Search

In order to predict the reliability of semiconductor lasers, an accelerated degradation test (ADT) was proposed as an element of reliability testing. Temperature-stressed ADT was applied for 8 Semiconductor lasers which used in space missions, and the degradation characteristics of output power of semiconductor lasers were studied. Then, a reliability model based multi-output Gaussian process regression (MOGP) was proposed to evaluate the lifetime and reliability for laser diodes. The advantage of the proposed MOGP based method is that it utilizes the output correlation between multiple degradation traces to make the outputs utilize each other's information and provide more accurate prediction than single modeling. Thereby improving the prediction accuracy. Furthermore, verifying applications and cases studies are discussed to prove the generality and practicability of the proposed reliability prediction model. Results show that the accuracy of the proposed MOGP based method is twice that of the SVM method.

As equipment becomes more and more complex, it is increasingly difficult to manually extract and select fault features manually based on expert experience or signal processing techniques. In addition, the shallow model such as BP neural network and SVM have trouble to deal with the complex mapping relationship with respect to the measured signal and the health condition of the equipment, who faces the problem of dimensional disaster. Combined with the advantages of deep confidence network (DBN) in features extraction and deal with high-dimensional and nonlinear samples, a fault feature extraction and diagnosis method based on deep confidence network for gearbox is investigated in this framework. The method uses the original time domain signal to train the deep confidence network and completes the intelligent diagnosis through deep learning. The preponderance is that it can take out the dependence on a great quantity of signal processing techniques and diagnostic experience, and accomplish the extraction of fault features and the intelligent diagnosis of health status with the characteristic of self-adaption. The method has no periodic requirements for time domain signals, and has strong versatility and adaptability. The experimental results of the fault diagnosis for the planetary gearbox demonstrated the feasibility and superiority of the presented method.

Maintenance simulation plays an important role in maintainability design and analysis of products. The immersive virtual maintenance simulation, with motion capture devices driving the virtual personnel to perform the same action as the operator does, can make the maintenance simulation more real and valuable. However, because of a lack of interaction approach between the operator and simulation system, the operator cannot drive the virtual personnel to interact with 3D entities in virtual environment, and consequently, interactions like grasping and carrying become impossible. Therefore, three is a great need to find a way to deal with the plight. In this paper, a speech recognition-based interaction approach is put forward for immersive virtual maintenance simulation to help the operator to control the interaction process with voice command. Based on analysis on maintenance action, an abstract mechanism for instruction language is developed. Next, the paper gives the simulation framework for speech recognition-based immersive virtual maintenance simulation system. Through Microsoft Speech SDK and Secondary Development Technology of DELMIA, a functional simulation system is built and used to carry out an engine maintenance simulation. The result of this application shows the efficiency and effectiveness of the speech recognition-based interaction approach proposed for immersive virtual maintenance simulation.

Virtual reality technology is a new way for people to visualize and interact with complex data through computers. Virtual reality technology based on simulation software is relatively mature. However, the simulation process is more complicated and the simulation of the interactivity cannot be satisfactory. Immersive virtual objects in the virtual scene are more interactive, and more in line with the laws of physics, and more consistent with the real world. The interactive features of virtual maintenance objects are expressed in a variety of ways. It uses modeling and provides constraint method to express the inherent attribute, object such as size, color, geometry. At the same time, we use software to correct the picture of the separation and the non-synchronization caused by the vertical relationship between screens. This paper uses a variety of techniques to achieve the acquisition and expression of virtual interactive features.

The real-time interaction between the user and virtual environment is now possible thanks to immersive virtual maintenance system, which greatly facilitates the maintainability design. At present, the methods to drive objects in virtual environment are mostly based on pose recognition, multi sensor or force, whose accuracy is not high enough, efficiency low and interaction unnatural. The development of Multi-Agent System (MAS) enlightened a new idea to achieve deep interaction of virtual maintenance. In this paper, we focus on the modeling and simulation of virtual maintenance by using multi-agent technology. Firstly, the framework of Multi-agent based virtual maintenance simulation system (MVMSS) is designed. Then virtual human, maintenance tools and device can be regarded as different agents and these agents are modeled with similar structures and the behavior selection mechanism for maintenance tools agents are presented. In order to demonstrate the feasibility and advantages of MVMSS, a maintenance case is simulated based on the proposed system and the traditional system respectively. The result shows that the MVMSS could achieve more natural, efficient and high-fidelity interaction.

As the products become more and more complex, it is very difficult for the quality analysis of the complex products. In order to facilitate analysis, complex products are usually divided into several zones reasonably. Products arranged in the same zone usually have some spatial correlation which can't be found in design stage until the operation stage, which usually has damage to the quality of products. Hence, it is important to express the correlation among products in the same zone accurately and clearly. However, the traditional physical models cannot express the correlation among products clearly. Although a product's function structure can be effectively and clearly described by using generalized directed graph (GDG) theory, it can't express the correlation between function and environment. In this paper, we extend generalized directed graph theory first, adding color and shape elements to express the information clearly. Then the modeling method of function structure is proposed for zonal products by using improved generalized directed graph (IGDG). Finally, a power screwdriver is modeled as an example. Comparing with the results of GDG, we can find that the correlation between function and environment can be expressed clearly by using IGDG. This provides effective information for the designers in the design and layout stage, avoiding the damage of the correlation on the quality of the zonal products.