Development of an intelligent model to categorise residential water end use events

Nguyen, Khoi; Zhang, Hong; Stewart, Rodney Anthony

doi:10.1016/j.jher.2013.02.004

Cited by 71 publications

(78 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Alternatively, an automatic prototype based on machine learning algorithms was proposed by Nguyen et al [13][14][15][16] to disaggregate and classify water consumption events. Unfortunately, the universal usability and compatibility of the tool is limited by the fact that the algorithms were trained with data originated from a specific water meter/data logger combination.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, all data were collected in the same geographical area from consumers sharing very similar water consumption habits and water appliances. Furthermore, the set of data employed for the training of the proposed machine learning tool has been obtained using Trace Wizard ® software, which has limited capabilities for disaggregating overlapped consumption events [13]. Following a similar approach, Piga et al [17] proposed an automated water and energy end use disaggregation, which has only been tested against electric energy data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Water End Use Disaggregation Based on Soft Computing Techniques

Pastor-Jabaloyes

Arregui

Cobacho

2018

Water

View full text Add to dashboard Cite

Disaggregating residential water end use events through the available commercial tools needs a great investment in time to manually process smart metering data. Therefore, it is extremely difficult to achieve a homogenous and sufficiently large corpus of classified single-use events capable of accurately describe residential water consumption. The main goal of the present paper is to develop an automatic tool that facilitates the disaggregation of the individual water consumptions events from the raw flow trace. The proposed disaggregation methodology is conducted through two actions that are iteratively performed: first, the use of an advanced two-step filter, whose calibration is automatically conducted by the Elitist Non-Dominated Sorting Genetic Algorithm NSGA-II; and second, a cropping algorithm based on the filtered water consumption flow traces. As a secondary goal, yet complementary to the main one, a semiautomatic massive classification process has been developed, so that the resulting single-use events can be easily categorized in the different water end uses in a household. This methodology was tested using water consumption data from two different case studies. The characteristics of the households taken as reference and their occupants were unequivocally dissimilar from each other. In addition, the monitoring equipment used to obtain the consumption flow traces had completely different technical specifications. The results obtained from the processing of the two studies show that the automatic disaggregation is both robust and accurate, and produces significant time saving compared to the standard manual analysis.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Water End Use Disaggregation Based on Soft Computing Techniques

Pastor-Jabaloyes

Arregui

Cobacho

2018

Water

View full text Add to dashboard Cite

show abstract

“…Nguyen et al [2013] proposed a novel activity recognition algorithm and compared it with TraceWizard©. They were able to improve labeling accuracy for activities such as dishwasher use, but not for activities such as garden watering and bathtub use.…”

Section: Analysis Of Consumptionmentioning

confidence: 99%

Water use signature patterns for analyzing household consumption using medium resolution meter data

Cardell-Oliver

2013

Water Resour. Res.

View full text Add to dashboard Cite

[1] Providers of potable water to households and businesses are charged with conserving water. Addressing this challenge requires accurate information about how water is actually being used. So smart meters are being deployed on a large scale by water providers to collect medium resolution water use data. This paper presents water use signature patterns, the first technique designed for medium resolution meters for discovering patterns that explain how households use water. Signature patterns are clusters (subsets) of water meter readings specified by patterns on volumes and calendar dates. Four types of signature pattern are introduced in this paper: continuous flow days; exceptional peak use days; programmed patterns with recurrent hours; and normal use partitioned by season and period of the day. Signature patterns for each household are calculated using efficient selection rules that scale for city populations and years of data collection. Data from a real-world, large-scale, smart metering trial are analyzed using water use signature patterns. The results demonstrate that water use behaviors are distinctive, for both individuals and populations. Signatures can identify behaviors that are promising targets for water conservation. Pattern discovery can be automated with an efficient and scalable computer program. By identifying relevant consumption patterns in medium resolution meter data, water use signature patterns can help to achieve the water conservation potential of large-scale smart metering.Citation: Cardell-Oliver, R. (2013), Water use signature patterns for analyzing household consumption using medium resolution meter data, Water Resour. Res., 49,[8589][8590][8591][8592][8593][8594][8595][8596][8597][8598][8599]

show abstract

“…The second approach achieves high accuracy and does not require human interaction once the system is operating but requires sensors to be attached to many water use devices in the home, which makes this technique cost-intensive, intrusive and can artificially influence water use behaviour [4,5]. The third approach (see [5][6][7]) overcomes the deficiencies of the first two, by only requiring a smart metre installed at the property boundary (i.e. replacement of traditional metre only).…”

Section: Introductionmentioning

confidence: 99%

“…The first prototype version of Autoflow was founded on an analytical procedure developed by the authors (see [5][6][7]) and achieved an average overall pattern recognition accuracy across all categories of 85%. However, higher recognition accuracies and the ability to recognise and adapt to new or altered end uses are needed before a commercially viable software tool can be developed.…”

Section: Introductionmentioning

confidence: 99%

Intelligent autonomous system for residential water end use classification: Autoflow

Nguyen

Stewart

Zhang

et al. 2015

Applied Soft Computing

Self Cite

View full text Add to dashboard Cite

a b s t r a c tOver half of the world's population will live in urban areas in the next decade, which will impose significant pressure on water security. The advanced management of water resources and their consumption is pivotal to maintaining a sustainable water future. To contribute to this goal, the aim of this study was to develop an autonomous and intelligent system for residential water end-use classification that could interface with customers and water business managers via a user-friendly web-based application. Water flow data collected directly from smart water metres connected to dwellings includes both single (e.g., a shower event occurring alone) and combined (i.e., an event that comprises several overlapping single events) water end use events. The authors recently developed an intelligent application called Autoflow which served as a prototype tool to solve the complex problem of autonomously categorising residential water consumption data into a registry of single and combined events. However, this first prototype application achieved overall recognition accuracy of 85%, which is not sufficient for a commercial application. To improve this accuracy level, a larger dataset consisting of over 82,000 events from over 500 homes in Melbourne and South-east Queensland, Australia, were employed to derive a new single event recognition method employing a hybrid combination of Hidden Markov Model (HMM), Artificial Neural Networks (ANN) and the Dynamic Time Warping (DTW) algorithm. The classified single event registry was then used as the foundations of a sophisticated hybrid ANN-HMM combined event disaggregation module, which was able to strip apart concurrently occurring end use events. The new hybrid model's recognition accuracy ranged from 85.9% to 96.1% for single events and 81.8-91.5% for combined event disaggregation, which was a 4.9% and 8.0% improvement, respectively, when compared to the first prototype model. The developed Autoflow tool has far-reaching implications for enhanced urban water demand planning and management, sustained customer behaviour change through more granular water conservation awareness, and better customer satisfaction with water utility providers.

show abstract

Development of an intelligent model to categorise residential water end use events

Cited by 71 publications

References 32 publications

Water End Use Disaggregation Based on Soft Computing Techniques

Water End Use Disaggregation Based on Soft Computing Techniques

Water use signature patterns for analyzing household consumption using medium resolution meter data

Intelligent autonomous system for residential water end use classification: Autoflow

Contact Info

Product

Resources

About