In response to the typical fault issues encountered during the operation of marine diesel engines, a fault diagnosis method based on a convolutional neural network (CNN), a temporal convolutional network (TCN), and the attention mechanism (ATTENTION) is proposed, referred to as CNN-TCN–ATTENTION. This method successfully addresses the issue of insufficient feature extraction in previous fault diagnosis algorithms. The CNN is employed to capture the local features of diesel engine faults; the TCN is employed to explore the correlations and temporal dependencies in sequential data, further obtaining global features; and the attention mechanism is introduced to assign different weights to the features, ultimately achieving intelligent fault diagnosis for marine diesel engines. The results of the experiments demonstrate that the CNN-TCN–ATTENTION-based model achieves an accuracy of 100%, showing superior performance compared to the individual CNN, TCN, and CNN-TCN methods. Compared with commonly used algorithms such as Transformer, long short-term memory (LSTM), Gated Recurrent Unit (GRU), and Deep Belief Network (DBN), the proposed method demonstrates significantly higher accuracy. Furthermore, the model maintains an accuracy of over 90% in noise environments such as random noise, Gaussian noise, and salt-and-pepper noise, demonstrating strong diagnostic performance, generalization capability, and noise robustness. This provides a theoretical basis for its practical application in the fault diagnosis of marine diesel engines.

To address the impact of load variations, external environmental changes, and the tuning of the parameters on Permanent Magnet Synchronous Motors (PMSMs) used in ships, this study proposes an Active Disturbance Rejection Control (ADRC) strategy for PMSMs, optimized by the Quantum-behaved Particle Swarm Optimization (QPSO) algorithm. First, based on the PMSM model, the study addresses the limited disturbance rejection capability of the traditional fal function in the Extended State Observer (ESO) of conventional ADRC. To improve the accuracy of the state observer, the faln function is introduced as a replacement for the traditional fal function. Second, due to the numerous parameters in ADRC, which are difficult to tune, the QPSO algorithm—known for its strong global search capabilities and fast convergence speed—is utilized for parameter optimization. Additionally, the position update formula within the optimization algorithm is revised and optimized. Finally, simulation experiments are conducted using the Matlab/Simulink platform, where practical conditions, such as load fluctuations and random noise, are incorporated. The simulation results demonstrate that, compared to PSO-ADRC control, IPSO-ADRC control, and ICFO-ADRC control, the proposed method offers a superior dynamic response. Specifically, the speed control accuracy is improved by 46.7%, torque ripple is reduced by 50.8%, and harmonic distortion decreases by 23.1%. These results highlight the significant advantages of this method in enhancing system robustness, dynamic response speed, and steady-state accuracy, making it particularly suitable for PMSM control systems in complex dynamic environments, such as those encountered on ships.