外文翻譯---用智能數(shù)據(jù)分析檢測建筑物中的能源異常消耗（英文）

1、Using intelligent data analysis to detect abnormal energy consumption in buildingsJohn E. Seem *Johnson Controls, Inc., 507 East Michigan Street, Milwaukee, WI 53202, USAReceived 31 October 2005; received in revised form

2、 11 March 2006; accepted 18 March 2006AbstractThis paper describes a novel method for detecting abnormal energy consumption in buildings based on daily readings of energy consumption and peak energy consumption. The meth

3、od uses outlier detection to determine if the energy consumption for a particular day is significantly different than previous energy consumption. For buildings with abnormal energy consumption, the amount of variation f

4、rom normal is determined using robust estimates of the mean and standard deviation. This new data analysis method will reduce operating costs by detecting problems that previously would have gone unnoticed. Also, operato

5、rs should save time by not having to manually detect faults or diagnose false alarms. The new data analysis method has successfully detected high-energy consumption in many buildings. This paper presents field test resul

6、ts for buildings that had the following problems: (1) chiller failure and a poor control strategy, (2) poor design of ventilating and air-conditioning equipment, and (3) improper operation of equipment following a change

7、 in the electrical panel. # 2006 Elsevier B.V. All rights reserved.Keywords: Energy consumption; Fault detection; Outlier analysis; Performance monitoring; Robust statistics1. IntroductionEnergy management and control sy

8、stems can collect and store massive quantities of energy consumption data. Facility operators can be overwhelmed with the quantity of data. For many operators, it is not possible to detect equipment, design, or operation

9、 problems because of data overload. Modern building management systems have two systems to help the operators with this data overload: alarm and warning systems and data visualization programs. Today, operators must sele

10、ct the thresholds for alarms and warnings. This is a difficult task. If the thresholds are too tight, then a number of false alarms are issued, and if the thresholds are too loose, then equipment or system failures can g

11、o undetected. The data visualization programs can help building operators detect and diagnose problems, but a large amount of time can be spent detecting problems.Also,theexpertise of building operators varies greatly. N

12、ew or inexperienced operators may have difficulty detecting faults and the performance of an operator can vary with the time of day or day of the week.The research community has developed a number of methods for detectin

13、g faults in buildings and heating, ventilating, and air-conditioning systems. Two major research efforts have been sponsored by the International Energy Agency: Annex 25 [1,2] and Annex 34 [3]. There are two basic approa

14、ches to fault detection and diagnostics in buildings: a component level (bottom-up) approach and a whole-building (top-down) approach. The component level approach looks for faults in individual systems such as variable-

15、air-volume boxes, air-handling units, chillers, or boilers. The whole-building approach looks for unusual behavior in high-level measure- ments such as the whole-building cooling, heating, or electrical consumption. Clar

16、idge et al. [4] describe an energy consumption report method that helps building operators and facility managers identify if the building systems are working properly. The report contains scatter plots of daily chilled w

17、ater energy consumption versus average daily temperature and daily hot water consumption versus average daily temperature for a 3- month period. For the last month, the scatter plot uses letters (M, T, W, H, F, S, U) to

18、identify the days of the week. The letters helps building operators identify outliers in energy consumption for a particular day. The report also contains two- and three- dimensional time series plots of chilled water co

19、nsumption andwww.elsevier.com/locate/enbuildEnergy and Buildings 39 (2007) 52–58* Tel.: +1 414 524 4677; fax: +1 414 524 5810. E-mail address: john.seem@gmail.com.0378-7788/$ – see front matter # 2006 Elsevier B.V. All r

20、ights reserved. doi:10.1016/j.enbuild.2006.03.0333. Outlier identification: GESD many-outlier procedureAn outlier is an observation that appears to be inconsistent with the majority of observations in a data set. For exa

21、mple, in the data set {1, 2, ?1, 0, 3, 2, 101, ?2}, the observation 101 appears to be an outlier. Data sets may contain more than one outlier. For example in the data set {1, 2, ?1, 0, 3, 2, 101, ?2, 96, 2, 0, ?209}, the

22、 observations 101, 96, and ?209 appear to be outliers. Barnet and Lewis [11] provide details on several common outlier identification methods. After comparing several popular outlier identification methods, Iglewicz and

23、Hoaglin [9] highly recommend the generalized extreme studentized deviate (ESD) many-outlier procedure that was proposed by Rosner [12] because it works well under a variety of conditions. The generalized ESD many-outlier

24、 procedure can identity the elements in a set that are outliers. Fig. 2 is a flow chart for determining one or more outliers from a set of n observations X 2 {x1, x2, x3, . . ., xn}. The user needs to specify the probabi

25、lity, a, of incorrectly declaring one or more outliers when no outliers exist and an upper bound, nu, on the number of potential outliers. Carey et al. [13] said the upper bound (nu) could be determined by finding the la

26、rgest integer that satisfies the following inequality: nu ? 0.5(n ? 1). Following are details on the numbered blocks in Fig. 2.? Block 1: Set nout = 0. This step is used to initialize the number of outliers to zero. ? Bl

27、ock 2: Compute average (¯ x) of elements in set X. The average is determined from¯ x ¼Pn j¼1 x jn (1)where xj is a member of set X and n equals the number of elements in set X. ? Block 3: Compute stan

28、dard deviation (s) of elements in set X. The standard deviation is determined froms ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pn j¼1ðx j ? ¯ xÞ2n ?

29、1s(2)? Block 4: s = 0. This block checks if the standard deviation of the elements in set X is zero. If the standard deviation equals zero, then the elements in set X all have the same value and there are no outliers in

30、the remaining elements in set X. (During field-testing of this method, several data sets had a standard deviation of zero.) To prevent a divide by zero in Block 6, execution goes to Block 10 when the standard deviation d

31、etermined in Block 3 equals zero. ? Block 5: Find ith extreme (xe,i) in set X. The extreme element, xe,i, is the element in set X that is furthest from ¯ x. Of all the elements in set X, the extreme element xe,i max

32、imizes the function jx j ? ¯ xj where xi is an element of set X. ? Block 6: Compute ith extreme studentized deviate Ri. The extreme studentized deviate is determined fromRi ¼ jxe;i ? ¯ xjs (3)where Ri is a

33、 normalized measure of how far the ith extreme is from the average value (¯ x) determined in Block 2. ? Block 7: Compute ith critical value li. Rosner [12] developed the following equation for determining the critic

34、al value:li ¼ ðn ? iÞtn?i?1; p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

35、ffiffiffiffiffi ffi ðn ? i þ 1Þðn ? i ? 1 þ t2 n?i?1; pÞ q (4)where tn?i?1,p is the Student’s t-distribution with (n?i?1) degrees and the tail area probability p is determined fromp ¼ a

36、2ðn ? i þ 1Þ (5)Abramowitz and Stegun [14] review equations for estimating the Student’s t-distribution.? Block 8: Ri > li. This block determines if the ith extreme studentized deviate, Ri, determined i

37、n Block 6 is greater than the ith critical value, li, determined in Block 7. ? Block 9: Set nout = i. This block sets the number of outliers, nout, equal to i. ? Block 10: Remove extreme element xe,i from set X. The extr

38、eme element xe,i is removed from set X and after removing the extreme element xe,i, the number of elements in Set X is n ? i. If i equals nu, then execution goes to Block 11; otherwise, return to the for loop on i.J.E. S