Home | Business News | Browse by Publication | A | ASHRAE Transactions

Data processing and data mining on energy consumption database of commercial buildings in Shanghai.

Article Excerpt
INTRODUCTION

In recent years, the energy consumed by buildings has been increasing rapidly in China both in the absolute quantity and the ratio in national total energy consumption, and it is certainly becoming a main field of energy saving in China.

A lot of work has been conducted by the professionals around the world on investigation, statistics and analysis on building energy consumption and the programming of building energy simulation software. Many researchers are dedicated in the statistics, analysis and modeling of building energy consumption using regression methodology. Lam et al. (1997) used the simulation computer program DOE-2 (LBNL 1991) to carry out a parametric study of a typical high-rise air-conditioned office building in Hong Kong. A total of 28 relative design parameters were found to correlate well with the predicted annual electricity consumption. Both linear and non-linear multiple regression techniques were used to develop regression models and energy equations for the prediction of annual electricity use. Twelve input design parameters were found to be the most significant design variables and were used in the energy prediction equations. General regression neural networks (GRNN) were adopted by Ben-Nakhi and Mahmoud (2004) to optimize HVAC thermal energy storage in public buildings as well as office buildings. The training database for the GRNN was generated using the building simulation software ESP-r. Three different buildings were investigated, and hourly outdoor temperatures and building cooling loads were the input and the output of the GRNN, respectively. The results showed that a properly designed NN is a powerful instrument for optimizing thermal energy storage in buildings, and it can work well when based only on outdoor temperature records. Chung et al. (2006) described a benchmarking process for energy efficiency by means of multiple regression analysis, with which the relationship between energy-use intensities (EUIs) and the explanatory factors (e.g., operating hours) is developed. Nine variables including building age, occupancy and type of energy system are adopted to establish the regression equation, and a benchmarking table is derived by removing the effect of variance in the significant explanatory factors using the multiple-regression model. Ghiaus (2006) adopted energy consumption and outdoor temperature recorded by a Building Energy Management Systems (BEMS) to assess the energy performance of the building, such as the heating load as a function of the outdoor temperature. The method was to use the range between the 1st and the 3rd quartile of the quantile-quantile (q-q) plot to check if the heating losses and the outdoor temperatures have the same distribution and, if yes, to perform the regression in this range of the q-q plot. The result was a model that conserves its prediction performance for data sets of the outdoor temperature different than those used for parameter identification. Pedersen (2007) provided an overview of the background for meteorological and sociological influences on thermal load and energy estimations. As his point of view, "regression analysis is mainly based on large amounts of metered load data, long-term weather characteristics and some information about the buildings being modeled. A statistical approach is most suitable for large development areas and long-term estimates of the expected load and energy demand." Freire (2008) adopted independent variables - heating, ventilation and air conditioning (HVAC) power, outdoor temperature, relative humidity and total solar radiation - to obtain the regression equations that were used to define a couple of linear Multiple-Input/Single-Output (MISO) models, since two main outputs were involved, indoor temperature and relative humidity. And validation procedures have shown very good agreement between the regression equations and the simulation tool for both winter and summer periods.

However in China there are few researchers working...

View this article FREE - Now for a Limited Time, try Goliath Business News
Free for 3 Days!