May 3, 2018

Industrial Fan Optimization & Measurements Analysis

Industry: Industrial Equipment | Product: pSeven  | Company: Bronswerk Heat Transfer

Objective

The Whizz-Wheel® fan by Bronswerk Heat Transfer (BHT) company is used in different applications as Air Cooled Condensers, Air Cooled Heat Exchangers and Cooling Towers.

In this use case, BHT’s engineers tested pSeven functionality in two fields:

  1. Whizz-Wheel design optimization to comply with BHT’s commitment to their customer: minimize the energy consumption of the fan to achieve the required airflow. The company also needs a quick way to create performance graphs for different design configurations in their search to extend the application areas of the Whizz-Wheel. The results are compared to those obtained by using the company’s traditional method.
  2. Analysis and assessment of fan performance measurements.

The Whizz-Wheel geometry

1. Whizz-Wheel Optimization

Challenges

There are many fan design parameters which can be changed to find the optimal Whizz-Wheel configuration for a certain process window. There might be 100 different Whizz-Wheel designs possible and it requires 200 data points for each design to create the performance graph. Making a fan performance graph for each possible design would require ~20.000 CFD simulations.

Solution

To find the minimum fan power, required to achieve the specified volume flow through the fan, BHT engineers have to optimize different parameters. The solution required the following steps:

  • Create the CAD model (PTC Creo Parametric)
  • Create the CFD model (FloEFD)
  • Setup the simulation workflow (pSeven)
  • Setup the optimization/ DoE workflow (pSeven)
  • Evaluate the results

To analyze relations between some design parameters and the performance output, Design of Experiments (DoE) tool in pSeven was applied. Bronswerk Heat Transfer company highly appreciated the capability of pSeven to provide the direct integration of company’s CAD and CFD software into the DoE study.

The dependency analysis in pSeven, which generates a model to describe the impact of each parameter on the output, showed that one parameter has a direct impact on the required fan power while the other parameters have a lot of interaction with each other.

Finding the best combination is therefore essential to obtain the minimum fan power. pSeven Model Explorer visualization tool was used to explore the influence of the parameters on the optimized goal. The pSeven Surrogate-Based Optimization algorithms quickly found the optimum configuration.

Model Explorer shows the influence of fan diameter, hub diameter, blade angle and number of blades on the required fan power

Benefit

Using traditional method, it took BHT more than 40 CFD simulations to find the optimum configuration. The optimized result was achieved with pSeven with ~50 engineering hours less. The optimized configuration used 5% less energy to reach the required airflow compared to the BHT’s traditional optimization method.

The amount of CFD simulations required to create the performance graph for each fan design can be drastically reduced when using DoE.

2. Analysis and Assessment of Fan Performance Measurements

Challenges

To measure the fan performance, BHT determines the flow by axial velocity measurements underneath the fan at several locations on the radius. To achieve an adequate quality of the measurement, the company takes >3000 measurements at each location, and engineers evaluate them once they are back in the office. BHT wants to validate the measurements and check the fan performance directly on-site.

Solution

The run-ready workflow with integrated Python scripts was created in pSeven based on one of the standard BHT’s output files of the measurement data. The output of the workflow is a bell curve for each measured location. The mean value of each measured location is used to create a velocity profile over the fan radius.

pSeven workflow for analysis and assessment of the measurement data

Benefit

The advantages of pSeven are the user-friendly interface to connect the different Python scripts and to visualize the results, as well as the advanced capabilities to find correlations between the environmental conditions and the measured fan performance. This helps to automate performance data analysis and investigate cooler performance under different weather conditions.

 

By Frank van Sikkelerus, R&D Engineer, Bronswerk Heat Transfer

LinkedIn
VK

Interested in the solution?

Click to request a free 30-day demo.

Request demo