Tim Probert, Deputy Editor
If you go into a power plant today, there is a good chance that it has a disparate collection of condition monitoring digital control systems (DCS). There may be, for example, a Siemens DCS, an ABB DCS, something different for its continuous emissions monitoring system, another system for water quality and so forth.
The A Hidd IWPP is an important facility and provides much needed electricity and water to Bahrain’s growing population
There are potentially hundreds of thousands of data values. It is not feasible for a power plant operator to get their arms around all that information. They cannot trend all those signals and look for anomalies. They need something that is more proactive and tells them where to look. Operators are good when they know where to focus their energy. They know their plants and their instrumentation well, but they need the help to find that ‘needle in a haystack’.
These systems are, in effect, islands of information. Each system will have its own historical archive. What US company InStep has achieved, is to provide a system that collects information from all these disparate systems, keep them in their original timestamp collected resolution, high fidelity, second-by-second data and store all the information in one platform.
Using one toolset, called eDNA (Enterprise Distributed Network Architecture), the user can have access to all this information and be able to correlate data across systems. Traditionally, without an enterprise historian infrastructure, one would not be able to trend data from, for example, the Siemens system against data from an ABB system. This, then, is a unified data repository that provides one version of the ‘truth’ for all the plant instrumentation and assets.
This technology, of course, has been widely utilized in Europe, in North America and Asia-Pacific, but in the Middle East there have been few early adopters, but in the next few years it is expected to grow rapidly. InStep’s first installation of eDNA in the Middle East is at the Al Hidd independent water and power plant (IWPP) in Bahrain.
The Al Hidd power plant is owned by the Hidd Power Company (HPC), a consortium of International Power, Sumitomo and GDF-Suez; it is a 1006 MW (gross), 937 MW (net) power generating and desalination facility with a water production capacity of 90 MIG/day of water. InStep’s complete solution is charged with monitoring the facility’s major gas turbines, associated heat recovery steam generators and desalination equipment.
Creating a centralized archive
InStep’s DNA collects the data from all the control systems throughout the power plant, transmission & distribution grid infrastructure, substations, or any other data producing devices. Hundreds of interfaces collect the data in real time from all these control systems either at one plant facility or a series of plants and bring all the data back to a centralized archive, where it is stored in a compressed format.
This archive is capable of storing potentially decades of high-resolution data, online and instantly accessible through a web-based client application for the enterprise. In other words, the user enters the application through a web browser, they can then trend the data, report on it and build graphical views that display clearly the information. All aspects of the package are based on a Microsoft platform, both on the client and the server side. The software also has a calculation engine for performing analytics.
This is an enterprise application. The idea is to get the information out in real-time to all the users within an organization, so instead of just having a look at their data from their DCS control system, which maybe only has a limited history, this allows everyone plant engineers, maintenance personnel, planning personnel, executives to have instantaneous access to this information either through a web browser, Intranet or through a Microsoft Excel add-in if preferable.
Sean Gregerson, sales manager of InStep said: “If you go and buy a digital control system today, it will come with a data historian, but these are not typically scalable they can only store a limited amount of data. At the Al Hidd power plant alone we are storing over 40 000 data points every second.”
eDNA provides information to plant operations personnel that was not previously available. Traditionally, they would not be able to obtain data about how their systems have changed over a longer period of time. They might have been able to trend data from the last 60-90 days, but not been able, for example, to compare plant performance during a day with similar weather conditions from the previous year.
Equipment degrades slowly, so it was not possible to look at parameters for a particular piece of equipment a year ago versus today. Nor would it be possible to get the information out to the common Microsoft desktop. So traditionally, users would have been operation type personnel, and they would have to sit at workstations and access this information.
With eDNA everyone across the enterprise can have access to secure information from their desktop, day-in, day-out. So this technology is used by hundreds of people throughout the organization.
At the Al Hidd power plant in Bahrain, for example, there are LCD monitors displaying the overview of the entire plant’s production eDNA data in the director of operation’s office. Other managers have tailor-made displays in their offices as per their different jobs, each displaying real-time data. Rumours that HPC’s managing director David Hadfield has a display on his yacht could not be confirmed!
InStep signed a deal with the Al Hidd power plant in August 2008. The initial implementation took one week on-site to install and configure the system and build application screens. Once the owners of Al Hidd had procured and install the necessary hardware, and begun preliminary project planning, InStep’s team installed the software, trained staff and then deployed the system. Since the installation, says Gregerson, eDNA has rapidly become a critical piece of their day-to-day operations.
InStep releases two major product updates per year, plus any patches when necessary, downloadable from a support website. Each customer has an annual support maintenance agreement, meaning that they purchase a software license and then an annual support maintenance agreement, which provides telephone and email support, and technical support updates for the package.
Once eDNA has been installed, the next step is Prism, a self-learning analytic application for monitoring the real-time health of critical assets in support of condition based maintenance (CBM). Prism uses artificial intelligence, pattern recognition and sophisticated data mining techniques to determine when a piece of equipment is performing poorly or is likely to fail.
Monitoring real-time equipment health
Take a pump, for example. A pump has various parameters that are measured by eDNA inlet pressure, outlet pressure, flow rate, motor current etc. Prism compares the known operational profiles with real-time operating data to detect the subtle changes in system behaviour that are often the early warning signs of pending equipment failure.
The tailor-made Prism ‘dashboard’ at the Al Hidd IWPP provides real-time monitoring of the facility’s major gas turbines, associated heat recovery steam generators and desalination equipment
An advanced alarm manager and email notification provide near real-time updates of how well a plant or system is functioning. Prism includes an advanced analysis application for identifying why an asset is not performing as expected.
Gregerson explains: “Prism says, ‘here are the ten thousand different ways I have seen this asset perform before.’ It might be on a hot day, a cold day, full load, half load, 30 per cent load it learns all the operational conditions and how these parameters for each independent asset correlate or interact with each other.
“Then, in real time, we monitor the system and we say out of these learned profiles which ones are we most like right now and how far are we deviating from the best known operational profile. With alarms and email notification, Prism offers an early warning system that can save huge amounts of money.”
Gregerson believes that while eDNA is a ‘no brainer’ investment, Prism is more of a return-on-investment (ROI) purchase that sees problems before they happen, to allow power plants to be more predictive in their maintenance regime. Predictive, rather than preventative, maintenance not only provides an early warning system, but also means that power plants can prioritize on certain assets and avoid performing unnecessary maintenance on others.
Prism was released two years ago and has been installed by utilities worldwide, including South Africa’s Eskom, Shanghai Electric Power of China, Japan’s Tokyo Electric Power Company (TEPCO), Korean Electric Power Corporation (KEPCO) of South Korea, Swedish Vattenfall, plus Exelon and Southern Company in the US, as well as, of course, by HPC in Bahrain.
A single identified fault has proven to save a power plant over $500 000; one InStep customer realized a five-to-one ROI in the first year of utilizing Prism. Gregerosn said: “The return on investment is certainly justified for a single plant, but it is exponential when trying to monitor a fleet of plants.
“We work very closely with our customers and the price information that we provide to potential clients is largely based on the results that our customers have achieved with the software from a savings perspective. Many of our customers have used the EPRI guidelines on how to calculate and justify savings.”
Together Prism and eDNA are providing the Al Hidd power plant with solutions that complement its state-of-the-art infrastructure and set a benchmark for Middle East energy facilities. InStep has plans to partner with companies in the region that are familiar with the technology and once more plants have adopted Prism and eDNA,the company plans to open an office in Bahrain.