Search

In recent years, a large number of building energy consumption prediction models, with various intended uses, have been proposed. The majority of these models have either taken a data-driven or a physical modeling approach, with each approach having its own strengths and limitations. Towards leveraging the strengths and reducing the limitations of each approach for improved prediction performance, this paper presents a hybrid machine-learning approach for occupant-behavior-sensitive energy consumption prediction. The proposed approach is composed of three constituent models: (1) a machine-learning model that learns the impact of outdoor weather conditions from simulation-generated data, (2) a machine-learning model that learns the impact of occupant behavior from real data, and (3) an ensemble model that predicts cooling energy consumption based on the outputs of the first two models. The simulation-generated data were created through simulating a set of reference buildings in EnergyPlus. The real data were collected from an office building in Pennsylvania. The proposed hybrid model was validated on an unseen real dataset. It achieved 0.73 kWh RMSE and 9.07% CV in hourly cooling energy consumption prediction, which indicates that the proposed approach is promising. [ABSTRACT FROM AUTHOR]

Since the Pennsylvania oil rush of 1859, petroleum has quickly become the dominant fuel of industrial society. The "Peak Oil" debate focused on whether or not there was an impending production crunch of cheap oil, and whilst there have been no shortages across the globe, a shift from conventional to unconventional oil liquids has occurred. One aspect of this shift was not fully explored in previous discussions–although of some importance in a low-carbon energy transition context: the extent to which the net-energy supply of oil products is affected by the use of lower quality energy sources. To fill this gap, this paper incorporates standard EROI (energy-return-on-investment) estimates and dynamic decline functions in the GlobalShift all-liquids bottom-up model on a global scale. We determine the energy necessary for the production of oil liquids (including direct and indirect energy costs) to represent today 15.5% of the energy production of oil liquids, and growing at an exponential rate: by 2050, a proportion equivalent to half of the gross energy output will be engulfed in its own production. Our findings thus question the feasibility of a global and fast low-carbon energy transition. We therefore suggest an urgent return of the peak oil debate, but including net-energy issues and avoiding a narrow focus on 'peak supply' vs 'peak demand'. [Display omitted] • Global gross and net-energy of oil liquids production is determined from 1950 to 2050. • Energy required for production is estimated to be 15.5% of the actual gross energy. • Oil liquids become a limit to a rapid and global low-carbon energy transition. • The peak supply vs. peak demand dispute needs to be re-examined. • Focus should be put instead on net-energy transition and wise energy consumption. [ABSTRACT FROM AUTHOR]