Content of review 1, reviewed on June 11, 2021
The main contribution of this manuscript is a multi-object optimization method for journal bearing. The structure and language of this article is clear. The method, termed as HOA in this manuscript, is a combination of genetic algorithm and sequential quadratic programming. The lubrication behaviour was obtained by an empirical model for time efficiency. From bearing design point-of-view, this is an interesting topic. However, the manuscript needs further explanations prior to its publication.
- The role of AI, as pointed out on p. 3 l. 23-25 needs further explanation, how the aforementioned methods relate to AI. Otherwise, it should be left out.
- To my understanding, the speed N, load W and diameter D are also used as fixed variables. This was not clear and should therefore be mentioned on page 8 in the context of the design variables.
- In page 9 line 30-32, the authors claimed that “Rapid convergence to a global minimum is achieved first using MOGA, followed by an accurate estimation of the minimum point using SQP”. For a better comparation, the optimization results or the consumed time of single MOGA and the HOA may be presented together.
- The viscosity of the lubricant is selected as a design variable in section 3. But in section 2, viscosity is a dependent variable of temperature. This leads to confusion. To my best knowledge, the inlet viscosity of the lubricant should be treated as a design variable.
- One of the key points for the multi-object optimization is how to deal with the conflicts between different state variables as shown in Figure 3. Please give a better clarification about this in your method.
- On page 14, it is stated that the optimum viscosity decreases with increasing load. At the same time, the L/D-ratio significantly changes between 5 and 6 kN. Thus, this conclusion might be misleading and physically doubtful, as a higher load-carrying capacity is achieved with higher viscosity. An extend discussion of this subject would help.
- Following your optimization, the So Number of your optimal solutions could be calculated to evaluate that it lays within the acceptable range (1 < So < 10)
Minor:
Rps (rounds per second) should be translated to an equivalent SI unit.
P 4. L 53 attuite angle should be attitude
Source
© 2021 the Reviewer.
Content of review 2, reviewed on August 13, 2021
I appreciate the way, the authors thoroughly evaluated and implemented my comments to improve the paper. The paper is now suitable for publication.
Source
© 2021 the Reviewer.
References
L., S. M., A., H. H. 2021. Multi-objective design optimization of hydrodynamic journal bearings using a hybrid approach. Industrial Lubrication and Tribology.
