In recent years there is a growing number of patients suffering from pneumonia, chronic obstructive pulmonary disease or asthma. In these situations, the application of Continuous Positive Airway Pressure (commonly known as CPAP) is indicated by clinicians. It is a noninvasive means of healthcare used widely. As a more advanced technique, the positive airway pressure may follow a time-cycled change between two preset pressure values. This technique is known as BPAP (Bi-level Positive Airway Pressure).
The devices are used mainly during sleep at home. Hence the aeroacoustic requirements are critical. In addition the devices must be portable and compact. Furthermore, the high frequency of pressure change required in BPAP devices poses additional demands on the design. Due to the complexity of the overall design problem, it may be solved efficiently by multi objective optimization.
The pressure head in these devices is generated by radial fans. For the aerodynamic optimization, we utilize a RANS solver. For the aeroacoustic optimization we use the Lattice-Boltzmann Method (LBM). Both are operating on a parametric geometric model of the fan and housing. For the propagation of sound waves into the far-field, we develop algorithmic strategies for using the Ffowcs Williams-Hawkings (FW-H) equation with the LBM. The constrained multi objective optimization is driven by a variant of the NSGA-II algorithm.
We outline the complete optimization procedure for a BPAP device. Our numerical results are compared with physical tests. To analyze the contribution of selected geometric features to the emitted sound pressure, we perform a sensitivity study. The new algorithmic arrangement has shown to drastically cut development costs and time.Copyright © 2016 by ASME