Prognostics, Health Management, and Control (PHMC)

Summary

Upcoming Event: May 2018
Industry Forum: Monitoring, Diagnostics, and Prognostics for Manufacturing Operations

Manufacturers need standards and guidance for how to effectively design, implement, verify, and validate monitoring, diagnostic, and prognostic technologies to enhance factory floor-level decision-making that directly impacts maintenance and control strategies. The requirement for standards and guidance becomes more critical as manufacturers become capable of collecting larger volumes of data, the health of their equipment changes, and their processes evolve to meet changing consumer demand. The Prognostics, Health Management, and Control (PHMC) project will deliver methods, protocols, and tools for robust sensing, diagnostics, prognostics, and control that enable manufacturers to respond to planned and un-planned performance changes thereby enhancing the efficiency of smart manufacturing systems. Early implementations of smart manufacturing technology have enabled manufacturers to use sensor-rich equipment and process the resulting data to provide decision-makers with information on many performance-related measures (e.g., machine status and utilization) and overall process health. There is increasing interest to leverage the same data to generate diagnostic (what happened) and prognostic (when something will happen) intelligence at the machine, process, and system levels. Complex system, sub-system, and component interactions within smart manufacturing systems make it challenging to determine the specific influences on process performance, especially during disruptions. The simultaneous operation of complex systems within the factory increases the difficulty to determine equipment and process degradation, and resolve failures due to ill- or undefined information flow relationships. No open standards exist that guide and manage sensing, prognostics and health management (PHM), and control at all levels (from the component to the system to the enterprise level). Some proprietary solutions exist that integrate some manufacturing systems, but they apply to systems from one vendor and are often expensive and inaccessible to many manufacturers. The goal of the project is to promote advanced sensing, PHM, and control from ISA 95 manufacturing1 levels 0 (production process) through 3 (manufacturing operations management). This will result in improved decision-making support and greater automation with a focus on vendor-neutral approaches and plug-and-play solutions.

Description

Objective:  Deliver methods, protocols, and tools for robust sensing, diagnostics, prognostics, and control that enable manufacturers to respond to planned (e.g., scheduled change-overs, new productivity targets) and un-planned (e.g., faults, failures) performance changes thereby enhancing the efficiency of smart manufacturing systems.

Technical Idea:  Manufacturers need standards and guidance to effectively design, implement, verify, and validate monitoring, diagnostic, and prognostic technologies to enhance factory floor-level decision-making. The technical idea is to advance the state of the art in monitoring and control for PHM as the foundation for improved decision-making support and further automation at levels 0 through 3 of the ISA 95 manufacturing hierarchy. This work will focus on developing measurement science (including test methods, performance metrics, reference data sets, and tools) to enable vendor-neutral approaches and plug-and-play solutions to address the following challenges for PHM monitoring and control:

• Standards for Critical Information: Identify the information required to make an informed decision with respect tosetting and updating control points. This information includes any necessary health, diagnostic, and prognostic datasince this information will influence the control strategy.
• Methodologies: Collect the necessary data at the appropriate times and appropriate manufacturing production levels(e.g., equipment, work-cell, line, factory) to minimize the collection of “big data.”
• Data Framework: Determine the appropriate structure, organization, and analysis of data to gain insightful diagnostic(e.g., what is going to fail) and prognostic intelligence (e.g., when will it fail).
• Decision-Making Guidelines: Feedback the collected intelligence to the manufacturing process at levels 0 through 3for the purpose of updating the control for optimal production.

The technical idea will address the challenges and needs to develop advanced, integrated PHM and control strategies capable of responding to performance changes. This includes changes caused by faults, scheduled maintenance, change-overs, or new productivity targets. Currently, this type of hierarchical, integrated PHM and control is lacking from ISA 95 levels 0 to 3 within a manufacturing facility.

The technical idea for this project is specifically broken out into three research thrusts: 1) Machine Tool Linear Axes Diagnostics and Prognostics, 2) Manufacturing Process and Equipment Monitoring, and 3) Health and Control Management for Robot Systems. Each of the three research thrusts promote the design, implementation, and assessment of metrics, test methods, and communications protocols. Likewise, each thrust will develop and validate reference data for public consumption. This data will include reference datasets (e.g., contextualized, output sensor data), use cases (e.g., a sample of common and atypical scenarios encountered by manufacturing), and test scenarios (derivative of use cases for assessment purposes) that would be used to implement protocols needed for sensing, PHM, and control.
 

Research Plan:  The technical idea feeds into the advancement of smart manufacturing. Smart manufacturing encompasses the collection and use of more data (e.g., controller, maintenance, performance, and inspection) from various manufacturing systems within a factory. The increased data promotes greater diagnostic and prognostic capabilities for manufacturing systems. Diagnosis and prognosis enhance traditional notions of control by informing manufacturers how best to utilize their equipment to achieve their performance targets. Many of the inefficiencies within a factory occur at the system level, such as those created by equipment degradation and failure (e.g., nonconformance to production schedule due to unexpected equipment faults and failures). These inefficiencies can be difficult to understand because manufacturing systems are inherently complex; many critical pieces of these systems rarely coordinate effectively with one another even though their operations are intimately connected. Understanding the complexities at the system level requires exploration of the lower levels of a process or piece of equipment. NIST’s research to address the measurement science barriers will further promote the advancement of integrated PHM and control strategies to overcome these inefficiencies. This overall objective will be accomplished through three main thrusts:

1. Machine Tool Linear Axes Diagnostics and Prognostics – Development of a sensor-based method to efficiently estimatethe degradation of linear axes. This research is supported by the development of a linear axis test bed (considered a component-level test bed) that uses data collected from a NIST-developed sensor suite to detect translational and angular changes due to axis degradation. This project will promote the optimization of maintenance scheduling and part quality through the development of diagnostics and prognostics of linear axes. Likewise, this research can also be leveraged to enhance machine tool linear axis calibration.
2. Manufacturing Process and Equipment Monitoring – Identification of high-value data sources and the most appropriateopportunities to collect data to avoid the challenges of big data. The focus is on having the right data at the right time to improve decision-making with respect to process and equipment performance. This research is supported by the development of a systems-level test bed of networked machine tools and sensors in an active manufacturing facility. The test bed provides a valuable testing and prototyping environment replete with rich data to support fundamental research, technology and standards development. Early efforts in this research thrust will focus on integrating heterogeneous shop-floor systems through the development and advancement of standards and protocols. Specifically, the task will integrate sensors, machine tool controllers, and production management systems. Initial standards research are focusing on the extension of MTConnect across manufacturing equipment and systems.
3. Health and Control Management for Robot Systems – Development of test methods, performance metrics, assessment protocols, and reference data sets to assess the health degradation of the robot system including how such degradation impacts key metrics of the robot system (e.g., end-effector accuracy). A PHM-focused robot systems test bed is being developed that will include industrial robotic arms and a range of sensing technologies. This is a work-cell-level test bed, placing it between the component-level and systems-level test beds outlined in the other two thrusts. This research thrust will further emphasize advanced sensing, condition monitoring, diagnostics, and prognostics, which will augment maintenance and control decisions. The initial focus of this effort is on monitoring and assessing the degradation of tool center position accuracy of the robot arm and, in turn, using this knowledge to update the work-cell’s control and maintenance strategies, appropriately. Emphasis will be placed on the health of the overall robot work-cell (e.g., arm, controller, sensors, end-effector) and the relationships between these work-cell components to determine how they influence arm performance. Initial standards efforts will focus on the development of guidelines presenting manufacturers with guidance to deploy PHM for robotics within their industrial arm robot work cells.

It is expected that the Machine Tool Linear Axes and Health and Control Management for Robot Systems research will evolve so that their outputs can be utilized by the systems-level Manufacturing Process and Equipment Monitoring research effort from the perspective of enhancing PHM and control. This integration will occur in concert with the development of control strategies that support dynamic production scheduling. These control architectures will allow equipment and processes to use the intelligence generated via sensing and analysis.

Major Accomplishments

• Summary Report on a Workshop on Advanced Monitoring, Diagnostics, and Prognostics for Manufacturing Operations
• Degradation Measurement of Robot Arm Position Accuracy Dataset
• Diagnostic and Prognostic Standards Gaps identified within Manufacturing Domains

• Comprehensive Review of Diagnostic and Prognostic Capabilities and Best Practices For Manufacturing

• Roadmapping Workshop on Measurement Science for Prognostics and Health Management of Smart Manufacturing Systems
• Measurement Science for Prognostics and Health Management of Smart Manufacturing Systems Roadmapping Workshop Report
• Case Studies that Identify the Needs, Priorities, and Constraints of Small- to Medium-sized Manufacturers with respect to Performance Measurement, Condition Monitoring, Diagnosis, and Prognosis
• Definition of a Use Case that is Representative of Common Manufacturing Operations to Support the Assessment of PHM Technologies