Pulper energy optimization using Computational Fluid Dynamics (CFD)
Pulper energy optimization using Computational Fluid Dynamics (CFD)
Task
Due to the dimensional change of the Andritz pulper, various problems occurred during operation. These problems have been investigated and solved.
Solution
Based on CFD simulations, a higher energy input as well as the use of internals could contribute to the solution.
Benefit
With the best practice approach an optimized operation behavior and a reduced energy consumption can be achieved. CFD simulations make it easy to visualize the problem and find a suitable solution.
Best Practice Approach
Due to the change to larger vessel dimensions of the Andritz pulper, adequate mixing conditions were not achieved. Consequently, the material couldn't be dispersed sufficiently and instead formed clusters on the surface of the suspension. The primary area where these clusters formed is situated on the suspension's surface, precisely between the two agitators. This problem of clustering is often referred to as the "iceberg problem". This accumulation can continue until the container overflows and the feed becomes clogged. The avoid this problem BST was instructed to overcome this issue and optimize the operating behavior. To achieve this goal the best practice approach, which are four steps has been performed (see Figure 1). At first primary measurements of the pulp have been conducted. After implementing the model and running of the CFD simulations adequate analysis were done to achieve proper solutions.
Figure 1: Best Practice approach
Measurement of pulp
To model the complex pulp flow behavior different challenges, must be met. Often the shear rate dependent behavior of pulp viscosity is not known. Different fiber types and different consistencies during the pulping process have a variety of non-Newtonian behavior. BST uses its inhouse large scale viscosimeter to measure the shear rate dependent viscosity of different samples from the pulping process. This viscosity behavior is analyzed and implemented in the simulation model. In Figure 2 this dependency of viscosity on the shear rate can be seen.
Figure 2: Shear rate dependent pulp viscosity
Pulper model
In the field of stock preparation, the pulper is a machine that consumes a high amount of energy. Conventional pulpers have an open tank for the feed inlet and a special agitator. The design of the agitator is chosen to achieve a proper distribution of the stock and wetbroke material. Figure 3 shows the geometrical arrangement of the agitators and the dimensions of the pulper. The majority of pulpers perform adequately when appropriately aligned with the geometry and power consumption requirements. Deviations from this alignment can potentially lead to a range of operational issues.
Figure 3: Pulper geometry
CFD Simulation
Conducting flow analysis of the pulp mixing within the pulper plays a crucial role in identifying problematic zones within the vat. Two cases were carried out for the simulation. In the first scenario, where the agitators were operated at only 50% of their maximum rotation speed, insufficient mixing was observed with the Andritz pulper. In the second case a higher rotation speed was simulated. In the figure 4 the velocity magnitude of both cases are shown.
Figure 4: Higher velocity magnitude due to higher power input
Due to higher velocities, higher strain rates and thus higher turbulence and better mixing behavior has been achieved. With this also the formation of stagnant areas can be avoided. This relationship is visually represented in 5. Notably, the strain rate in the higher speed case is considerably higher. This increase in turbulence extends to the surface of the suspension and thus prevents agglomeration of the material.
Figure 5: Higher speed case show more areas with high strain rates (right) compared to the 50% speed case with more areas with low strain rates (left)
Based on the simulation results, potentially problematic zones with low strain rates, weak vortices and low pulp velocities can be identified. The quality of the mixing behavior can be directly related to these zones.
Analysis
Table 1 below shows the results of the simulated scenarios. It is worth noting that the power density at the higher speed is four times higher than at 50%.
Table 1: Summary of the two investigated cases
Additional inquiries like “deflectors”, “breakers” and “cones” (see figure 6) should be conducted to improve the flow direction and distribution. Due to these internals, the operating speed can be set to a lower level and still achieve sufficient mixing. Together with an adjustment of the speeds and the use of internals, energy savings can be achieved.
Figure 6: Optimization possibilities with internals; Deflector to stop the accumulation at problematic areas; Smoother flow deflection.
