Safety Monitoring Method for the Uplift Pressure of Concrete Dams Based on Optimized Spatiotemporal Clustering and the Bayesian Panel Vector Autoregressive Model

Round 1

Reviewer 1 Report

The article by Cheng L. et al. describes a mathematical system for monitoring water levels in hydropower dams and controlling seepage. The article contains a good and comprehensive introduction. The existing mathematical models and their accompanying calculations are also thoroughly shown by the authors. The authors also describe in detail a flowchart of the uplift pressure safety monitoring method. In the end, the authors test the model on data obtained from a real hydropower station and compare the results with those predicted by other methods. In general, the article is very detailed, which is its undoubted advantage. The only drawback of the article is that the authors could have added more supporting references to the literature.

Specific comments are as follows.

Line 47: “domestic and international scholars”. The words "domestic and international" seem unnecessary, as this article is in an international journal for readers from all over the world.

Line 48: “spatiotemporal distribution models, principal component analysis, multioutput machine learning models and panel data models”. It would be helpful if each model was accompanied by a reference to a review or an article describing it.

Lines 119-123: “Common distance measurement methods mainly include Euclidean distance, Manhattan distance, Mahalanobis distance and so on. In contrast to the distance metric, the value of the similarity metric is inversely proportional to the difference. The most commonly utilized similarity methods include the Pearson correlation coefficient and Bharbyian distance.” Supporting references would be helpful. The same applies to the following text describing the three methods.

Lines 139, 342: “Hossein Kamalzadeha and Abbas Ahmadi et al. proposed”, “The method provided by the Gibbs sampler”. References are needed.

Line 154: “First, two time series, Q = [q1, q2, …, qm] and C = [c1, c2, …, cn], and The permutation is constructed into an m×n matrix. Each point (i,j) in the matrix represents the distance measure of and.” There are one or two errors in the text here. Please check.

Lines 220-235, 282-287, 306-328. It would be helpful if the authors added references to the literature.

Line 362: “The minimum number of points in the neighbourhood radius is .” The sentence seems unfinished.

Figure 3 contains Chinese characters that should be replaced with English text.

Table 2. There is no reference nor discussion of this table in the text of the article.

Table 4. Are the authors sure that the accuracy of measurements is such that it is possible to write values with such precision as 5 digits after the point? It seems that the numbers should be rounded.

The English language is fine, but the article needs editing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In the article authors proposes a safety monitoring model that considers space-time information on the uplift pressure of a dam foundation. It is a good study. However, I would like to suggest some revisions, which are remarked on the text, as given below:

The title of a scientific study should be clear and more understandable.  I would like to make a revision on the title as given below:

“Safety Monitoring Method for the Uplift Pressure of Concrete Dam based on Optimized Spatiotemporal Clustering and the Bayesian Panel Vector Autoregressive Model”

What is the scientific conclusion of this study?  It must be given by one or two sentences in Abstract.

I think it would be better if the authors mention to the name of the dam project.  Additionally its location, the name of river on which the dam was constructed and a brief of foundation materials such as its geology, structural features, groundwater level etc., should be given here. It can be a good reference for the forthcoming studies, which will be performed by other scientists.

The locations of some points such as UP8, UP9, .....UP, 15 and UP16 are not showed on the Figure 3. They can be out of order.  Their position should be mentioned here.

I know that LLC test, which is used to check the unit root of the panel data for uplift pressure, are mentioned in section 2.3. The others are IPS test and ADF-Fisher test.  it would be better if authors a brief definition about tests here generally (without theoretical base), comparatively.

The authors suggest a forthcoming studies on finding more effective methods to determine the most accurate hyperparameters.  What does it mean hyperparameters here? Please explain it by one or two sentences. The outline of next studies can be given in a brief form.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf