By Chris Held and Chip S. Whitworth
|Built on the site of Tampa Electric’s former Gannon Power Station, H.L. Culbreath Bayside Power Station began operation in 2003. As opposed to the coal-fired Gannon Station, Bayside uses natural gas, which has played a significant role in helping reduce Tampa Electric’s overall emissions. Photo courtesy: Tampa Electric|
The Bayside Power Station is a combined-cycle facility that provides approximately 1,800 megawatts (MW) of natural gas-fueled electricity, and 200 MW of simple-cycle natural gas-fueled electricity which is enough power to meet the needs of 350,000 average-sized U.S. homes. The plant features seven GE gas turbines – all 7FA.03 units featuring Dry Low NOx 2.6 (DLN 2.6) combustor technology. The gas turbines began commercial service in 2003 and 2004, replacing older, coal-fired units in order to meet increased Tampa Electric customer demand for more reliable, cleaner and cost-effective energy.
In May 2015, combustion issues were remotely detected by GE in two of Bayside’s gas turbines. To help determine the best way to address the problem, the utility leveraged the expertise of GE Power Generation Services’ Monitoring & Diagnostics (M&D) Center in Atlanta, which oversees GE’s globally installed base of gas and steam turbines.
In the Atlanta Power Generation Services’ M&D Center, a team of more than 20 engineers analyze more than 38,000 operational alarms per year, assisting operators 24 hours a day in enhancing their asset reliability and performance by diagnosing problems ranging from failed sensors to gas turbine compressor damage. Drawing on the experience of millions of hours of fleet operational data, the M&D Center team has developed dozens of physics-based proprietary algorithms that provide early warnings of more than 60 different failure mechanisms. This depth of expertise, as well as operational excellence in the execution of downstream processes, generated GE’s customers and operators of monitored assets savings estimated at more than $100 million in 2014.
|Tampa Electric team members oversee construction at Bayside in 2009. Photo courtesy: Tampa Electric|
Armed with this expertise, GE, including the Atlanta M&D Center, has been supporting the Bayside Power Station under a 13-year contractual services agreement since 2003. In 2015, Tampa Electric also began participating in a remote monitoring and diagnostics pilot project spearheaded by GE’s Power Generation Services business and GE Intelligent Platform’s Smart Signal predictive diagnostics software and services team. Like the M&D Center, Smart Signal enhances asset performance management by identifying impending equipment problems early to avoid unexpected shutdowns.
As part of this cross-business M&D initiative, GE holds a weekly conference call with Tampa Electric team members to provide an update on the Bayside Power Station equipment’s performance, with the Atlanta M&D Center covering the station’s gas and steam turbines and the Smart Signal team covering most of the balance of plant data as well as some turbine and generator data.
That collaborative relationship between utility and service provider paid off in May 2015 when GE’s engineers detected combustion anomalies in two of the gas turbines at the Bayside Power Station.
On May 21, one of GE’s monitoring analytics triggered an alert to the Atlanta M&D Center for elevated pressure pulsations in the combustion chambers (referred to as combustion dynamics) on Unit 2D. Then, on May 28, 2015, one of GE’s proprietary analytics detected an anomaly in the exhaust temperature readings of Unit 1B. Because the anomalies involved the gas turbines’ combustion systems, the Atlanta M&D Center took the lead. Here, the Atlanta M&D Center has the visualization and analysis tools to evaluate the combustion dynamics and other operational data, and has staff experienced in data analysis and the configuration, operation and service of power generation equipment.
The GE Remote Tuning group is also based out of the Atlanta M&D Center. This group works with sites to remotely adjust the fuel flows to the different fuel nozzles in a gas turbine, thereby tuning the unit to achieve optimal emissions, combustion dynamics and flame stability. The monitoring and diagnostics specialists work closely with the Remote Tuning team to provide complete solutions to combustion operability issues.
In response to the first alert on Unit 2D, the M&D combustion dynamics specialist reviewed the data and the tuning history of the unit and determined that the unit needed a seasonal retune to increase flame stability and reduce the combustion dynamics. The alert on Unit 1B was triggered by a shift in some of the exhaust thermocouples, and the M&D combustion expert traced the exhaust temperature anomaly to Combustion Can #12. Both of these anomalies were highlighted in GE’s weekly report emailed to Bayside on June 1.
Upon seeing the report, Bayside’s team collected NOx emissions data over the following days (the Atlanta M&D Center does not have direct visibility to NOx data for these units) and presented the data to GE during the GE-Bayside meeting on June 3. The GE combustion expert reviewed Bayside’s NOx data the same day it was provided. The NOx data from Unit 2D confirmed the M&D Center’s initial conclusions, the GE tuning group was contacted and a remote seasonal retune was scheduled at the most convenient time for the plant operator.
The NOx data from Unit 1B was evaluated along with a deeper investigation into the unit’s operational, tuning and inspection histories. By the end of the day, GE’s investigation concluded the temperature anomaly traced to Combustor 12 was not concerning enough to merit action at that time. However, the analysis did reveal that the unit was operating with a relatively lean fuel-to-air ratio, similar to Unit 2D, so the M&D Center’s combustion experts recommended that Unit 1B also be retuned. This tuning was also completed by the GE remote tuners.
The tuning was accomplished in approximately six hours per unit and successfully increased the flame stability, thereby reducing the risk of unit trips due to lean blow-outs. The tuning also reduced the elevated combustion dynamics on Unit 2D, which-if not corrected-could have contributed to premature failure of the combustion hardware and resulted in potential forced outages. This fast turnaround was particularly noteworthy because the Bayside plant typically experiences high demand during the hot summer months of May through September, so a trip or forced outage could have had significant operating implications for Tampa Electric.
|Bayside achieved its latest record for generation in July 2015 with 967,561 megawatt-hours. Photo courtesy: Tampa Electric|
According to Tiller Mills, maintenance specialist with Bayside Power Station, “GE Power Generation Services responded with a sense of urgency to assist us in resolving the combustion dynamics anomaly. GE’s detailed reports and collaboration with Smart Signal and our station’s technical staff has proven to be successful combination. The collaboration and data sharing expedited the analysis and judgment to tune the units in a planned manner; to minimize down time, equipment failure and trips; and to operate within our environmental compliance procedures.”
The remote tuning project offers an example of how investing in GE’s M&D technology and services to cover its power generation assets enabled Tampa Electric to take proactive action to help keep its Bayside plant operating during the critical summer months, resulting in significant customer benefits.
All tuned and ready to go, the Bayside Power Station had its two biggest months ever in July and August 2015 for generation-and it did so by a significant margin. The new record, set in July, is 967,561 megawatt-hours (MWh). Close behind was August, with 938,875 MWh. The previous record, from July 2009, was 906,269 MWh.
“We waited until the end of August to celebrate July’s milestone because there was a good chance that we would break a new record,” said Bayside Power Station Director Scott Cannon.
“We’re also on track to set an all-time third quarter and possibly annual record for generation.”
The Power of Predictive Maintenance
Every day, GE collects more than 30,000 operating hours of data from a fleet of more than 1,500 gas turbine and generator assets, supplementing a 40-terabyte database representing more than 100 million fleet operating hours. At GE Power Generation Services’ Monitoring & Diagnostics (M&D) Center in Atlanta, Georgia, more than 50 M&D engineers analyze more than 25,000 operational alarms per year, assisting customers in enhancing their asset reliability and performance 24 hours a day, 365 days a year.
The data-driven insights drawn from this volume of power generation “Big Data” resulted in operator savings estimated at $70 million in 2014, up from $53.9 million in 2013.
The team diagnoses problems ranging from failed sensors to gas turbine compressor damage. Drawing on the experience of hundreds of thousands of resolved alarms, the team has developed dozens of physics-based proprietary algorithms that provide early warning of more than 150 different potential failure mechanisms.
The wealth of physical understanding combined with intelligent application of statistical methods has enabled the team to continually improve the algorithms, thereby increasing the probability of detection while reducing false alarms. GE announced that the data-driven insights drawn from this volume of power generation “big data” have translated to operator savings estimated at $70 million in 2014, up from $53.9 million in 2013.
Among the conditions monitored are the inlet temperature of a compressor, the thermal performance of a gas turbine, combustion exhaust temperature, dynamic tones of the combustion system-which was a key area of scrutiny in the GE-TECO retuning project-and the vibration levels of a rotor and temperature of bearings. On a GE gas turbine unit there are more than 100 physical sensors and 300 virtual sensors.
“Our monitoring and diagnostics team and capabilities play a key role in helping power plants operate at high levels of performance and reliability,” said Justin Eggart, general manager, fleet management for GE’s Power Generation Services business. “Our team takes a holistic approach to what we call ‘predictive maintenance,’ which focuses on helping customers sidestep operational barriers before they occur, no matter what type of equipment they are managing.”
The ability to foresee and forestall issues is at the very heart of predictive maintenance. GE’s PredictivityTM line of solutions help operators harness massive volumes of data in order to help them make more informed operational and business decisions.
In 2014, the Atlanta M&D Center served 502 sites in 58 countries, providing assistance more than 8,000 times. Additional regional support is provided from GE’s M&D sites in Scotland, France and India as well as its newest service facilities in China, Dubai and Saudi Arabia.
“We have operational data flowing in continuously from the fleet, and we are in constant contact with operator sites as well as our broader engineering teams,” GE’s Eggart said. “Oftentimes, the collaboration between our engineers and the operators leads to the fastest and most complete resolution to the issue at hand, and we appreciate those interaction opportunities that our services enable.”
In addition to GE units, this remote monitoring can be applied across an operator’s entire fleet. Through “device-agnostic” predictive solutions, the M&D Center monitors technology and equipment not only from GE, but also other suppliers. GE has made substantial improvements in reducing trip rates since the M&D Center opened in 1996. For example, trips per thousand hours for the combined 7F and 9F monitored fleet of gas turbines are down approximately 25 percent since 2009.
Chris Held is an engineering manager at GE Power & Water. Chip Whitworth is manager of Engineering and Maintenance at Tampa Electric.