Key Engineering Materials Vols. 439-440

Abstract: It is one of the effective way to apply JIT (just in time) mode in the management of discrete job shop to improve the productivity of job shop. A kind of JIT management mode for discrete job shop is proposed in this paper. The machining tools such as cutters and frocks are distributed with materials in JIT mode, so that the inventory of machining tools is reduced, and productivity is improved. The process of distribution plan under this mode is studied. The JIT distribution management information system is developed based on the proposed studies, and is applied in a traditional machining job shop with satisfactory result.

Abstract: Aiming at the characteristics of complicated organization behavior, diversified business process and changeable functions in the construction of networked manufacturing platform, this paper proposed an integrated modeling method for networked manufacturing platform combined with DEMO(Dynamic Essential Modeling of Organization） and NAM (Norm Analysis Method), which making use of the two method’s characteristics sufficiently, to support system modeling with full consideration of organization behavior criterion, function subdivision and business process refinement. This integrated modeling method provides an effective tool to design and develop networked manufacturing platform and achieved good effect in application.

Abstract: In order to improve the sensitive to illegal manipulations of image hashing, a novel robust image hashing algorithm based on fractional Fourier transform (FRFT) for detecting and localizing image tampering is proposed in this paper. The framework of generating an image hashing includes three steps: preprocessing, feature extracting and post processing. The robust hashing sequence is obtained by FRFT coefficients of image blocks. The security of proposed algorithm is totally depended on the orders of FRFT which are saved as secret keys. Experimental results show that the proposed scheme is robust against perceptually acceptable modifications to the image such as JPEG compression, mid-filtering, and rotation, more importantly the tampered place can be identified accurately.

Abstract: To reduce risks of Virtual Enterprise （VE）, a novel Distributed Decision Making （DDM） model for VE risk management was presented. This model had two levels, namely, the top-model and the base-model, which described the decision processes of the owner and the partners, respectively. The owner allocated funds to each member of the VE, the partners then selected their own optimal risk control actions to reduce their risks with the constraint given by the allocated budget. When number of members in VE, the number of risk factors and the number of actions increased, the size of the search space would be very huge, the brief solutions of the models are also discussed.

Abstract: The scope of emerging energy is broad, and the development scale and stage of each kind of energy is also irregular. In order to choose the priority development fields of emerging energy, this paper introduces the particle swarm (PS) optimization algorithm into the neural network (NN) training based on an overall situation stochastic optimization thought, establishes the PS-BP neural network model, which optimizes the initial weight value of BP neural network using PS first, then uses the neural network to complete the study of given accuracy. The simulation results indicated that the improved PS-BP algorithm to be able to solve the slow convergence rate and easy to fall into local minimum of learning network weight and the threshold value of conventional BP algorithm effectively, has the quick convergence rate and the high evaluating precision.

Abstract: On the basis of wavelet denoising and its better time-frequency characteristic, this paper presents an effective vibration signal denoising method for food refrigerant air compressor. The solution of eliminating strong noise is investigated with the combination of soft threshold and exponential lipschitza. The good denoising results show that the presented method is effective for improving the signal noise ratio and builds the good foundation for further extraction of the vibration signals.

Abstract: The paper analyzes the mode and method for unstructured information, compares the similarities and differences between ontology and vocabulary and taxonomy, and discusses the position and the function of ontology for unstructured information from the two aspects of standard system of semantic net and the concept characteristic of ontology. Based on the ontologies DB, the annotation of unstructured information is obtained. Information retrieval which consists of concept-matched retrieval and rule-based reasoning retrieval is proposed with the help of information annotation. The experiment results show that rule-based retrieval has better recall and shorter retrieval time than fact-based retrieval and relationship-based retrieval.

Abstract: To build a real-time alerting system of road damages caused by flood in Dehong autonomous prefecture, Yunnan Province. Base on the Yunnan Traffic system network, to grab data from WebPages and collect information related to road damage in Dehong autonomous prefecture, and then combine with the expert analyzing system with the function of artificial data modifying and self- learning, so as to predict more than 70% of the road damages from floods and save time for rescuing.

Abstract: In this paper, a missing data classification algorithms based on rough technique is proposed, and the complexity of the algorithms is analyzed, finally a missing data classification experiment with a typical dataset is conducted. The result of experimentation shows the algorithms not only can effectively improve the accuracy and efficiency of classification while enormously reducing the number of attributes, but also have the good performance on noises control.

Abstract: This paper addresses digit-writing hand gesture recognition using ego-motion detection algorithm. In training stage, according to the manual labeled feature points, the ego-motion detection algorithm is constructed and the corresponding average feature is obtained. In recognition stage, the interesting digit-writing hand gesture region is firstly segmented by skin and movement cues. Secondly, the models are fitted to the image that includes the hand gesture, and the relative features are extracted. Thirdly, the classification is done by comparing the extracted features and average features.