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.
The authors have addressed the consideration of the reviewers. It can be accepted currently. Congrats!
[# PeerJ Staff Note - this decision was reviewed and approved by Jyotismita Chaki, a PeerJ Section Editor covering this Section #]
The manuscript is now written in professional and clear English, with well-structured sections and a logical flow. Figures and tables are relevant, clearly labeled, and of good quality. The introduction has been strengthened by incorporating recent works related to slurry rheology, transport mechanisms, and online characterization methods, thereby improving the scientific context and motivation for the study.
The research question is clearly articulated and addresses an important knowledge gap in online prediction of slurry mass concentration for coal preparation processes.
The methods are described in sufficient detail to allow replication, particularly the ARI-ICA modeling framework, the adaptive node-generation strategy, and the data acquisition process. The authors have effectively integrated multi-source data—optical intensity and differential pressure signals—which enhances the robustness of the design.
The justification for selecting these modalities is now clearly explained, including why other sensor options were not adopted.
The results are statistically sound, well-controlled, and convincingly validated. The comparison with BP, SVR, and ELM models is clear and demonstrates meaningful improvements in accuracy and robustness offered by the ARI-ICA approach.
The conclusions are appropriately aligned with the research objectives and remain within the scope of the presented results.
Hyperparameter selection criteria and the rationale for the chosen parameter ranges have been clarified, improving transparency and reproducibility.
The manuscript titled “Study on Slurry Mass Concentration Prediction Method Driven by Multi-Source Data Fusion” presents a technically sound and novel approach for online prediction of slurry mass concentration in coal preparation systems. Recognizing the limitations of traditional offline and single-modality online measurements, the authors propose a bimodal soft-sensing model that fuses infrared optical intensity and differential pressure signals. The improved nonlinear ARI-ICA algorithm—incorporating adaptive node generation and regularization decay—enhances prediction robustness, feature sparsity, and interpretability under limited data conditions.
The author has revised the manuscript as per my comments. So it can be accepted in this form.
The author has revised the manuscript as per my comments. So it can be accepted in this form.
The author has revised the manuscript as per my comments. So it can be accepted in this form.
The author has revised the manuscript as per my comments. So it can be accepted in this form.
Accpeted
Accpeted
Accpeted
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** When preparing your next revision, please ensure that your manuscript is reviewed either by a colleague who is proficient in English and familiar with the subject matter, or by a professional editing service. PeerJ offers language editing services; if you are interested, you may contact us at [email protected] for pricing details. Kindly include your manuscript number and title in your inquiry. – PeerJ Staff
How does multi-source data fusion improve slurry concentration prediction?
What challenges limit real-time slurry mass concentration measurement accuracy?
What factors influence transmitted light intensity in slurry measurement systems?
How does pressure difference measurement complement optical methods for prediction?
How does ARI-ICA modeling reduce overfitting in small datasets?
Can dynamic weight sampling improve neural network convergence and prediction accuracy?
What role does regularization decay play in model generalization performance?
How does Figure 4 validate sensor integration in measurement systems?
What benefits arise from integrating physical equations with data-driven models?
How does ARI-ICA compare to BP, SVR, and ELM models?
How does Figure 11 depict accuracy differences among prediction models?
What does Figure 12 reveal about error distribution in ARI-ICA predictions?
How can real-time monitoring optimize the coal preparation process energy efficiency?
How does Figure 1 illustrate the slurry concentration's role in operations?
What insights does Figure 3 provide on infrared transmission measurement?
The manuscript is written in professional English, with well-structured sections and logical flow.
Figures and tables are relevant, of good quality, and properly labeled.
The introduction is strong but would benefit from the inclusion of more recent works addressing slurry rheology, transport, and online characterization.
The research question is clearly stated, filling an important knowledge gap in online slurry mass concentration prediction.
Methods are described with sufficient detail for replication, particularly the ARI-ICA framework and data acquisition system.
Experimental design integrates multi-source data effectively (optical and differential pressure signals), enhancing robustness.
Results are robust, statistically sound, and well-controlled.
The comparison with BP, SVR, and ELM models is convincing, showing clear advantages of the ARI-ICA approach.
Conclusions are appropriately linked to the original research question and limited to supporting results.
The article “Study on Slurry Mass Concentration Prediction Method Driven by Multi-Source Data Fusion” presents a novel approach for online prediction of slurry mass concentration in coal preparation processes. Recognizing the limitations of conventional offline and online measurement methods, the authors propose a bimodal soft-sensing model integrating infrared light intensity and differential pressure signals. An improved nonlinear ARI-ICA model is developed, introducing adaptive node generation and progressive regularization decay to improve prediction robustness and interpretability. Experimental validation demonstrates that the proposed method achieves higher accuracy and stability than conventional neural networks and SVR, making it highly relevant for real-time intelligent monitoring and control in industrial coal preparation. The manuscript is technically sound and requires minor revision before publication, focusing on improving language clarity, refining scientific writing, and strengthening the justification of model choices.
1. The introduction is strong but would benefit from the inclusion of more recent works addressing slurry rheology, transport, and online characterization.
2. Some figure captions could be expanded to be more self-explanatory, ensuring that figures are understandable even without referring back to the main text.
3. The choice of the ARI-ICA model over other advanced machine learning methods (e.g., deep learning architectures, ensemble models) should be explained more clearly. Why was ARI-ICA deemed more suitable for this slurry prediction task?
4. The rationale for selecting infrared light intensity and differential pressure as the two modalities should be discussed in greater detail. Were other sensor modalities considered?
5. The authors should clarify the criteria used to set the hyperparameters in the ARI-ICA framework beyond grid search, and explain why those final ranges were selected.
Captions for a few figures could be improved by adding more information. The introduction is solid but would benefit from a few additional references to recent slurry transport CFD studies for a broader context. The structure of the manuscript is appropriate.
The experimental methodology is clearly explained, and the proposed ARI-ICA model approach is the original contribution. Research questions are well defined and relevant.
The result would enhance the literature. The work could be further strengthened by including Uncertainty quantification. Conclusions are well stated.
A significant issue in coal preparation is addressed in this well-written manuscript: the precise online prediction of slurry mass concentration by multi-source data fusion. Since intelligent control and real-time sensing are crucial for energy optimization and process efficiency, the topic is relevant today. To forecast slurry mass concentration, the authors have integrated optical and pressure difference observations with an enhanced ARI-ICA model. This is both innovative and practically significant.
Comments
o Although the introduction gives useful background information, it should incorporate more recent slurry transport studies based on CFD.
o All trials are conducted in a lab-scale environment, despite the study's claims of industrial relevance. Please provide justification for scalability to plant-scale conditions (such as fluctuating turbulence intensity, particle size dispersion, or industrial noise).
o The size of the dataset (around 160 samples) is modest for a machine-learning application. The model's potential generalization to bigger, more diverse datasets should be discussed by the authors.
o There is no quantification of uncertainty. Claims would be more robust if Figures 7–10 included confidence ranges for the RMSE/MAE or error bands.
o For Equations (1–9), please define all constants clearly (e.g., absorbance coefficient, mean free path). Currently, variables appear without full physical meaning.
o Addition of the implications of the findings of the current study would enhance the readability of the manuscript.
1. Authors should modify the abstract section, incorporating the ranges of different parameters studied.
2. The conclusion section should be modified by quantifying the main findings of the results and discussion section.
3. Grammatical and typological errors should be rectified in the manuscript.
-
-
This paper proposes an advanced method for predicting slurry mass concentration in coal preparation by fusing light intensity and differential pressure signals and using a specialized non-linear model.
1. The paper argues that existing online techniques for volumetric concentration have reduced accuracy due to variations in particle true density. How does the proposed method uniquely address this specific issue? Is the particle's true density also a time-varying parameter that requires online prediction?
2. What are the specific features extracted from the raw light intensity and differential pressure signals?
3. The study is for coal preparation. What is the range of variation for key parameters in the experimental data?
4. What are the specific architectures and configurations of the traditional neural networks and support vector regression models used as benchmarks?
5. Considering the goal of real-time monitoring and process optimization control, what is the prediction response time of the proposed ARI-ICA method compared to the traditional methods? Is it fast enough for practical, closed-loop control?
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.