Anomaly Detection Method for Multivariate Time Series Data of Oil and Gas Stations Based on Digital Twin and MTAD-GAN

Round 1

Reviewer 1 Report

In this manuscript, the authors propose an anomaly detection method for multivariate time series data. The method is based on a combination of digital twin and a generative deep neural network approach. An extensive comparison with other methods show a considerable improvement.

The manuscript is indeed interesting. The context is novel and the exploitation of digital twin is appreciated. There are some issues the authors could address in order to improve the quality of the manuscript:

- The Related Work section offers an overall view of related studies. However, it is not straightforward to grasp pros and cons of each approach w.r.t. the proposed one. I suggest the authors to include a table highlighting these aspects.
- As for the anomaly detection context, the authors could consider to add to the reference the following work [https://doi.org/10.1016/j.future.2020.08.010].
- Eq. 4, 6 and 7 could be formatted better.
- The authors claim to introduce a multivariate time series anomaly detection method. However, the experiments seem to use training data; also, in Table 3 authors even report the train time for the proposed approach. Please clarify.
- A brief introduction of all methods used for comparison in Section 4.3 should be provided.
- Acronyms should be uniformed throughout the whole manuscript. Also, some typos are present, I suggest the authors to carefully read the manuscript and fix these.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

this paper proposes a digital twin-driven MTAD-GAN(Multivariate Time Series Data Anomaly Detection with GAN) oil and gas station anomaly detection method by combining mechanism of knowledge graph attention and temporal hawkes attention to judge the abnormal samples by a given threshold. The topic addressed in the manuscript is potentially interesting and the manuscript contains some practical meanings, however, there are some issues which should be addressed by the authors:

 -The readability and presentation of the study should be further improved. Please make sure throughout the text to include spaces before adding a citation (e.g., “Transfer learning[30]” should be “Transfer learning [30]”, “Figure 3.During the” should be “Figure 3. During the”,…

- The related work section needs a major revision in terms of providing more accurate and informative literature review and the pros and cons of the available approaches and how the proposed method is different comparatively.

-In section 3.2.5, can you discuss the applicability of hyper-parameters optomizers.

-"Discussion" section should be added in a more highlighting, argumentative way. The author should analysis the reason why the tested results is achieved.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors successfully addressed my concern.

Author Response

Thank you for your decision and constructive comments on our manuscript.