No abstract

Intelligent fault diagnosis methods based on deep learning becomes a research hotspot in the fault diagnosis field. Automatically and accurately identifying the incipient micro-fault of rotating machinery, especially for fault orientations and severity degree, is still a major challenge in the field of intelligent fault diagnosis. The traditional fault diagnosis methods rely on the manual feature extraction of engineers with prior knowledge. To effectively identify an incipient fault in rotating machinery, this paper proposes a novel method, namely improved the convolutional neural network-support vector machine (CNN-SVM) method. This method improves the traditional convolutional neural network (CNN) model structure by introducing the global average pooling technology and SVM. Firstly, the temporal and spatial multichannel raw data from multiple sensors is directly input into the improved CNN-Softmax model for the training of the CNN model. Secondly, the improved CNN are used for extracting representative features from the raw fault data. Finally, the extracted sparse representative feature vectors are input into SVM for fault classification. The proposed method is applied to the diagnosis multichannel vibration signal monitoring data of a rolling bearing. The results confirm that the proposed method is more effective than other existing intelligence diagnosis methods including SVM, K-nearest neighbor, back-propagation neural network, deep BP neural network, and traditional CNN.

In this article, the operation of a large containership main engine was investigated with emphasis at slow steaming conditions. A cycle mean value approach implemented in the MATLAB/Simulink environment was adopted to simulate the two-stroke marine diesel engine due to the fact that it combines simplicity with adequate prediction accuracy. For accurately representing the compressor performance when the engine operates at low loads, the extension of the compressor map at the low rotational speed region was carried out based on a non-dimensional parameters method incorporating a novel way of calculating the compressor isentropic efficiency. The compressor map extension method results were validated using a corrected similarity laws approach. The engine steady state operation for various loads was simulated and the predicted engine performance parameters were validated using shop trial measurements. Furthermore, the engine transient operation in the load region below 50% was studied and the simulation results including the compressor operating points trajectory are presented and discussed. Based on the obtained results, the influence of the activation/deactivation of the installed electric driven blowers and the turbocharger cut-out on the engine operation was analysed

Even though ghost imaging (GI), an unconventional imaging method, has received increased attention by researchers during the last decades, imaging speed is still not satisfactory. Once the data-acquisition method and the system parameters are determined, only the processing method has the potential to accelerate image-processing significantly. However, both the basic correlation method and the compressed sensing algorithm, which are often used for ghost imaging, have their own problems. To overcome these challenges, a novel deep learning ghost imaging method is proposed in this paper. We modified the convolutional neural network that is commonly used in deep learning to fit the characteristics of ghost imaging. This modified network can be referred to as ghost imaging convolutional neural network. Our simulations and experiments confirm that, using this new method, a target image can be obtained faster and more accurate at low sampling rate compared with conventional GI method.

Abstract:In this article, the operation of a large two-stroke marine diesel engine including various cases with turbocharger cut-out was thoroughly investigated by using a modular zero-dimensional engine model built in MATLAB/Simulink environment. The model was developed by using as a basis an in-house modular mean value engine model, in which the existing cylinder block was replaced by a more detailed one that is capable of representing the scavenging ports-cylinder-exhaust valve processes. Simulation of the engine operation at steady state conditions was performed and the derived engine performance parameters were compared with the respective values obtained by the engine shop trials. The investigation of engine operation under turbocharger cut-out conditions in the region from 10% to 50% load was carried out and the influence of turbocharger cut-out on engine performance including the in-cylinder parameters was comprehensively studied. The recommended schedule for the combination of the turbocharger cut-out and blower activation was discussed for the engine operation under part load conditions. Finally, the influence of engine operating strategies on the annual fuel savings, CO2 emissions reduction and blower operating hours for a Panamax container ship operating at slow steaming conditions is presented and discussed. OPEN ACCESSEnergies 2015, 8 5739

No abstract

This study focuses on the development of an extended MVEM capable of predicting the engine performance parameters (thermodynamic, flow and mechanical) of two-stroke marine engines at varying settings of injection timing and turbine area. The extension employed mapping of a number of the engine parameters carried out based on a zero-dimensional model. Both the zero-dimensional and the mean value engine models were developed in MATLAB/Simulink environment following the same modular approach and their accuracy was validated against experimental data from shop trials. Subsequently, the zero-dimensional model was used for engine parametric simulation by changing the start of fuel injection and the turbocharger turbine area. By analyzing the derived results, the relationships between the investigated engine parameters were established and the appropriate corrections were applied in the MVEM. The extended MVEM was benchmarked against the zero-dimensional model and MVEM at steady and transient conditions and the derived results were analysed and discussed revealing the advantages and limitations of the investigated modelling approaches. Based on the obtained results, the proposed extension methodology improves the MVEM prediction capability without considerably increasing the complexity and the execution time and therefore, it can be employed for the engine performance prediction in control system design investigations overcoming limitations of the MVE

No abstract