LIFE CONSUMPTION ASSESSMENT OF CYCLING HRSGS AS THE UNIT OPERATIONS PROGRESS

摘要

The combined cycle plants installed in the mid-nineties to early 2000s were mainly designed to operate at base load with a limited number of planned starts. However, with deregulation of the power industry and the resultant volatility in the natural gas market, operation of many plants shifted to almost daily cycling. This operational mode change caused much concern among the owners and operators of combined cycle plants because not only was it not possible to operate according to OEM operating manuals, but it was also not clear what kind of damage could occur if the units were ramped up faster than recommended. The plants built later for cycling operations did not fare much better because the OEM life cycle calculations were based on very specific operating procedures that were often too difficult to maintain or too impractical to follow. This paper describes various historical damaging mechanisms encountered in Heat Recovery Steam Generators (HRSGs) during operation and the impact of cycling, including the effect of these damaging mechanisms on various HRSG components. Based on available analysis techniques, correlations can be evolved to estimate the life consumption or damage factor of various HRSG components. Using these correlations and the expected number and type of life cycles, it is possible to calculate the life of the most critical components of any HRSG. The correlations can be used to develop dynamic software which can estimate the life consumption of critical components on a continual basis. The dynamic software can use the data directly from the plant Data Acquisition system, thus eliminating the tedious task of manual data transmission. The necessity to strictly adhere to the operating procedures may be reduced, as long there are some safeguards to avoid exceeding the very high transients. The damage factor and the life consumption thus calculated can be used on a continual basis to gauge the economic benefits of the aggressive operations as the unit is operating. The calculations can also be used to foresee when focused inspection is necessary and to determine which components need repair and when. The paper will also present examples of usage of such dynamic software using the actual plant data.