Skoglund et al.
10.3389/fther.2023.1282028
background) and the steam cycles (the foreground), can be evaluated numerically, but in this case, the size of the steam turbine cycle is not fi xed and can be optimized. To fi nd the potential for power co-generation, the steam fl ows through different turbine stages were therefore optimized to identify the integrated heat cascades that correspond to ideal integration of steam turbine cycles. In order to keep the programming model linear, a simpli fi ed steam cycle was assumed with turbine extractions limited to the existing steam levels of the mill, and an assumed fi xed value for the isentropic ef fi ciency of the turbine. For each mill and capture integration con fi guration, the heat cascades of the processes and the steam cycle are optimized to minimize primary hot utility (i.e., fuel use) and maximize power generation according to an objective function where these two objectives are combined. For formulation of the objective function, two different optimization scenarios were considered: “ Minimized fuel use ” and “ Maximized power co-generation. ” In the fi rst scenario, minimized fuel is the main objective, and always prioritized over power generation. In this scenario, power generation would only be achieved if there is suf fi cient heat from the recovery boiler to generate all steam necessary for process heating as well as steam for turbine operation. To include the power generation potential from available high-temperature excess heat, power generation is, however, also included in the objective function, but with a lower weight than fuel use. In the second scenario, maximized power generation is the main objective and prioritized over minimized fuel use. However, only co-generation of power in back-pressure steam turbines are considered. To avoid solutions where fuel is used for power-only generation, fuel use is considered in the objective function also in this scenario, but now with a lower weight than power generation. The weights of hot utility and power generation in the objective function can be chosen rather arbitrarily as long the relative weights ensure that one of the objectives is prioritized over the other according to the chosen optimization scenario. The energy targeting computations were performed using the in-house Matlab- based tool Mat4PI (Morandin, 2017). The integration with the steam turbine cycle can also be visualized using split GCCs, e.g., as shown in Figure 5. In this case, the stream data for the integrated carbon capture and pulping process (base case mill con fi guration without lignin extraction) was used to construct the GCC for the background process, while the steam turbine cycle is represented as the foreground GCC. For further information about energy targeting for integration of steam turbine CHP (see Harvey et al., 2023). All in all, three different mill con fi gurations and two scenarios for the optimization trade-off between fuel use and electricity generation were investigated. These mill con fi gurations and scenarios are brie fl y summarized in Table 5.
FIGURE 4 Split GCCs illustrating the potential heat integration between the carbon capture and liquefaction process (foreground) and the kraft pulping process, including the high-temperature heat from the recovery boiler (background).
be a better representation of a future scenario in which further steps have been taken to improve heat recovery in the mill. In our analysis, two different cases were assumed for the mill processes, where the fi rst corresponds to the current mill con fi guration and normal operation, and the second assumes that lignin extraction has been implemented in the mill. In both cases, high temperature heat from combustion of black liquor in the recovery boiler is considered to be a process-inherent heat source (i.e., the steam from the recovery boiler is not seen as utility steam), since the processing of black liquor in the recovery boiler is a necessary step in the recovery of chemicals. Note, however, that the amount of heat available from the recovery boiler will be lower when lignin is assumed to be extracted from the black liquor. The mill con fi guration without lignin extraction is also used as a base case for the energy targeting study, and serves as a reference when comparing the energy targets for the mill with the integrated carbon capture process to the mill without capture. The optimal heat cascades were established assuming a minimum temperature difference for heat exchange ( Δ T min ) of 10 K. The process streams representing the heat sources and sinks of the mill (see Section 12) form the so-called background process. The process streams related to the carbon capture and liquefaction process form the foreground process. In our analysis, the potential matching of the heat cascades for the foreground and background processes was evaluated numerically. This provides an indication of how well the carbon capture plant can be heat integrated with the mill. The GCCs representing the respective heat cascades can also be plotted in the same fi gure in order to visualize the potential integration between the two processes. For the base case mill con fi guration without lignin extraction, the split GCCs for the mill and the capture processes are shown in Figure 4. Next, it was estimated how the integration of the carbon capture and liquefaction process would affect the potential for steam turbine power generation in the mill. Again, the matching of the heat cascades for the pulp mill and carbon capture process (the
3 Results
3.1 Minimum heating and cooling requirements
Figure 4 shows the integration between the mill (without lignin extraction) and the capture and liquefaction process. The GCC of the carbon capture and liquefaction process is shown as the
Frontiers in Thermal Engineering
08
frontiersin.org
Made with FlippingBook interactive PDF creator