Improvement of Dynamic Characteristics of Purpose-Built Vehicles Using Semi-Active Suspension System
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors1.What is the semi-active damper used in this work? It should be declared. LQR controller is hard to utilized on the semi-active damper.
2.The performance under real road excitation is adviced.
3.Are the results got from simulation or experiment? If experiment, the explanation of the experiment should be provided.
Comments on the Quality of English Language1.What is the semi-active damper used in this work? It should be declared. LQR controller is hard to utilized on the semi-active damper.
2.The performance under real road excitation is adviced.
3.Are the results got from simulation or experiment? If experiment, the explanation of the experiment should be provided.
Author Response
Comment 1: What is the semi-active damper used in this work? It should be declared. LQR controller is hard to utilized on the semi-active damper.
Response 1: The damper targeted in this paper is the MR damper, which can perform LQR control. As a basis for this many previous studies on LQR control of the MR damper in Semi-Active control have beed published. The following papers are related papers.
- Semi-active Suspension Control with PSO Tuned LQR Controller Based on MR Damper
- Experimental performance evaluation of an equipment isolation using MR dampers
- Differential Flatness Based LQR Control of a Magnetorheological Damper in a Quarter Car Semi-active Suspension System
- Response of a quarter car model with optimal magnetorheological damper parameters
Comment 2: The performance under real road excitation is adviced.
Response 2: Since the unmanned PBV is currently under development, we plan to conduct futher verification by conducting experiments on real roads in the future.
Comment 3: Are the results got from simulation or experiment? If experiment, the explanation of the experiment should be provided.
Response 3: The results of this study were based on simulation. To prevent misunderstanding, the word ‘experiment’ was replaced with ‘simulation’ in Session 4.
Good point, thank you very much. Thanks to this, I think the completeness of the thesis has improved.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis article mainly discusses the control effect of using the PID controller and LQR controller on the Cargo, Passenger, and Bed model, proving that (1) the system model and (2) the LQR controller is better than the PID controller. My comments are as follows:
1. Some figures, such as Figs. 2 and 4 and Figs. 18-22 are not mentioned in this article.
2. In Tables 13-15, the definition of maximum error needs to be clarified.
3. In line 500, "Subsequently, a velocity-control logic was devised via multiple regression analysis to ensure stability when driving over speed bumps of various shapes." Where is it?
4. Section 4, "4. Experimental Results and Analysis" is obviously a simulation and no experiment.
5. In the Appendix, the controller parameters of the PID controller and the Q and R of the LQR controller are mentioned, but the K value of the LQR controller is not mentioned. How much is it? How are the parameters of the PID controller determined? How are the parameters Q and R of the LQR controller determined? Why use a PID controller instead of only a PI controller? If only a PI controller is used, will the results be different? One of the key points of using I-control is to reduce the steady-state error. In this study, is the discussion of steady-state error necessary? In addition, will D-control be used to amplify the impact of speed-​​bump in this study?
Author Response
Comment 1: Some figures, such as Figs. 2 and 4 and Figs. 18-22 are not mentioned in this article.
Response 1: To complement what you said, Fig.2 is added to line 114, Fig.4 is added to line 181, Fig.18 is added to line 383, fig.19 is added to line 404, fig. 20-22 added the corresponding content to line 430.
Comment 2: In Tables 13-15, the definition of maximum error needs to be clarified.
Response 2: The definition of maximum error has been added at line 457.
Comment 3: In line 500, "Subsequently, a velocity-control logic was devised via multiple regression analysis to ensure stability when driving over speed bumps of various shapes." Where is it?
Response 3: The content is explained starting from the next paragraph, but considering the location and meaning of the sentence, it was deemed unnecessary and was removed.
Comment 4: Section 4, "4. Experimental Results and Analysis" is obviously a simulation and no experiment.
Response 4: As you mentioned, in order to prevent misunderstandings, the word experiment was replaced with simulation in Session 4.
Comment 5: In the Appendix, the controller parameters of the PID controller and the Q and R of the LQR controller are mentioned, but the K value of the LQR controller is not mentioned. How much is it? How are the parameters of the PID controller determined? How are the parameters Q and R of the LQR controller determined? Why use a PID controller instead of only a PI controller? If only a PI controller is used, will the results be different? One of the key points of using I-control is to reduce the steady-state error. In this study, is the discussion of steady-state error necessary? In addition, will D-control be used to amplify the impact of speed-​​bump in this study?
Response 5: The K gain value in LQR control has been presented in appendix (a8) and (a9) and added to column 293, considering the points mentioned. In addition, as stated in the text, tuning was performed to reduce transition time and overshoot for both PID and LQR. In the case of LQR, tuning was performed based on the relationship between Q and R that each affect LQR control and the model and function-based simulation results using Matlab. In PID control, D control is essential to reduce overshoot, and I control must be performed to accurately follow the reference value for stabilization when passing through speed bumps. In particular, when comparing PID control and PI control, the PI controller was larger from a peak-to-peak perspective, resulting in poor ride comfort, so PID control was adopted.
I am very pleased that the completeness and accuracy of this paper have improved thanks to your precise comments and feedback. Thank you very much for reviewing this paper.
