Advanced Materials Research Vol. 1039

Abstract: In view of the problem of complex cabling and the difficulty of real time monitoring of the workshop, the wireless network technology, the LabVIEW software and Internet technology are used in this paper to realize the hardware system network, the real time monitoring of temperature and humidity, dust, noise in the workshop, and the web publishing and remote monitoring of the system respectively. In this way, the informatization management level of workshop has been improved, and thus the workshop intelligent control ability can be improved.

Abstract: A three layer BP NN is created to design the cam. The data of cam contour ,which can be measured by CMM, has been used for the training where back propagation method is used.Advantage of solving nonliner problems gives BP netwok the ability to make out a more demand curve of cam. Taking advantage of its learning ability,NN model fits the actual cam contour gradually until the error fulfil the demand.Cam contour is ploted by Matlab,the result of which is better than cubic curve fitting,especially in the aspects of precise and velocity.

Abstract: For the purpose of recycling end of life product, Selective disassembly methods are most common used for dismantling the old product. Only some parts which valuable for remanufacturing or reuse are dismantled. In addition, the optimal disassembly sequence which created automatic by the computer will help to decrease the disassembly cost and increase the whole revenue of recycling process. However, the disassembly model are still cannot be composed by the computer automatic, that is, some of the work need be done manually. Especially, the priority information among the parts should be analyzed by the engineer. In this paper, an automatic method is presented by comparing the feature of the parts. And then, by extracting and analyzing mates in the assembly, the adjacent information is obtained. Adjacency information and priority information respectively expressed by adjacency matrix and influence matrix, which can be used to depict the hybrid graph model in matrix-form, achieving the automatically creation of disassembly hybrid graph model.

Abstract: Intelligent predictive maintenance (IPdM) is a maintenance strategy that makes maintenance decisions automatically and dynamically based on Artificial Intelligence and Data mining techniques through condition monitoring of machines, equipment and production processes. IPdM system consists of the following main modules: sensor and data acquisition, signal and data processing, feature extractions, maintenance decision-making, key performance indicators, maintenance scheduling optimization and feedback control and compensation. Among them, the most important part of IPdM is maintenance decision-making, which includes diagnostics and prognostics. This paper proposes a framework of intelligent faults diagnosis and prognosis system (IFDaPS) and discuss some key technologies for implement IPdM policy in manufacturing and industries. A case study focus on the vibration signals collected from the sensors mounted on a pressure blower for critical components monitoring. We decompose the pre-processed signals into several signals using Wavelet Packet Decomposition (WPD), and then the signals are transformed to frequency domain using Fast Fourier Transform (FFT). The features extracted from frequency domain are used to train Artificial Neural Network (ANN). Trained ANN model is able to identify the fault of the components and predict its Remaining Useful Life (RUL). The case study demonstrates how to implement the proposed framework and intelligent technologies for IPdM and the result indicates its higher efficiency and effectiveness comparing to traditional methods.

Abstract: This paper addresses a single-machine large-scale rescheduling problems with efficiency and stability subject to machine breakdown. Partial rescheduling (PR) strategy is used to cope with the computational complexity. Two kinds of objective functions of PR sub-problem, where the global dual objectives are reflected to an extent, are designed respectively for the procedural PR horizon and the terminal PR horizon. The PR problem is solved by a branch and bound algorithm. Lower bound and upper bound procedures as well as dominance rules are developed for the branch and bound algorithm. An extensive experimentation was conducted. The computational results show that the branch and bound can solve PR sub-problems with certain scales and the partial rescheduling procedure developed can greatly improve the stability of schedule with little sacrifice in efficiency and provide a reasonable trade-off between solution quality and computational cost.

Abstract: This paper proposes a two-level robust optimization model in the context of job shop scheduling problem. The job shop scheduling problem optimizes the makespan under uncertain processing times, which are described by a set of scenarios. In the first-level optimization, a traditional stochastic optimization model is conducted to obtain the optimal expected performance as a standard performance, on which a concept of bad-scenario set is defined. In the second-level optimization, a robustness measure is given based on bad-scenario set. The objective function for the second robust optimization model is to combine expected performance and robustness measure. Finally, an extensive experiment was conducted to investigate the advantages of the proposed robust optimization model. The computational results show that the two-level model can achieve a better compromise between average performance and robustness than the existing robust optimization models.

Abstract: According to the problem of being hard to achieve precise management of production in the actual production environment, RFID tags are used to identity production materials and RFID readers are placed at some important workstations and warehouse channels, to gain products information in real-time, automatically and accurately. On the basis, the paper uses J2EE programming to design the real-time scheduling management system of WIP (work-in-process), and realizes the closed-loop control of production. The system has been successfully applied to Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, and satisfactory results are obtained.

Abstract: 3D vision based quality inspection has been widely applied in manufacturing industry. Product quality is retrieved from the point cloud obtained using 3D vision methods. Generally, three sorts of quality inspection methods can be selected according to the specific requirements. This paper studied a combining quality inspection method for the quality inspection of a plastic molded part with multiple geometry shapes. Only incomplete point cloud is available because of the characteristics of the part material. Shape fitting and template matching methods are applied for deformation detection with respect to different shapes. Experiment result shows the proposed method can accomplish the quality inspection task for the part with multiple geometry shapes.

Abstract: This paper describes two methods for the industrial quality inspection: Supervised classification algorithm Chi-Square Automatic Interaction Detector (CHAID) and unsupervised clustering algorithm Self-Organizing Map (SOM). The classification and clustering are modelled in IBM software SPSS. Models’ functioning is illustrated on a wheel assembly geometric features inspection. The classifying accuracies are compared for the two methods. CHAID has shown better classifying ability than SOM, while SOM can be used to improve quality of predictor values, and therefore classifiers accuracy.

Abstract: With the proliferation of radio frequency identification (RFID) systems, existing two dimensional (2-D) location algorithms cannot meet the manufacturing demand anymore. In this paper, an efficient degradation particle swarm optimization (DPSO) algorithm is proposed to solve the three dimensional (3-D) location problems in passive tag RFID systems. Performance evaluation shows this method can approach the actual target tag position with acceptable deviation and stability which can meet the newly generated production demand.