(719g) Monitoring the Stress and Oxide Layer Evolution Profile in a Supercritical Pulverized Coal Boiler

With renewable sources of energy becoming a larger contributor to the U.S. power grid, traditional sources of power generation such as supercritical pulverized coal-fired (SCPC) power plants are forced to cycle their load in response to the availability of these intermittent renewable power sources. These frequent load changes can greatly upset processes within SCPC power plants that are designed to be operated primarily at base load. The boiler, specifically, undergoes drastic shifts in its thermal profile such that some heat exchange components can experience large temperature swings, causing fatigue damage, or much higher temperatures than they were designed for, leading to creep damage. Increased damage to the boiler can lead to more unplanned shutdowns, therefore, the health of the boiler under load-following conditions must be monitored to avoid such damaging operation.

Work has been done in this area to monitor the degree of fouling in the superheaters of coal boilers,[1], [2] which can cause hotspots. Estimation of the stresses on the boiler tubes based on the measured tube temperatures has been documented[3]. However, these data-driven models still rely on having temperature measurements at the location where the stresses are being calculated, which means that any local effects of hotspots on the stress profile can be missed in the analysis. On the other hand, first-principle computational fluid dynamics (CFD) models provide enough detail to identify local effects,[4], [5] but are typically too computationally expensive to be applied to dynamic on-line monitoring.

To accurately monitor boiler tube health, very detailed knowledge of the overall state of the boiler must be available at many points along the boiler. This includes determination of the process conditions on both the inner (steam) and outer (gas) sides of the tube, tube inner and outer wall temperatures, and the through-wall thermal profile of the tube. Even in highly-instrumented boilers, many of these variables are unmeasurable or sparingly measured due to the extreme operating conditions of SCPC boilers. While first-principles dynamic modeling of the entire boiler can be used to estimate these variables, the current condition of the tubes is also important to consider. If fouling has accumulated on either the inside (from oxide scaling) or outside (from slag and/or ash deposits) of the tubes, it can affect the dynamic thermal profile of the tube walls or create localized heating that can lead to further creep damage. If the tubes have any corrosion, it can also have a large impact on the dynamics of the tube temperatures both from decreased heat transfer efficiency and thinner tube thickness, which can also make the tubes far more likely to fail. Thus, the ability to capture the impact of these different factors is vital to monitoring the overall boiler health.

In this work, a dynamic, distributed-parameter, first-principles process model of an SCPC boiler has been developed for the model-based condition monitoring. The model accounts for the drastic change in the transport and physical properties as the boiler transitions from the supercritical to subcritical regime and vice-versa during load following. The boiler model also includes a 1-D model for the superheater tubes accounting for the effects of fouling on the inner and outer tube surfaces so that the transient impact of this fouling on the tube stresses can be determined. In addition to its impact of added heat transfer resistance, oxide layer formation inside the tubes due to reaction between the tube metal and the steam leads to tube thinning and eventual spalling as well as loss of material structural properties, which has a large effect on the stress threshold that can be withstood by the tubes. Thus a more detailed model for oxide layer formation and its impact on the thermal profile is included. The thermal profiles of the tubes are used to calculate the thermal and mechanical stresses in the tangential, radial, and axial directions as well as the equivalent stress. Tube life consumption due to cycling operation is based on the European standard EN 12952-3 for the design and calculation of water-tube boilers, [7] while the tube rupture time is estimated based on the material properties[8].

One difficulty in monitoring the stress and oxide layer growth profiles in boiler tubes is that they cannot be measured online. Stress is not likely to affect any variable that is measured, and the impact of the oxide layer on any measured variable may not be reflected in any of the measured variables below a threshold, thus leading to lack of observability. Since a minimum error covariance estimate of the entire system is not possible, observable and unobservable modes are extracted and partitioned. As measurements of oxide layers may be available during plant shut down (i.e. offline), earlier estimates of state variables as well as stress and oxide layer growth profiles can be improved. This is similar to fixed-point smoothing problems but with missing measurements of the system. A surrogate model based optimization approach is developed for improved estimation of the system during load changes. The estimation’s sensitivity to the process and measurement model discrepancies is evaluated.

References

[1] M. Trojan, D. Taler, and S. Wielgus, “On-line monitoring of the fouling of the boiler heating surfaces,” Thermal Science, vol. 23, pp. 1289–1300, Jan. 2019, doi: 10.2298/TSCI19S4289T.

[2] D. Taler, M. Trojan, P. Dzierwa, K. Kaczmarski, and J. Taler, “Numerical simulation of convective superheaters in steam boilers,” International Journal of Thermal Sciences, vol. 129, pp. 320–333, Jul. 2018, doi: 10.1016/j.ijthermalsci.2018.03.005.

[3] T. Sobota, “Improving Steam Boiler Operation by On-Line Monitoring of the Strength and Thermal Performance,” Heat Transfer Engineering, vol. 39, no. 13/14, pp. 1260–1271, Sep. 2018, doi: 10.1080/01457632.2017.1363641.

[4] P. Madejski and D. Taler, “Analysis of temperature and stress distribution of superheater tubes after attemperation or sootblower activation,” Energy Conversion and Management, vol. 71, pp. 131–137, Jul. 2013, doi: 10.1016/j.enconman.2013.03.025.

[5] L. Pang, S. Yi, L. Duan, W. Li, and Y. Yang, “Thermal Stress and Cyclic Stress Analysis of a Vertical Water-Cooled Wall at a Utility Boiler under Flexible Operation,” Energies, vol. 12, p. 1170, Mar. 2019, doi: 10.3390/en12061170.

[6] “Water-tube boilers and auxiliary installations - Part 3: Design and calculation of pressure parts,” British Standards Institution, London, UK, BS EN 12952-3:2001, May 2002.

[7] ECCC, “ECCC Data Sheets 2005.” D G Robertson & S R Holdsworth ETD Ltd., Sep. 02, 2005.