Cooperative Control for Multi-Module Charging Systems of Ultracapacitors

Round 1

Reviewer 1 Report

In the paper titled “Cooperative Cascade Charging of Autonomous Rail Rapid Transit (ART) Vehicles” authors describe proposed decentralized control method designed for the multi –module charging system in order to eliminate the current imbalance among modules. Presented scientific problem was formulated properly and the new control scheme was presented, simulated and experimentally verificated. The mathematical analysis of the described problem is good and does not raise any objections. I have some questions to the elaborated case:

• Is it the full hardware setup photo? You write that there are three power converters while the section (d) in the figure 8 looks like the only buck converter. I think, the other converters may be hidden under the controller board.
• How does the buck converters number limitation look like?
• How does the communication delays between buck converters and synchronizations process influence the control process?

In conclusion, I think that this is a good, valuable paper and could be improved by the real model experiments. Nevertheless, the low-scale model is sufficient to present the formulated problem.

Author Response

Responses to Reviewer’s Comments

Comment A “In the paper titled “Cooperative Cascade Charging of Autonomous Rail Rapid Transit (ART) Vehicles” authors describe proposed decentralized control method designed for the multi –module charging system in order to eliminate the current imbalance among modules. Presented scientific problem was formulated properly and the new control scheme was presented, simulated and experimentally verificated. The mathematical analysis of the described problem is good and does not raise any objections. I have some questions to the elaborated case:”

Response: Thanks for the positive comments. We have carefully revised our paper according to your comments, and we hope this improved manuscript would be found satisfactory.

Comment B “Is it the full hardware setup photo? You write that there are three power converters while the section (d) in the figure 8 looks like the only buck converter. I think, the other converters may be hidden under the controller board.”

Response: Thanks for the comment. Figure 8 shows the full hardware setup photo with three power converters. In the revised manuscript, we've annotated three converters in the figure. The details are given as follows:

Figure 8. The hardware setup of the three-charger testbed: (1) power source; (2) measurement board; (3) three buck converters; (4) ultracapacitor; (5) PXI platform; (6) Labview.

The experiment platform is shown in Fig. 8 which consists of (1) direct current (DC) power source, (2) measurement board, (3) three buck converters, (4) ultracapacitor, (5) PXI platform, and (6) Labview. The charging system consists of one micro-controller, three buck converters, and one ultracapacitor. In this paper, we choose the DSP2808 as the control chip, the controllable DC power supply as the power source, and a 100 F ultracapacitor of Maxwell as the load. The PXI platform is used to observe and record the waveform of the ultracapacitor voltage and three chargers' currents. (Section 4.3. Page 11)

Comment C “How does the buck converters number limitation look like?”

Response: Thanks for the comment. In the proposed cooperative charging method, each module communicates with its neighbors to make a control decision. In small-scale applications where the number of modules is limited, one single micro-controller can be applied to control the modules, where the inter-module communication is realized through the signal flow in the software design. In this case, the number of charging modules is limited by the hardware resources of the micro-controller (e.g., the number of IO ports). When the number of charging modules is large and multiple micro-controllers are applied, each module’s controller only communicates with its neighbors to generate the control signal, implying that the new plug-in module will not affect the other existing ones, i.e., the proposed method benefits from a good scalability. In the revised manuscript, we provide a scalability issue discussion in Section 4.6. The details are given as follows:

Scalability: The method proposed in this paper is a cooperative control method. Each charger only communicates with adjacent modules. Therefore, this method has a good scalability. The number of chargers will not affect the stability of the whole system. The proposed method is suitable for multi-charger applications. (Section 4.6.3. Page 14)

Comment D “How does the communication delays between buck converters and synchronizations process influence the control process?”

Response: Thanks for the comment. In the experiment, we apply one micro-controller to control three charging modules. The communications among modules are just signal flows in the software design. Thus, the communication delays and synchronization issues are avoided. We agree with the reviewer that when the proposed method is realized with multiple micro-controllers, the communication delay and synchronization issues exist and will affect the performance of the charging system. This is an interesting topic, and will be considered in our future work. In the revised manuscript, we provide a communication delays issue discussion in Section 4.6. The details are given as follows:

Communication delays: The proposed method can be applied in small-scale or large-scale scenarios. In small-scale application, only one micro-controller is needed. The micro-controller can provide the port needed for control without delay problem. In large-scale application where multiple micro-controllers are required, the communication delays between modules will affect the overall performance of the system. (Section 4.6.6. Page 14)

Time delay and synchronization issues of multiple micro-controllers and measurement noises will be considered in our future work. (Section 5. Page 15)

Comment D “In conclusion, I think that this is a good, valuable paper and could be improved by the real model experiments. Nevertheless, the low-scale model is sufficient to present the formulated problem.”

Response: Thanks for the suggestion. Due to the hardware restrictions, we cannot conduct the high-power experiments in the lab. In the revised manuscript, in order to illustrate the effectiveness of the proposed method in high-power applications, we re-conduct the simulations with large charging currents. The simulation results are shown in the figures below.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors,

First of all the Reviewer want to thank the Authors for the work performed. The paper discusses the cascade cooperating charging system for Autonomous Rail Rapid Transit (ART) Vehicles. In the Reviewer's opinion the paper requires significant improvements and therefore may not be published in the present form.

In general, the article presents a concept of control system, which ensures balanced current distribution for a supercapacitor based energy container under charging process for a ART vehicle. In practice the voltage level of the DC bus of traction network is about 1500 V and for currents equal to thousands of amperes. Since in real application of the bus voltage of VRT vehicles is reaches the level from several hundreds to several thousands of volts the power storage system are realized as an complex array of power cells. In such a case the balancing of charging current becomes an extremely important issue, ensuring a proper operation of the energy system. The selected operation conditions of investigated device were 24 V input voltage, 3 amperes and 2.7 V output voltage for three charging modules and one single supercapacitor on the output. In respect to the aforementioned, in reality the investigated system is a single multiphase asynchronous buck converter with a supercapacitor on the output. Taking into account the aforementioned, the proposed charging system is designed rather for supplying of a low voltage electronic circuitry instead of a high power ART vehicle. Moreover, the concept of a multiphase buck converter is well known since years and therefore is nothing novel in general. All of this implies a relatively poor scientific soundness of the proposed paper. In the present form, it may be published rather as an conference paper than a journal paper with impact factor. In lines 296-299 the Authors wrote: "The charging scheme proposed in this paper aims at the communication of different charging modules. Therefore, it will not increase the application difficulty due to the increase of current and it can be used in the practical charging scenario." Therefore, the Reviewer is wondering why the Authors did not conduct such tests, especially since the title of the paper indicates an application for ART vehicles. Inferring from the title, the reader could expect experimental results from a real vehicle rather than from a classic low-voltage multiphase step-down converter. Tis is an serious lack in the paper. The experimental part should be enhanced by appropriate results related to a charging system operating with a multi-cell power storage system, which ensures close to real operating conditions.

In respect to the above, the experimental results doesnt match the basic topic of the proposed paper and therefore, in the Reviewer's opinion, may not be accepted.

Sincerely Yours,

the Reviewer.

Author Response

Responses to Reviewer’s comments

Comment A “First of all the Reviewer want to thank the Authors for the work performed. The paper discusses the cascade cooperating charging system for Autonomous Rail Rapid Transit (ART) Vehicles. In the Reviewer's opinion the paper requires significant improvements and therefore may not be published in the present form.”

Response: Thanks for reviewing our manuscript and providing constructive comments. We have carefully revised our paper according to your comments, and we hope this improved manuscript would be found satisfactory.

Comment B “In general, the article presents a concept of control system, which ensures balanced current distribution for a supercapacitor based energy container under charging process for a ART vehicle. In practice the voltage level of the DC bus of traction network is about 1500 V and for currents equal to thousands of amperes. Since in real application of the bus voltage of VRT vehicles is reaches the level from several hundreds to several thousands of volts the power storage system are realized as an complex array of power cells. In such a case the balancing of charging current becomes an extremely important issue, ensuring a proper operation of the energy system. The selected operation conditions of investigated device were 24 V input voltage, 3 amperes and 2.7 V output voltage for three charging modules and one single supercapacitor on the output. In respect to the aforementioned, in reality the investigated system is a single multiphase asynchronous buck converter with a supercapacitor on the output. Taking into account the aforementioned, the proposed charging system is designed rather for supplying of a low voltage electronic circuitry instead of a high power ART vehicle.”

Response: Thanks for the comment. We agree that a large charging current is required in the charging process of ART vehicles. This paper focuses on a new control method rather than real vehicle experiments. In this paper, we propose a cooperative cascade charging method for onboard supercapacitors of ART vehicles. Due to the hardware restrictions, we verify the effectiveness of the proposed method using a scale-down laboratory prototype. In order to avoid confusions, we change the title of this paper to “A Cooperative Cascade Charging Method of Autonomous Rail Rapid Transit (ART) Vehicles”. Moreover, we provide the simulation results with high charging currents to illustrate the effectiveness of the proposed method in high-power applications. The details are as follows.

A Cooperative Cascade Charging Method of Autonomous Rail Rapid Transit (ART) Vehicles (Title, Page 1)

A laboratory prototype is built to verify the effectiveness of the proposed charging method. Both simulation and experiment results verify that the proposed method can suppress the current imbalance effectively when compared with the classical decentralized method. (Abstract, Page 1)

By selecting the corresponding module in Simulink, connecting each module with reference to the mathematical model, and adjusting the parameters, we build the Simulink block diagram, where the reference voltage is set to 810 V, and the three charging modules charge the capacitor with a total current of 900 A [19]. (Section 4.2. Page 11)

[19] P. Newman, K. Hargroves, S. Davies-Slate, D. Conley, M. Verschuer, M. Mouritz, and D. Yangka, “The trackless tram: Is it the transit and city shaping catalyst we have been waiting for?” Journal of Transportation Technologies, vol. 09, pp. 31–55, 01 2019.

Comment C “Moreover, the concept of a multiphase buck converter is well known since years and therefore is nothing novel in general. All of this implies a relatively poor scientific soundness of the proposed paper. In the present form, it may be published rather as an conference paper than a journal paper with impact factor.

Response: Thanks for the comment. The parallel multi-module charging system can be treated as a multi-phase buck converter indeed, and we agree that there are extensive studies in this topic. The contribution of this paper is that we propose a distributed cooperative control method for the modular charging systems. Compared with the classical centralized control method, the proposed method benefits from a good scalability and fault-tolerance capability (as shown in Fig. 11). When compared with the decentralized control method, the proposed method can effectively suppress the current imbalance (as shown in Fig. 7). In the revised manuscript, we discuss the superiority of the proposed method, which are provided as follows.

Superiority: By analyzing the experiment results in Fig. 7, the current imbalance can be restrained by the cooperative charging method compared with the decentralized control method. The stability and lifetime of the entire charging system are guaranteed. As shown in Fig. 11, compared with the classical centralized control method, the proposed method benefits from a good scalability and fault-tolerance capability. (Section 4.6.2. Page 14)

Comment D  In lines 296-299 the Authors wrote: "The charging scheme proposed in this paper aims at the communication of different charging modules. Therefore, it will not increase the application difficulty due to the increase of current and it can be used in the practical charging scenario." Therefore, the Reviewer is wondering why the Authors did not conduct such tests, especially since the title of the paper indicates an application for ART vehicles.”

Response: Thanks for the comment. The contribution of this paper is to propose a new cooperative control method for multi-module charging system. Due to the hardware restrictions, we verify the effectiveness of the proposed method using a laboratory prototype. It is necessary to discuss how to extend the proposed method to high-power applications in the real scenarios. Thus, we provide a paragraph to discuss the practicability of the proposed method. These statements have been refined as follows.

In the experiment, we set the current of each scale-down charging module as 1 A. In practical applications, a large current is needed to charge the vehicles. In order to get a large current, there are two ways: (i) increasing the current of each charging module, and (ii) using more charging modules. In the first case, the output current of each charging module can be increased by regulating the reference current in the cooperative charging method (assuming the current is still within the limit range of charging modules). In the second case, since the proposed method is a distributed method, increasing the number of charging modules will not increase the design complexity. Thus, the proposed method is desired in both cases, and can be applied in high-power applications. (Section 4.6.5. Page 14)

Comment E “Inferring from the title, the reader could expect experimental results from a real vehicle rather than from a classic low-voltage multiphase step-down converter. This is an serious lack in the paper. The experimental part should be enhanced by appropriate results related to a charging system operating with a multi-cell power storage system, which ensures close to real operating conditions.

In respect to the above, the experimental results doesn’t match the basic topic of the proposed paper and therefore, in the Reviewer's opinion, may not be accepted.”

Response: Thanks for the comment. Our response in regard to this comment includes the following comments.

• We agree that the title may be misunderstanding. The title of this paper has been changed to "A Cooperative Cascade Charging Method of Autonomous Rail Rapid Transit (ART) Vehicles".
• This paper focuses on the current balancing of multiple charging modules during the charging process, where the supercapacitor cells are just loads. Thus, one cell or multiple cells won’t affect the current balancing during the charging process (though the charging rate will be different). We agree with the reviewer, however, that both the current synchronization of charging modules and cell balancing of supercapacitors are necessary in the practical applications. This will be considered in our future work.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper provides a distributed approach for charging of super capacitors of the autonomous rail rapid transit vehicles via the distributed charging modules. Overall, the paper has an interesting topic and provides a technically sound approach. Here are my comments:

1. The paper claims that “there is no failure of single point for the proposed system”. So is the proposed system robust if the leader for the consensus algorithm fails? What if there is a communication link failure that makes the communication graph disconnected? If the paper is robust to these single points of failures please demonstrate it in the case studies. Otherwise, please correct this claim and explain its limitations.
2. In the experimental results how did the modules communicate? What was the communication setup/protocol, etc.
3. What happens if a module reaches to its current limit due to failure of another module and thus cannot deliver the requested current? Would the proposed approach result in other modules to increase their current generation to compensate for the module that has reached its limit?
4. What is the communication medium, protocol, etc. in the case studies.
5. How is the performance of the proposed approach under measurement noises?
6. Paper needs English editing, e.g.
   1. In conclusion “can guarantees” should be “can guarantee”
   2. Under section 4.1.2, the paper starts by “With loss of generality” which seems to be “without loss of generality”?
   3. Line 213, “Chosing” should be “choosing”

Author Response

Responses to Reviewer’s comments

Comment A “The paper provides a distributed approach for charging of super capacitors of the autonomous rail rapid transit vehicles via the distributed charging modules. Overall, the paper has an interesting topic and provides a technically sound approach. Here are my comments:”

Response: Thanks a lot for your time and efforts handling our manuscript. We have carefully revised our paper according to the comments from you, and we hope this improved manuscript would be found satisfactory.

Comment B “1. The paper claims that “there is no failure of single point for the proposed system”. So is the proposed system robust if the leader for the consensus algorithm fails? What if there is a communication link failure that makes the communication graph disconnected? If the paper is robust to these single points of failures please demonstrate it in the case studies. Otherwise, please correct this claim and explain its limitations.”

Response: Thanks for the suggestion. There are two prerequisites for the effectiveness of the proposed method: (i) the reference pins to at least one charging module; (ii) the communication graph of modules is connected. If the leader (pinned module) fails, the charging modules cannot receive the correct reference current. If the communication graph is disconnected, the isolated module cannot receive the reference current information neither. In the revised manuscript, we provide a related statement in Section 2.2 to explain the limitation. The details are as follows.

There are two prerequisites for the effectiveness of the proposed method: (i) the reference pins to at least one charging module; (ii) the communication graph of modules is connected. (Section 2.2. Page 6)

Comment C “2. In the experimental results how did the modules communicate? What was the communication setup/protocol, etc.”

Response: Thanks for the comment. In the experiment, we apply one micro-controller to control three charging modules, i.e., there are three control procedures in the single micro-controller, where the inter-module communications are just signal flows in the software design. In the revised manuscript, we added the relevant description in Section 4.3. The details are as follows.

In the experiment, we apply one micro-controller to control three charging modules, i.e., there are three control procedures in the micro-controller. Each control procedure collects the current of the corresponding charging module and then communicates with each other through the signal flow to make a control decision. The communications among modules are signal flows in the software design. (Section 4.3. Page 11)

Comment D “3. What happens if a module reaches to its current limit due to failure of another module and thus cannot deliver the requested current? Would the proposed approach result in other modules to increase their current generation to compensate for the module that has reached its limit?”

Response: Thanks for the comment. When the current of a module reaches the limit, we limit the current through the program design to ensure safety. In the revised manuscript, we explained it in Section 4.5. The details of the revisions are given as follows:

Due to the failure of other modules, the current of single module may be too high. When the current of a module reaches the limit, we limit its value through the program design to ensure safety. (Section 4.5. Page 13)

Comment E “4. What is the communication medium, protocol, etc. in the case studies.”

Response: Thanks for the comment. In the experiment, we apply one micro-controller to control three charging modules. The communications among modules are just signal flows in the software design.

Comment F “5.  How is the performance of the proposed approach under measurement noises?”

Response: Thanks for the comment. In the experiment, the measured analog signal is filtered by the operational amplifier TL074ID. The influence of communication delay and measurement noises will be further considered in the future work. We explained it in the revised manuscript. The details are as follows.

The measured analog signal is filtered by the operational amplifier TL074ID. (Section 4.3. Page 11)

Time delay and synchronization issues of multiple micro-controllers and measurement noises will be considered in our future work. (Section 5. Page 15)

Comment G “6. Paper needs English editing, e.g.

1. In conclusion “can guarantees” should be “can guarantee”
2. Under section 4.1.2, the paper starts by “With loss of generality” which seems to be “without loss of generality”?
3. Line 213, “Chosing” should be “choosing””

Response: Thanks for the suggestion. We have revised the inappropriate English expressions in the manuscript and rechecked the full text.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Dear Authors,

thank You for the improvements made. Additional comments are included in the attached file.

Sincerely Yours,

The Reviewer.

Comments for author File: Comments.pdf

Author Response

Responses to Reviewer’s comments

Comment A.II “The Reviewer want to thank to the Authors for the improvements made. Unfortunately the title of the presented paper should precise match to the experimental content of the manuscript. Therefore, the title of the paper should be revised once time more. Detailed comments are present in the comments listed below.”

Response: Thanks again for reviewing our manuscript and providing constructive comments. We have carefully revised our paper according to your comments. We change the title of this paper to “Cooperative Control for Multi-Module Charging Systems of Ultracapacitors” and revise the manuscript correspondingly. We hope the revised manuscript would be found satisfactory.

Comment B.II “First of all, the Reviewer thanks to the Authors for the response obtained. A single output capacitor allows for verification of current balance. This issue becomes more complex in case of an array of connected supercapacitors. From this point-of-view a critical issue becomes a proper voltage balancing of each single cell, which has to be realized using a proper current balancing. In respect to the aforementioned, the current balancing technique should taking into account the relation of voltages between particular cells. In the investigated laboratory setup the issue of voltage balancing does not occur. In such a case the realized experimental tests is a huge simplification and omits very important issues in electric energy containers based on supercapacitors. This phenomenon is essential especially in charging system dedicated for Autonomous Rail Rapid Transit (ART) Vehicles. Therefore, improving the simulation tests only by increasing of the voltage and current level, without multiplication of power cells is not sufficient. Thus, the conducted experimental tests do not correspond to the reality of power charging systems.”

Response: Thanks for the comment. We agree that the low-power laboratory experiment setup may not well support the application of the proposed method in real ART scenarios. In the revised manuscript, the ART application has been removed from both the title and the main content of the paper. Detailed revisions can be found in the title, abstract and introduction of the revised manuscript.

Comment C.II “First of all, the Reviewer thanks to the Authors for the response obtained. The improvement made in the title, in the Reviewer’s opinion, is not sufficient. From the practical point-of-view the investigated system does not correspond to a VRT charging system. It remains still a distributed cooperative control method for a single multiphase-buck converter. The mentioned application at the introduction seems occur only for improvement of the improving the attractiveness of the proposed control system, and is not really supported by simulation either experimental tests. In the Reviewers opinion the proved application in proposed manuscript is related to a distributed cooperative control method for a multiphase-buck converter equipped in a supercapacitor on the output, not for a VRT charging system. Therefore, the proposed title still does not match to the experimental content of the paper.”

Response: Thanks for the constructive comments. We agree with the reviewer that the paper is focused on the distributed cooperative control for the multiphase converter with a supercapacitor as the load. For the consistency throughout the paper, we have revised the title, abstract, introduction sections, which are provided as follows.

Cooperative Control for Multi-Module Charging Systems of Ultracapacitors (Title, Page 1)

Ultracapacitors have lately received great attention for energy storage due to their small pollution, high power density, and long lifetime. In many applications, ultracapacitors need to be charged with a high current, where a multi-module charging system is typically adopted. (Abstract, Page 1)

As an emerging energy storage device, ultracapacitors have been widely used in many high-power applications, including catenary-free trams [1,2], elevators [3], and electric vehicles [4-6]. Compared with traditional batteries, ultracapacitors have its merits such as faster charging and discharging, greater power density, and longer service lifetime [7-9].

In practical applications, the ultracapacitors need to be charged and discharged frequently [10]. There is a great demand for the charging speed of the ultracapacitor and high current charging is adopted generally. For instance, in the catenary-free tram applications, onboard supercapacitors need to be recharged at each station where the charging time is limited to 30 seconds [11]. In order to satisfy the fast charging requirement, the multi-module charging system is typically adopted [12,13].

In practice, a decentralized control method is applied to manage the multi-module charging system, which means that charging modules work independently [14-16]. Due to the differences of the charging modules, this method may cause charging current imbalance among charging modules, resulting in different lifetimes of each charging module. Extensive centralized control methods have also been proposed to manage multiple modules, i.e., the currents of all modules are collected and processed in a central node [17]. The drawback of the centralized methods is that the single point failure degrades the reliability of the multi-module charging system [15]. (Section 1. Page 1)

Comment D.II “First of all, the Reviewer thanks to the Authors for the response obtained. The introduced discussion is still in the area of hypothetical considerations because any tests has not been made to confirm the thesis made. Therefore, the divagations have a poor experimental background rather than confirmed properties. In such a case the scientific impact of the discussion made is none.”

Response: Thanks for the comment. We agree with the reviewer the discussion without experimental support is difficult to convince readers. This part of the discussion has been removed from the revised manuscript.

Comment E.II “First of all, the Reviewer thanks to the Authors for the response obtained. In the Reviewer opinion the title of the manuscript should be focused on a concept of a distributed cooperative control method for a multiphase-buck converter equipped in a supercapacitor on the output instead of a VRT vehicle charging system. Therefore, the Reviewer suggests to add sufficient experimental tests or change the title of the manuscript in order to met the experimental content presented in the manuscript.”

Response: Thanks for the comment. We change the title of this paper to “Cooperative Control for Multi-Module Charging Systems of Ultracapacitors” and revise the corresponding presentations in the revised manuscript.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors have addressed my comments. I do not have any additional comments. Thank you.

Author Response

Responses to Reviewer’s comments

Comment A “The authors have addressed my comments. I do not have any additional comments. Thank you.”

Response: Thank you very much for reviewing our manuscript and providing valuable comments in the previous round.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Dear Authors,

thank You for the improvements made. In the Reviewers opinion the preented paper may be almost published in the present form. The only remark is related to the simulational data. It should be consistent with the experimental tests i.e. the level of voltage and current. After this modification an another review is not necessary and the paper may be published.

Sincerely Yours,

The Reviewer.

Author Response

Responses to Reviewer’s comments

Comment A “Dear Authors,

thank you for the improvements made. In the Reviewers opinion the preented paper may be almost published in the present form. The only remark is related to the simulational data. It should be consistent with the experimental tests i.e. the level of voltage and current. After this modification an another review is not necessary and the paper may be published.”

Response: Thank you very much for reviewing our manuscript and providing valuable comments. Following your suggestions, we have modified the simulation to be consistent with experiment results. The simulation results are shown in the figures below.

Author Response File: Author Response.pdf