روش و مطالعه موردی تجزیه و تحلیل عدم قطعیت کمی در موجودی مصرف انرژی ساختمانی

Collections of building energy consumption data are often uncertain due to unavoidable measurement errors, random errors, non-representativeness of sample data, etc. On the basis of the quantitative uncertainty and Monte Carlo uncertainty propagation methods, the uncertainty of building energy consumption data is quantified. A real case study is conducted on the electricity and gas consumptions of buildings in Ma’anshan city in China in 2009. The results show that the electricity consumption distributions of four kinds of buildings fit Weibull distribution, gamma distribution, normal distribution and lognormal distribution respectively. The total energy consumption of buildings in the city at the cumulative probability of 97.5% is 16.6% higher than that obtained using the conventional method. The uncertainty of random sampling error in total energy consumption is about 14%. The sensitivity analysis results can provide information about the main sources that can help in reducing the uncertainty of the overall energy consumption inventory. This kind of quantification of uncertainty in energy consumption inventories could assist the decision-makers in determining the likelihood of complying with energy reduction objectives, and framing more scientific energy-saving strategies that may reduce building energy consumption.

The use of quantitative methods for characterizing uncertainty has been widely recognized and recommended for environmental modeling and assessment applications [1], [2], [3], [4], [5] and [6]. For instance, United States National Research Council (NRC) called for new efforts on mobile source emissions estimation and quantification of uncertainty in emission models [2]. Heijungs [3] and [4] reviewed several approaches to treat different types of uncertainty in LCA (life cycle assessment) and discussed the qualitative techniques that are available to address uncertainty. Macdonald and Strachan [5] demonstrated the partial application of uncertainty analysis by reviewing sources of uncertainties and incorporation of uncertainty analysis in Esp-r. Zhong [6] introduced the current methodological framework for quantifying uncertainty in emission inventories and conducted a case study on NOx emission inventory from power plants using uncertainty analysis. In recent years, the use of uncertainty and sensitivity techniques is largely popularized in different branches of disciplines. There have been some researches focused on the investigation of uncertainties and/or sensitivity in input parameters for building design support and in prediction of energy consumption. Hopfe and Hensen [7] conducted a case study of uncertainty analysis on design parameters in building simulation. Wang [8] investigates uncertainties in energy consumption due to actual weather and building operational practices using a simulation-based analysis of a medium-size office building. Macdonald [9] quantifies the effects of uncertainty in building simulation with respect to annual energy consumption, peak loads and the internal temperature. Brohus [10] presented a new approach for the prediction of building energy consumption. The approach quantifies the uncertainty of building energy consumption by means of stochastic differential equations to predict the performance of the systems and the level certainty for fulfilling design requirements under random conditions. Heiselberg et al. [11] applied sensitivity analysis to identify the important design parameters to change in order to reduce the primary energy consumption. Most of these researches almost concern the uncertainty analysis and sensitivity analysis of input parameters on energy consumption of whole-buildings, in which, the number of uncertain parameters is much larger. This study addresses the building energy consumption at city level. Nearly all analyses on energy and environmental control technologies at early phases of research involve uncertainties. Energy consumption inventories are a vital stage in energy-saving decision making. For example, energy consumption inventories are used for: (a) characterization of temporal energy usage; (b) energy-saving budgeting for regulatory and compliance purposes; and (c) prediction of the trends of energy consumption. The energy consumption, such as electricity and gas, is increasing rapidly with the rapid development of economy in Ma’anshan city of Anhui province recently. Therefore, the inventories reflecting the energy consumptions of different buildings in the city could be very helpful in assisting building energy conservation programs at city level. However, there may be uncertainty in the resulting building energy consumption data because of random sampling errors, measurement errors, and/or possibly non-representative samples. If random errors and biases in energy consumption inventories are not quantified, they may lead to erroneous judgment on trends of total energy consumption, source apportionment, compliance, etc. Uncertainty analysis and sensitivity analysis can provide information about the reliability and influence of uncertain parameters on total energy consumption. The methods could enable the decision-makers to determine the likelihood of complying with energy reduction objectives, and to frame more scientific energy-saving strategies that may reduce building energy consumption.

Based on the basic information on the energy consumptions of office buildings, large public buildings, small public buildings and residential buildings, data uncertainties and their propagation are analyzed. A generic method to quantify the uncertainty in energy consumption inventories is proposed and validated. The procedure of uncertainty analysis, uncertainty propagation approaches and the method of sensitivity analysis to identify key sources of uncertainty are presented. The quantitative uncertainty analysis method is used to analyze energy consumption intensity distributions of four types of buildings and 90 percent of confidence bound is obtained. Their distributions fitted Weibull distribution, gamma distribution, normal distribution and lognormal distribution respectively. The distribution of total electricity consumption, total gas consumption and total energy consumption of buildings in the city are predicted using Monte Carlo simulation. It is found that the cumulative probability is about 37% for the total energy consumption of buildings in the city having the value predicted using the mean energy intensity, while the value of total energy consumption at the cumulative probability of 97.5% is 16.6% higher than the value predicted using conventional methods. The uncertainty of random sampling error in total energy consumption is about 14%. The sensitivity analysis results indicate clearly that the most effective way to reduce uncertainty in the energy consumption inventory is to reduce uncertainties in electricity consumptions of the small public buildings and residential buildings. The uncertainty analysis and sensitivity analysis approaches applied in energy consumption inventory enable decision-makers to assess an energy consumption budget will be met and to indicate the direction to get a more precise prediction of energy consumption. The methods proposed here also allow decision-makers to assess the quality of the data and to decide on whether it is needed and how to reduce the uncertainties in the energy consumption inventory.