Jing Bing Zhang

Data management for wireless sensor networks: research issues and challenges

Wireless sensor networks (WSNs) as the basic infrastructure for pervasive computing environment are data-centric in nature. The research issues and challenges relating to sensor data management are much different from those of traditional database and wireless communication. This paper presents a concise review of the latest technologies, research issues, and challenges facing data management for WSNs.

Service-oriented architecture and technologies for automating integration of manufacturing systems and services

Enterprise application integration (EAI) has been pursued strenuously by researchers and industrial practitioners for decades. With the advent of Web Services and Service-Oriented Architecture (SOA), realizing enterprise integration, even automating inter-enterprise interactions become feasible. Virtualization is the fundamental for implementing an SOA based on Web Service technologies. In this paper, we first review the Web Service-Resource Framework (WS-RF) with emphasis on the concepts of Web Services and stateful resource modeling. Then we outline a Service-Oriented architecture that shows how various types of manufacturing resources are virtualized as Web Services; what technologies are used in Web Services register, discovery and transport, and in business process automation. We adopt the WS-Resource approach, for illustration purpose, to virtualize a category of resource-Automatic Guided Vehicle (AGV). Different projections are defined based on WS-Resource properties upon virtualization so as to offer different perspectives to the same resources to meet a variety of business requirements

Event-driven service oriented framework for integrative serviceability management of networked manufacturing systems

Manufacturing companies face frequently changing market demands, time-to-market pressure, continuously emerging new technologies and global competition. To quick respond to ever-changing environment, the intra-enterprise integration of key technology applications is one of the fundamental requirements of the management and supervisory control systems at plant level. This paper presents an event-driven service oriented framework within which main services provided by manufacturing systems at different levels can be effectively managed in a holistic manner. The framework is derived from the analysis of fundamental requirements of three key services of manufacture systems, namely DRAM (dynamic resource allocation and management), RCAR (real-time control and auto-recovery), PMDP (performance monitoring, diagnosis and prognosis). The conceptual designs of the main components of the framework are presented, which include information data model, event triggering and message exchange mechanism as well as Shop-floor Service Bus that enables service oriented architecture.

A Comparative Study of Dynamic Learning Rate BPN and Wavelet Neural Networks

This paper presents an improved back propagation network (iBPN) with dynamic learning rates to accelerate the network learning and convergence speed. As compared with conventional BPN, the improved BPN is able to approximate the complex non-linear functions with higher efficiency and accuracy. Wavelet neural network (WNN) is a comparatively novel universal tool for functional approximation, and is effective in solving the inherent problems of poor convergence or even divergence encountered in other kinds of neural networks. This paper, through a comparative study, shows that iBPN has the same generalization performance as wavelet neural networks. While WNN shows the highest efficiency, it lacks consistency. In contrast, results obtained from iBPN are highly consistent and are quite comparable with those obtained from WNN.

Event-based communications for equipment supervisory control

Event-based approaches are becoming more and more popular for building large scale distributed systems and applications, especially in the Internet environment. In this paper, the concept and techniques of event-based communications are first discussed. A hybrid approach of event-based communications for real-time manufacturing supervisory control is then proposed together with an event-based communication agent (ECA) framework to provide extendable and reliable communications, as well as efficient development method in multi-OS, multiple types of network protocol environment. A case study involving an industrial implementation for a large-scale distributed supervisory control system based on this approach is then discussed

Using neural nets in rule discovery for intelligent equipment maintenance, diagnosis and prognosis

This paper presents a framework of using artificial neural networks technologies in knowledge discovery to support intelligent equipment maintenance, diagnosis and prognosis in automated manufacturing environment. It covers in details one of the key components of the framework, namely, the knowledge discovery engine, and its related technologies such as artificial neural networks. The paper also describes some practical considerations of using the framework in solving real-life industrial problems, including results of applying the techniques and algorithms developed.

A structured methodology for the development of anticipative event management systems with self-recovery capability

Event management is imperative for manufacturing enterprises. This paper presents the concept of anticipative event management (AEM) as a new approach to manufacturing event management that focuses on prediction of the events that are going to happen and the actions that can be taken when the events occur. A structured methodology that is based on a three-dimensional event classification system and an AEM framework is proposed for the rapid development of AEM systems with intelligent self-recovery capability. The paper also presents the conceptual prototype that applies the methodology to the development of an AEM system.

Evaluation of enabling RFID technology on the operation visibility in MRO plant

This paper presents a hardware simulation study for an MRO shop and highlights the reduction in process time due to the introduction of RFID technology. Use of this technology will not only result in less number of process steps, but also enable the operation planner to get real time updates on the location and processing of various components scattered within the workshop. Each RFID tag will store its (read workshop component) own history, expiry date, time stamps. The RFID tag can be an electronic substitute for a physical piece of paper that accompanies a particular ‘job’ logging its details for the different processes that it undergoes. Thus, by scanning the tag, one can easily trace back the history of that job; and by making available this information at real time, a complete operation visibility is attained.

Consortium based negotiation for distributed multi-equipment execution control

The execution control of a complex multi-equipment system poses a challenging problem to the conventional approaches that adopt centralized or hierarchical control architecture. Distributed execution control was therefore introduced as an alternative approach. A distributed autonomous multi-equipment system is perceived in be more robust and responsive to the dynamic operating environment. Nevertheless, the efficiency of the system lies in its ability to communicate, cooperate and negotiate effectively among the peers. This paper proposes a new negotiation approach specifically designed for the dynamic and time-critical environment of a multi-equipment system that requires high-speed decision-making capability.

On-line fault diagnosing techniques for a class of manufacturing equipment

This research develops an intelligent on-line monitoring and diagnosis system for a class of key manufacturing equipment. Important issues concerning the fault diagnosis system are discussed in this paper, including wide-band sensor, high-speed data acquisition for on-line monitoring, advanced signal processing techniques to suppress severe interference in manufacturing environment, and computational intelligence techniques to recognize fault patterns. Successful implementation of the intelligent monitoring and diagnosis system has provided decision-making support for equipment serviceability and condition-based maintenance.