thermal load corresponding to the working posture of a multi-tool excavator. The Grubbs criterion is used to eliminate the outliers in the clustering samples, and unbiased estimation is performed on the clustering samples to obtain the prediction model. The results show that the cross-validation of the sample data under the specific sample characteristics of the thermal load model was carried out. Compared with other methods, the prediction accuracy of the thermal load model based on the Kalman filter is higher, the adaptability is strong, and the maximum prediction deviation percentage is stable within 5%. This study has value as a reference for random cycle thermal load analyses of low-frequency electromechanical products.

Prediction and Optimization of Thermal Loads in Buildings with Different Shapes by Neural Networks and Recent Finite Difference Methods

This study aimed to estimate the heating load (HL) and the cooling load (CL) of a residential building using neural networks and to simulate the thermal behavior of a four-layered wall with different orientations. The neural network models were developed and tested using Multi-Layer Perceptron (MLP) and Radial Basis (RB) networks with three algorithms, namely the Levenberg-Marquardt (LM), the Scaled Conjugate Gradient (SCG), and the Radial Basis Function (RB). To generate the data, 624 models were used, including six building shapes, four orientations, five glazing areas, and five ways of distributing glazing. The LM model showed the best accuracy compared to the experimental data. The L-shape facing south with windows on the east and south sides and a 20% window area was found to be the best shape for balancing the lighting and ventilation requirements with the heating and cooling loads near the mean value. The heating and cooling loads for this shape were 22.5 kWh and 24.5 kWh, respectively. The simulation part used the LH algorithm coded in MATLAB to analyze the temperature and heat transfer across the wall layers and the effect of solar radiation. The maximum and minimum percentage differences obtained by HAP are 10.7% and 2.7%, respectively. The results showed that the insulation layer and the wall orientation were important factors for optimizing the thermal comfort of a building. This study demonstrated the effectiveness of neural networks and simulation methods for building energy analysis.

Temporal graph attention network for building thermal load prediction

Machine learning models have seen widespread application in predicting building thermal loads. Yet, these existing models generally predict the thermal load for a single thermal zone or an entire building, overlooking buildings with multiple thermal zones. Moreover, interactions between thermal zones are seldom accounted for these methods. Addressing this gap, our study introduces a graph-based approach that employs a Temporal Graph Attention Network (TGAT) model for building thermal load prediction. This model accommodates both the spatial and temporal dependence across different thermal zones. Specifically, the TGAT model initially utilizes graph attention neural networks (GATs) to identify the spatial relationships between thermal zones. The spatial representation thus obtained is subsequently processed via Gated Recurrent Units (GRUs) to ascertain temporal dependencies. Lastly, a fully connected layer is engaged to decode the representation for the final output. Our model incorporates 11 features as inputs to depict the characteristics of outdoor environment and thermal zones. Testing the proposed TGAT model in a small office virtual environment for heating load prediction revealed that a configuration of four GAT attention heads achieved optimal result for predicting the ensuing hour’s heating load with a six-hour lookback. Additionally, we scrutinized and discussed the efficacy of feature selection and the GAT module. This research thus establishes a novel groundwork for forthcoming studies on zone-based building energy management optimization.

Integrative soft computing approaches for optimizing thermal energy performance in residential buildings.

As is known, early prediction of thermal load in buildings can give valuable insight to engineers and energy experts in order to optimize the building design. Although different machine learning models have been promisingly employed for this problem, newer sophisticated techniques still require proper attention. This study aims at introducing novel hybrid algorithms for estimating building thermal load. The predictive models are artificial neural networks exposed to five optimizer algorithms, namely Archimedes optimization algorithm (AOA), Beluga whale optimization (BWO), forensic-based investigation (FBI), snake optimizer (SO), and transient search algorithm (TSO), for attaining optimal trainings. These five integrations aim at predicting the annual thermal energy demand. The accuracy of the models is broadly assessed using mean absolute percentage error (MAPE), root mean square error (RMSE), and coefficient of determination (R2) indicators and a ranking system is accordingly developed. As the MAPE and R2 reported, all obtained relative errors were below 5% and correlations were above 92% which confirm the general acceptability of the results and all used models. While the models exhibited different performances in training and testing stages, referring to the overall results, the BWO emerged as the most accurate algorithm, followed by the AOA and SO simultaneously in the second position, the FBI as the third, and TSO as the fourth accurate model. Mean absolute error (MAPE) and Considering the wide variety of artificial intelligence techniques that are used nowadays, the findings of this research may shed light on the selection of proper techniques for reliable energy performance analysis in complex buildings.

Simplified heat-transfer calculation for the trapezoidal cross-section wall in Gobi solar greenhouses and greenhouse thermal-load prediction

Simplified heat-transfer calculation for the trapezoidal cross-section wall in Gobi solar greenhouses and greenhouse thermal-load prediction

Adaptive thermal load prediction in residential buildings using artificial neural networks

Accurate prediction of thermal load in buildings is essential for efficient energy planning. In this study, we investigate the application of Artificial Neural Networks (ANNs) to predict thermal load and indoor temperature evolution in residential buildings. We propose a flexible and adaptive model that is retrained on a daily basis to deliver updated hour-by-hour predictions for the following 24 h. To strike a balance between prediction accuracy and computational efficiency, we employ various design choices, including feature selection using Pearson's correlation coefficient (PCC), dynamic architecture, down-sampling, and specific state resetting and weight initialization. The results demonstrate the efficacy of our proposed model, as indicated by low root mean square error (RMSE) and mean bias error (MBE) values. For zone temperature, the average RMSE and MBE are 0.3 °C and 0.18 °C (summer) and 0.5 °C and 0.2 °C (winter), respectively. Furthermore, the average RMSE and MBE for thermal load predictions are 12 W/m2 and -2.4 W/m2 (winter) and 10 W/m2 and -0.6 W/m2 (summer), respectively. These performance metrics establish our model as a valuable tool for optimizing heating and cooling systems, resulting in energy savings and cost reductions. Our findings emphasize the potential of ANNs for precise thermal load and indoor temperature predictions, offering practical implications for building operators, engineers, and researchers involved in energy management.

Machine Learning Method Based on Symbiotic Organism Search Algorithm for Thermal Load Prediction in Buildings

This research investigates the efficacy of a proposed novel machine learning tool for the optimal simulation of building thermal load. By applying a symbiotic organism search (SOS) metaheuristic algorithm to a well-known model, namely an artificial neural network (ANN), a sophisticated optimizable methodology is developed for estimating heating load (HL) in residential buildings. Moreover, the SOS is comparatively assessed with several identical optimizers, namely political optimizer, heap-based optimizer, Henry gas solubility optimization, atom search optimization, stochastic fractal search, and cuttlefish optimization algorithm. The dataset used for this study lists the HL versus the corresponding building conditions and the model tries to disclose the nonlinear relationship between them. For each mode, an extensive trial and error effort revealed the most suitable configuration. Examining the accuracy of prediction showed that the SOS–ANN hybrid is a strong predictor as its results are in great harmony with expectations. Moreover, to verify the results of the SOS–ANN, it was compared with several benchmark models employed in this study, as well as in the earlier literature. This comparison revealed the superior accuracy of the suggested model. Hence, utilizing the SOS–ANN is highly recommended to energy-building experts for attaining an early estimation of the HL from a designed building’s characteristics.

A decomposition-ensemble prediction method of building thermal load with enhanced electrical load information

Accurate thermal load forecasting is essential for intelligent building energy management. The lack of meteorological data and the variability of thermal load often result in low prediction accuracy of thermal load. This paper proposes a novel decomposition-ensemble prediction model to predict thermal load accurately when the meteorological parameters are missing. The model combines the variational mode decomposition technology and the gated recurrent unit algorithm, in which the information of electrical load is extracted and considered to assist the thermal load prediction. The two datasets of an office building and a museum building are employed to train and test the model. The case results demonstrate that the variational mode decomposition method can effectively decrease the prediction uncertainty for thermal load series. The advantages of the proposed model have better prediction performance than the basic and hybrid models.

Building thermal load prediction using deep learning method considering time-shifting correlation in feature variables

Building thermal load prediction is of great significance for energy conservation in HVAC systems. Due to the visible and complicated time delay between influencing factors and building thermal load, ensuring the consistency of time variation between feature variables and load in prediction is challenging. In this paper, the time-shifting correlation of feature variables to building thermal load was quantified, and the bidirectional network structure was introduced to develop prediction models to solve the forecasting delay issue of algorithms. The performance of four load prediction models was compared using measured data obtained from a railway station in Tibet to discuss the superiority of bidirectional network structure in building thermal load prediction based on deep learning method. The results show that long short-term memory (LSTM) and gated recurrent unit (GRU) models have significant prediction delays, while bidirectional long short-term memory (Bi-LSTM) and bidirectional gated recurrent unit (Bi-GRU) models do not, which indicates the bidirectional network structural characteristics can better capture the trend of thermal load variations in time. Moreover, Bi-GRU has the best performance, with all prediction relative errors being less than 1%, which demonstrates that Bi-GRU has a great advantage in building thermal load prediction. In addition, the performance of seven feature variable sets was compared to further demonstrate the convincingness of time-shifting correlation analysis between parameters. The results show that the proposed quantitative time-shifting correlation analysis method can evidently solve the time-dependent problem of feature variables in building thermal prediction. Finally, we also discussed the impact of prediction horizon on prediction accuracy, which illustrates the 15-min forecast interval is more suitable for ultra-short-term building load prediction.

Room thermal load prediction based on analytic hierarchy process and back-propagation neural networks

Room thermal load prediction based on analytic hierarchy process and back-propagation neural networks

Hybrid model for exhaust systems in vehicle thermal management simulations

AbstractUsing Vehicle Thermal Management (VTM) simulations to predict the thermal load experienced by components is a popular method within the automotive industry. The VTM simulation approach is fast becoming equivalent to conducting thermal load tests with prototypes for vehicles powered by internal combustion engines. This is especially true in the early development phase of the vehicle. The accuracy of the VTM simulations plays a pivotal role at them being accepted as an eventual replacement for physical testing. The correct prediction of thermal loads in VTM simulations depends on a multitude of different parameters, but the modelling of the exhaust system plays a central role in it. This is because the exhaust gas, and with it the exhaust system, is the primary source of heat in a vehicle powered by an internal combustion engine. The developed approach not only needs to be accurate but also modular enough to allow for different exhaust configurations to be tested. It also needs to be capable of integration into any VTM simulation workflow while maintaining an industrially acceptable turnaround time. This paper explores a new methodology to achieve these requirements. A 1D/3D hybrid approach to exhaust system modelling is presented. In this, the components that have an enthalpy change of the exhaust gas, such as the turbocharger, have been modelled as 1D and simple components such as pipes have been modelled in 3D. This has the advantage of combining the speed of 1D simulations with the spatial accuracy of 3D simulations. The method uses a unique three-code co-simulation technique for full vehicle VTM simulations. The coupling is between a 3D CFD software, a 1D simulation tool, and a Finite Element based thermal solver. The methodology was validated against experimental data for multiple loadcases. The results show good agreement with experiment within acceptable tolerances.