All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
As the authors addressed all previous comments correctly, it can be accepted.
[# PeerJ Staff Note - this decision was reviewed and approved by Xiangjie Kong, a PeerJ Section Editor covering this Section #]
The authors incorporated the comments accordingly.
N/A
N/A
N/A
Please consider the reviewers comments.
The paper "Intelligent Control Strategy for Industrial Furnaces based on Yield Classification Prediction using a GCM Machine Learning Model" proposes a hybrid deep network model combining a grey relative correlation, a convolutional neural network, and a multilayer perceptron model (GCM) for categorizing production processes and predicting yield classifications in industrial furnaces. It aims to improve yield, production efficiency, and energy conservation by adjusting operating parameters based on the predicted yield classification.
Well defined but needs improvement.
Valid, but needs improvement.
By addressing these areas, the paper can significantly improve its contribution to the field.
1. The paper could benefit from a more detailed explanation of the data collection process, including the selection of variables and the rationale behind it.
2. A suggestion is to provide a clearer description of the dataset used and its relevance to the application.
3. The description of the GCM model and its components could be more detailed. It's important to explain how each component of the model contributes to the overall prediction accuracy. Clarifying these aspects can make the methodology more replicable and understandable for readers.
4. The paper mentions the comparison with other methods, a more comprehensive analysis of how the proposed GCM model outperforms existing models would strengthen the paper. This could include a discussion of the advantages and limitations of the GCM model in comparison to traditional methods.
5. It would be beneficial to include a discussion on the practical implications of the findings, how they contribute to the existing body of knowledge, and their potential impact.
6. The paper would benefit from a section discussing the limitations of the current study and potential areas for future research.
7. Improving the overall structure and clarity of the paper would make it more accessible.
8. Ensuring that the language is clear, professional, and free of ambiguities is crucial.
9. Figures are not clear, need to include clear high-resolution figures.
This paper proposes a hybrid deep learning model combining gray relative correlation, convolutional neural network, and multilayer perceptron (GCM) to predict yield classification in industrial lime kilns. The model is trained on over 1 year of operational data from a sleeve lime kiln. Gray correlation analysis is first used to select relevant input variables. The preprocessed data is then fed into the CNN-MLP model for classification. Experiments compare performance of GCM against 5 other methods, with GCM achieving highest accuracy of 88.27%. Based on predicted yield class, an intelligent control strategy adjusts operating parameters to optimize yield, efficiency, and costs. Applying this on sample data estimates a potential 1090 t/100 days yield increase.
Review:
BASIC REPORTING :
1. Language is clear and professional
2. Provides adequate context and cites relevant literature
3. Structure follows academic standards
4. Figures are clear and described well
5. Raw data provided as supplemental files
EXPERIMENTAL DESIGN
1. Primary research collecting substantial industrial data
2. Clearly defines research objective to predict yield classification and how it addresses a knowledge gap in intelligent control
3. Technically sound investigation with rigorous methodology
4. Sufficient details provided to replicate experiments
VALIDITY OF FINDINGS
1. Impact is practical yields, efficiency and costs optimization in lime kilns
2. Findings appropriately qualified and limited to available data
3. Statistical analysis sound; data appears robust
4. Conclusions supported by results
MAJOR COMMENTS
1. Very relevant topic linking machine learning and industrial intelligent control
2. Hybrid GCM model with gray correlation feature selection is novel and effective
3. Control strategy has direct real-world optimization impact
MINOR COMMENTS
1. Expand description and rationale for 5 yield classification levels
2. Provide more details on model optimization approaches
3. Discuss limitations of findings and assumptions made
The paper makes a valuable contribution in applying state-of-the-art machine learning for industrial intelligence. I recommend acceptance after the authors address the minor comments above.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.