Anomaly ranking of District Heating Substations - Revision history

Slawek at 18:41, 23 September 2018

2018-09-23T18:41:02Z

Slawek at 21:01, 27 September 2017

2017-09-27T21:01:17Z

Slawek at 20:56, 27 September 2017

2017-09-27T20:56:10Z

Ececal at 20:51, 27 September 2017

2017-09-27T20:51:17Z

Ececal at 20:50, 27 September 2017

2017-09-27T20:50:19Z

Ececal: In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.

2017-09-27T16:27:42Z

In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.

Ececal: In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.

2017-09-27T16:25:52Z

In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.

New page

{{StudentProjectTemplate
|Summary=In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.
|References=M. Goldstein and S. Uchida, "A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data", PLOS ONE, vol. 11, no. 4, p. e0152173, 2016.

P. Arjunan, H. Khadilkar, T. Ganu, Z. Charbiwala, A. Singh and P. Singh, "Multi-User Energy Consumption Monitoring and Anomaly Detection with Partial Context Information", Proceedings of the 2nd ACM International Conference on Embedded Systems for Energy-Efficient Built Environments - BuildSys '15, 2015.

D. Araya, K. Grolinger, H. ElYamany, M. Capretz and G. Bitsuamlak, "An ensemble learning framework for anomaly detection in building energy consumption", Energy and Buildings, vol. 144, pp. 191-206, 2017.

S. Rayana and L. Akoglu, "Less is More", ACM Transactions on Knowledge Discovery from Data, vol. 10, no. 4, pp. 1-33, 2016.

Huang, Huaming, "Rank Based Anomaly Detection Algorithms" (2013). Electrical Engineering and Computer Science - Dissertations.Paper 331.
|Prerequisites=Artificial Intelligence and Learning Systems courses; good knowledge of data mining; programming skills for implementing machine learning algorithms
|Supervisor=Ec
}}
District heating system is a common way to distribute heat through underground pipelines for residential and commercial requirements. Faults are common in district energy systems due to the high number of substations and instrumentation components. Also, the standard energy-metering instrumentation is designed for low cost and billing, not for automated fault detection. Large variations in building dynamics, building subsystems, human behaviour and the environment make the system complex to model and analyse.

Anomaly detection refers to the process of detecting abnormal events that do not conform to expected patterns. However, it is hard to differentiate actual faults from the changes in energy consumption due to seasonal variations and changes in personal profiles such as holidays etc.

In practice, multiple anomaly detection tools are used to continuously raise alarms for different application domains. These alarms include both true positives and false alarms. Operators act on these alarms for diagnosis and deeper root cause
analysis and take appropriate maintenance actions to mitigate the anomalous behaviours. Given the scale and scope of the district heating substations, the operators can be overwhelmed with the large number of alarms at any given instant. It is therefore necessary to prioritize and rank these alarms by their severity. In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.

Ensemble methods are meta-algorithms that combine several machine learning techniques into one predictive model in order to improve prediction. Ensemble learning for anomaly detection aims to combine results from different detectors with varying outputs to achieve better anomaly detection performance. It generally follows following steps:

• Model Creation: This is the individual methodology or algorithm which is used to create the corresponding component of the ensemble.

• Normalization: Different methods may create outlier scores which are on very different scales. In some cases,
the scores may be in ascending order, whereas in others, they may be in descending order. In such cases, normalization is important in being able to combine the scores meaningfully, so that the outlier scores from different components are roughly comparable.

• Model Combination: This refers to the final combination function, which is used in order to create the outlier score.

Objectives:

-Applying and comparing state-of-the-art unsupervised anomaly detection methods

-Implementing novel ensemble method by combining multiple detectors

-Implementing novel anomaly ranking schema in order to aggregate results from different detectors

-Testing anomaly-ranking framework on district heating and heat pump datasets.

We have collaboration with 2 companies within this project i.e. HEM and Öresundskraft.

← Older revision		Revision as of 18:41, 23 September 2018
Line 36:		Line 36:
	-Testing anomaly-ranking framework on district heating and heat pump datasets.		-Testing anomaly-ranking framework on district heating and heat pump datasets.

−	We have collaboration with 2 companies within this project i.e. HEM and Öresundskraft.	+	We have collaboration with 2 companies within this project, i.e., HEM and Öresundskraft.

← Older revision		Revision as of 21:01, 27 September 2017
Line 21:		Line 21:

	In practice, multiple anomaly detection tools are used to continuously raise alarms for different application domains. These alarms include both true positives and false alarms. Operators act on these alarms for diagnosis and deeper root cause analysis and take appropriate maintenance actions to mitigate the anomalous behaviours. Given the scale and scope of the district heating substations, the operators can be overwhelmed with the large number of alarms at any given instant. It is therefore necessary to prioritize and rank these alarms by their severity. In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.		In practice, multiple anomaly detection tools are used to continuously raise alarms for different application domains. These alarms include both true positives and false alarms. Operators act on these alarms for diagnosis and deeper root cause analysis and take appropriate maintenance actions to mitigate the anomalous behaviours. Given the scale and scope of the district heating substations, the operators can be overwhelmed with the large number of alarms at any given instant. It is therefore necessary to prioritize and rank these alarms by their severity. In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.
		+
		+	Examples of techniques to try are e.g. ensemble methods or learning-to-rank.

	It is also interesting to look at ways to categorise different anomalies, based on different criteria.		It is also interesting to look at ways to categorise different anomalies, based on different criteria.

@@ Line 1: / Line 1: @@
 {{StudentProjectTemplate
-|Summary=Anomaly ranking algorithm in order to monitor district heating substations.
+|Summary=Implementing anomaly ranking algorithm to monitor district heating substations.
 |Keywords=anomaly detection, self monitoring, data mining, learnin-to-rank
 |References=M. Goldstein and S. Uchida, "A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data", PLOS ONE, vol. 11, no. 4, p. e0152173, 2016.

@@ Line 20: / Line 20: @@
 Anomaly detection refers to the process of detecting abnormal events that do not conform to expected patterns. However, it is hard to differentiate actual faults from the changes in energy consumption due to seasonal variations and changes in personal profiles such as holidays etc.
-In practice, multiple anomaly detection tools are used to continuously raise alarms for different application domains. These alarms include both true positives and false alarms. Operators act on these alarms for diagnosis and deeper root cause
+In practice, multiple anomaly detection tools are used to continuously raise alarms for different application domains. These alarms include both true positives and false alarms. Operators act on these alarms for diagnosis and deeper root cause analysis and take appropriate maintenance actions to mitigate the anomalous behaviours. Given the scale and scope of the district heating substations, the operators can be overwhelmed with the large number of alarms at any given instant. It is therefore necessary to prioritize and rank these alarms by their severity. In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.
 Objectives:
@@ Line 37: / Line 28: @@
 -Applying and comparing state-of-the-art unsupervised anomaly detection methods
--Implementing novel ensemble method by combining multiple detectors
+-Implementing novel anomaly ranking schema, for example based on aggregating results from different detectors
--Implementing novel anomaly ranking schema in order to aggregate results from different detectors
+-Categorisation of anomalies
 -Testing anomaly-ranking framework on district heating and heat pump datasets.
 We have collaboration with 2 companies within this project i.e. HEM and Öresundskraft.

@@ Line 1: / Line 1: @@
 {{StudentProjectTemplate
 |Summary=In this project, we aim to propose a novel anomaly ranking algorithm in order to monitor district heating substations.
 |References=M. Goldstein and S. Uchida, "A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data", PLOS ONE, vol. 11, no. 4, p. e0152173, 2016.
@@ Line 11: / Line 12: @@
 Huang, Huaming, "Rank Based Anomaly Detection Algorithms" (2013). Electrical Engineering and Computer Science - Dissertations.Paper 331.
 |Prerequisites=Artificial Intelligence and Learning Systems courses; good knowledge of data mining; programming skills for implementing machine learning algorithms
-|Supervisor=Ec
+|Supervisor=Ece Calikus, Sławomir Nowaczyk
 }}
 District heating system is a common way to distribute heat through underground pipelines for residential and commercial requirements. Faults are common in district energy systems due to the high number of substations and instrumentation components. Also, the standard energy-metering instrumentation is designed for low cost and billing, not for automated fault detection. Large variations in building dynamics, building subsystems, human behaviour and the environment make the system complex to model and analyse.
@@ Line 40: / Line 43: @@
 -Testing anomaly-ranking framework on district heating and heat pump datasets.
- We have collaboration with 2 companies within this project i.e. HEM and Öresundskraft.
+We have collaboration with 2 companies within this project i.e. HEM and Öresundskraft.