ASHRAE Journal - February 2021 - 44

TECHNICAL FEATURE

Acknowledgments
The authors appreciate the support from the Beijing
Municipal Natural Science Foundation (No. 8172032)
and from the Fundamental Research Funds for the
Central Universities (FRF-GF-17-B35).

FIGURE 9 Accumulated heat data before and after abnormal data detection.
(a) Raw data collected from heat meter with a small amount of outliers.
(b) Detection result after using Z-score method.
(c) Detection result using k-nearest-neighbor method after implementing Z-score
method.

A.

8
Accumulated Heat (kWh)

A simpliﬁed and efﬁcient algorithm-a detection algorithm of abnormal data based on k-nearest neighbor-
was proposed in this article. The main conclusions are as
follows:
* The inﬂuences of the number of intervals, N, and
abnormal number threshold, k, in the algorithm on the
detection results were analyzed. It can be seen from the
results that selecting N and k in the transition region
can correctly detect abnormal data points to the greatest
extent and reduce false detection.
* This algorithm was used to detect ﬂow rate and temperature data from a real thermal station. For the three
groups of test data, the accuracy of the k-nearest neighbor method is 99.5%, 97.5% and 99.6%, respectively. The
actual data showed that the algorithm can effectively
identify abnormal points.

6
4
2
0
0

25

50

B.

44

ASHRAE JOURNAL

ashrae.org

FEBRUARY 2021

100

125

150

175

4.0
3.5
3.0
2.5
2.0
1.5

0

25

50
75
100
125
Minutes (In Thousands)

150

175

0

25

50

150

175

References
C.
Accumulated Heat (kWh)

1. Tsinghua University Building Energy Research Center
(THUBERC). 2017. " District energy system in China, " pp. 11 - 12.
International Energy Agency.
2. National Bureau of Statistics of the People's Republic of China.
2017. China Statistical Yearbook, pp. 823. Beijing: China Statistics
Press.
3. Pedro, D., C. Paulo, V. Renato, K. Wolfgang. 2016. " Building
automation systems: concepts and technology review. " Computer
Standards & Interfaces 45(3):1 - 12.
4. H. Yu, J. Zhang, L. Zhao, X. Li. 2015. " Wireless data acquisition
system development and application on HVAC equipment. " Procedia
Engineering 121(1):2006 - 2013.
5. Yu, Z., H. Fariborz, M. Benjamin M. 2016. " Advances and
challenges in building engineering and data mining applications
for energy-efﬁcient communities. " Sustainable Cities and Society
25:33 - 38.
6. Zug, S., A. Dietrich, J. Kaiser. 2011. " Fault-handling in
networked sensor systems. " In Fault Diagnosis in Robotic and Industrial
Systems, pp. 20 - 40. Hong Kong: iConcept Press.
7. Grubbs, F.E. 1969. " Procedures for detecting outlying
observations in samples. " Technometrics 11(1):1 - 21.
8. Eskin, E. 2000. " Anomaly detection over noisy data using
learned probability distributions. " Proceedings of the 17th International
Conference on Machine Learning, pp. 255 - 262.
9. Edwin, K., N. Raymond. 1998. " Algorithms for mining
distance-based outliers in large datasets. " Proceedings of the
International Conference on Very Large Data Bases, pp. 392 - 403.
10. Ramaswamy, S., R. Rastogi, K. Shim. 2000. " Efﬁcient

75

Minutes (In Thousands)

Accumulated Heat (kWh)

Conclusion

3.50
3.25
3.00
2.75
2.50
2.25
2.00
1.75

75

100

125

Minutes (In Thousands)
algorithms for mining outliers from large data sets. " ACM SIGMOD
Record 29(2):427 - 438.
11. Bai, M., X. Wang, J. Xin, G. Wang. 2016. " An efﬁcient algorithm
for distributed density-based outlier detection on big data. "
Neurocomputing 181:19 - 28.
12. J. Tang, Z. Chen, A. Fu. 2002. " Enhancing Effectiveness of
Outlier Detections for Low Density Patterns. " Proceedings of the 6th
Paciﬁc-Asia Conference on Advances in Knowledge
Discovery and Data Mining, pp. 535 - 548.
13. Papadimitriou, S., H. Kitagawa, P. Gibbons.
2003. " Loci: Fast outlier detection using the
local correlation integral. " Proceedings of the
https://bit.ly/3aC5wVH
19th International Conference on Data Engineering,
Rate this Article
pp.315 - 326.
https://www.bit.ly/3aC5wVH http://ashrae.org
ASHRAE Journal - February 2021

Table of Contents for the Digital Edition of ASHRAE Journal - February 2021
