Hot Gas Path - Condition Based Maintenance (CBM): Control Level (Part 2/7)

Introduction:

The output from the instrumentation is acquired and used locally as input to the unit’s control station. The control station consists of a PC equipped with original equipment manufacturer (OEM) software that provides the operator with a series of Windows, UNIX, etc. based displays which present the measured and calculated data, alarm status, etc.

1. Maintenance Intervals

There are different factors that influence equipment life and they are responsible for the component degradation mechanisms. Each of these factors has an effect on gas turbine maintenance intervals and component parts’ lives, and will vary depending on each unit’s operation.

Each OEM has developed its own conventions such as ‘factored hours’ or ‘equivalent operating hours’ (EOH) to define maintenance intervals. It is not readily apparent from the OEM formulations which specific component location, for example, first row blade tip, leading edge, trailing edge, platform, etc. is indeed the weak link driving the maintenance action. This underlying uncertainty can result in higher repair fallout rates or possible lost production time. In other cases, the maintenance interval may be too conservative, leading to premature parts retirement. This approach could be improved with a more machine specific condition based approach to determining the timing of maintenance. This approach relies on recording operational data and component condition.

?

2. Operational Data

Control systems play a critical role in the intermediate step of data collection, conditioning and analysis within the automated CBM infrastructure. Advanced process control systems include analysis algorithms that enable them to diagnose and report malfunctions to the CBM system, which ultimately supports system-wide asset management.

Data manipulation technology is critical for CBM success:

• Information technologies for data manipulation:- On-line data acquisition: The data acquisition module represents the module that provides system access to digitized sensor or transducer data. It may represent a specialized data acquisition module that has analogue feeds from sensors, or it may collect and consolidate sensor signals from a data bus.- Data processing and validation:- Data fusion: which is a formal framework used to express the convergence data from different sources in which is expressed the means and tools for the merging of data that has originated from different sources.- Data mining tools: which provide new insights into wear and failure mechanisms in engine components:
• Advanced control algorithms.- Predictive control algorithms for combustion instability.- Adaptive controller.- Closed-loop steam cooling control.- Active control technologies for enhanced performance and reliability and reduced emission.- Fault tolerant engine control (smart sensor and actuators).- Closed-loop optimization.

The idea is to produce useful information from these sources and make it available to operational, engineering, and management personnel in their day-to-day work. Data fusion provides one consistent source for validated data for reports, planning, analysis, monitoring and diagnostics. According to the system architecture shown in the Figure.

3. Business Decisions

Key data are read from various source systems by the plant information system (which can then perform a number of operations on these data) and are loaded into the plant data base at each plant.

By means of the plant information systems, data are routed through the system, where processing tools can transform them into useful information. Interactive displays make that information available wherever it is needed. The process information can be stored in a nearly limitless amount of data. With the plant information system, personnel throughout the enterprise gain unprecedented access to real-time information, helping them to make better business decisions. The plant information system is an information management infrastructure that turns operational data into a powerful corporate asset, acquiring data from the power plant or process, usually from automated control systems or other sources.

?

4. Data Fusion

This data fusion platform must be open and flexible, to allow the multi- level and multifaceted process dealing with the automatic detection, association, correlation, estimation, and combination of data and information from single and multiple sources. The resulting information is more useful to the user when fusion is performed than simply delivering the raw data.

Data fusion users at the plant connect to the plant database using data fusion client applications running on a digital operating system via the plant local area network (LAN) for data analysis and reporting. Users at other locations can also access the plant databases at the various plants via the corporate wide area network (WAN). A sub-set of the data at each plant is periodically loaded into the corporate data fusion data warehouse by the corporate information system. This corporate information system also rolls up plant data into the monthly and year-to-date values most often used for corporate-level reports.

A data fusion system could be developed as a client-server application with interconnections through the WAN.

It has the capability to have several hundred users at power plants and at the corporate office. A possibility is to have a web-based version (web data fusion system) which can be run over the intranet and provide integration with plant data historians. The web data fusion system is based on a complete system of inter- faces, web services and client tools used to create a real-time user interface on the web.

To be Continued...