Loose Belt Fault Detection and Virtual Flow Meter Development Using Identified Data-driven Energy Model for Fan Systems
Round 1
Reviewer 1 Report
The paper presents an approach to identify an energy model of existing VFD-motor-fan systems using available experimental data. The authors focus on the applications of this approach in loose belt fault detection and virtual airflow meter development for optimal control. The proposed methodology is data-driven and utilizes fan head, speed, system power input, and temporarily measured airflow rate to identify the fan head, fan efficiency, and drive efficiency curves without the need for measuring shaft power. The following issues should be addressed:
1. Introduction:
The introduction adequately provides an overview of the main points and contributions of the paper. However, the organization of the section could be improved to ensure a more logical flow of information. Additionally, the relationship between the development of the virtual airflow meter and loose belt fault detection should be clarified to enhance the coherence of the paper.
2. Methodology:
The methodology section adequately describes the data-driven approach used to identify the energy model of VFD-motor-fan systems. However, section 2 lacks sufficient detail and discussion. It would be beneficial to provide more specific explanations of the steps involved in the methodology, as well as their significance, to enhance the clarity and rigor of the paper.
Moreover, Figure 2 should be modified to visually represent the relationships between the steps more explicitly, including the main identification equations.
3. Results and Discussion:
The discussion in section 4.1 requires a more concise analysis of the obtained results. It would be valuable to discuss the importance of multi-step identification and the limitations of using single-step models. Additionally, comparing the proposed methodology with simple machine learning models may further highlight the value of the presented approach.
4. Conclusion:
The conclusion effectively summarizes the main findings and contributions of the paper. However, it would be beneficial to provide a concise recap of the methodology and emphasize the practical implications of the identified energy model in loose belt fault detection and virtual airflow meter development.
5. Formatting:
Formatting improvements are needed, including correcting the chapter numbering and addressing inconsistent font sizes in the figures. These modifications should be implemented to enhance the overall formal and academic style of the paper.
There are some grammar and typo errors in the paper. Please check the paper and revise them.
Author Response
See attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
- The references are quite old, and only 5/20 references are from 2017 and newer.
- The literature review is not well described. Previous works on this problem are not mentioned. For example, recent studies on problem similar to this research using AI have been published, such as ANN: https://doi.org/10.1016/S0360-1323(01)00028-2. Therefore, the authors should elaborate more on the literature review and then emphasize the novelty of this work, which is mentioned but not supported with a thorough literature review.
- What is “1” in equation (12), “DP” in Fig. 1 and page 8,
- “Relative fan speed” in fig. 3b is not defined.
- Fig. 3b: How does the averaging of the oscilation of the power data affect the uncertainty of the data?
- “As discussed previously, the fan speed not only directly impacts the fan head curve and system power input curve based on the affinity laws but also indirectly impacts the system power input through the fan speed-related drive efficiency” (page 10): This sentence is incomprehensible. Please revise.
- Fig. 4: Please explain why the data are so scattered, even for the 100% case.
- Figs. 5-9: Many data points fall outside the distribution. Why do you keep them? How do they affect the regressions?
- Figs 6-9: What are the R^2 values? In Fig. 6, why H~Q, not ~Q^2 or ~Q^3 as for typical centrifugal fans?
- Fig. 8a: Many data points of the efficiency is >1.0? How are they valid?
- Fig. 10b: It seems that the error is much more than 4% as you stated. Please add lines for 5%, 10%, and 20% discrepancies and add more discussions.
- Fig. 15: Although strong discrepancies between the measured and the prdicted flow rates are seen, only 4% is stated? Please validate this error by estimating the total error which is a product of measuring devices, averaging data…
Author Response
See attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
Interesting paper which seeks to drive energy efficiency in buildings. However, many sections are not clearly tied to a defined gap in the literature:
The abstract should be structured to follow requirements of a scientific paper.
The introduction section should be reviewed as the structure is not clear. Current literature in the area should be included as well as the gaps the study seeks to bridge.
The main position of the paper in the context of the gaps in the literature should be addressed and correctly tied to a defined set of questions.
Are current studies in the area adequate to address the current problems the paper seeks to achieve?
The number of sections captioned 1.1 should be checked to ensure the flow of the paper.
There should be a section on literature review as the are many studies addressing similar issues.
The position of the paper on loose belt faults as a factor of speed is not consistent with the literature and should be checked.
The paper presents no section on methodology and methods thus affecting the quality of the study. To what extent is the paper experimental? To what extent is the study quantitative or mixed?
For example, “In summary, the fan energy model can be defined by the fan head-airflow curve and 140 the fan efficiency curve, while the drive system model can be defined by the drive efficiency-fan speed curve”. How this objective intends to fill a gap is not clear in the paper.
Again, “First, an identification approach is developed to identify the fan head and efficiency curves and drive efficiency curves using available fan head, speed, and system power input as well as temporarily measured airflow rate” What the authors seek to address is not consistently tied to a defined approach.
Also, “Finally, an identification approach is developed to identify the fan efficiency curve and the drive efficiency curve without measuring fan shaft power”. How does the curve relate to loose belt?
The experimental setup should be illustrated in the form of pictures to improve the quality of the paper.
How Figures A and B relate to the detection of the loose belt should clearly linked to ensure the flow of the paper.
Also, “To intentionally eliminate the direct impacts of the variable fan speed, the affinity laws, defined by Equations (4), were applied to convert the raw fan performance data at all fan speeds to the fan performance data at the full design speed” How was this done and to what extent does that establish loose belt detection?
The paper is poorly referenced with many statements ignored, this should be corrected. For example, line 376-380 etc.
“Overall, the fan head curve defined by Equation (18), the equivalent fan efficiency curve defined by Equation (19), and the equivalent drive efficiency curve defined by Equation (20)” How equations 18-20 were established should be justified as the steps are not clear.
So, the fan speed (ω), motor speed, and VFD output frequency (f) are proportional and exchangeable in this paper. How this statement is grounded in science should be justified.
The analysis section is mixed with the data extraction and should be corrected.
The discussion is too brief as the conclusions and should be corrected.
The current structure of the paper should be improved as it does not follow a scientific approach.
The paper may not be effective for teaching and learning as there are many missing steps to tie in all the equations and how they were applied.
N/A
Author Response
See attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
This revised manuscript can be accepted for publication.
Author Response
See attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
Dear authors, many thanks for your kind responses. All of my questions have been answered. The quality of the manuscript was much improved.
I think that it can now be considered for publication with this journal.
Author Response
See attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors have improved the paper however there remain issues concerning the structure and presentation.
The 'Introduction' section requires extensive review.
The paper should have sections to indicate the literature review.
The methodology section should be improved to match the gaps.
The discussion section is too brief and should be expanded.
The conclusions should tie all the gaps the study investigated.
N/A
Author Response
See attachment
Author Response File: Author Response.pdf