Now showing 1 - 10 of 27
  • Publication
    Modelling User Navigation
    This paper proposes the use of neural networks as a tool for studying navigation within virtual worlds. Results indicate that the network learned to predict the next step for a given trajectory. The analysis of hidden layer shows that the network was able to differentiate between two groups of users identified on the basis of their performance for a spatial task. Time series analysis of hidden node activation values and input vectors suggested that certain hidden units become specialised for place and heading, respectively. The benefits of this approach and the possibility of extending the methodology to the study of navigation in Human Computer Interaction applications are discussed.
      97
  • Publication
    Lessons learned from the use of the SlideWiki OpenCourseWare platform in different learning contexts
    The SlideWiki platform is fostering open education by supporting the collaborative authoring, sharing, reusing and remixing of open educational content online. This paper presents the lessons learned from piloting the SlideWiki platform in different learning contexts and scenarios. In particular, we describe the learning scenarios considered in each pilot and how these scenarios have been implemented with the use of SlideWiki. We then describe the methodology followed in each pilot, in terms of the authoring, teaching and learning activities performed by educators and learners using SlideWiki. Finally, the lessons learned from each pilot are discussed, focusing on the challenges faced in each pilot, how these challenges have been addressed, as well as the best practices that have emerged from the pilots regarding the collaborative authoring and sharing of open educational content.
      276
  • Publication
    Quantification and characterization of energy flexibility in the residential building sector
    (International Building Performance Association, 2019-09-04) ; ; ; ; ;
    Demand response can enable residential consumers to take advantage of control signals and/or financial incentives to adjust the use of their resources at strategic times. These resources usually refer to energy consumption, locally distributed electricity generation, and energy storage. The building structural mass has an inherent potential either to modify consumption or to be used as a storage medium. In this paper, the energy flexibility potential of a residential building thermal mass for the winter design day is investigated. Various active demand response strategies are assessed using two flexibility indicators: the storage efficiency and storage capacity. Using simulation, it is shown that the available capacity and efficiency associated with active demand response actions depend on thermostat setpoint modulation, demand response event duration, heating system rated power and current consumption.
      152
  • Publication
    Towards Robustness of Data-Driven Predictive Control for Building Energy Flexibility Applications
    Identifying physics-based models of complex dynamical systems such as buildings is challenging for applications such as predictive and optimal control for demand side management in the smart grid. Data-driven predictive control using machine learning algorithms show promise as a more scalable solution when considering the greater building stock. The robustness of these algorithms for different climate data, building types, quality and quantity of data, is still not yet well understood. The objective in this study is to investigate model identification and the resultant accuracy for these various contexts using the `separation of variables' technique (DPC-En) and the consequent performance implications of the data-driven controller. The DPC-En controller is tested using a closed-loop simulation testbed of a `large office' archetype building. The results show that the technique is relatively robust to missing data and different climate types and delivers promising results using limited training data without the need for disruptive excitation measures. This work contributes to enabling a greater proportion of the diverse building stock to be utilised for demand side management by harnessing their inherent energy exibility potential.
      65
  • Publication
    Data-Driven Predictive Control for Commercial Buildings with Multiple Energy Flexibility Sources
    Data-Driven Predictive Control, representing the building as a cyber-physical system, shows promising potential in harnessing energy flexibility for demand side management, where the efforts in developing a physics-based model can be significant. Here, predictive control using random forests is applied in a case study closed-loop simulation of a large office building with multiple energy flexibility sources, thereby testing the suitability of the technique for such buildings. Further, consideration is given to the feature selection and feature engineering process. The results show that the data-driven predictive control, under a dynamic grid signal, is capable of minimising energy consumption or energy cost.
      107
  • Publication
    Data-Driven Predictive Control for Commercial Buildings with Multiple Energy Flexibility Sources
    Data-Driven Predictive Control, representing the building as a cyber-physical system, shows promising potential in harnessing energy flexibility for demand side management, where the efforts in developing a physics-based model can be significant. Here, predictive control using random forests is applied in a case study closed-loop simulation of a large office building with multiple energy flexibility sources, thereby testing the suitability of the technique for such buildings. Further, consideration is given to the feature selection and feature engineering process. The results show that the data-driven predictive control, under a dynamic grid signal, is capable of minimising energy consumption or energy cost.
      94
  • Publication
    Evaluation of Machine Learning Algorithms for Demand Response Potential Forecasting
    This paper focuses on the ability of machine learning algorithms to capture the demand response (DR) potential when forecasting the electrical demand of a commercial building. An actual sports-entertainment centre is utilised as a testbed, simulated with EnergyPlus, and the strategy followed during the DR event is the modification of the chiller water temperature of the cooling system. An artificial neural network (ANN) and a support vector machine (SVM) predictive model, are utilised to predict the DR potential of the building, due to the significant amount of execution time of the EnergyPlus model. The data-driven models are trained and tested based on synthetic databases. Results demonstrate that both ANN and SVM models can accurately predict the building electrical power demand for the scenarios without or with daily DR events, whereas both predictive models are not accurate in forecasting the electrical demand during the rebound effect.
      77
  • Publication
    Feature Assessment in Data-Driven Models for Unlocking Building Energy Flexibility
    Data-driven approaches are playing an increased role in building automation. This can, in part, be attributed to building operation and energy management system data becoming more readily accessible. A particular application is models to allow predictive control harnessing building energy flexibility, which is of interest to different stakeholders including; energy utilities, aggregators and end-users. Given the possibility of thousands of data features, feature selection becomes a critical part of the model development process. This paper considers various filter, wrapper and embedded methods applied in conjunction with three predictors in addressing the problem of constructing a suitable data-driven model to facilitate predictive control and provision of energy flexibility in a large commercial building. The feature selection algorithms are generally shown to significantly reduce model evaluation time and, in some cases, increase model accuracy. A random forest model with embedded feature selection was found to be the optimal solution in terms of model accuracy.
      197
  • Publication
    Automated Bridge Deck Evaluation through UAV Derived Point Cloud
    Imagery-based, three-dimensional (3D) reconstructions from Unmanned Aerial Vehicles (UAVs) hold the potential to provide a safer, more economical, and less disruptive approach for bridge inspection. This paper describes a methodology using a low-cost UAV to generate an imagery-based, dense point cloud for bridge deck inspection. Structure from motion (SfM) is employed to create a three-dimensional (3D) point cloud. Outlier data are removed through a density-based filtering method. Next, the unsupervised learning algorithm k-means and an object-based region growing algorithm are compared for accuracy with respect to bridge deck extraction. Last, an automatic pavement evaluation method is proposed to estimate the deck’s pavement condition. The procedure is demonstrated through an actual case study, in which a 3D point cloud of 16 million valid points was generated from 212 images. With that data set, the region growing method successfully extracted the deck area with an F-score close to 95%, while the unsupervised learning approach only achieved 76%. In the last, to evaluate the surface condition of the extracted pavement, a polynomial surface fitting method was designed to evaluate and visualise the damages.
      108