Real-Time Classification of Diesel Marine Engine Loads Using Machine Learning

Shahid, Syed Maaz; Ko, Sung H.; Kwon, Sungoh

doi:10.3390/s19143172

Cited by 12 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Information about the relative load is needed to optimize the parameter adjustments during the exploitation of an ICE. Shahid et al [37] studied relative engine load classification using measured signals depending on the degree of crankshaft rotation. An ANN was trained using processed data for five grades of relative load and classified with an accuracy of 99.4% using a support vector machine.…”

Section: Existing Methods Of Damage Recognitionmentioning

confidence: 99%

Application of Machine Learning to Classify the Technical Condition of Marine Engine Injectors Based on Experimental Vibration Displacement Parameters

Monieta,

Kasyk

2023

Energies

View full text Add to dashboard Cite

The article presents the possibility of using machine learning (ML) in artificial intelligence to classify the technical state of marine engine injectors. The technical condition of the internal combustion engine and injection apparatus significantly determines the composition of the outlet gases. For this purpose, an analytical package using modern technology assigns experimental test scores to appropriate classes. The graded changes in the value of diagnostic parameters were measured on the injection subsystem bench outside the engine. The influence of the operating conditions of the fuel injection subsystem and injector condition features on the injector needle vibration displacement waveforms was subjected to a neural network (NN) ML process and then tested. Diagnostic parameters analyzed in the amplitude, frequency, and time–frequency domains were subjected after a learning process to recognize simulated various regulatory and technical states of suitability and unfitness with single and complex damage of new and worn injector nozzles. Classification results were satisfactory in testing single damage and multiple changes in technical state characteristics for unfitness states with random wear injectors. Testing quality reached above 90% using selected NNs of Statistica 13.3 and MATLAB R2022a environments.

show abstract

Section: Existing Methods Of Damage Recognitionmentioning

confidence: 99%

Application of Machine Learning to Classify the Technical Condition of Marine Engine Injectors Based on Experimental Vibration Displacement Parameters

Monieta,

Kasyk

2023

Energies

View full text Add to dashboard Cite

show abstract

“…Liang et al [13] proposed a self-organizing comprehensive real-time state evaluation model based on operating condition (OC) classification and identification in the field of oil pump units, which can realize OC identification on real-time monitoring data. Shahid et al [14] proposed an artificial neural network real-time internal combustion engine load classification method based on sensor signals, which can be used to identify five types of load states, with an accuracy of 99.4%. Jiang et al [15] proposed a multifactor working condition identification method based on a one-dimensional convolutional neural network (CNN) and a long-and shortterm network (LSTM) and completed the identification of 12 working conditions of a diesel engine.…”

Section: Introductionmentioning

confidence: 99%

Decoupling Identification Method of Continuous Working Conditions of Diesel Engines Based on a Graph Self-Attention Network

et al. 2022

View full text Add to dashboard Cite

Complex and changeable working conditions are important factors affecting the accuracy and robustness of diesel engine fault diagnosis models. Working condition identification can provide a basic reference for the unit operation state, which is of great significance for fault diagnosis. At present, most working condition identification models take power as the identification parameter, divide the power parameter into several discrete intervals, and obtain the power interval of the current state through the classification model. However, describing the working condition only by power will lead to the coupling of speed and load parameters, and the working condition parameters should be continuous variables. In this paper, a continuous working condition model decoupled by speed and load parameters is proposed, and the working condition identification model is established based on a graph self-attention network (GSAN). A large number of experimental data of 32 working conditions under normal and typical fault simulations was obtained on a diesel engine experimental bench, which was used for training and testing models. Under the condition of untrained working conditions, the 𝑅2 𝑎𝑑𝑗 coefficients of the proposed method are 96.70% and 97.27% for normal and typical fault experimental data respectively, demonstrating the efficiency of the proposed approach.

show abstract

“…Conventional FD methods mainly consist of analytic model-based, knowledge-based and data-driven approaches [4]. For the diesel engine, early studies of FD have been researched, such as those on artificial neural networks (ANN) [5], the support vector machine (SVM) [6], variational mode decomposition, and Bayesian networks (BN) [7][8][9][10][11].…”

Section: Introductionmentioning

confidence: 99%

Multi-branch convolutional neural networks with integrated cross-entropy for fault diagnosis in diesel engines

Zhao

Mao

Zhang

et al. 2021

Meas. Sci. Technol.

View full text Add to dashboard Cite

Fault diagnosis based on deep learning has become a hot research topic because of the successful application of deep learning in other fields. Due to variable operating conditions and a harsh operating environment, it is extremely difficult to effectively diagnose some typical faults of diesel engines. When operating conditions and environmental factors change, the performance of deep learning models also become extremely unstable. In order to solve these problems, this paper proposes a novel deep learning model, called multi-branch convolutional neural networks (MBCNNs) with an integrated cross-entropy. MBCNN can be embedded in the proposed model and simultaneously equipped with four auxiliary classifiers. The proposed model is trained on two different datasets separately, which consist of six diesel engine faults. The trained network model was compared with other methods to prove the superiority of this network model. Meanwhile, by adding Gaussian white noise, the performance of the MBCNN in different noise environments is investigated. The final results show that the MBCNN with integrated cross-entropy can effectively diagnose different diesel engine faults under variable operating conditions.

show abstract

Real-Time Classification of Diesel Marine Engine Loads Using Machine Learning

Cited by 12 publications

References 21 publications

Application of Machine Learning to Classify the Technical Condition of Marine Engine Injectors Based on Experimental Vibration Displacement Parameters

Application of Machine Learning to Classify the Technical Condition of Marine Engine Injectors Based on Experimental Vibration Displacement Parameters

Decoupling Identification Method of Continuous Working Conditions of Diesel Engines Based on a Graph Self-Attention Network

Multi-branch convolutional neural networks with integrated cross-entropy for fault diagnosis in diesel engines

Contact Info

Product

Resources

About