Energy Performance Prediction Research Articles

Net zero energy analysis and energy conversion of sustainable residential building in Muscat, Oman

The building sector is the predominant consumer of primary energy globally. The building sector accounts for around 40% of global energy production.Net Zero Energy Buildings (NZEBs) are highly suggested by energy experts as an effective option to alleviate the strain on primary energy sources caused by the building sector. The disparity between energy performance predictions provided during the design phase and the actual energy performance of residential buildings is mostly attributed to a limited comprehension of the components that influence energy consumption and the constraints of whole building simulation software. The objective of this research was to perform a comparison analysis of the expected and actual energy consumption of a prototype net-zero energy house built at the University of Technology and Applied Sciences in Muscat. The Hourly Analysis Programme (HAP V4.2) was utilised to forecast the energy consumption of a Net Zero Energy Building (NZEB) at HCT, taking into account the availability of an Energy Recovery Ventilator (ERV) and the absence of an ERV. The newly built house underwent a one-month testing phase to fulfil many duties according to competition regulations. One of the main goals was to generate on-site energy through photovoltaic panels, producing an amount proportional to the energy consumed by the house. Upon comparing the actual energy consumption data with the simulated result, it was noticed that the actual energy demand of the house was around 20% lower than the prediction made by the simulation tool.

Nonlinear impedance characteristics of a gas hydraulic energy absorber and performance prediction for railway transportation of valuable goods

It is meaningful to investigate the nonlinear impedance characteristics of gas hydraulic energy absorbers (GHEAs) and their potential application in modern railway transportation of valuable goods. Fluid-mechanics-based full parametric modelling of a GHEA used in a high-speed train is performed, both simulations with the developed mathematic model and physical experiments are carried out on the static and dynamic characteristics of the GHEA. The agreement between the results of the simulation and the experiment with tolerable deviations validates the nonlinear mathematic model. Longitudinal train dynamics (LTD) modelling and simulation of a Chinese rapid freight train with the validated GHEA model or the conventional MT-2 friction-type energy absorber model is finally conducted. The results show that when the train uses GHEAs instead of MT-2 energy absorbers, the coupler forces and vehicle accelerations of the train are significantly reduced, and the vehicles operate more steadily in situations such as train startup, emergency braking and shunting operations; thus, gas hydraulic energy absorber is protective and would be alternative for equipping rapid freight trains for the transportation of valuable goods. The obtained nonlinear mathematic model in terms of the structural, oil and gas property parameters of the GHEA is useful and meaningful in further parameter design optimization of GHEAs and optimal specification of GHEAs for rapid freight trains.

Utilizing domain knowledge: Robust machine learning for building energy performance prediction with small, inconsistent datasets

Machine learning (ML) applications often require large datasets, a requirement that can pose a major challenge in fields where data is sparse or inconsistent. To address this issue, we propose a novel approach that combines prior knowledge with data-driven methods to significantly reduce data dependency. This study represents a disentangled system compositionality knowledge by the method of Component-Based Machine Learning (CBML) in the context of energy-efficient building engineering. In this way, CBML incorporates semantic domain knowledge within the structure of a data-driven model. To understand the advantage of CBML, we conducted a case experiment to assess the effectiveness of this knowledge-encoded ML approach in scenarios with sparse data input (1 % - 0.0125 % sampling rate) and several typical ML methods. Our findings reveal three key advantages of this approach over traditional ML methods: 1) It significantly improves the robustness of ML models when dealing with extremely small and inconsistent datasets; 2) It allows for efficient utilization of data from diverse record collections; 3) It can handle incomplete data while maintaining high interpretability and reducing training time. These features offer a promising solution to the challenges associated with deploying data-intensive methods and contribute to more efficient real-world data usage. Additionally, we outline four essential prerequisites to ensure the successful integration of prior knowledge and ML generalization in target scenarios and open-sourced the code and dataset for community reproduction.

FS-DBoost: cross-server energy efficiency and performance prediction in cloud based on transfer regression

FS-DBoost: cross-server energy efficiency and performance prediction in cloud based on transfer regression

Development of a rapid assessment tool for integrating thermal comfort in early design stage of energy-efficient office buildings

Comprehending and predicting energy performance of existing buildings and new constructions is crucial towards decarbonization. Instead of utilizing simulation software, multiple regression models are utilized to predict building energy consumption while ensuring indoor thermal comfort to speed up this process. However, previous predictive models prioritize reducing energy demand, with limited focus on thermal comfort. This study aims to support decision-making during retrofitting and new construction planning though developing a prediction model. An air-conditioned office building served as a reference building for simulation. 21 design parameters were analyzed, including aspects of weather, building envelope, internal loads, ventilation, and temperature settings. Stepwise regression results unveiled the crucial variables in the final model, with 8, 9, 13, and 6 variables remaining for peak cooling load, annual cooling load, overheating hours (WE), and Environmental Quality Index for thermal comfort (EQITC) in the perimeter zones, and 5, 6, 8, and 5 variables for the core zones, respectively. Furthermore, insights into the important variables regarding cooling load and thermal comfort were respectively provided. Weather- and envelope-related variables, such as cooling degree-days, global solar radiation, solar heat gain coefficient (SHGC), and U-value, have the highest impacts on cooling load. For thermal comfort, variables including temperature setpoint, occupant activity level, and factors related to window sunlight transmission performance, such as SHGC, window area ratio, and overhang projection ratio, proved to be influential. Overall, this study provided accurate models for assessing optimal strategies for energy efficiency and thermal comfort during the early design phases, advancing building performance practices.

Quantitative analysis of energy loss and performance prediction of regenerative compressor

The intricate internal flow structure within the regenerative compressor leads to substantial energy loss, presenting a long-standing challenge in accurately predicting its performance under various operating conditions. This paper establishes a performance prediction model that considers the impact of blade side exit angle and generalizes energy loss coefficient to accurately predict the performance of regenerative compressor under different operating conditions. Through numerical simulation analysis, the internal flow structure of the regenerative compressor is visualized, and the distribution and variation patterns of internal energy loss are studied. The magnitudes of loss rates for various components are quantitatively analyzed, revealing the underlying energy loss mechanism. A link between loss coefficients and the operating condition parameters of the regenerative compressor is established. Finally, the prediction model is applied to various similar compressor models for performance prediction, and compared with numerical simulation results. The results indicate that the prediction model can effectively predict the performance of various regenerative compressor models under different operating conditions, with maximum errors in total pressure and efficiency of 3.8% and 3.4%, and at rated conditions of 1.13% and 1.78%, respectively.

Energy Performance Prediction Model for Mixed-Flow Pumps by Considering the Effect of Incoming Prerotation

Abstract Mixed-flow pump is one of the most broadly applied sorts of power equipment in the field of petrochemical and water conservancy. The effect of inlet prerotation on the energy characteristics and operational stability of a pump is a significant consideration. The aim of this study is to analyze the relationship between inlet prerotation and the total energy consumption of a mixed flow pump by developing a predictive model. The impact of prerotation on the pumping performance and energy conversion for a typical mixed-flow pump has been investigated by a combined approach of theoretical derivation, numerical simulation, and experimental verification. Validation of the numerical methods was achieved by comparing the results to the experimentally obtained data. A prediction model was developed for head and power, which incorporated inlet prerotation. The study utilized a mathematical model and numerical simulation to compute the head and power output of a mixed-flow pump for a wide range of inlet prerotation angles. The results of the two methods were highly consistent. Moreover, the effects of prerotation on the flow structure of the mixed-flow pump were analyzed. It was found that prerotation led to an increase in the incoming flow angle, resulting in unstable flow patterns causing secondary flows and low-pressure vortex in the impeller flow path. This induced a rise in energy consumption of the impeller. The prediction model and analysis of the internal flow structure provide a theoretical foundation for predicting the hydraulic performances of mixed-flow pumps under prerotation conditions and improving their stability of operation.

Prediction of energy performance of residential buildings using regularised neural models

Human habitats are one of the major consumers of energy. Therefore, in the current age of increasing carbon dioxide footprints, analysing energy efficiency of a building is important and is the subject of the current study. Machine-learning-based artificial neural network (ANN) approaches are used in the current study to investigate building energy performance. Eight parameters – relative compactness, surface area, wall area, roof area, overall height and orientation of the building, as well as the glazing area and its distribution – are selected as the input parameters and heating and cooling loads (CLs) as the output parameters. The network prediction capability was checked by comparing the predictions of the ANN architecture with the benchmark test case. A well-trained and validated ANN is used to predict 96 conditions by varying glazing area and glazing area distribution. The ANN is found to capture the physics efficiently. This study revealed that there is a significant potential to improve the energy efficiency of a building and the maximum saving in the CL can be as high as 20.67% for a fraction of the glazing areas equal to 0.15 if the glazing area distribution is 32.5% in the north and 22.5% each in the east, south and west.

Energy performance prediction of pump as turbine (PAT) based on PIWOA-BP neural network

The greenhouse effect and the depletion of fossil energy sources have emerged as significant challenges impacting development of society. The development of micro hydropower stations is crucial in reducing carbon emissions and achieving carbon neutrality. Among them, pump as turbine (PAT) technology is considered one of the most promising alternatives in small and micro hydropower. However, pump manufacturers don't provide performance curves in turbine conditions, leading to issues such as difficulty in selecting a model and inaccuracy in calculating the turbine's high-efficiency zone when employing PAT in practice. This study proposes a new solution to these problems by introducing a new PIWOA-BPNN model. The model integrates the enhanced WOA algorithm with the parameter-optimized multilayer BPNN and employs the innovative spiral guidance vector p to direct the random search within the WOA algorithm, significantly enhancing the prediction capability of the neural network while reducing time cost. Additionally, the internal parameters of the turbine are analyzed, and a network model incorporating the geometric parameters of the impeller as auxiliary input is proposed. Several experimental tests demonstrate that the method offers greater accuracy and reliability compared to current mainstream algorithms. The prediction accuracy of the proposed model is approximately 4 %, meeting engineering requirements.

Urban building energy performance prediction and retrofit analysis using data-driven machine learning approach

Stakeholders such as urban planners and energy policymakers use building energy performance modeling and analysis to develop strategic sustainable energy plans with the aim of reducing energy consumption and emissions from the built environment. However, inconsistent energy data and the lack of scalable building models create a gap between building energy modeling and traditional planning practices. An alternative approach is to conduct a large-scale energy usage survey, which is time-consuming. Similarly, existing studies rely on traditional machine learning or statistical approaches for calculating large-scale energy performance. This paper proposes a solution that employs a data-driven machine learning approach to predict the energy performance of urban residential buildings, using both ensemble-based machine learning and end-use demand segregation methods. The proposed methodology consists of five steps: data collection, archetype development, physics-based parametric modeling, machine learning modeling, and urban building energy performance analysis. The devised methodology is tested on the Irish residential building stock and generates a synthetic building dataset of one million buildings through the parametric modeling of 19 identified vital variables for four residential building archetypes. As a part of the machine learning modeling process, the study implemented an end-use demand segregation method, including heating, lighting, equipment, photovoltaic, and hot water, to predict the energy performance of buildings at an urban scale. Furthermore, the model's performance is enhanced by employing an ensemble-based machine learning approach, achieving 91% accuracy compared to the traditional approach's 76%. Accurate prediction of building energy performance enables stakeholders, including energy policymakers and urban planners, to make informed decisions when planning large-scale retrofit measures.

A digital twin approach for maritime carbon intensity evaluation accounting for operational and environmental uncertainty

Maritime industry has set ambitious goals to drastically reduce its greenhouse gas emissions through stipulating and enforcing a number of energy assessment measures. Unfortunately, measures like the EEDI, EEXI, SEEMP and CII do not account for the operational and environmental uncertainty of operations at sea, even though they do provide a first means of evaluating the carbon footprint of ships. The increasing availability of high-frequency operational data offers the opportunity to quantify and account for this uncertainty in energy performance predictions. Current methods to evaluate and predict energy performance at a whole energy system level do not sufficiently account for operational and environmental uncertainty. In this work, we propose a digital twin that accurately predicts the fuel consumption and carbon footprint of the hybrid propulsion system of an Ocean-going Patrol Vessel (OPV) of the Royal Netherlands Navy under the aggregate effect of operational and environmental uncertainty. It combines first-principle steady-state models with machine learning algorithms to reach an accuracy of less than 5% MAPE on both mechanical and electrical propulsion, while bringing a 40% to 50% improvement over a model that does not utilise machine learning algorithms. Results over actual voyage intervals indicate a prediction accuracy of consumed fuel and carbon intensity within 2.5% accounting for a confidence interval of 95%. Finally, the direct comparison between mechanical and electrical propulsion showed no clear energy-saving benefits and a strong dependency of the results on each voyage’s specific operational and environmental conditions.

Building energy performance prediction: A reliability analysis and evaluation of feature selection methods

Building energy performance prediction: A reliability analysis and evaluation of feature selection methods

A comprehensive model considering multiple types of occupant behavior for building energy performance prediction and simulation – taking a university campus as an example

Occupant behavior, encompassing occupant numbers and actions, significantly contributes to the uncertainty of building energy performance simulation. By characterizing the regularity of occupant behavior, it becomes feasible to evaluate its impact on building energy consumption and mitigate disparities between simulated and measured outcomes. This paper presents a comprehensive model that integrates two key aspects of occupant behavior to predict and simulate energy performance in campus buildings. The model introduces two distinct categories of equipment, namely environment-related equipment (e.g. lighting, air conditioning) and time-related equipment (e.g. water dispenser, printer), to account for the diverse factors that influence occupant behavior. The impact of occupant perception on the environment is considered to enhance simulation accuracy. Furthermore, this study compares the proposed model with models that consider only occupant number or occupant energy use behavior individually to assess its effectiveness in capturing the complexities of occupant behavior in energy performance simulations.

Prediction of building energy performance using mathematical gene-expression programming for a selected region of dry-summer climate

Developing energy-efficient buildings considering building design parameters can help reduce buildings' energy consumption. The energy efficiency of residential buildings is considered a top priority for the energy policies of a country. Thus, this study utilizes gene-expression programming (GEP) to estimate the energy performance of residential buildings. The energy consumption evaluations were carried out using the Etotect energy simulation software. Eight building parameters with 768 data points were considered to generate the database for the heating load (HL) and cooling load (CL), including relative compactness, surface area, wall area, roof area, overall building height, glazing orientation, glazing area, and distribution of glazing area. Different GEP predictive models with varying parameters for building HL and CL were developed, and the best-performing prediction model was selected. In addition, several statistical indices were utilized to measure the accuracy of the proposed GEP models. The results revealed that GEP14 gives the most robust prediction model for HL having R2-value greater than 0.9 for both the training and validation set. Likewise, R2-value >0.9 is achieved for best- CL (GEP11). Furthermore, the mean absolute error (MAE) values for both the predictive HL’ and CL’ by prediction models were relatively less for both the training and testing databases. The sensitivity and parametric analysis reveal that the overall height (Ho), roof area (Af), and glazing area (Ag) were the most influential parameters for both predictive models. Thus, GEP results demonstrate the robust performance in predicting the building energy.

A reinforcing transfer learning approach to predict buildings energy performance

PurposeThe purpose of this paper is to propose a novel data-driven approach for predicting energy performance of buildings that can address the scarcity of quality data, and consider the dynamic nature of building systems.Design/methodology/approachThis paper proposes a reinforcing machine learning (ML) approach based on transfer learning (TL) to address these challenges. The proposed approach dynamically incorporates the data captured by the building management systems into the model to improve its accuracy.FindingsIt was shown that the proposed approach could improve the accuracy of the energy performance prediction compared to the conventional TL (non-reinforcing) approach by 19 percentage points in mean absolute percentage error.Research limitations/implicationsThe case study results confirm the practicality of the proposed approach and show that it outperforms the standard ML approach (with no transferred knowledge) when little data is available.Originality/valueThis approach contributes to the body of knowledge by addressing the limited data availability in the building sector using TL; and accounting for the dynamics of buildings’ energy performance by the reinforcing architecture. The proposed approach is implemented in a case study project based in London, UK.

Traditional Machine Learning Models for Building Energy Performance Prediction: A Comparative Research

A large proportion of total energy consumption is caused by buildings. Accurately predicting the heating and cooling demand of a building is crucial in the initial design phase in order to determine the most efficient solution from various designs. In this paper, in order to explore the effectiveness of basic machine learning algorithms to solve this problem, different machine learning models were used to estimate the heating and cooling loads of buildings, utilising data on the energy efficiency of buildings. Notably, this paper also discusses the performance of deep neural network prediction models and concludes that among traditional machine learning algorithms, GradientBoostingRegressor achieves better predictions, with Heating prediction reaching 0.998553 and Cooling prediction Compared with our machine learning algorithm HB-Regressor, the prediction accuracy of HB-Regressor is higher, reaching 0.998672 and 0.995153 respectively, but the fitting speed is not as fast as the GradientBoostingRegressor algorithm.

Meteorological-Data-Based Modeling for PV Performance Optimization

Developing a sustainable and reliable photovoltaic (PV) energy system requires a comprehensive analysis of solar profiles and an accurate prediction of solar energy performance at the study site. Installing the PV modules with optimal tilt and azimuth angles has a significant impact on the total irradiance delivered to the PV modules. This paper proposes a comprehensive optimization model to integrate total irradiance models with the PV temperature model to find the optimal year-round installation parameters of PV modules. A novel integration between installation parameters and the annual average solar energy is presented, to produce the maximum energy output. The results suggest an increase in energy yields of 4% compared to the conventional scheme, where tilt angle is equal to the latitude and the PV modules are facing south. This paper uses a real-time dataset for the NEOM region in Saudi Arabia to validate the superiority of the proposed model compared to the conventional scheme, but it can be implemented as a scheme wherever real-time data are available.

A Novel Energy Performance Prediction Approach towards Parametric Modeling of a Centrifugal Pump in the Design Process

Traditional centrifugal pump performance prediction (CPPP) employs the semi-theoretical and semi-empirical approaches; however, it can lead to many prediction errors. Considering the superiority of deep learning when applied to nonlinear systems, in this paper, a method combining hydraulic loss and convolutional neural network (HLCNN) is applied to CPPP. Head and efficiency were selected as two variables for demonstrating the energy performance of the centrifugal pump in order to reflect the prediction ability of the proposed model. The evaluation results indicate that the predicted head and efficiency are accurate, compared with the experimental results. Furthermore, the HLCNN prediction model was compared against machine learning methods and the computational fluid dynamic method. The proposed HLCNN model obtained a better AREmean, root mean square error, sum of squares due to error, and mean absolute error for centrifugal pump energy performance. The research revealed that the HLCNN model achieves accurate energy performance prediction in the design of centrifugal pumps, reducing the development time and costs.

The impact of urban morphology on the building energy consumption and solar energy generation potential of university dormitory blocks

Urban morphology is a major factor affecting building energy consumption and solar potential in the urban block. The aim of this research was to evaluate the impact of urban morphology on both building energy consumption and solar energy generation potential for university dormitory blocks in Wuhan. This paper proposed a classification method for dormitory blocks, calculated the building energy consumption and solar energy generation potential of 55 blocks, and analyzed the correlation between urban morphology and three energy performance indicators: Energy Use Intensity (EUI), Solar Energy Generation Intensity (SEGI) and Net Energy Use Intensity (NEUI). Multiple regression models were used to predict energy performance by urban morphological parameters. The results indicate that different block types could lead to up to 12.25% difference in EUI, and 35.85% significant difference in building NEUI with the deployment of photovoltaic (PV) panels. The EUI is mainly affected by three morphological parameters, which are Average length of block, Shape factor and Building density; while the SEGI and NEUI are mainly affected by Average height of block, Shape factor and Sky view factor. This study could serve as guidelines for planners and policymakers in campus planning and architectural design to improve building energy conservation at block-scale.

Impact of green retrofits on heating energy consumption of apartment buildings based on nationwide energy database in South Korea

In South Korea, apartment buildings account for 63% of residential buildings and approximately 40% of apartment buildings were built before 2000, which leads to the growing demand for green retrofits of old apartment buildings. In this study, heating energy consumption of apartment complexes in Seoul and Busan was analyzed using a nationwide database, in order to evaluate the current energy status of apartment complexes. The energy consumption trend with building age were analyzed by classifying the heating bill data into four groups by construction year of apartment complexes. Based on these data, a simulation model of an old apartment building was developed to predict energy performance improvement through green retrofits. The results clearly showed that heating energy consumption increased with building age regardless of heating degree days in both regions. In particular, a significant difference in energy consumption was found in apartment complexes constructed before and after 2000, which implied the necessity of green retrofits for old apartments. The results of green retrofit simulation showed that heating energy consumption was reduced by up to 31.6% and 32% in Seoul and Busan, respectively. The method and results of this study can be applied to establish a green retrofit strategy based on a nationwide database of residential energy.