Framework for Dynamic Modelling of the Dam and Reservoir System Reduced Functionality in Adverse Operating Conditions

Round 1

Reviewer 1 Report

Specific comments provided in attached PDF. Below are my general comments.

General

• Overall the paper was a bit difficult to follow, and requires some work to make the key contributions and concepts clear. What makes this work an advancement over other similar work? The only thing that was obvious after reading was the use of generic 'functionality indicators' to instigate failure scenarios in the modelling. However, these are still applied on a case-by-case basis to the different types of components within the system.
• Major English edit recommended.
• I would caution the authors against the presentation of inflows as a disturbance… Dam Systems are constructed on rivers, to control and manage inflows. Inflows are an input across the system boundary, but even extreme inflows aren't technically a disturbance. These are flow control systems.
• Please clarify upfront within the paper what the key scientific contributions are.

 

Methodology

• General comment - the methodology is hard to follow partly due to lack of up front definitions of key concepts/terms, and partly due to confusing terminology for key parameters.
• Use of the terms "failure prone parameters" and "functionality indicators" is confusing. The failure prone parameters suggested in the paper appear to be failure outcomes - how is "penstock leakage" "failure prone"? What is the difference between a functionality indicator and a failure prone parameter? Eg. Penstock leakage is a "failure prone parameter" but also could be an indicator of functionality. These concepts, and others within the paper, are not clearly defined in the text or easily understandable. It may be useful to re-think the names of these concepts to something more self-explanatory to prevent confusion. Adding definitions and explanations of how they're used within the model or as results indicators would be helpful to the reader.
• What is the difference between what the authors are calling the "control unit" sub-system that determines how the system is operated, and the "system operating rules" which are also listed within the "input scenarios" block? Operations is an important part of the system structure, not an input.. Perhaps step (7) refers to external demands - eg. Irrigation, power demand? If so this should be clarified.
• Broad crested weir equation suggested for gated spillways - this is not appropriate for the use case, which as the text suggests is "flow confined under a gate". Real-world gates have complex rating curves and may transition from weir to large orifice to small orifice flow depending on the head. So representing gate flows using a single equation and single coefficient may not be adequate for operating conditions outside of the normal range.
• Subsystems (at least maintenance/control) are being described as "units". This is a possible source of confusion as "units" in hydropower typically refers to generating units - suggest referring to these as sub-systems if that's what they are (and others).
• Variable time steps are mentioned in the methodology, but no guidance given as to why this might be useful or how it might be implemented. Case study appears to be hourly, and not variable. Please clarify whether this capability has been added and how.
• The sub-system descriptions struggle to link the model structure back to the physical reality of a generic DRS - in particular what variables are representations of the real system, and what ones are modelling conveniences.
• Table 1 - equation provided for Qleak does not imply how Qleak itself would be computed - penstock ruptures could look a variety of different ways and result in many different leakage outcomes.
• Subsystems being described as "units" is unclear ("units" in hydropower typically refers to generating units) - suggest referring to these as sub-systems if that's what they are.
• Dynamic resilience mentioned in Methodology but does not appear to be utilized in results/case study.
• How does the model decide what "functionality indicator" values are depending on the disturbance inputs? In some cases there may be complex engineering analyses required to determine how a component of the system is going to respond to an earthquake or other disturbance. I realize this is something we have to approximate with these complex system models but the limitations of the analysis should be commented on in the methodology, not just the conclusion.
• Is any probabilistic analysis proposed? Dam owners are interested in evaluating the likelihood of various events or sequences of events… How will this methodology work towards that goal?

 

Case Study

• Better description of the features of this system would be helpful. Author refers to another paper but doesn't provide enough detail to understand the case study as a stand-alone document. Without looking at that document, I do not know whether the DRS case study has a gated or a free overflow spillway.
• Can some more details on hydrologic inputs be provided? Are you just routing single hydrographs or are you routing longer term stochastic/historical inflows? It's not clear.
• Is the model capable of simulating variable flow releases from turbines and gates due to operational decisions/demand/release constraints? If so, there is no description of how this is implemented either in methodology or case study. Is the position of gates explicitly modelled? 
• Table 3 - adding a header over AS1 and AS2 identifying these columns as scenarios would be helpful for clarity.
• Similar to my point from methodology, I had confusion when reading the Case study as to whether operations are represented as part of the system structure or as an input?
• Scenario generation - the case study implies that just a few scenarios were investigated, with various (randomly picked?) disturbances implemented at different (random?) times. The methodology and discussion indicates that these are somehow generated - implying an algorithm was used. It is not clear how the disturbance scenarios and timing are generated. If they are just randomly chosen, as it seems they are in the case study, they can only realistically represent one possible sequence of events of an extremely high number of possibilities. Please provide some clarity on this and the potential limitations throughout the methodology and case study.
• How are failure impacts (functionality indicators) determined from the inputs?

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors,
The article "Framework for dynamic modelling of the dam and reservoir system reduced functionality in adverse operating conditions" presents an interesting approach to the presented topic.
The work consists of 5 chapters, which structure is arranged in a logical whole. A strong point of the study is an attempt to decompose the system supervising the operation of the device. The content itself is prepared very carefully and does not require much input for improvement.

Despite this, the article has several shortcomings that should be taken into account in order to improve its attractiveness to the reader:
- Figure 6 as presented is unreadable - it is not clear what they are about.
- Table 3 presents failure rates that relate to mathematical relationships by their number in the text. It would be better if the table included the designations of these indicators, and the numbers could appear next to them in an additional column.
- In Figure 7, I propose to make the axis descriptions in black, they will be more visible.
- I suggest to better describe in the text the two bottom figures located in figure 7. Without knowing the topic, it is very difficult to understand what they represent.

Conclusion: 
In order to continue the publication process further, the reviewer would like to encourage the authors to follow the comments/suggestions. A thorough revision of the paper is required. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I am generally OK with the responses and changes.