Intelligent Temperature Control of a Stretch Blow Molding Machine Using Deep Reinforcement Learning
Round 1
Reviewer 1 Report
The manuscript is organized and well-written. All of the findings are clearly presented and covered. It would be interesting if the authors included an explanation of their key results and standpoint of view in the conclusion section. This will improve understanding of the study's contribution.
Another minor point is that each work cited in the introduction to the current study needs to have the main contributions and advancements better explained.
Author Response
Point 1: The manuscript is organized and well-written. All of the findings are clearly presented and covered. It would be interesting if the authors included an explanation of their key results and standpoint of view in the conclusion section. This will improve understanding of the study's contribution.
Response 1:
The author would like to express the appreciation to the reviewer for the valuable feedback on the manuscript. The author have carefully considered the suggestions and have made the necessary revisions.
Specifically, the author have incorporated an explanation of the key results and the standpoint in the conclusion section of the paper. Additionally, the author have further emphasized in the conclusion section that our study offers a concrete and feasible approach to shorten the training time of the algorithm. Through the utilization of the blower in the heating box as a disturbance source, we have successfully simulated temperature variations resulting from the alternating day and night environment. This deliberate manipulation has proven to be an effective means of reducing the required training time. We believe that this practical solution will be valuable for researchers and practitioners seeking to develop algorithms with improved training efficiency.
Once again, the author sincerely thank the reviewer for their insightful comments, which have undoubtedly enhanced the quality and contribution of our research.
Point 2: Another minor point is that each work cited in the introduction to the current study needs to have the main contributions and advancements better explained.
Response 2:
We would like to express our gratitude to the reviewer for providing valuable feedback on our manuscript. We kindly invite the reviewer to refer to the revised version of the manuscript, where we have incorporated these improvements. Once again, we sincerely appreciate the reviewer's valuable feedback.
Reviewer 2 Report
While the approach may be useful, this paper needs major revisions.
1) The control system is very difficult to follow because it is not tightly linked to a specific type of injection stretch blow molding (i.e., ABS or RHB). There are considerable differences between the heating configurations and temperature ranges in these processes. Moreover, there are several different configurations for the ABS process. Which machine did you use, which configuration did you use, and what was the specific design of that configuration?
2) The author exhibits a poor understanding of PET. When heated using IR radiation, the preforms are not amorphous. Control of crystal size is critical during cooling the preforms. PET's behavior also is very sensitive to humidity levels.
3) IR heating of polymers varies with the polymer, its additive package, and the preform thickness/design. These factors vary the wavelengths at which absorption the preforms absorb IR radiation.
4) There is a significant amount of knowledge about IR heating of PET for ISBM, but most of this knowledge is not published. You might also check discussions of IR heating that are associated with thermoforming.
5) Clearly explain the source of the temperature differences in the blow molding facility.
The English requires only minor revisions.
Author Response
Point 1: The control system is very difficult to follow because it is not tightly linked to a specific type of injection stretch blow molding (i.e., ABS or RHB). There are considerable differences between the heating configurations and temperature ranges in these processes. Moreover, there are several different configurations for the ABS process. Which machine did you use, which configuration did you use, and what was the specific design of that configuration?
Response 1: We sincerely appreciate the reviewer's valuable feedback on our manuscript. In the revised version of the paper, we have provided additional clarification regarding the specific type of injection stretch blow molding (ISBM) process targeted in our study. We have specified that our research focuses on the second stage of the ISBM process, namely the stretch blow molding process. We acknowledge that there are variations in heating configurations and temperature ranges across different ISBM processes.
Regarding the machine used in our study, we employed a machine developed by CHUMPOWER, a Taiwanese company, for our experiments. The heating system's structure is depicted in Figure 4 of the paper, and we have included explanations of this configuration in the revised manuscript. Once again, we sincerely appreciate the reviewer's insightful comments, which have helped us improve the quality and accuracy of our paper.
Point 2: The author exhibits a poor understanding of PET. When heated using IR radiation, the preforms are not amorphous. Control of crystal size is critical during cooling the preforms. PET's behavior also is very sensitive to humidity levels.
Response 2: We sincerely appreciate the reviewer's insightful comments on our manuscript. We have carefully considered the concerns raised and have made revisions. We kindly invite the reviewer to refer to the revised version of the manuscript. Once again, we sincerely appreciate the reviewer's valuable feedback, which has helped us refine our understanding of PET and improve the quality of our paper.
Point 3: IR heating of polymers varies with the polymer, its additive package, and the preform thickness/design. These factors vary the wavelengths at which absorption the preforms absorb IR radiation.
Response 3: We would like to express our gratitude to the reviewer for their valuable suggestions and feedback. Regarding the variation in IR heating of polymers, we acknowledge that it is influenced by various factors such as the specific polymer used, the additive package incorporated, as well as the thickness and design of the preforms. The production process of PET bottles consists of two stages: injection molding and stretch blow molding. In this study, we focused on the temperature control parameter adjustment of the heating system during the second stage of stretch blow molding, assuming the availability of preforms. The supply of preforms was provided by a collaborating company, and the thickness and design of the preforms were specified by the customer, which were beyond our control. Once again, we sincerely appreciate the reviewer's valuable feedback.
Point 4: There is a significant amount of knowledge about IR heating of PET for ISBM, but most of this knowledge is not published. You might also check discussions of IR heating that are associated with thermoforming.
Response 4: We sincerely appreciate the reviewer's valuable feedback.
Point 5: Clearly explain the source of the temperature differences in the blow molding facility.
Response 5:
The production process of PET bottles consists of two stages: injection molding and stretch blow molding. In this study, we focused on the temperature control parameter adjustment of the heating system during the second stage of stretch blow molding. Taking general stretch blow molding equipment as an example, it includes the preform infeed system, transfer system, heating system, molding system, bottle discharge system, etc. Accurate temperature control plays a vital role in PET bottle forming technology. Failure to maintain the heating temperature of PET preforms within an optimal range can result in substandard product outcomes, as the heated PET preforms may not possess the necessary qualities for producing satisfactory end products. In certain seasons, specific regions encounter substantial variations in ambient temperature, where the disparity between morning and evening temperatures can exceed 20℃. In such scenarios, the on-site operators of the blow molding equipment often need to adjust the infrared heating lamps in the heating system several times. The effectiveness of these adjustments heavily relies on the expertise and experience of the operators.
In addition, the heating system is composed of multiple heating boxes. Each heating box is equipped with several infrared heating lamps on one side, and a reflector made of aluminum alloy on the other side. The reflector of the heating box can efficiently reflect heat energy to the PET preform. It not only ensures the heating effect of the PET preform, but also reduces energy consumption. When the PET preform is heated by the infrared lamp, the heat energy will be transferred from the outside of the preform to the inside. In order to avoid excessive temperature difference between the inside and outside of the preform, a blower is usually installed in the heating box as a cooling element, so that the temperature in the heating box can be uniform and excess heat energy can be taken away. This will prevent the temperature on the outside of the preform from rising too quickly, thereby producing a product with a more uniform thickness and closer to the ideal. However, the heating box is a semi-open structure, which is easily affected by the interference of ambient temperature changes. It has become an automatic production challenge for the bottle manufacturers. Furthermore, with the development trend of environmental protection and green manufacturing, preforms made of recycled materials are gradually used, and their requirements for temperature control are more precise and stricter.
Round 2
Reviewer 2 Report
Revised paper: The revisions addressed many issues that were in the original version of the paper.
Note: As molded, preforms do not always have uniform thickness - so that when stretched, the thickness in the stretch blow molded bottle is uniform. This issue was not addressed in lines 50-60 and will affect future work.
