Advanced ML-Based Ensemble and Deep Learning Models for Short-Term Load Forecasting: Comparative Analysis Using Feature Engineering

Round 1

Reviewer 1 Report

In the manuscript, there is no explanation of what bagging is and why the authors use bagging in their model among diverse ensemble approaches. In addition, although bagging ensemble can use a variety of machine learning models, there is no mention of why they select LR and SVR as weak learners.

If this point is complemented, the soundness of the paper will be improved.

Author Response

Dear Editor,

The authors are grateful for providing major revisions and resubmissions. We hereby attached the response file to the first reviewer below.

Best Regards,

Authors

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper proposes a short-term load forecasting approach using bagging approach to ensemble multiple machine learning algorithms. The MSE loss is used as the loss function for the training process. Comparative analysis is performed against the popular deep learning forecasting algorithms such as DNN and LSTM. Experimental results confirmed the outperformance of the proposed approach. Overall, this is an interesting research that may attract attention from the readers in the energy sector. However, the current version of the manuscript requires strong revision for the following reasons:

 

The literature review is rather weak. A lot of recent load forecasting papers are not sufficiently discussed.

 

How is cross-validation performed? More details are needed.

 

How are the inputs of the prediction model determined? Any computation of ACF and PACF?

 

More detail regarding the determination of the seasonal patterns in the time-series data need to be provided.

 

Please cite the following two papers and discuss how signal decomposition approach can assist with load forecasting:

 

Li, H., Deng, J., Feng, P., Pu, C., Arachchige, D. D., & Cheng, Q. (2021). Short-Term Nacelle Orientation Forecasting Using Bilinear Transformation and ICEEMDAN Framework. Frontiers in Energy Research, 697.

Li, H., Deng, J., Yuan, S., Feng, P., & Arachchige, D. D. (2021). Monitoring and Identifying Wind Turbine Generator Bearing Faults Using Deep Belief Network and EWMA Control Charts. Frontiers in Energy Research, 770.

 

Author Response

Dear Editor,

We hereby attached the response file to the second reviewer below. We thank you for encouraging us to resubmit our manuscript.

Kind Regards,

Author

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The reviewer believes the authors have addressed all comments and made great improvement to the manuscript.