Lithium-Ion Battery State-of-Health Prediction for New-Energy Electric Vehicles Based on Random Forest Improved Model

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

·         The abstract is too long and many the introduction part is general. Some introduction text can be removed especially the first 2 sentences.

·         Line 44, there are 2 hyphens. Please remove one of them.

·         Any machine learning methods especially recent ones were reported for similar approach? If yes, what are their limitations that lead to your study?

·         Last paragraph of introduction is not needed. Can remove it unless ore present in the flowchart format.

·          The introduction of 2.1 section is too long. Those introduction should be in section 1 – Introduction. For section 2, please outline the stepwise procedures accurately.

·         Where did you get the data for this study?

·         The setup and procedures to use RF are unclear.

·         Define all parameters in Eq 5 and 6. Also, others.

·         How did you verify the prediction? Did you run the experiments at other values that were initially not used in RF?

·         For results comparisons, the true values were initially adopted/employed for RF?? If yes, did you run other points that were never used for RF to prove the validity?

·         Fig 6 can be removed. Showing the fitted equations are good enough.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript can be accepted at its current stage. The following are few comments:

1. Abbrevations to be added at the end of the manuscript.

2. List of contributions to be added in the introduction

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The article focuses on addressing environmental pollution in urban transportation by promoting green transportation and low-carbon environmental protection through new energy electric vehicles. Lithium-ion batteries (LIB) are highlighted as a predominant power source for electric vehicles, and maintaining the state of health (SOH) of LIB is crucial for their stable operation. The paper presents an improved prediction approach for estimating LIB SOH using experimental data from lithium-iron phosphate (LiFePO4) batteries. This approach utilizes a random forest (RF) model and incorporates a particle swarm optimization (PSO) algorithm to adaptively optimize key RF model parameters. Results demonstrate that the proposed approach outperforms other models in predicting LIB SOH, emphasizing its potential for enhancing the quality of new energy electric vehicles and promoting green transportation.

 

Comments for Manuscript Modification:

 

Clearly articulate the significance of the research at the beginning of the article, emphasizing the importance of accurate SOH prediction for the success of new energy electric vehicles and green transportation.

 

Provide a brief introduction to the challenges and drawbacks of traditional fuel vehicles in terms of environmental pollution to set the context for the importance of green transportation.

 

Simplify the language in the abstract to ensure that readers from various backgrounds can grasp the main findings and contributions of the study.

 

In the methodology section, consider including a flowchart or diagram illustrating the steps involved in the proposed SOH prediction approach using the RF model and the PSO algorithm.

 

Organize the equations in the modeling section clearly and concisely, emphasizing the main equations necessary to understand the model and its optimization process.

 

Explain the rationale behind choosing the specific five major parameters for optimization using the PSO algorithm and how these parameters affect the accuracy of SOH prediction.

 

Provide a more detailed explanation of the experimental data collection process, including the setup, conditions, and equipment used, to ensure transparency and reproducibility.

 

Discuss potential limitations or uncertainties associated with the proposed SOH prediction approach, and suggest avenues for future research to address these limitations.

 

Include a section that discusses the broader implications of accurate SOH prediction for LIB in electric vehicles, such as its impact on battery longevity, vehicle efficiency, and environmental benefits.

 

Highlight the practical applications and potential industry relevance of the proposed methodology, including its adaptability to different types of batteries and electric vehicle models.

 

Consider providing a brief comparison of the computational requirements and resource constraints associated with the RF model and the PSO algorithm to help readers understand the computational complexity of the approach.

 

Address any ethical or safety considerations related to the implementation of the proposed methodology, especially if it involves adapting parameters using machine learning algorithms.

 

Include a brief section on data availability and code availability to encourage transparency and collaboration within the research community.

 

Conclude the manuscript by summarizing the key findings, contributions, and the practical significance of the research in promoting green transportation and enhancing the performance of new energy electric vehicles.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The article discusses the growing concern over environmental pollution caused by traditional fuel vehicles in the context of urban transportation development. It highlights the emergence of new energy electric vehicles, particularly those powered by lithium-ion batteries (LIB), as a solution to address environmental issues. Maintaining the state of health (SOH) of LIB is crucial for the stable operation of electric vehicles, and accurately predicting SOH is essential for improving vehicle quality and reducing energy consumption. The article presents an improved prediction approach that uses experimental data from lithium-iron phosphate (LiFePO4) battery charge-discharge cycles. This approach combines a random forest (RF) model with a particle swarm optimization (PSO) algorithm to adaptively optimize key parameters of the RF model. The results demonstrate the effectiveness of this approach in predicting SOH, outperforming other models.

 

Comments for Manuscript Modification:

 

1. Provide a more concise and focused introduction to the article's objectives and contributions.

2. Clarify the specific environmental issues associated with traditional fuel vehicles and how new energy electric vehicles can address them.

3. Clearly define the acronym SOH (State of Health) upon its first use in the article.

4. Describe the experimental setup and data collection process in more detail to ensure reproducibility.

5. Explain the rationale for choosing lithium-iron phosphate (LiFePO4) batteries as the basis for experimentation.

6. Provide a brief overview of the random forest (RF) model and its relevance to predicting SOH.

7. Elaborate on the particle swarm optimization (PSO) algorithm and how it optimizes the RF model parameters.

8. Include a table or chart summarizing the results for a more accessible understanding of the model comparisons.

9. Discuss the practical implications of the research findings and potential applications in the electric vehicle industry.

10. Consider expanding the conclusion section to discuss future research directions and limitations of the proposed methodology.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf