Reliable energy forecasting helps managers to prepare future budgets for their buildings. Therefore, a simple, easier, less time consuming and reliable forecasting model which could be used for different types of buildings is desired. In this paper, we have presented a forecasting model based on five years of real data sets for one dependent variable (the daily electricity consumption) and six explanatory variables (ambient temperature, solar radiation, relative humidity, wind speed, weekday index and building type). A single mathematical equation for forecasting daily electricity usage of university buildings has been developed using the Multiple Regression (MR) technique. Data of two such buildings, located at the Southwark Campus of London South Bank University in London, have been used for this study. The predicted test results of MR model are examined and judged against real electricity consumption data of both buildings for year 2011. The results demonstrate that out of six explanatory variables, three variables; surrounding temperature, weekday index and building type have significant influence on buildings energy consumption. The results of this model are associated with a Normalized Root Mean Square Error (NRMSE) of 12% for the administrative building and 13% for the academic building. Finally, some limitations of this study have also been discussed.
The building sector consumes about 40% of the world’s primary energy. Seasonal climatic conditions have a significant effect on the energy consumption in buildings. One of the famous methods used for decoding this seasonal variation in buildings energy consumption is the “Degree Days Method”. Data has been widely published for the heating and cooling degree days of different countries. Unfortunately, there is very limited and outdated published data for the heating and cooling degree-days of Pakistan. In this study, yearly average heating and cooling degree-days for different regions of Pakistan are established by using 30 year long-term measured data for different base temperatures. The data is presented in tables and figures whereas heating and cooling degree-day maps of Pakistan have been developed.
Smart grid plays a vital role in energy management systems. It helps to mitigate the demand side management of electricity by managing the microgrid. In the modern era, the concept of hybrid microgrids emerged which helps the smart grid management of electricity. Additionally, the Internet of Things (IoT) technology is used to integrate the hybrid microgrid. Thus, various policies and topologies are employed to perform the task meticulously. Pakistan being an energy deficient country has recently introduced some new policies such as Energy Wheeling Policy (EWP), Energy Import Policy (EIP), and Net Metering/Distributed Generation Policy (NMP) to manage the electricity demand effectively. In addition, the Energy Efficiency and Conservation Act (EECA) has also been introduced. In this paper, we present the overview and impact of these policies in the context of the local energy market and modern information and communication mechanisms proposed for smart grids. These new policies primarily focus on energy demand–supply for various types of consumers such as the demand for bulk energy for industrial ventures and the distributed production by consumers. The EWP deals with obtaining power from remote areas within the country to ease the energy situation in populated load centers and the EIP highlights energy import guidelines from foreign countries. The NMP deals with the integration of renewable energy resources and EECA is more focused on the measures and standardization for energy efficiency and conservation. The benefits and challenges related to EWP, NMP, and EIP have also been discussed concerning the present energy crisis in Pakistan. The generalized lessons learned and comparison of a few aspects of these policies with some other countries are also presented.
15 June 2020 in IEEE Transactions on Nanotechnology
Higher order intensity of light introduces third order nonlinear susceptibility in metallic nanostructures. Circuit theory is often used to represent complex phenomena and adequately models the frequency dependent linear resonance in metallic nanostructures with great accuracy. In this article, we use the linear circuit approach and propose a compact impedance model to represent resonance in metallic nanoparticle with cubic nonlinearity. The model uses full-wave dipole equation by employing all time-dependent fields and introduces the concept of nonlinear radiation impedances. Analytical expressions for the nonlinear radiation (internal and external) impedances are derived and used for the first time to get the close-form expression of nonlinear (NL) scattering cross-sectional area using voltages, currents and circuit elements. The effects of nanoparticle parameters i.e. radius, intensity and dielectric function of surrounding medium on cubic nonlinearity are analyzed using impedance model. The validated close-form impedance model preserves the macroscopic properties to give intuitive understanding of NL behavior necessary in facilitating next-generation nanophotonics applications.
Smart Grids (SG) have emerged as one of the complex cyber-physical systems integrating information and communication technologies to existing power system infrastructure for reliable power delivery. However, this integration makes the system highly susceptible to cyber-attacks, and brings forth the challenge of dealing with big data by the expanded size of SG. To this end, an autoencoder-based feature extraction technique is employed for obtaining discriminant features while preserving the primordial properties of the system. The reconstructed features are then provided as input to binary support vector machine classifier to detect cyber-intrusions, outliers and attacks in SG. Various IEEE test cases are used in the simulation. The performance comparison shows that our proposed scheme outperforms the existing schemes in capturing prominent features leading to improved detection accuracy of the classifier.
12 March 2020 in ASME Journal of Engineering for Sustainable Buildings and Cities
Energy Usage Intensity (EUI, kWh/m2/year) of a building category helps energy managers and building owners in evaluating the energy performance of their buildings. Banks are energy-intensive buildings but there is limited research published in this highly energy-intensive building category. In this paper, the energy consumption of 98 bank buildings located in different cities of the AJK region of Pakistan has been critically analyzed and their EUI value has been established. Data were collected through site visits, interviews, and questionnaires. Electricity consumption of each bank branch was normalized using Cooling Degree Days method and respective annual normalized EUIs were calculated. It was found through analysis that three major factors that drive electricity consumption in bank buildings in Pakistan are; building's area, the number of workstations and the number of Automatic Teller Machines. The EUI of banks in Pakistan has been found to be 222kWh/m2/yr which is comparable with the same of eight other countries.