eMaintenance

Essential Electronic Tools for Efficiency

Diego Galar, Uday Kumar

• Publisher: Academic Press
• Year: 2017
• Print ISBN: 9780128111536
• eBook ISBN: 9780128111543
• Topic: Negocios y empresa
• Subtopic: Gestión industrial

eMaintenance: Essential Electronic Tools for Efficiency enables the reader to improve efficiency of operations, maintenance staff, infrastructure managers and system integrators, by accessing a real time computerized system from data to decision. In recent years, the exciting possibilities of eMaintenance have become increasingly recognized as a source of productivity improvement in industry. The seamless linking of systems and equipment to control centres for real time reconfiguring is improving efficiency, reliability, and sustainability in a variety of settings.The book provides an introduction to collecting and processing data from machinery, explains the methods of overcoming the challenges of data collection and processing, and presents tools for data driven condition monitoring and decision making. This is a groundbreaking handbook for those interested in the possibilities of running a plant as a smart asset.- Provides an introduction to collecting and processing data from machinery- Explains how to use sensor-based tools to increase efficiency of diagnosis, prognosis, and decision-making in maintenance- Describes methods for overcoming the challenges of data collection and processing

Chapter 1

Sensors and Data Acquisition

Abstract

The diagnostic and prognostic needs a broad range of organizations may be met in a cost-effective manner and with minimal risk by leveraging three critical areas. The first is a framework that can efficiently integrate reusable algorithms. The second is communications, in particular, wireless technologies. The third is an infrastructure for the system-level integration of distributed intelligent system elements. System-level integration provides the foundation for robust systems capable of generating actionable diagnostics and prognostic plans, hardware platforms, sensor modules, database information, and intelligent devices in an automated system.

Keywords

Electronics; Energy harvesting; Microsensor; Microelectromechanical system; Sensor Fusion; WINS

1.1. Sensors in Maintenance and the Need to Integrate Information

The diagnostic and prognostic needs a broad range of organizations may be met in a cost-effective manner and with minimal risk by leveraging three critical areas. The first is a framework that can efficiently integrate reusable algorithms. The second is communications, in particular, wireless technologies. The third is an infrastructure for the system-level integration of distributed intelligent system elements. System-level integration provides the foundation for robust systems capable of generating actionable diagnostics and prognostic plans, hardware platforms, sensor modules, database information, and intelligent devices in an automated system.

The integration of these three factors, together with advanced prognostic algorithms, is the basis for a new machinery maintenance paradigm. This paradigm employs targeted, task-specific sensors and algorithms integrated in a framework that is readily expanded and adapted to changing operational needs. Open industry-standard systems and interfaces are fundamental to this framework. Such systems can readily be integrated with a plant's equipment and coupled with various IT, operational planning, finance, and control systems.

Various components and functional capabilities of diagnostic and prognostic systems may be organized as a framework of system components or building blocks. Such a framework provides a scheme for identifying and organizing essential information about a system and a structure for defining standard interfaces between system elements and their functional requirements.

The requirements for capturing machinery health information, interpreting the data, and acting on the results of the analysis may be arranged in a hierarchy (Table 1.1). This hierarchy ranges from simply capturing data from a particular process or machine at the most basic level to interpreting sampled data to detecting a fault that has occurred and predicting what fault or faults will occur. Higher levels of the hierarchy provide a range of capabilities to automatically react to novel faults or to react before an anticipated failure. Higher levels, beginning with diagnosis and prognosis, typically require the integration of multiple data sources, as well as knowledge of the process equipment and operating state or context (Discenzo et al., 2001a).

Table 1.1

Hierarchy of Intelligent Machines

1. Data acquisition

2. Monitor

3. Detect

4. Diagnose

5. Prognosis

6. Prognostics and control

7. System-level prognosis and control

8. Dynamic optimization/multiobjective control

9. Adaptive/Reconfigurable Order of increasing complexity/cost/economic benefit

Adapted from Discenzo, F.M., Loparo, K.A., Chung, D., Twarowski, A., 2001a. Intelligent Sensor Nodes Enable A New Generation of Machinery Diagnostics and Prognostics. Virginia Beach, Virginia, April 2–5, 2001.

Throughout the working world, there is a move to intelligent devices and distributed intelligence. Intelligent components may be found at the structural level (e.g., smart materials), at the sensor level, or at the device level (e.g., embedded intelligence). There are no international standards for sensors, but this should happen in the near future. IEEE 1451 establishes a standard for a transducer (sensor/actuator) interface. It seeks to move data acquisition, distributed sensing, and control to an open system by establishing a framework and data elements for “smart” transducers. Included in this standard is a specification to facilitate sensor identification, calibration, documentation, sensor replacement, and network integration (O'Mara, 2000). Research in self-validating sensors at the University of Oxford over the last 12 years has been directed at defining intelligent sensors, which dynamically sense their own condition and provide information on the quality or validity of the sensor value returned (Henry, 2000). This is an important emerging area, with a standard proposed at the draft level by British Standards Institution for data quality metrics (Wood, 2000). Information on data quality becomes critical as we move to higher levels of the hierarchy of intelligent machines: automatic control, decision-making, and autonomous machines (Discenzo et al., 2001a).

Data critical to establishing the health of equipment may be captured and stored in an application-specific manner to accommodate essential memory or timing constraints. Preferably, machinery data should be organized in an open, industry standard format, such as that defined by Manufacturers Information Management Open Systems Alliance (MIMOSA). The data format and definitions are the result of many years of work by an international team of MIMOSA sponsors and members. This standard is open and accessible to the public (Discenzo et al., 2001a).

Figure 1.1 Open system architecture for condition-based maintenance. Adapted from Adapted from Discenzo, F.M., Loparo, K.A., Chung, D., Twarowski, A., 2001a. Intelligent Sensor Nodes Enable A New Generation of Machinery Diagnostics and Prognostics. Virginia Beach, Virginia, April 2–5, 2001.

More recently, a group of industry and academic partners teamed together to develop an Open Systems Architecture for Condition-Based Maintenance (OSA-CBM). This program is part of the Dual Use Science & Technology (DU&ST) program with joint industry–government funding (BAA 98–023). This effort, building on the work of MIMOSA, has resulted in an operational framework for machinery diagnostics in an open-system, layered framework, as shown in Fig. 1.1 (Discenzo et al., 2001b).

The integration of intelligent sensors and self-validating sensors in an open operational framework as specified by MIMOSA–OSA-CBM promotes the development and deployment of distributed intelligent sensors across a broad range of applications (Discenzo et al., 2001a).

1.1.1. Sensors Put Intelligence Into Maintenance

It seems like everything we buy today is intelligent. Thermostats adjust themselves based on weather conditions. Cars alert us to upcoming maintenance needs. Alarm systems can be armed or disarmed from a mobile phone. Wristbands track fitness progress.

Thanks to the Internet of Things, the same technologies that have revolutionized the way we do everyday things can now enhance the way we manage maintenance in the plant...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Chapter 1. Sensors and Data Acquisition
• Chapter 2. Data Collection
• Chapter 3. Preprocessing and Features
• Chapter 4. Data and Information Fusion From Disparate Asset Management Sources
• Chapter 5. Diagnosis
• Chapter 6. Prognosis
• Chapter 7. Maintenance Decision Support Systems
• Chapter 8. Actuators and Self-Maintenance Approaches
• Index