• Energy Research

Abstract Practical building operations usually deviate from the designed building operational performance due to the wide existence of operating faults and improper control strategies. Great energy saving potential can be realized if inefficient or faulty operations are detected and amended in time. The vast amounts of building operational data collected by the Building Automation System have made it feasible to develop data-driven approaches to anomaly detection. Compared with supervised analytics, unsupervised anomaly detection is more practical in analyzing real-world building operational data, as anomaly labels are typically not available. Autoencoder is a very powerful method for the unsupervised learning of high-level data representations. Recent development in deep learning has endowed autoencoders with even greater capability in analyzing complex, high-dimensional and large-scale data. This study investigates the potential of autoencoders in detecting anomalies in building energy data. An autoencoder-based ensemble method is proposed while providing a comprehensive comparison on different autoencoder types and training schemes. Considering the unique learning mechanism of autoencoders, specific methods have been designed to evaluate the autoencoder performance. The research results can be used as foundation for building professionals to develop advanced tools for anomaly detection and performance benchmarking.

Accurate building energy prediction is vital to develop optimal control strategies to enhance building energy efficiency and energy flexibility. In recent years, the data-driven approach based on machine learning algorithms has been widely adopted for building energy prediction due to the availability of massive data in building automation systems (BASs), which automatically collect and store real-time building operational data. For new buildings and most existing buildings without installing advanced BASs, there is a lack of sufficient data to train data-driven predictive models. Transfer learning is a promising method to develop accurate and reliable data-driven building energy prediction models with limited training data by taking advantage of the rich data/knowledge obtained from other buildings. Few studies focused on the influences of source building datasets, pre-training data volume, and training data volume on the performance of the transfer learning method. The present study aims to develop a transfer learning-based ANN model for one-hour ahead building energy prediction to fill this research gap. Around 400 non-residential buildings’ data from the open-source Building Genome Project are used to test the proposed method. Extensive analysis demonstrates that transfer learning can effectively improve the accuracy of BPNN-based building energy models for information-poor buildings with very limited training data. The most influential building features which influence the effectiveness of transfer learning are found to be building usage and industry. The research outcomes can provide guidance for implementation of transfer learning, especially in selecting appropriate source buildings and datasets for developing accurate building energy prediction models.

Abstract With the advances of information technologies, today's building automation systems (BASs) are capable of managing building operational performance in an efficient and convenient way. Meanwhile, the amount of real-time monitoring and control data in BASs grows continually in the building lifecycle, which stimulates an intense demand for powerful big data analysis tools in BASs. Existing big data analytics adopted in the building automation industry focus on mining cross-sectional relationships, whereas the temporal relationships, i.e., the relationships over time, are usually overlooked. However, building operations are typically dynamic and BAS data are essentially multivariate time series data. This paper presents a time series data mining methodology for temporal knowledge discovery in big BAS data. A number of time series data mining techniques are explored and carefully assembled, including the Symbolic Aggregate approXimation (SAX), motif discovery, and temporal association rule mining. This study also develops two methods for the efficient post-processing of knowledge discovered. The methodology has been applied to analyze the BAS data retrieved from a real building. The temporal knowledge discovered is valuable to identify dynamics, patterns and anomalies in building operations, derive temporal association rules within and between subsystems, assess building system performance and spot opportunities in energy conservation.

Massive amounts of building operational data are collected and stored in modern buildings, which provide rich information for in-depth investigation and assessment of actual building operational performance. However, the current utilization of big building operational data is far from being effective due to the gaps between building engineering and advanced big data analytics. Data mining is a promising technology for extracting previously unknown yet potentially useful insights from big data. This paper aims to explore the potential application of advanced data mining techniques for effective utilization of big building operational data. A case study of mining the operational data of an educational building for performance improvement is presented. Decision tree, clustering analysis, and association rule mining are adopted to analyze the operational data. The results show that useful knowledge can be extracted for identifying typical building operation patterns, detecting operation deficiencies, and spotting energy conservation opportunities. Practical application:The current utilization of big building operational data in the building industry is rather limited due to the lack in experience of using advanced big data analytics. This study presents a data mining-based method for analyzing massive building operational data. The case study results validate the efficiency and effectiveness of the method proposed. It can help building professionals to discover valuable insights into building operation patterns and thereby developing strategies for improving building energy efficiency. The method can be fully realized using the open-source software R, which provides great flexibilities in its integration with building automation systems.

Abstract Uncertainties exist widely at the planning and design stages of district cooling systems, which have significant impacts on the design optimization. This paper therefore proposes a design method for district cooling systems by quantifying the uncertainties, which is so-called uncertainty-based design optimization method. Uncertainties in the outdoor weather, building design/construction and indoor conditions are considered. The application of the uncertainty-based design optimization method is examined in several aspects: the performance assessment, system sizing, configuration selection and technology integration. With the performance distribution at different risk levels, the design of district cooling systems can be determined by the stakeholders based on the compromise between quantified risk and benefit. Sensitivity analysis is conducted to identify influential variables with uncertainties for the cooling loads of district cooling systems. Results show that the uncertainties in the indoor condition are the most important and the uncertainties in building design/construction have the least impact.

Abstract The advance in information technology has enabled the real-time monitoring and controls over building operations. Massive amounts of building operational data are being collected and available for knowledge discovery. Advanced data analytics are urgently needed to fully realize the potential of big building operational data in enhancing building energy efficiency. Data mining (DM) technology, which is renowned for its excellence in discovering hidden knowledge from massive datasets, has attracted increasing attention from the building industry. The rapid development in DM has provided powerful mining methods for extracting insights in various knowledge representations. Gradual pattern mining is a promising technique for identifying interesting patterns in big data. The knowledge discovered is represented as gradual rules, i.e., ‘the more/less A, the more/less B’. It can bring special interests to building energy management by highlighting co-variations among building variables. This paper investigates the usefulness of gradual pattern mining in analysing massive building operational data. Together with the use of decision trees, motif discovery and association rule mining, a comprehensive mining method is developed to ensure the quality and applicability of the knowledge discovered. The method is validated through a case study, using the real-world data retrieved from an educational building in Hong Kong. It shows that novel and valuable insights on building operation characteristics can be obtained, based on which fault detection and optimal control strategies can be developed to enhance building operational performance.

Abstract Variable speed drivers (VSDs) are commonly used for enhancing energy efficiency in building central cooling systems. However, VSDs often consume about 4–8% of the converted energy. Moreover, the initial and maintenance costs of VSDs for extremely large and high voltage chillers could be extremely high. This study proposes to use thermal energy storage (TES) to enhance energy efficiency of extremely large constant speed chillers. A new model predictive control method is proposed to control the charging/discharging of TES and on/off of chillers to achieve high efficiency. The proposed method partially decouples the demand side and the supply side, so that the large chillers are either operated in high efficiency or turned off. The method can also solve the problem of frequent chiller tripping due to too low load in winter conditions. The proposed optimal control strategy has been validated on a dynamic platform built based on the existing chiller plant in a high-rise commercial building. Validation tests were conducted in both summer and winter conditions based on real operation data. Results show that the proposed method could improve the efficiency of chillers by 3.10% and 22.94% in summer and winter conditions, respectively.

Abstract Short-term building cooling load prediction is the essential foundation for many building energy management tasks, such as fault detection and diagnosis, demand-side management and control optimization. Conventional methods, which heavily rely on physical principles, have limited power in practice as their performance is subject to many physical assumptions. By contrast, data-driven methods have gained huge interests due to their flexibility in model development and the rich data available in modern buildings. The rapid development in data science has provided advanced data analytics to tackle prediction problems in a more convenient, efficient and effective way. This paper investigates the potential of one of the most promising techniques in advanced data analytics, i.e., deep learning, in predicting 24-h ahead building cooling load profiles. Deep learning refers to a collection of machine learning algorithms which are powerful in revealing nonlinear and complex patterns in big data. Deep learning can be used either in a supervised manner to develop prediction models with given inputs and output (i.e., cooling load), or in an unsupervised manner to extract meaningful features from raw data as model inputs. This study exploits the potential of deep learning in both manners, and compares its performance in cooling load prediction with typical feature extraction methods and popular prediction techniques in the building field. The results show that deep learning can enhance the performance of building cooling load prediction, especially when used in an unsupervised manner for constructing high-level features as model inputs. Using the features extracted by unsupervised deep learning as inputs for cooling load prediction can evidently enhance the prediction performance. The findings are enlightening and could bring more flexible and effective solutions for building energy predictions.

Abstract Today's building automation system (BAS) provides us with a tremendous amount of data on actual building operation. Buildings are becoming not only energy-intensive, but also information-intensive. Data mining (DM) is an emerging powerful technique with great potential to discover hidden knowledge in large data sets. This study investigates the use of DM for analyzing the large data sets in BAS with the aim of improving building operational performance. An applicable framework for mining BAS database is proposed. The framework is implemented to mine the BAS database of the tallest building in Hong Kong. After data preparation, clustering analysis is performed to identify the typical power consumption patterns of the building. Then, association rule mining is adopted to unveil the associations among power consumptions of major components in each cluster. Lastly, post-mining is conducted to interpret the rules. 457 rules are obtained in association rule mining, of which the majority can be easily deduced from domain knowledge and hence be ignored in this study. Four of the rules are used for improving building performance. This study shows that DM techniques are valuable for knowledge discovery in BAS database; however, solid domain knowledge is still needed to apply the knowledge discovered to achieve better building operational performance.