September-October 2017 NPJ

Plant Performance Optimization By Eric Mino, GE Hitachi Nuclear Energy. Eric Mino Eric Mino joined GE in 2005 and has held a variety of positions at GE Hitachi Nuclear Energy including Nuclear Controls and Systems Upgrades Manager, Innovation Leader and Vice President of Asset Management Services. He currently serves as Vice President of Digital Nuclear Services. Eric holds a Bachelor of Science in Facilities and Plant Engineering from the Massachusetts Maritime Academy and a Master of Business Administration from the University of North Carolina Wilmington. An interview by Newal Agnihotri, Editor of Nuclear Plant Journal on August 7, 2017 at the American Nuclear Society Utility Working Conference in Amelia Island, Florida. Data Analytics & Internet-of-Things 1. How can the data analytics optimize the plant performance? For assessing power uprate potentials or just getting the most out of the plant and its assets in their current operating state, essentially, that’s where optimiza- tion focused analytics can help to identify opportunities. We have established three categories of analytics to help organize our solutions and continued develop- ment: Asset Performance Management, Operational Optimization and Business Optimization. Business Optimization analytics fo- cus on decisions related to which genera- tion asset should I de- ploy today or tomor- row? Would I have a mix of wind, nuclear, fossil and hydro? If you have nuclear as a baseload, what’s the strategy for the non-nuclear genera- tion assets so that you optimize your profits? Business level is re- ally decision support across an organizations fleet of genera- tion types. With Operations Optimization, that’s where we’re looking at the analytics to help with a specific plants total performance. Whether it be for traditional efficiency and operational metrics, or non-traditional metrics around human performance, resources, and work scope optimization. For the traditional metrics like megawatt output, capacity factor and minimizing unplanned down time, Operation Optimizations analytics are centered on leveraging real time or near time data to expose opportunities that are not easily seen or involve complex calculations that were historically done in paper reports. As with historical power uprate and margin recapture strategies, historical data would be gathered and then assessed by engineering teams. The results would take weeks to months and would provide the justification to reduce a specific amount of the original conservatism designed into the plant, because the engineers now had actual operating data versus design assumptions. With analytics that are able to run continuously, sites can now see real time or near time additional opportunities to reduce conservative assumptions. We call this decision support, where multiple analytics can be combined to not only create the operational opportunity, but also the assumed financial value, which helps to pre-justify any changes or design analysis that may be required to take advantage of the opportunity. This is where we see GE’s Predix TM platform helping us as we start to move analytics from paper into real-time usable analytics that continually gather data, continually do that math, and show where real-time margins are, and how you could start to further optimizing the plant. Achieving this real data-based, math-based conservatism based on continuous and actual operating conditions is where the industry wants us to start moving towards. The next level of this optimization for the plant occurs when you start feeding in asset level data your true asset health. Applying a similar philosophy of cascaded analytics you can begin to see how major systems are operating in relation to their optimal performance levels as well. Are they 10% low? Is one of my system is not performing up to the maximum? Am I wasting money on a system upgrade because the overall plant performance is really being limited by other systems? As we start to look at the whole plant through this type of lens, then you can begin to see the tie into the life extension strategies and decision support around 80 year life extensions. Today many of the systems get upgraded based on individual systems performance and the site team’s experience. Achieving a single pane of glass across the whole plant allows teams to more easily assess the bigger picture. For example if you take the condenser system, with the feed water system with the rod control system, and you look at how each of those has been impacting your operational performance, and then you start benchmarking them against others either in your fleet or in the industry, your decisions become that much more informed and impactful. This is where we want to go, simplifying and 34 NuclearPlantJournal.com Nuclear Plant Journal, September-October 2017