Investigating pump cavitation based on audio sound signature recognition using artificial neural network

Arendra, Anis; Sabarudin, Akhmad; Winarso, Kukuh; Herianto, -

doi:10.1088/1742-6596/1569/3/032044

Cited by 3 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In actual engineering, it is usually considered that the head of the centrifugal pump drops by 3% as the basis for the onset of cavitation (Arendra et al, 2020). In this paper, the cavitation development process is divided into three stages according to the head drop ratio: non-cavitation (NPSH a = 7.49 m), cavitation inception (head drop 3%, NPSH a = 1.89 m) and severe cavitation (head drop 6%, NPSH a = 1.76 m).…”

Section: Resultsmentioning

confidence: 99%

Recognition of cavitation characteristics in non-clogging pumps based on the improved Lévy flight bat algorithm

Lang,

Ni,

Chen

et al. 2023

Front. Energy Res.

View full text Add to dashboard Cite

The performance and operational stability of non-clogging pumps can be affected by cavitation. To accurately identify the cavitation state of the non-clogging pump and provide technical references for monitoring its operation, a study was conducted on the optimization of Elman neural networks for cavitation monitoring and identification using the Improved Lévy Flight Bat Algorithm (ILBA) on the basis of the traditional Bat Algorithm (BA). The ILBA employs multiple bats to interact and search for targets and utilizes the local search strategy of Lévy flight, effectively avoiding local minima by taking advantage of the non-uniform random walk characteristics of large jumps. The ILBA algorithm demonstrates superior performance compared to other traditional algorithms through simulation testing and comparative calculations with eight benchmark test functions. On this basis, the optimization of the weights and thresholds of the Elman neural network was carried out by the improved bat algorithm. This leads to an enhancement in the accuracy of the neural network for identifying and classifying cavitation data, and the establishment of the ILBA-Elman cavitation diagnosis model was achieved. Collect pressure pulsation signals at the tongue of the non-clogging pump volute through cavitation tests. Through the cavitation feature extraction method based on Variational Mode Decomposition (VMD) and Multi-scale Dispersion Entropy (MDE), the interference signal can be effectively suppressed and the complexity of the time series can be measured from multiple angles, thereby creating a cavitation feature data set. The improved cavitation diagnosis model (ILBA-Elman) can realize the effective identification of the cavitation characteristics of non-clogging pumps through a variety of algorithm comparison experiments.

show abstract

Section: Resultsmentioning

confidence: 99%

Recognition of cavitation characteristics in non-clogging pumps based on the improved Lévy flight bat algorithm

Lang,

Ni,

Chen

et al. 2023

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…A considerable number of studies have focused on investigating cavitation in kinetic pumps, as well as water turbines, as revealed by the literature (Al-Obaidi and Towsyfyan, 2019;Bordoloi and Tiwari, 2017;Čdina, 2003;Durdu et al, 2021;Kan et al, 2022;Panda et al, 2018). Recently, researchers have attempted to identify cavitation by utilizing machine learning models (Arendra et al, 2020;Bordoloi and Tiwari, 2017;Matloobi and Riahi, 2021;Panda et al, 2018;Wang, et al, 2019;Wang et al, 2020) . However, given the unpredictable nature of cavitation, an accurate numerical estimation of the resulting noise and vibration is not feasible.…”

Section: Introductionmentioning

confidence: 99%

“…As a result, determining the 3% decrease in Hm value becomes difficult. To address this issue, some studies have utilized artificial neural networks or other machine learning algorithms to predict cavitation in (Arendra et al, 2020;Matloobi and Riahi, 2021;Wang et al, 2020;Wang et al, 2019;Yong et al, 2009). Unlike these previous studies, the present study focuses on predicting the NPSH, noise, and vibration levels associated with the 3% decrease in Hm using an artificial neural network model.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based Prediction of NPSH, Noise, and Vibration Levels in Radial Pumps Under Cavitation Conditions

ORHAN,

KURT,

KIRILMAZ

et al. 2024

Tekirdağ Ziraat Fakültesi Dergisi

View full text Add to dashboard Cite

Cavitation, a physical phenomenon that detrimentally affects pump performance and reduces pump life, can cause wear on pump elements. Various engineering methods have been developed to identify the initiation and full development of the cavitation process. One such method is the determination of the net positive suction head (NPSH) through a 3% decrease in total head (Hm) at a constant flow rate. In radial pumps, commonly used in agricultural irrigation and industry, cavitation conditions result in a sudden drop in the Hm-Q curve, making it challenging to detect the 3% Hm value drop. This study differs from others in the literature by modelling NPSH, noise, and vibration levels using three machine learning models, specifically artificial neural networks (ANN), support vector machines (SVM), and decision tree regression (DTR). The best-performing model predicts NPSH, noise, and vibration levels corresponding to a 3% decrease in Hm level. The present study determined the NPSH values of a horizontal shaft centrifugal pump at different flow rates and constant operating speed, and the vibration and noise levels were measured for these NPSH values. For each of the NPSH, noise, and vibration levels, ANN, SVM and DTR models were created. The performances of these models were evaluated using criteria such as root mean squared error (RMSE), Mean Absolute Error (MAE) and mean absolute percentage error (MAPE). In addition, Taylor and error box diagrams were created. The ANN model and DTR yielded high accuracy predictions for NPSH values (R2 = 0.86 and R2 = 0.8, respectively). The ANN model provided the best prediction performance for noise and vibration levels. By entering the level of 3% drop in the Hm value of the pump as external data input to the ANN model, NPSH3, noise, and vibration levels were determined. The ANN models can be effectively employed to determine NPSH3, noise, and vibration levels, particularly in radial flow pumps, where detecting 3% reductions in manometric height value is challenging.

show abstract

A novel application of deep transfer learning with audio pre-trained models in pump audio fault detection

Anvar

Mohammadi

2023

Computers in Industry

View full text Add to dashboard Cite

Investigating pump cavitation based on audio sound signature recognition using artificial neural network

Cited by 3 publications

References 9 publications

Recognition of cavitation characteristics in non-clogging pumps based on the improved Lévy flight bat algorithm

Recognition of cavitation characteristics in non-clogging pumps based on the improved Lévy flight bat algorithm

Machine Learning-Based Prediction of NPSH, Noise, and Vibration Levels in Radial Pumps Under Cavitation Conditions

A novel application of deep transfer learning with audio pre-trained models in pump audio fault detection

Contact Info

Product

Resources

About