As part of the 1990's business re-engineering process, many of Honeywell's top customers have identified potential financial rewards through better utilisation and more efficient management of existing assets - people, plant and equipment. In response, Honeywell has developed @sset.MAX which it says is a balanced suite of software applications and services that aim to maximise the effectiveness of these assets and extend Hi-Spec Solutions' expertise in process control and optimisation into new areas of business performance.
Industry is under constant pressure to reduce operational expenses while complying with regulations, improving safety, reducing emissions, and increasing process availability. At the same time corporate and in-plant support staff is being reduced, forcing the move from specialists to generalists. Those who remain are under pressure for constant improvement.
To address these business goals, Honeywell Hi-Spec Solutions has developed the Equipment Health Management (EHM) Alert Manager and Virtual Testing Services. These decision support tools are intended to automate and integrate existing human expertise in the areas of maintenance, reliability, optimisation, and control engineering.
EHM Alert Manager
Typically, the notification of a problem originates in the control room when a control loop is unable to maintain its set point. The cry goes out to test the sensors and check the valves. Perhaps the valve positioner is not responding, differential pressure cells are plugged creating sensor error or a pump is in cavitation. Problem resolution might begin by pulling process flow diagrams, loop drawings, manufacturers maintenance manuals, test and inspection procedures and the like. Data types vary from standard process documentation to periodic maintenance and work-order records, to on-line continuous measurements. Users say that just gathering the data and documents they need for troubleshooting ranges from several hours to several days. This is an inefficient use of a valuable and limited resource.
Today there are experts that are monitoring, analysing and repairing different pieces of critical plant equipment. Analysis tools range from the experts' own knowledge of the process and specific pieces of equipment, to Excel, Matlab and other software tools. Equipment checks are initiated by a variety of sources ranging from time or usage based intervals, to control operator requests, to full blown on-line predictive monitoring equipment. Once a request is initiated, a defined test or ad hoc troubleshooting process begins.
The EHM Alert Manager automates the existing scenario into a decision support system that engages any person or application either on-line or off-line, in the plant or remote, into the problem resolution process. People and applications play roles in the problem resolution process, generating symptoms and resolving them into faults. The expertise of client personnel, Honeywell engineers and other industry sources are used to configure a list of symptoms for each asset you choose and map those symptoms into a list of faults. Typical assets include rotating and reciprocating pumps and compressors, columns, heat exchangers, motors, control valves, instruments and so on. The user also assigns priorities and defines who (or what) to notify (control alarm, pager, e-mail, work order, etc). For the first time, the troubleshooting process is automated and seemingly disparate equipment tests are related into a defined decision support environment.
For example, let us say the control operator receives a control alarm warning of a high bearing temperature. If you are using the EHM Alert Manager, a symptom is generated - high bearing temperature. A rule-based system establishes the priority of the symptom and automatically initiates the appropriate action. The Alert Manager, upon receipt of the symptom high bearing temperature, automatically checks for the presence or absence of other symptoms. For example, the EHM Alert Manager queries the on-line vibration monitoring system for a high vibration symptom and also directs the operator to pull a lubricant oil sample for off-line analysis. The Alert Manager sends an e-mail notification to the lubricant oil laboratory and vibration specialists. If the high vibration symptom is present and the oil analysis indicates contaminated lubricant oil, then the Alert Manager concludes the bearing is failing and issues a recommended action using the work order management system.
As a learning system, the Alert Manager will increasingly deliver value over time. One important aspect of a learning system is it may actually get 'smarter' when you transition employees. The new employee is trained on an existing system picking up where the last engineer left off. The departing engineer's tools and applied knowledge have been captured and automated. Upon entering the new position, the new engineer brings along a different set of experiences and expertise. To continue this example, perhaps the new engineer is fluent with Matlab and replaces formulas developed using Excel by the previous employee with advanced algorithms. End users could benefit in several tangible ways, including reduced training time for employees in new positions, earlier fault diagnosis through improved efficiency of existing detection methods, reduced exposure to process upsets, increased asset availability and perhaps most importantly, the capture of knowledge. Furthermore, captured knowledge is distributed to all users of the system in a very easy to use manner with increasing levels of automation over time.
Another feature of the EHM Alert Manager is the Asset Explorer. Asset Folders automatically gather data related to the failing asset such as bitmaps, manual test histories, work order histories, OEM drawings, service contacts, transmitter test histories, vibration analysis data, etc. The asset folders are sorted using a variety of methods including; tree view, by asset type, by priority of symptom or fault, etc. The Asset Explorer alone provides significant value, by greatly reducing the time engineers spend on nonvalue-added clerical tasks. The Asset Explorer is tightly integrated with computer based maintenance management systems such as MAXIMO and SAP's PM enabling users to generate, track and close work orders directly from an Asset Folder.
EHM Virtual Testing
EHM Virtual Testing is an advanced solution that automatically determines relationships between process variables. The relationships are generated with process data gathered during training periods in which the asset condition is known (for example, after an overhaul). Virtual Test analyses the data and learns the relationships between process variables, creating a statistical model of the asset based on empirical data. The operation of the asset is then compared to the Virtual Test model to determine if the process performance of the asset has changed. When the asset's performance deviates significantly from the baseline, a performance degradation symptom will be passed to the EHM Alert Manager system, from where fault diagnosis will be initiated.
Honeywell says that EHM Virtual testing can be applied to all assets, provided that the performance of the asset can be characterised through process measurements. Virtual Testing is able to tolerate incomplete instrumentation and can distinguish between individual sensor errors and actual performance degradation of the asset itself.
@sset.MAX Equipment Health Management allows clients to capture the knowledge, tools and problem solving methods that exist in their plant today and transform them into a learning system where the collaboration between humans and computers is optimised. Honeywell says that this will release plant staff from routine hit and miss equipment analysis, and enable them to refocus their efforts on business optimisation.
Honeywell Hi-Spec Solutions
(011) 463 5524
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version