Accurate and timely fault diagnosis for the diesel engine is crucial to guarantee it works safely and reliably, and reduces the maintenance costs. A novel diagnosis method based on variational mode decomposition (VMD) and kernel-based fuzzy c-means clustering (KFCM) is proposed in this paper. Firstly, the VMD algorithm is optimized to select the most suitable K value adaptively. Then KFCM is employed to classify the feature parameters of intrinsic mode functions (IMFs). Through the comparison of many different parameters, the singular value is selected finally because of the good classification effect. In this paper, the diesel engine fault simulation experiment was carried out to simulate various faults including valve clearance fault, fuel supply fault and common rail pressure fault. Each kind of machine fault varies in different degrees. To prove the effectiveness of VMD-KFCM, the proposed method is compared with empirical mode decomposition (EMD)-KFCM, ensemble empirical mode decomposition (EEMD)-KFCM, VMD-back propagation neural network (BPNN), and VMD-deep belief network (DBN). Results show that VMD-KFCM has advantages in accuracy, simplicity, and efficiency. Therefore, the method proposed in this paper can be used for diesel engine fault diagnosis, and has good application prospects.

Fault diagnosis can improve the safety and reliability of diesel engines. An end-to-end method based on a multi-attention convolutional neural network (MACNN) is proposed for accurate and efficient diesel engine fault diagnosis. By optimizing the arrangement and kernel size of the channel and spatial attention modules, the feature extraction capability is improved, and an improved convolutional block attention module (ICBAM) is obtained. Vibration signal features are acquired using a feature extraction model alternating between the convolutional neural network (CNN) and ICBAM. The feature map is recombined to reconstruct the sequence order information. Next, the self-attention mechanism (SAM) is applied to learn the recombined sequence features directly. A Swish activation function is introduced to solve “Dead ReLU” and improve the accuracy. A dynamic learning rate curve is designed to improve the convergence ability of the model. The diesel engine fault simulation experiment is carried out to simulate three kinds of fault types (abnormal valve clearance, abnormal rail pressure, and insufficient fuel supply), and each kind of fault varies in different degrees. The comparison results show that the accuracy of MACNN on the eight-class fault dataset at different speeds is more than 97%. The testing time of the MACNN is much less than the machine running time (for one work cycle). Therefore, the proposed end-to-end fault diagnosis method has a good application prospect.

Engine fault detection is critical to enhancing the reliability of modern equipment. However, it is challenging to obtain a large number of high-quality labeled data for engines, which is not conducive to improving the training accuracy of deep learning methods. Therefore, this article proposes a fault detection method combining adaptive recursive variational mode decomposition (ARVMD) and component energy distribution spectrum (CEDS). The article first introduces recursive mode into VMD. Then, the mode number is dynamically selected according to the energy distribution of the power spectral density (PSD) to extract the intrinsic mode functions (IMFs) continuously. The quadratic penalty term is optimized correspondingly using SNR. The decomposition results of artificial and real signals demonstrated that ARVMD has higher SNR and efficiency than VMD. Next, the center frequency and unit bandwidth energy of IMF are used to construct CEDS. Fault diagnosis is realized by the CEDS correlation ranking of various faults. Finally, two case studies are performed to illustrate the effectiveness of the proposed method. The results show that the proposed ARVMD-CEDS method provides an efficient and effective solution for single-channel engine fault diagnosis.

Transfer learning (TL) is a powerful approach that enhances the generalizability of cross-domain fault diagnosis. However, the challenge of acquiring high-quality mechanical fault signals limits its application. This article introduces the extreme class imbalance problem in the cross-domain diagnosis, restricting the label space of the target domain while relaxing the restrictions of unsupervised learning. The study proposes a novel generative TL method called fast sparse neural style, which employs sparse representation to capture the domain-invariant fault features as well as the Gram matrix to measure the domain-specific features. Fault features and domain features are proven to be separable in mechanical signals and are fused in the data generation process. Compared to other methods through various cross-domain diagnostic tasks on a piston aero-engine, the proposed method has obvious advantages in tasks with substantial inter-domain differences, demonstrating the potential and research value of generative TL.

Knock is an abnormal combustion phenomenon in gasoline engines. Strong knocks will reduce the efficiency and durability of engine, while with slight knocks engines can run on a high-efficiency state. It is necessary to detect knock and control the state of knock in order to improve the thermal efficiency of engine. This paper proposes a novel approach for detecting engine knocks in various intensities based on vibration signal of engine block using variational mode decomposition (VMD) and semi-supervised local fisher discriminant analysis (SELF). Since the quadratic penalty of recursive VMD has a great influence on decomposition results, the approach establishes the connection between the quadratic penalty and the stop condition by analyzing a large amount of data and quantifies the relationship by polynomial fitting, which reduces the complexity and subjectivity of recursive VMD. A multilevel SELF is developed for solving the problem that SELFs sometimes may not find ideal embedding space under large scale dimensionality reduction. This method adopts multi embedding spaces, with gradually decreasing dimension, to reduce the dimensionality of original data by a series of small steps. Verifications show the proposed approach can achieve high classification accuracy in knock detection and is able to identify the intensity of knock. This research contributes to the field of engine abnormality detection and can be implemented on vibration-based faults diagnosis area.

Increasingly energy and environmental crises put forward higher request on diesel engine. It promotes the development of diesel engine, while the complexity of structure is much higher, which leads to higher probability of faults. In order to recognize the states of engine in harsh environments effectively, variational mode decomposition (VMD) and expectation maximization (EM) are introduced into this paper to analyze multi-channel vibration signals. To select the decomposition level of VMD adaptively, a novel power spectrum segmentation based on scale-space representation is proposed for the optimization of VMD and results show this approach can discriminate different frequency components in high noise circumstance accurately and efficiently. To improve the adaptability and accuracy of EM, a feature selection approach based on genetic algorithm (GA) is introduced to preprocess original data and a cross validation method is used for selecting cluster number adaptively. Combined with these approaches, a diesel engine state recognition scheme based on multi-channel vibration signals using optimized VMD and EM is proposed. Compared with existing method, this scheme shows great advantages in accuracy and efficiency, and could be applied in actual engineering.