Search results

Energy management is critical to data centres (DCs) majorly because they are high energy-consuming facilities and demand for their services continue to rise due to rapidly increasing global demand for cloud services and other technological services. This projected sectoral growth is expected to translate into increased energy demand from the sector, which is already considered a major energy consumer unless innovative steps are used to drive effective energy management systems. The purpose of this study is to provide insights into the expected energy demand of the DC and the impact each measured parameter has on the building's energy demand profile. This serves as a basis for the design of an effective energy management system.

This study proposes novel tunicate swarm algorithm (TSA) for training an artificial neural network model used for predicting the energy demand of a DC. The objective is to find the optimal weights and biases of the model while avoiding commonly faced challenges when using the backpropagation algorithm. The model implementation is based on historical energy consumption data of an anonymous DC operator in Cape Town, South Africa. The data set provided consists of variables such as ambient temperature, ambient relative humidity, chiller output temperature and computer room air conditioning air supply temperature, which serve as inputs to the neural network that is designed to predict the DC’s hourly energy consumption for July 2020. Upon preprocessing of the data set, total sample number for each represented variable was 464. The 80:20 splitting ratio was used to divide the data set into training and testing set respectively, making 452 samples for the training set and 112 samples for the testing set. A weights-based approach has also been used to analyze the relative impact of the model’s input parameters on the DC’s energy demand pattern.

The performance of the proposed model has been compared with those of neural network models trained using state of the art algorithms such as moth flame optimization, whale optimization algorithm and ant lion optimizer. From analysis, it was found that the proposed TSA outperformed the other methods in training the model based on their mean squared error, root mean squared error, mean absolute error, mean absolute percentage error and prediction accuracy. Analyzing the relative percentage contribution of the model's input parameters based on the weights of the neural network also shows that the ambient temperature of the DC has the highest impact on the building’s energy demand pattern.

The proposed novel model can be applied to solving other complex engineering problems such as regression and classification. The methodology for optimizing the multi-layered perceptron neural network can also be further applied to other forms of neural networks for improved performance.

Based on the forecasted energy demand of the DC and an understanding of how the input parameters impact the building's energy demand pattern, neural networks can be deployed to optimize the cooling systems of the DC for reduced energy cost.

The use of TSA for optimizing the weights and biases of a neural network is a novel study. The application context of this study which is DCs is quite untapped in the literature, leaving many gaps for further research. The proposed prediction model can be further applied to other regression tasks and classification tasks. Another contribution of this study is the analysis of the neural network's input parameters, which provides insight into the level to which each parameter influences the DC’s energy demand profile.

Energy demand is an important economic index, and demand forecasting has a significant role when devising energy development plans for cities or countries. GM(1,1) model has become popular because it needs only a few data points to construct a time-series model without statistical assumptions. Several methods have been developed to improve prediction accuracy of the original GM(1,1) model by only estimating the sign of each residual. This study aims to address that this is too tight a restriction for the modification range.

Based on the predicted residual, this study uses the functional-link net (FLN) with genetic-algorithm-based learning to estimate the modification range for its corresponding predicted value obtained from the original GM(1,1) model.

The forecasting ability of the proposed grey prediction model is verified using real energy demand cases from China. Experimental results show that the proposed prediction model performs well compared to other grey residual modification models with sign estimation.

The proposed FLNGM(1,1) model can improve prediction accuracy of the original GM(1,1) model using residual modification. The distinctive feature of the proposed model is to use an FLN to estimate sign and modification range simultaneously for the predicted value based on its corresponding predicted residual obtained from the residual GM(1,1) model.

This paper aims to use artificial neural networks to develop models for forecasting energy demand for Australia, China, France, India and the USA.

The study used quarterly data that span over the period of 1980Q1-2015Q4 to develop and validate the models. Eight input parameters were used for modeling the demand for energy. Hyperparameter optimization was performed to determine the ideal parameters for configuring each country’s model. To ensure stable forecasts, a repeated evaluation approach was used. After several iterations, the optimal models for each country were selected based on predefined criteria. A multi-layer perceptron with a back-propagation algorithm was used for building each model.

The results suggest that the validated models have developed high generalizing capabilities with insignificant forecasting deviations. The model for Australia, China, France, India and the USA attained high coefficients of determination of 0.981, 0.9837, 0.9425, 0.9137 and 0.9756, respectively. The results from the partial rank correlation coefficient further reveal that economic growth has the highest sensitivity weight on energy demand in Australia, France and the USA while industrialization has the highest sensitivity weight on energy demand in China. Trade openness has the highest sensitivity weight on energy demand in India.

This study incorporates other variables such as financial development, foreign direct investment, trade openness, industrialization and urbanization, which are found to have an important effect on energy demand in the model to prevent underestimation of the actual energy demand. Sensitivity analysis is conducted to determine the most influential variables. The study further deploys the models for hands-on predictions of energy demand.

This paper aims to select the best scenario for energy demand forecast of residential and commercial sectors in Iran by using particle swarm optimization algorithm.

In this study, using variables affecting energy demand of residential and commercial sectors in Iran, the future status of energy demand in these sectors is predicted. Using the particle swarm optimization algorithm, both linear and exponential forms of energy demand equations were studied under 72 different scenarios with various variables. The data from 1968 to 2011 were applied for model development and the appropriate scenario choice.

An exponential model with inputs including total value added minus that of the oil sector, value of made buildings, total number of households and consumer energy price index is the most suitable model. Finally, energy demand of residential and commercial sectors is estimated up to the year 2032. Results show that the energy demand of the sectors will achieve a level of about 1,718 million barrels of oil equivalent per year by 2032.

To the best of our knowledge in this study a suitable model is selected for energy demand forecast of residential and commercial sectors by evaluation of various models with different variables as inputs.

This paper aims to study the present situation of Tianjin industrial energy consumption carbon emissions and put forward constructive suggestions for future energy-saving emission reduction work.

Using the energy consumption data form the Tianjin’s Industrial Energy Efficiency Guide (TJBS, 2009-2013) and Tianjin’s Statistical Yearbook (NBS, 2006-2012), some models were able to predict the future with a high degree of accuracy.

With an average error of 3.06 per cent for the logistic regression model and an average error of 2.03 per cent for the gray model, the R2 for the energy elasticity model is 0.99158. It also indicated that between 2008 and 2012, the energy consumption per unit of industrial added value decreased by approximately 33.61 per cent. These results show that energy-saving efforts and the optimization of the industrial structure have increased the energy efficiency of Tianjin.

The authors think that their contribution refers to a combination between methodology of forecasting and industrial energy consumption.

The 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.

This 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.

It 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.

The 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.

This 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.

Electricity demand forecasting has always been a key issue, and inaccurate forecasts may mislead policymakers. To accurately predict China’s electricity demand up to 2030, this paper aims to establish a cross-validation-based linear model selection system, which can consider many factors to avoid missing useful information and select the best model according to estimated out-of-sample forecast performances.

With the nine identified influencing factors of electricity demand, this system first determines the parameters in four alternative fitting procedures, where for each procedure a lot of cross-validation is performed and the most frequently selected value is determined. Then, through comparing the out-of-sample performances of the traditional multiple linear regression and the four selected alternative fitting procedures, the best model is selected in view of forecast accuracy and stability and used for forecasting under four scenarios. Besides the baseline scenario, this paper investigates lower and higher economic growth and higher consumption share.

The results show the following: China will consume 7,120.49 TWh, 9,080.38 TWh and 11,649.73 TWh of electricity in 2020, 2025 and 2030, respectively; there is hardly any possibility of decoupling between economic development level and electricity demand; and shifting China from an investment-driven economy to a consumption-driven economy is greatly beneficial to save electricity.

Following insights are obtained: reasonable infrastructure construction plans should be made for increasing electricity demand; increasing electricity demand further challenges China’s greenhouse gas reduction target; and the fact of increasing electricity demand should be taken into account for China’s prompting electrification policies.

The changes brought by Industry 4.0 go beyond transformations in the industrial environment. The increasingly frequent digitization and robotization of activities is not only restricted to the industrial environment, but also to people's daily routine. People spend a large part of their time inside buildings, and maintaining adequate Indoor Environmental Quality (IEQ) is an essential factor for a healthy and productive environment. In this sense, the purpose of this study is to verify how the Internet of Things (IoT) is being used to improve the indoor environment, through sensors that instantly measure the conditions of the environment.

The aim of this paper is to verify, through a literature review, how IoT is being used for building control (for energy saving purposes) and to monitor IEQ conditions inside buildings, in order to provide a better environment for occupants, in terms of health and comfort. By combining keywords in databases, PRISMA method was used to select the articles for analysis, and 91 articles were analyzed.

The main findings in this research are: (1) the main purpose for applying IoT inside buildings is to reduce energy consumption; (2) there is an interest in developing low-cost sensoring devices with a learning approach; (3) Machine Learning methods are mainly used for energy saving purposes and to learn about occupants' behavior inside buildings, focusing on thermal comfort; (4) sensors in the IoT era are a requirement to help improve people's comfort and well-being.

Studies directly correlating IoT and IEQ are limited. This paper emphasises the link between them, through the presentation of recent methods to control the built environment.

The purpose of this paper is to evaluate the risks of building defects associated with rapid advancement of “green” construction technologies. It identifies the methods adopted by the sector for the determination of pre-construction defects that are framed within the context of, traditional; scientific; and professional design approaches. These are critically evaluated and utilised in attempts to mitigate defects arising from diffusing low carbon construction innovations.

The paper takes the form of an evaluative literature review. Polemic in orientation, the paper critically compares two periods of time associated with rapid advancement of innovation. The first, the post-Second World War housing boom is synonymous with a legacy of substandard buildings that in many cases rapidly deteriorated, requiring refurbishment or demolition shortly after construction. The second, is today’s “green” technology “shift” with its inherent uncertainty and increased risk of latent building defects and potential failure to deliver meaningful long-term performance. Central to this is an exploration of the drivers for innovation, and subsequent response, precautionary measures initiated, and the limitations of institutionalised systems to identify and mitigate defects. Similarities and differences between these historical periods frame a discussion around the theoretical approaches to defects and how these may be limited in contemporary low carbon construction. A conceptual framework is presented with the aim of enhancing the understanding for obviation of defects.

Sufficient commonality exists between the periods to initiate a heightened vigilance in the identification, evaluation and ideally the obviation of defects. Design evaluation is not expressly or sufficiently defect focused. It appears that limited real change in the ability to identify defects has occurred since the post-war period and the ability to predict the performance of innovative systems and materials is therefore questionable. Attempts to appraise defects are still embedded in the three principle approaches: traditional; scientific; and professional design. Each of these systems have positive characteristics and address defect mitigation within constrains imposed by their very nature. However, they all fail to address the full spectrum of conditions and design and constructional complexities that lead to defects. The positive characteristics of each system need to be recognised and brought together in an holistic system that offers tangible advantages. Additionally, independent design professionals insufficiently emphasise the importance of defect identification and holistic evaluation of problems in design failure are influenced by their professional training and education. A silo-based mentality with fragmentation of professional responsibility debases the efficacy of defect identification, and failure to work in a meaningful, collaborative cross professional manner hinders the defect eradication process.

Whilst forming a meaningful contribution to stimulate debate, further investigation is required to tangibly establish integrated approaches to identify and obviate defects.

The structured discussion and conclusions highlight areas of concern for industry practitioners, policy makers, regulators, industry researchers and academic researchers alike in addressing and realising a low carbon construction future. The lessons learned are not limited to a UK context and they have relevance internationally, particularly where rapid and significant growth is coupled with a need for carbon reduction and sustainable development such as the emerging economies in China, Brazil and India.

The carbon cost associated with addressing the consequences of emerging defects over time significantly jeopardises attempts to meet legally binding sustainability targets. This is a relatively new dimension and compounds the traditional economic and societal impacts of building failure. Clearly, blindly accepting this as “the cost of innovation without development” cannot be countenanced.

Much research has been undertaken to evaluate post-construction defects. The protocols and inherent complexities associated with the determination of pre-construction defects have to date been largely neglected. This work attempts to rectify this situation.

The work set out to design and develop an overheating risk tool using the UKCP09 climate projections that is compatible with building performance simulation software. The aim of the tool is to exploit the Weather Generator and give a reasonably accurate assessment of a building's performance in future climates, without adding significant time, cost or complexity to the design team's work.

Because simulating every possible future climate is impracticable, the approach adopted was to use principal component analysis to give a statistically rigorous simplification of the climate projections. The perceptions and requirements of potential users were assessed through surveys, interviews and focus groups.

It is possible to convert a single dynamic simulation output into many hundreds of simulation results at hourly resolution for equally probable climates, giving a population of outcomes for the performance of a specific building in a future climate, thus helping the user choose adaptations that might reduce the risk of overheating. The tool outputs can be delivered as a probabilistic overheating curve and feed into a risk management matrix. Professionals recognized the need to quantify overheating risk, particularly for non‐domestic buildings, and were concerned about the ease of incorporating the UKCP09 projections into this process. The new tool has the potential to meet these concerns.

The paper is the first attempt to link UKCP09 climate projections and building performance simulation software in this way and the work offers the potential for design practitioners to use the tool to quickly assess the risk of overheating in their designs and adapt them accordingly.