Initiatives, procedures improve operating reliability
New Harquahala Generating Co LLC
Owned by MachGen Holdings LLC
Managed by Competitive Power Ventures, Operated by NAES Corp
1080-MW, gas-fired, three-unit, 1 × 1 combined cycle located in Tonopah, Ariz
Plant manager: Dean Motl
Key project participants: Jacob Mattingly, operations manager
The plant, consisting of three combined-cycle units, began commercial operation in September 2004; it sells capacity and energy into the WECC market. From the beginning of a contested transfer of care, custody, and control, facility personnel have been committed to achieve best-in-industry operating statistics. Last year marked the pinnacle of a process that showed continuous positive outcomes.
Plant personnel have shown unwavering determination to improve operating reliability, by implementing or dramatically improving several key programs and policies. The effort was made to ensure that the substantial investments made to correct equipment shortcomings were complemented by the best possible operating standards and practices available.
The facility began by completely revamping all operating procedures. Along with the common goals of ensuring the procedures are accurate and consistent, keen attention was paid to making the procedures concise, visible, durable, and ultimately attractive to use. To achieve this, two official versions of each procedure were created—one long form, one short.
The short form cut out all content not directly related to an operational action. In completed format, the “user” version or short form of each procedure begins page one with system start up directives. To keep the procedures short and maximize utility, graphics and pictures were incorporated as often as possible.When completed the final revisions of each procedure were printed in color on ineffaceable paper and placed in a sturdy binding. Lastly, the procedures were given a visible and user friendly storage arrangement that paid consideration to ensuring they could be obtained expediently during plant evolutions (Fig 104).
To complete this task, responsibility for the final outcome relied strictly on the operations manager and the most senior operator. A large portion of the work was delegated to the individual operators and they were given guidance documents along with one-on-one coaching to ensure a consistent outcome.It wasn’t enough to rely on individual procedure pre-start requirements. To truly establish a consistency with plant startup preparations, personnel generated a site-specific and comprehensive platform for all pre-start readiness efforts. This ultimately condensed down to a one page “pre-start checklist” that must be filled out and submitted to management with each scheduled startup (Fig 105).
The form was created with versatility enough to cover any startup scenario. It predicates some of the essential actions to be taken prior to issuing a startup command for the GTs, based on a layout that begins 24 hours before, and extends to, the actual time of startup. As with most power production facilities, the oncoming shift of employees responsible for the startup is typically not the same group responsible for preparations. Since the advent of the pre-start checklist, each operating crew can now expect repeatable and reliable plant operating conditions after completing shift turnover prior to an actual startup.
Also incorporated throughout the pre-start checks are specific performance validations completed on sensitive and critical plant equipment. For example, all major valves are loop- checked and stroked to ensure reliability prior to start. This habit of checking equipment functionality came full circle when the OEM unveiled a starting-reliability upgrade option that included most of the content in manual our pre-start checklist. The plant opted out of the upgrade simply because it was merely an automated version of our checklist.
Mated with the pre-start checklist was the creation and implementation of a “startup narrative” checklist to cover any startup scenario. It was written to chronologically depict all of the critical actions of a successful startup in a concise and abbreviated format (Fig 106).
Ideally, this narrative does not directly dictate operator action but is simply used as a tool by the CROs to help prevent any procedural oversight. Requiring the use of this tool has aided in the effort to improve consistency of startup performance and methodology. More importantly it is quick and easy to use for employees who want some reassurance that they didn’t miss anything—especially when a startup includes a couple of unexpected surprises.
Developing some of these new procedural resources allowed the facility to recognize and incorporate mitigating efforts for some of the known system- or machine-specific issues. In particular, this plant is faced with a design that incorporates an igniter or spark plug for each combustor known to negatively impact reliability. It is a fleet-wide problem that is all but guaranteed to negatively impact operating statistics. Not settling for the occasional “fired abort,” personnel worked together to come up with a trending analysis to predict an igniter failure before it occurs.
As part of the pre-start checks, flashback temperature trends from the last startup are analyzed. If the “signature” of the trend has changed, the igniter is replaced. This last year alone the facility has detected two damaged igniters before they could have caused a failed start. Specific trend templates were developed in the DCS and training documents were made to ensure that everyone involved is looking for the same symptoms (Fig 107). The moral of the story here is not to wait for the OEM to validate solutions to its known equipment deficiencies but rather use site resources to minimize the harmful outcomes.
The facility was able to fully implement these procedural improvements before the end of 2009, which allowed for a clear assessment of their impact by evaluating outcomes of 2010. Along with the familiar plant operating statistics presented at the beginning of this article, this facility methodically tracked and recorded any incident that negatively affects power-generation schedules.For years the facility has been producing and indexing generation-incident reports (GIR). The requirements to produce GIRs range from full-load trips to the smallest possible load shedding protective function and cover every possible startup interruption. Not only do these reports ensure effective corrective actions are taken to prevent a similar recurrence but they provide a powerful assessment tool to determine how well a plant is truly doing.
With the help of the aforementioned improvements our plant was able to have a year-over-year decrease in GIRs between 2009 and 2010 of 81 to 10. Additionally, most of the 10 incidents in 2010 were very minor. Site personnel are extremely encouraged by these accomplishments and are looking forward to maintaining a facility with the highest possible reputation. As the saying goes, “The only thing harder than getting to the top is staying there.”
In 2010, on top of doing major planned outages, the facility realized a 90.6% availability factor, 99.7% reliability factor, and an unprecedented 100% starting reliability (table). This is a far cry from a facility that suffered owner bankruptcy during construction, EPC contractor abandonment, required over half of all I/O wires to be remade, and received a plant-wide batch of instrumentation transmitters later discovered to include manufacturing defects (to name a few).
Volumes of literature can be produced to cover the list of discrepancies and solutions related to this facility’s effort to improve reliability. With the support of a very cooperative asset manager, we did the heavy lifting that is ultimately responsible for the practically flawless statistics of 2010.
This facility has successfully and dramatically improved the standards of operating reliability by implementing industry leading programs and procedures. By updating and reformatting procedures, adding to operating readiness requirements, ensuring consistency and continuity of operations, and diligently tracking generation upsets, the facility has effectively improved its standing as a reliable power provider.The work completed to achieve recent successes was several years in the making and relied almost entirely on onsite management and personnel. Without the methodical and determined approach to complete and implement these powerful tools the facility would surely have not achieved a remarkable 100% starting reliability.
For those wondering how this sort of effort is completed start to finish, remember the words of Henry Ford, “No job is too big if broken down in to small-enough parts.” Therein is the secret to developing first-class programs for a generation facility.