Condition classification of heating systems valves based on acoustic features and machine learning

Potočnik, Primož; Olmos, B.; Vodopivec, Lučka; Susič, Egon; Govekar, Edvard

doi:10.1016/j.apacoust.2020.107736

Cited by 17 publications

(7 citation statements)

References 41 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After analyzing the sound pressure at various frequency bands, they observed that there was a clear and marked increase in pressure level at 4 kHz at the time of inception. Potocňik et al 36 used extraction of spectral acoustic features for identifying inception of cavitation in valves.…”

Section: Introductionmentioning

confidence: 99%

Acoustic Analysis of Vortex-based Cavitation Devices: Inception and extent of cavitation

Ranade¹,

Sarvothaman

Ranade

2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Hydrodynamic cavitation (HC) is a process of generation, growth, and collapse of gas-/vapor-filled cavities leading to intense shear and localized hot spots. It is essential to identify the inception and extent of cavitation for ensuring appropriate operation of HC devices and processes. In this work, we demonstrate for the first time, usefulness of acoustic data acquired using an everyday mobile phone for characterizing inception and extent of cavitation. Acoustic data from vortex-based cavitation devices for a range of operating pressure drop (0−390 kPa) was obtained. Systematic methodology for identifying relevant acoustic features is presented. "Audio" and "DSP" Toolboxes of MATLAB were used for processing acoustic data. Three specific trends of extracted features with respect to the flow rate/ pressure drop across the HC device were observed. All three trends clearly identified inception of cavitation between 50 and 80 kPa pressure drop across the HC device. An attempt is made to connect features extracted from acoustic signals with the extent of cavitation in terms of perpass performance of HC device. The "flatness" was found to capture the influence of the HC device scale on performance (in other words, extent of cavitation) reasonably well. The methodology is quite general and will be applicable for any cavitation device. The presented results will be useful for online identification of inception and extent of hydrodynamic cavitation.

show abstract

Section: Introductionmentioning

confidence: 99%

Acoustic Analysis of Vortex-based Cavitation Devices: Inception and extent of cavitation

Ranade¹,

Sarvothaman

Ranade

2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…In another work [42], a prediction model of secondary supply temperature is built using indoor temperature and building thermal inertia to achieve more refined control. On the other hand, Potočnik et al [43] focuses on assessing the quality and condition of valves installed in district heating systems. A method for classification of valve sounds is proposed, based on acoustic features and Machine Learning models.…”

Section: Review Of Recent Research Articlesmentioning

confidence: 99%

Opportunities for Machine Learning in District Heating

et al. 2021

View full text Add to dashboard Cite

The district heating (DH) industry is facing an important transformation towards more efficient networks that utilise significantly lower water temperatures to distribute the heat. This change requires taking advantage of new technologies, and Machine Learning (ML) is a popular direction. In the last decade, we have witnessed an extreme growth in the number of published research papers that focus on applying ML techniques to the DH domain. However, based on our experience in the field, and an extensive review of the state-of-the-art, we perceive a mismatch between the most popular research directions, such as forecasting, and the challenges faced by the DH industry. In this work, we present our findings, explain and demonstrate the key gaps between the two communities and suggest a road-map ahead towards increasing the impact of ML research in the DH industry.

show abstract

“…Tiwari et al [31] extracted 6 statistical features from the time domain pressure data, and then fed these features as input into the neural network to classify blockage and cavitation. Potocnik et al [32] extracted spectral and psychoacoustic features from the valve acoustic data and then took these features to be input of a variety of machine learning algorithms to classify the cavitation, flow noise, whistling and rattling.…”

Section: Introductionmentioning

confidence: 99%

A multi-task learning for cavitation detection and cavitation intensity recognition of valve acoustic signals

Sha,

Faber,

Gou

et al. 2022

Preprint

View full text Add to dashboard Cite

With the rapid development of smart manufacturing, data-driven machinery health management has received a growing attention. As one of the most popular methods in machinery health management, deep learning (DL) has achieved remarkable successes. However, due to the issues of limited samples and poor separability of different cavitation states of acoustic signals, which greatly hinder the eventual performance of DL modes for cavitation intensity recognition and cavitation detection. Also different task were performed separately conventionally. In this work, a novel multi-task learning framework for simultaneous cavitation detection and cavitation intensity recognition framework using 1-D double hierarchical residual networks (1-D DHRN) is proposed for analyzing valves acoustic signals. Firstly, a data augmentation method based on sliding window with fast Fourier transform (Swin-FFT) is developed to alleviate the small-sample issue confronted in this study. Secondly, a 1-D double hierarchical residual block (1-D DHRB) is constructed to capture sensitive features from the frequency domain acoustic signals of valve. Then, a new structure of 1-D DHRN is proposed. Finally, the devised 1-D DHRN is evaluated on two datasets of valve acoustic signals without noise (Dataset 1 and Dataset 2) and one dataset of valve acoustic signals with realistic surrounding noise (Dataset 3) provided by SAMSON AG (Frankfurt). Our method has achieved state-of-the-art results. The prediction accurcies of 1-D DHRN for cavitation intensitys recognition are as high as 93.75%, 94.31% and 100%, which indicates that 1-D DHRN outperforms other DL models and conventional methods. At the same time, the testing accuracies of 1-D DHRN for cavitation detection are as high as 97.02%, 97.64% and 100%. In addition, 1-D DHRN has also been tested for different frequencies of samples and shows excellent results for frequency of samples that mobile phones can accommodate.

show abstract

Condition classification of heating systems valves based on acoustic features and machine learning

Cited by 17 publications

References 41 publications

Acoustic Analysis of Vortex-based Cavitation Devices: Inception and extent of cavitation

Acoustic Analysis of Vortex-based Cavitation Devices: Inception and extent of cavitation

Opportunities for Machine Learning in District Heating

A multi-task learning for cavitation detection and cavitation intensity recognition of valve acoustic signals

Contact Info

Product

Resources

About