The Experts below are selected from a list of 334746 Experts worldwide ranked by ideXlab platform

Lei Xie - One of the best experts on this subject based on the ideXlab platform.

  • fault detection in dynamic systems using the kullback leibler divergence
    Control Engineering Practice, 2015
    Co-Authors: Lei Xie, Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk
    Abstract:

    Abstract This paper proposes detecting incipient fault conditions in complex dynamic systems using the Kullback–Leibler or KL divergence. Subspace identification is used to identify dynamic models and the KL divergence examines changes in probability density functions between a reference set and online data. Gaussian Distributed Process variables produce a simple form of the KL divergence. Non-Gaussian Distributed Process variables require the use of a density-ratio estimation to compute the KL divergence. Applications to recorded data from a gearbox and two distillation Processes confirm the increased sensitivity of the proposed approach to detect incipient faults compared to the dynamic monitoring approach based on principal component analysis and the statistical local approach.

  • detecting abnormal situations using the kullback leibler divergence
    Automatica, 2014
    Co-Authors: Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk, Lei Xie
    Abstract:

    This article develops statistics based on the Kullback-Leibler (KL) divergence to monitor large-scale technical systems. These statistics detect anomalous system behavior by comparing estimated density functions for the current Process behavior with reference density functions. For Gaussian Distributed Process variables, the paper proves that the difference in density functions, measured by the KL divergence, is a more sensitive measure than existing work involving multivariate statistics. To cater for a wide range of potential application areas, the paper develops monitoring concepts for linear static systems, that can produce Gaussian as well as non-Gaussian Distributed Process variables. Using recorded data from a glass melter, the article demonstrates the increased sensitivity of the KL-based statistics by comparing them to competitive ones.

Jaap Geluk - One of the best experts on this subject based on the ideXlab platform.

  • fault detection in dynamic systems using the kullback leibler divergence
    Control Engineering Practice, 2015
    Co-Authors: Lei Xie, Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk
    Abstract:

    Abstract This paper proposes detecting incipient fault conditions in complex dynamic systems using the Kullback–Leibler or KL divergence. Subspace identification is used to identify dynamic models and the KL divergence examines changes in probability density functions between a reference set and online data. Gaussian Distributed Process variables produce a simple form of the KL divergence. Non-Gaussian Distributed Process variables require the use of a density-ratio estimation to compute the KL divergence. Applications to recorded data from a gearbox and two distillation Processes confirm the increased sensitivity of the proposed approach to detect incipient faults compared to the dynamic monitoring approach based on principal component analysis and the statistical local approach.

  • detecting abnormal situations using the kullback leibler divergence
    Automatica, 2014
    Co-Authors: Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk, Lei Xie
    Abstract:

    This article develops statistics based on the Kullback-Leibler (KL) divergence to monitor large-scale technical systems. These statistics detect anomalous system behavior by comparing estimated density functions for the current Process behavior with reference density functions. For Gaussian Distributed Process variables, the paper proves that the difference in density functions, measured by the KL divergence, is a more sensitive measure than existing work involving multivariate statistics. To cater for a wide range of potential application areas, the paper develops monitoring concepts for linear static systems, that can produce Gaussian as well as non-Gaussian Distributed Process variables. Using recorded data from a glass melter, the article demonstrates the increased sensitivity of the KL-based statistics by comparing them to competitive ones.

Jeff Kramer - One of the best experts on this subject based on the ideXlab platform.

  • model based analysis of obligations in web service choreography
    Advanced Industrial Conference on Telecommunications, 2006
    Co-Authors: Howard Foster, Sebastian Uchitel, Jeff Magee, Jeff Kramer
    Abstract:

    In this paper we discuss a model-based approach to the analysis of service interactions for coordinated web service compositions using obligation policies specified in the form of Message Sequence Charts (MSCs) and implemented in the Web Service Choreography Description Language (WSCDL). The approach uses finite state machine representations of web service compositions (implemented in BPEL4WS) and service choreography rules, and assigns semantics to the Distributed Process interactions. The move towards implementing web service choreography requires design time verification of these service interactions to ensure that service implementations fulfill requirements for multiple interested partners before such compositions and choreographies are deployed. The described approach is supported by a suite of cooperating tools for specification, formal modeling, animation and providing verification results from choreographed web service interactions.

  • using a rigorous approach for engineering web service compositions a case study
    IEEE International Conference on Services Computing, 2005
    Co-Authors: Howard Foster, Sebastian Uchitel, Jeff Magee, Jeff Kramer
    Abstract:

    In this paper we discuss a case study for the UK Police IT Organisation (PITO) on using a model-based approach to verifying Web service composition interactions for a coordinated service-oriented architecture. The move towards implementing Web service compositions by multiple interested parties as a form of Distributed system architecture promotes the ability to support 1) early verification of service implementations against design specifications and 2) that compositions are built with compatible interfaces for differing scenarios in such a collaborative environment. The approach uses finite state machine representations of Web service orchestrations and Distributed Process interactions. The described approach is supported by an integrated tool environment for providing verification and validation results from checking designated properties of service models.

  • compatibility verification for web service choreography
    International Conference on Web Services, 2004
    Co-Authors: Howard Foster, Sebastian Uchitel, Jeff Magee, Jeff Kramer
    Abstract:

    In this paper we discuss a model-based approach to verifying Process interactions for coordinated Web service compositions. The approach uses finite state machine representations of Web service orchestrations and assigns semantics to the Distributed Process interactions. The move towards implementing Web service compositions by multiple interested parties as a form of Distributed system architecture motivates the need for supporting compatibility verification of activities and transactions in all the Processes. The described approach is supported by a suite of cooperating tools for specification, formal modeling and providing verification results from orchestrated Web service interactions.

Lihui Wang - One of the best experts on this subject based on the ideXlab platform.

  • adaptive Distributed Process planning and executions for multi tasking machining centers with special functionalities
    Proceedings of the 25th International Conference on Flexible Automation and Intelligent Manufacturing, 2015
    Co-Authors: Mohammad Givehchi, Azadeh Haghighi, Lihui Wang
    Abstract:

    Today, the dynamic market requires manufacturing firms to possess a high degree of adaptability to deal with shop-floor uncertainties. Specifically targeting SMEs active in the metal cutting sector ...

  • a novel function block based integration approach to Process planning and scheduling with execution control
    International Journal of Manufacturing Technology and Management, 2007
    Co-Authors: Lihui Wang, Weiming Shen
    Abstract:

    In today's decentralised business environment, manufacturing enterprises are implementing advanced Distributed manufacturing planning and control strategies to adapt to and win the fluctuating global market. Within the context, this paper presents a novel approach to the integration of Process planning, scheduling and execution in dynamic machining shop floors. Based on the concept of Distributed Process planning (DPP), function blocks are adopted as information carriers to optimise and specialise the nonlinear Process plans (NLPP) progressively throughout the three planning stages: supervisory planning, execution control and operation planning, with scheduling functionality integrated seamlessly. Architecture of DPP and function block based integration are proposed, followed by a system functional design using IDEF0 methodology. The advantages of adopting function block technology are demonstrated and recent research results as well as future directions are pointed out in this paper.

  • embedding machining features in function blocks for Distributed Process planning
    International Journal of Computer Integrated Manufacturing, 2006
    Co-Authors: Lihui Wang, W Jin, Hsi-yung Feng
    Abstract:

    IEC 61499 function blocks, being emerging industrial Process measurement and control standards, are chosen as data Processing and execution control elements within the context of a Distributed Process planning (DPP) system. A two-layer structure, consisting of supervisory planning (SP) and operation planning (OP), is identified to separate the generic Process data from those machine-specific ones. Within DPP, function blocks are used for data encapsulation, machining parameter optimization, Process monitoring, scheduling integration, and CNC control. From SP to OP, a function block evolves from meta function block, through object function block, to execution function block. The detailed design of a meta function block is depicted in this paper. A function block designer has been developed to help Process planners define basic function block types and generate composite function block networks. It is expected that the function block-enabled DPP approach and the associated algorithms can largely improve the...

  • Enriched machining feature-based reasoning for generic machining Process sequencing
    International Journal of Production Research, 2006
    Co-Authors: Lihui Wang, Ningxu Cai, Hsi-yung Feng, Z. Liu
    Abstract:

    This paper presents an enriched machining feature (EMF)-based reasoning approach to generic machining Process sequencing for Distributed Process planning (DPP). An EMF is represented by combining its machining volume with surface, geometric and volume features, as well as other technological information needed to machine the feature. The information embedded in the EMF is retrieved progressively for machining sequence generation. Following an introduction of EMF and its representation scheme, the problems in determining machine-independent feature groups (set-ups) in DPP and their machining sequences to be followed for a given part are investigated. Based on the EMF concept, five reasoning rules are formulated and the algorithms developed. As the set-ups and sequences are generated based on manufacturing constraints and datum references but separated from specific resources, they are generic and applicable to machine tools with varying configurations and capabilities. This approach is further validated th...

  • architecture design for Distributed Process planning
    Journal of Manufacturing Systems, 2003
    Co-Authors: Lihui Wang, Hsi-yung Feng, Ningxu Cai
    Abstract:

    Abstract Today's machining shop floors, characterized by a large variety of products in small batch sizes, require dynamic Process planning capabilities that are responsive and adaptive to the rapid changes of production capacity and functionality. To meet the requirement, this research proposes a new methodology for dynamic and Distributed Process planning. The primary focus of this paper is on the architecture of a new approach using function blocks. The secondary focus is given to the other supporting technologies—machining features and agents. Different from conventional methods, this approach uses a two-layer structure—supervisory planning and operation planning. It is expected that the new architecture can improve the system performance in a dynamic environment.

Jiusun Zeng - One of the best experts on this subject based on the ideXlab platform.

  • fault detection in dynamic systems using the kullback leibler divergence
    Control Engineering Practice, 2015
    Co-Authors: Lei Xie, Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk
    Abstract:

    Abstract This paper proposes detecting incipient fault conditions in complex dynamic systems using the Kullback–Leibler or KL divergence. Subspace identification is used to identify dynamic models and the KL divergence examines changes in probability density functions between a reference set and online data. Gaussian Distributed Process variables produce a simple form of the KL divergence. Non-Gaussian Distributed Process variables require the use of a density-ratio estimation to compute the KL divergence. Applications to recorded data from a gearbox and two distillation Processes confirm the increased sensitivity of the proposed approach to detect incipient faults compared to the dynamic monitoring approach based on principal component analysis and the statistical local approach.

  • detecting abnormal situations using the kullback leibler divergence
    Automatica, 2014
    Co-Authors: Jiusun Zeng, Uwe Kruger, Xun Wang, Jaap Geluk, Lei Xie
    Abstract:

    This article develops statistics based on the Kullback-Leibler (KL) divergence to monitor large-scale technical systems. These statistics detect anomalous system behavior by comparing estimated density functions for the current Process behavior with reference density functions. For Gaussian Distributed Process variables, the paper proves that the difference in density functions, measured by the KL divergence, is a more sensitive measure than existing work involving multivariate statistics. To cater for a wide range of potential application areas, the paper develops monitoring concepts for linear static systems, that can produce Gaussian as well as non-Gaussian Distributed Process variables. Using recorded data from a glass melter, the article demonstrates the increased sensitivity of the KL-based statistics by comparing them to competitive ones.