Improvement of Energy Efficiency and Control Performance of Cooling System Fan Applied to Industry 4.0 Data Center
Round 1
Reviewer 1 Report
This paper proposes a control method to improve energy efficiency and performance of cooling fan used for cooling using fuzzy controllers. The authors claim that this work improves the energy efficiency and performance of the cooling system using a FAN. The reviewer’s comments are listed below:
1. As mentioned in Section 4, PI controller is limited in performance due to fixed gain. Why the fuzzy system is designed to output the input of PI controller, not to the PI gains of PI controller.
2. Some minor points:
Line 205: What is LM? Should be NM?
Line 217: Figure 5? Two “Figure 5”s in this manuscript, the first one is in Line 197.
Line 253: PI controller 2 outputs … Should be PI controller 1?
Line 288~289: The hot side temperature was lower…cold side was 3.18 ℃. However, from Figure 12, ΔT cannot be equal to 3.18℃.
Specific Comments:
1. In the design of the PI controller, anti-windup is essential to the integral term because the integral operation will lead to the saturation of the cooling fan, resulting in no control response. Therefore you should give more description about Figure 5. How can you obtain Equation 17 (PI controller) from Figure 5.
2. In your Fuzzy system, three variables, error, changing error and output, are transferred through membership function. How do you decide the parameters of membership functions?
3. In Table 1 the fuzzy rules seem to give the control value of fan speed, but in FPI, the fuzzy system is to output the input of PI controller. Is it consistent? Have you taken into account the possibility to reduce the fuzzy rules? Please explain more detail about the design of fuzzy rules.
4. The fan control signal is PWM. It is recommended that you should plot the control result in every experiment for the evaluation of control performance.
5. How do you evaluate the fan power consumption? The results show that the controlled variable ΔT is oscillating in every controller, resulting in fan speed oscillation. Why do you plot the fan power consumption using unit of mW instead of W?
Author Response
English language and style
( ) Extensive editing of English language and style required
(x) Moderate English changes required
( ) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style
Reply
à
The English part was revised from the reviewer’s comment to the whole part of the paper through the specialized agency.
Yes Can be improved Must be improved Not applicable
Does the introduction provide sufficient background and include all relevant references? ( ) ( ) (x) ( )
Reply
à
We have added the contributions and references of the paper to improve the introduction.
ADD 1)
In this paper, we propose an FPI controller that controls the input value of the PI controller using fuzzy control, a VFPI controller that adjusts the fuzzy control gain of the FPI controller, and a VFPI_VL controller that adjusts the output limit of the VFPI controller. The limitation of the performance improvement by the fixed gain value of the conventional PI controller can be solved by adjusting the input value using the fuzzy control and the control performance can be improved by adjusting the gain value of the fuzzy control and the limit value of the controller output. The control method proposed in this paper is applied to the control system using FAN. The cooling system using the FAN changes the power consumption according to the operating time and speed of the FAN. The method proposed in this paper optimally controls the operation time and speed of the FAN according to the operating conditions, thereby reducing the power consumption of the FAN. The controller proposed in this paper can be applied to various variable speed drive systems, and it is expected to improve speed control performance and reduce power consumption.
ADD 2)
Reference
Khooban, M. H.; Soltanpour, M. R.; Abadi, D. N. M.; Esfahani, Z. Optimal Intelligent Control for HVAC System. Journal of Power Technologies. 2012, 92, 192-200
Khooban, M. H.; Adadi, D. N. M.; Alfi, A.; Siahi, M. Optimal Type-2 Fuzzy Controller For HVAC System. Automatika. 2014, 55, 69-78.
Khooban, M. H.; Niknam, T. A new intelligent online fuzzy tuning approach for multi-area load frefuency control: Self Adatpvie Modified Bat Algorithm. International Journal Electrical Power and Energy Systems. 2015, 71, 254-261.
Khooban. M. H.; Naghash-Almasi, O.; Niknam, T.; Sha-Sadeghi, M. Intelligent robust Pi adaptive control strategy for speed control of EV(s). The Institution of Engineering and Technology(IET). 2016, 10, 433-441.
Is the research design appropriate? ( ) (x) ( ) ( )
Reply
à
In order to improve the contents of the research design, we added the formula and flowchart, and improved the description of the design.
ADD 1)
| (8) |
(9) | |
(10) | |
(11) | |
(12) | |
(13) | |
(14) |
ADD 2)
Figure 11 shows the flow chart for the PWM control that is the speed control signal of the FAN. Set the reference temperature (Set_Tem) and calculate the temperature difference ( by measuring the temperature of the hot side () and cold side () of the thermoelectric element. This temperature difference is used to maintain the PWM through the Main controller or to calculate a new PWM signal. The control ends when the set time is over or when the user enters the stop.
Figure 11. Flow chart for control
Are the methods adequately described? ( ) ( ) (x) ( )
Reply
à
We have revised the content and added references to improve the adequateness of the method.
CHANGE 1)
Fuzzy control is a control method that uses the ambiguity of the boundary. And the fuzzy membership function shows the degree of membership of the input value. In this paper, membership range of membership function is used as follows. The required control amount according to the input value is set as Large, Medium, Small and Zero according to the size, and it is divided into 7 part as positive and negative according to the control direction. The parameter of the membership function is used as a percentage of input value. For fast control speed, the range of NL, NM, PM, PL is set large, ZE is smallest, and NS and PS are medium. For this, We have designed membership function as follow. the range of ZE is -0.2 to 0.2, PS and NS are | 0-0.6 |, NM and PM are | 0.2-1 |, NL and PL are | above 0.6 |. In this paper, 49 rules were used. The number of rules affects computation time and system performance. If the number of rules is large, the calculation time increases but the control performance is the best[45-47]. Therefore, in this paper, we use 49 rules, which shows the best control performance because the change in temperature is not fast.
CHANGE 2)
Figure 7. The general structure of the fuzzy controller
CHANGE 3)
(17) |
ADD 1)
Reference
Farah, N. S. Y.; Talib, M. H. N.; Ibrahim, Z.; Rasin, Z. and Rizman, Z.I. Experimental Investigation of Different Rules Size of Fuzzy Logic Controller for Vector Control of Induction Motor. Journal of Fundamental and Applied Sciences. 2018, 10, 1696-1717.
Bajpai, D. and Mandal, A. Comparison of Different Rules Based Fuzzy Logic Controller for PMSM Drivers. Journal of Electrical and Electronics Engineering(IOSR-JEEE). 2015, 10, 30-37.
Lazi, J. M.; Ibrahim Z. Sulaiman, M. Patakor, F. A and Isa, S. N. M. Fuzzy Logic Speed Controller with Reduced Rule Base for Dual PMSM Drivers. International Journal of Electrical and Computer Engineering. 2011, 5, 623-628
Are the results clearly presented? ( ) (x) ( ) ( )
Reply
à
We added pictures and supplemented the descriptions to improve the clarity of the results.
CHANGE 1)
Figure 12 shows the configuration of the experimental device to test the performance of the proposed method. Arduino mega2580 is used as the main controller, and DS18B20 temperature sensor is used for temperature of hot side and cold side of thermoelectric device. DHT22 temperature / humidity sensor was used for ambient air temperature and IRF520 Power MOSFET for PWM control of thermoelectric device and FAN. TEC1-12708 is used as the thermoelectric element and NF-S12 FXL is used as the cooling fan. Details of the parts used are shown in Table 2. The sampling period for the experiment is 1 [sec], and the switching frequency for PWM control is 980 [Hz].
The process for performance testing was as follows.
1. Keep ambient temperature and humidity constant for experiment.
2. Operate the experimental device until the temperature of the hold side and cold side of the thermoelectric device is stabilized.
3. When the temperature of the thermoelectric element stabilizes, data is acquired at the set time intervals.
4. Repeat steps 1 to 3 for all control methods.
CHANGE 2)
When the hot side was cooled, the temperature of the hot side remained uniformly up to 28.81 ℃, so that the cold side temperature could be cooled up to 1.56 ℃.
ADD 1)
PWM signal
Figure 15. Temperature control by PI controller
ADD 2)
PWM signal
Figure 17. Temperature control by FPI controller
ADD 3)
PWM signal
Figure 19. Temperature control by VFPI controller
ADD 4)
PWM signal
Figure 20. Temperature control by VFPI-VL controller
ADD 5)
Table 12. Comparison of average voltage, current and consumption power
PI | FPI | VFPI | VFPI-VL | |
PWM average | 242.1 | 224.7 | 224.4 | 207.2 |
Are the conclusions supported by the results? ( ) (x) ( ) ( )
Comments and Suggestions for Authors
This paper proposes a control method to improve energy efficiency and performance of cooling fan used for cooling using fuzzy controllers. The authors claim that this work improves the energy efficiency and performance of the cooling system using a FAN. The reviewer’s comments are listed below:
1. As mentioned in Section 4, PI controller is limited in performance due to fixed gain. Why the fuzzy system is designed to output the input of PI controller, not to the PI gains of PI controller.
Reply
à
In this paper, we added reason and reference to adjust the input value instead of gain value of PI controller.
ADD 1)
In order to solve the disadvantages of PI controller with fixed gain, methods of adjusting gain value have been proposed[42-44]. The PI controller uses the proportional gain and the integral gain. In order to automatically adjust the gain value, the PI controller has to calculate the two values according to the operation state, which increases the calculation time. To solve this problem, a high performance CPU is required. Therefore, in this paper, we propose a method to adjust the input value of PI controller to solve the problem of fixed gain of PI controller. When the input value is adjusted, only one controller is used. Therefore, the calculation time is faster than the method of adjusting the gain value.
ADD 2)
Reference
Chang, S. H. and Chen, P. Y. Self-tuning gains of PI controllers for current control in a PMSM. IEEE Conference on Industrial Electronics and Applications. 2010, 1282 – 1286.
Ximei, Z. and Xianfeng, S. Neural-network-based self-tuning PI controller for Permanent Magnet Synchronous Motor. 2011 International Conference on Electrical Machines and Systems. 2011, 1 – 4.
Mokrani, L. and Abdessemed, R. A fuzzy self-tuning PI controller for speed control of induction motor drive, IEEE Conference on Control Applications. 2003, 2, 785 – 790.
2. Some minor points:
Line 205: What is LM? Should be NM?
Reply
à
LM of the paper is typo. According to the opinion of review, it was corrected to NM as follows.
CHANGE 1)
Fuzzy control is a control method that uses the ambiguity of the boundary. And the fuzzy membership function shows the degree of membership of the input value. In this paper, membership range of membership function is used as follows. The required control amount according to the input value is set as Large, Medium, Small and Zero according to the size, and it is divided into 7 part as positive and negative according to the control direction. The parameter of the membership function is used as a percentage of input value. For fast control speed, the range of NL, NM, PM, PL is set large, ZE is smallest, and NS and PS are medium. For this, We have designed membership function as follow. the range of ZE is -0.2 to 0.2, PS and NS are | 0-0.6 |, NM and PM are | 0.2-1 |, NL and PL are | above 0.6 |. In this paper, 49 rules were used. The number of rules affects computation time and system performance. If the number of rules is large, the calculation time increases but the control performance is the best[45-47]. Therefore, in this paper, we use 49 rules, which shows the best control performance because the change in temperature is not fast.
Line 217: Figure 5? Two “Figure 5”s in this manuscript, the first one is in Line 197.
Reply
à
Figure 5 has been used twice. The numbers in Figure 5 and Figure 6 have been integrated into one.
CHANGE 1)
(a) The error (e) membership function
(b) The changing error (ce) membership function
(c) Output membership function
NL : Negative Large, NM : Negative Medium, NS : Negative Small, ZE : Zero
PL : Positive Large, PM : Positive Medium, PS : Positive Small
Figure 6. The membership function of fuzzy control
Line 253: PI controller 2 outputs … Should be PI controller 1?
Reply
à
The PI controller 2 used in the paper is a typo. The contents of the paper have been revised as follows.
CHANGE 1)
PI controller 1 outputs the output gain (F_Gain) of the fuzzy controller with the error (e) and the error change value (ce) as inputs.
Line 288~289: The hot side temperature was lower…cold side was 3.18 ℃. However, from Figure 12, ΔT cannot be equal to 3.18℃.
Reply
à
There was a problem with the contents of the paper. The contents of the paper were reviewed and revised as follows.
CHANGE 1)
When the hot side was cooled, the temperature of the hot side remained uniformly up to 28.81 ℃, so that the cold side temperature could be cooled up to 1.56 ℃.
Specific Comments:
1. In the design of the PI controller, anti-windup is essential to the integral term because the integral operation will lead to the saturation of the cooling fan, resulting in no control response. Therefore you should give more description about Figure 5. How can you obtain Equation 17 (PI controller) from Figure 5.
Reply
à
Based on the reviewer's comments, we added the description of Figure 5, modified Figure 5 to derive Equation 24, and added Equations (8) to (14).
ADD 1)
In Figure 5, is before the limit, and is the PI controller output after the limit. The error value() between these two values is calculated by Equation (8). If this value() is generated, it can be determined that the saturation of the output value of the PI controller has occurred. Since the saturation phenomenon of the output value of the PI controller is generated by the integral control, the input value of the integral control is reduced by the Equation (9) using the output value error (), as the result the integral output is adjusted. The control amount of PI controller is given by Equation (12) and the change of control amount() can be expressed as Equation (14) by using the current output u(k) and previous output .
ADD 2)
(8) | |
(9) | |
(10) | |
(11) | |
(12) | |
(13) | |
(14) |
CHANGE 1)
(24) |
2. In your Fuzzy system, three variables, error, changing error and output, are transferred through membership function. How do you decide the parameters of membership functions?
Reply
à
Added a paper on how to set the parameters of the fuzzy membership function.
ADD 1)
Fuzzy control is a control method that uses the ambiguity of the boundary. And the fuzzy membership function shows the degree of membership of the input value. In this paper, membership range of membership function is used as follows. The required control amount according to the input value is set as Large, Medium, Small and Zero according to the size, and it is divided into 7 part as positive and negative according to the control direction. The parameter of the membership function is used as a percentage of input value. For fast control speed, the range of NL, NM, PM, PL is set large, ZE is smallest, and NS and PS are medium. For this, We have designed membership function as follow. the range of ZE is -0.2 to 0.2, PS and NS are | 0-0.6 |, NM and PM are | 0.2-1 |, NL and PL are | above 0.6 |.
3. In Table 1 the fuzzy rules seem to give the control value of fan speed, but in FPI, the fuzzy system is to output the input of PI controller. Is it consistent? Have you taken into account the possibility to reduce the fuzzy rules? Please explain more detail about the design of fuzzy rules.
Reply
à
Table 1 is also used in the field of temperature control, inverter switching control, and gain control of PID as well as speed control in fuzzy control. The rule base in Table 1 is based on ZE, and when the input value changes by one step, the control amount also changes by one step, and the control amount is proportional to the change of the input value. In this paper, fuzzy control is used to control the input value of PI controller. The control value of the input value of the PI controller using the fuzzy controller is controlled to be proportional to the magnitude of the error and error. For this, the rule base shown in Table 1 is used. We have added the contents and references to the paper as follows. In addition, in this paper, we use 49 rules, and we have added additional explanation to this paper.
ADD 1)
Rule base in Table 1 is used for temperature control, PWM control of inverter for wind power generation and fuzzy control for controlling gain of PI D controller as well as motor speed control[45-47]. If the level increases, the control variable is also increased by one level. This method can adjust the control amount in proportion to the change of the input value.
ADD 2)
Reference
Basu, Srismrita. Realization of Fuzzy Logic Temperature Controller. International Journal of Emerging Technology and Advanced Engineering. 2012, 2, 151-155.
Rohin, M. and Adel, M. S. A Rule-Based Fuzzy Logic Controller for a PWM Inverter in a Stand Alone Wind Energy Conversion Scheme. IEEE Transactions on INDUSTRY APPLICATIONS. 1996, 32, 57-65.
Shi, D.; Gao, G.; Gao, Z. and Xiao p. Application of Expert Fuzzy PID method for Temperature Control of Heating Furnace. 2012 International Workshop on Information of Electronics Engineering. 2012, 29, 257-261.
ADD 3)
In this paper, 49 rules were used. The number of rules affects computation time and system performance. If the number of rules is large, the calculation time increases but the control performance is the best[48-50]. Therefore, in this paper, we use 49 rules, which shows the best control performance because the change in temperature is not fast.
ADD 4)
Farah, N. S. Y.; Talib, M. H. N.; Ibrahim, Z.; Rasin, Z. and Rizman, Z.I. Experimental Investigation of Different Rules Size of Fuzzy Logic Controller for Vector Control of Induction Motor. Journal of Fundamental and Applied Sciences. 2018, 10, 1696-1717.
Bajpai, D. and Mandal, A. Comparison of Different Rules Based Fuzzy Logic Controller for PMSM Drivers. Journal of Electrical and Electronics Engineering(IOSR-JEEE). 2015, 10, 30-37.
Lazi, J. M.; Ibrahim Z. Sulaiman, M. Patakor, F. A and Isa, S. N. M. Fuzzy Logic Speed Controller with Reduced Rule Base for Dual PMSM Drivers. International Journal of Electrical and Computer Engineering. 2011, 5, 623-628
4. The fan control signal is PWM. It is recommended that you should plot the control result in every experiment for the evaluation of control performance.
Reply
à
We added PWM signal picture and explanation to it.
ADD 1)
Figure 15 (a) shows the temperature change and Figure 15 (b) shows the PWM signal for controlling the FAN according to temperature. In case of Arduino, PWM is used from 0 to 255, PWM 0 means duty ratio 0 [%], and PWM 255 means duty ratio 100 [%][54].
ADD 2)
PWM signal
Figure 15. Temperature control by PI controller
ADD 3)
PWM signal
Figure 17. Temperature control by FPI controller
ADD 4)
PWM signal
Figure 19. Temperature control by VFPI controller
ADD 5)
PWM signal
Figure 20. Temperature control by VFPI-VL controller
5. How do you evaluate the fan power consumption? The results show that the controlled variable ΔT is oscillating in every controller, resulting in fan speed oscillation. Why do you plot the fan power consumption using unit of mW instead of W?
Reply
à
The voltage and current were measured input value of the FAN by using the sensor . The sensor used in this case was the Voltage & Current sensor shown in Table 2.In addition, the ΔT changes temperature depending on the influence of the thermoelectric element, which increases or decreases the speed of the FAN. In addition, the capacity of the FAN used in the paper is about 1.44 [W], and the unit of mW is used to clearly show the power consumption comparison for each control method. The unit of power consumption was modified to W.
CHANGE 1) Table 12. Comparison of average voltage, current and consumption power
PI | FPI | VFPI | VFPI-VL | |
Voltage (V) | 11.6 | 11.63 | 11.63 | 11.63 |
Current (A) | 0.20374 | 0.10268 | 0.09001 | 0.07388 |
Consumption power (W) | 2.3696 (100%) | 1.19668 (50.5%) | 1.0507 (44.3%) | 0.86177 (32.6%) |
PWM average | 242.1 | 224.7 | 224.4 | 207.2 |
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper presents an FPI control approach for the Cooling System Fan. The presented idea is interesting, and the simulations validate the effectiveness of the proposed scheme. However, there are some issues to be improved:
1- For the first time, the authors must define abbreviations such as FPI, VFPI-VL, VFPI and ....
2- As there are a large number of variables, parameters, and sets, the authors should provide nomenclatures in the first section of the paper. It helps the reader to follow the paper conveniently.
3- In the introduction section, the literature review should be expanded. Moreover, the references are old. Please review the below references in the introduction section:
-- Optimal intelligent control for HVAC systems. Journal of Power Technologies. 2012 Oct 1;92(3):192-200.
-- Optimal Type-2 Fuzzy Controller For HVAC Systems. Automatika: Journal for Control, Measurement, Electronics, Computing & Communications. 2014 Jan 1;55(1).
-- Intelligent robust PI adaptive control strategy for speed control of EV (s). IET Science, Measurement & Technology. 2016 Aug 1;10(5):433-41.
-- A new intelligent online fuzzy tuning approach for multi-area load frequency control: Self Adaptive Modified Bat Algorithm. International Journal of Electrical Power & Energy Systems. 2015 Oct 1;71:254-61.
4- Please highlight the contributions of the paper in the last paragraph of the introduction section.
5- Please check the Fig. 7. it has some problems.
6- How the PI Controller 1 in Fig. 10 can update the membership functions of the fuzzy system!!!
7- Section 5 is Experimental results?! or simulation results?!
8- How actually the switching control signal is generated under the approach? Try to describe from the flow chart.
9- The control signals for each simulation are NOT given. This makes it more difficult to understand the benefit of the proposed approach. The author must give the control signals obtained for simulations.
10- What's the computation effort of the proposed method compared to other references?
11- what is the rate of sampling? Whether the inverter is operating with a fixed switching frequency or variable switching frequency under the proposed approach?
12- There is no mention of how the experimental results were carried out, the authors should clearly describe the experimental setup, what hardware was used and so on!!
Author Response
English language and style
( ) Extensive editing of English language and style required
(x) Moderate English changes required
( ) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style
Reply
à
The English part was revised from the reviewer’s comment to the whole part of the paper through the specialized agency.
Yes Can be improved Must be improved Not applicable
Does the introduction provide sufficient background and include all relevant references? ( ) ( ) (x) ( )
Reply
à
We have added the contributions and references of the paper to improve the introduction.
ADD 1)
In this paper, we propose an FPI controller that controls the input value of the PI controller using fuzzy control, a VFPI controller that adjusts the fuzzy control gain of the FPI controller, and a VFPI_VL controller that adjusts the output limit of the VFPI controller. The limitation of the performance improvement by the fixed gain value of the conventional PI controller can be solved by adjusting the input value using the fuzzy control and the control performance can be improved by adjusting the gain value of the fuzzy control and the limit value of the controller output. The control method proposed in this paper is applied to the control system using FAN. The cooling system using the FAN changes the power consumption according to the operating time and speed of the FAN. The method proposed in this paper optimally controls the operation time and speed of the FAN according to the operating conditions, thereby reducing the power consumption of the FAN. The controller proposed in this paper can be applied to various variable speed drive systems, and it is expected to improve speed control performance and reduce power consumption.
ADD 2)
Reference
Khooban, M. H.; Soltanpour, M. R.; Abadi, D. N. M.; Esfahani, Z. Optimal Intelligent Control for HVAC System. Journal of Power Technologies. 2012, 92, 192-200
Khooban, M. H.; Adadi, D. N. M.; Alfi, A.; Siahi, M. Optimal Type-2 Fuzzy Controller For HVAC System. Automatika. 2014, 55, 69-78.
Khooban, M. H.; Niknam, T. A new intelligent online fuzzy tuning approach for multi-area load frefuency control: Self Adatpvie Modified Bat Algorithm. International Journal Electrical Power and Energy Systems. 2015, 71, 254-261.
Khooban. M. H.; Naghash-Almasi, O.; Niknam, T.; Sha-Sadeghi, M. Intelligent robust Pi adaptive control strategy for speed control of EV(s). The Institution of Engineering and Technology(IET). 2016, 10, 433-441.
Is the research design appropriate? ( ) ( ) (x) ( )
Reply
à
In order to improve the contents of the research design, we added the formula and flowchart, and improved the description of the design.
ADD 1)
| (8) |
(9) | |
(10) | |
(11) | |
(12) | |
(13) | |
(14) |
ADD 2)
Figure 11 shows the flow chart for the PWM control that is the speed control signal of the FAN. Set the reference temperature (Set_Tem) and calculate the temperature difference ( by measuring the temperature of the hot side () and cold side () of the thermoelectric element. This temperature difference is used to maintain the PWM through the Main controller or to calculate a new PWM signal. The control ends when the set time is over or when the user enters the stop.
Figure 11. Flow chart for control
Are the methods adequately described? ( ) ( ) (x) ( )
Reply
à
We have revised the content and added references to improve the adequateness of the method.
CHANGE 1)
Fuzzy control is a control method that uses the ambiguity of the boundary. And the fuzzy membership function shows the degree of membership of the input value. In this paper, membership range of membership function is used as follows. The required control amount according to the input value is set as Large, Medium, Small and Zero according to the size, and it is divided into 7 part as positive and negative according to the control direction. The parameter of the membership function is used as a percentage of input value. For fast control speed, the range of NL, NM, PM, PL is set large, ZE is smallest, and NS and PS are medium. For this, We have designed membership function as follow. the range of ZE is -0.2 to 0.2, PS and NS are | 0-0.6 |, NM and PM are | 0.2-1 |, NL and PL are | above 0.6 |. In this paper, 49 rules were used. The number of rules affects computation time and system performance. If the number of rules is large, the calculation time increases but the control performance is the best[45-47]. Therefore, in this paper, we use 49 rules, which shows the best control performance because the change in temperature is not fast.
CHANGE 2)
Figure 7. The general structure of the fuzzy controller
CHANGE 3)
(17) |
ADD 1)
Reference
Farah, N. S. Y.; Talib, M. H. N.; Ibrahim, Z.; Rasin, Z. and Rizman, Z.I. Experimental Investigation of Different Rules Size of Fuzzy Logic Controller for Vector Control of Induction Motor. Journal of Fundamental and Applied Sciences. 2018, 10, 1696-1717.
Bajpai, D. and Mandal, A. Comparison of Different Rules Based Fuzzy Logic Controller for PMSM Drivers. Journal of Electrical and Electronics Engineering(IOSR-JEEE). 2015, 10, 30-37.
Lazi, J. M.; Ibrahim Z. Sulaiman, M. Patakor, F. A and Isa, S. N. M. Fuzzy Logic Speed Controller with Reduced Rule Base for Dual PMSM Drivers. International Journal of Electrical and Computer Engineering. 2011, 5, 623-628
Are the results clearly presented? ( ) ( ) (x) ( )
Reply
à
We added pictures and supplemented the descriptions to improve the clarity of the results.
CHANGE 1)
Figure 12 shows the configuration of the experimental device to test the performance of the proposed method. Arduino mega2580 is used as the main controller, and DS18B20 temperature sensor is used for temperature of hot side and cold side of thermoelectric device. DHT22 temperature / humidity sensor was used for ambient air temperature and IRF520 Power MOSFET for PWM control of thermoelectric device and FAN. TEC1-12708 is used as the thermoelectric element and NF-S12 FXL is used as the cooling fan. Details of the parts used are shown in Table 2. The sampling period for the experiment is 1 [sec], and the switching frequency for PWM control is 980 [Hz].
The process for performance testing was as follows.
1. Keep ambient temperature and humidity constant for experiment.
2. Operate the experimental device until the temperature of the hold side and cold side of the thermoelectric device is stabilized.
3. When the temperature of the thermoelectric element stabilizes, data is acquired at the set time intervals.
4. Repeat steps 1 to 3 for all control methods.
CHANGE 2)
When the hot side was cooled, the temperature of the hot side remained uniformly up to 28.81 ℃, so that the cold side temperature could be cooled up to 1.56 ℃.
ADD 1)
PWM signal
Figure 15. Temperature control by PI controller
ADD 2)
PWM signal
Figure 17. Temperature control by FPI controller
ADD 3)
PWM signal
Figure 19. Temperature control by VFPI controller
ADD 4)
PWM signal
Figure 20. Temperature control by VFPI-VL controller
ADD 5)
Table 12. Comparison of average voltage, current and consumption power
PI | FPI | VFPI | VFPI-VL | |
PWM average | 242.1 | 224.7 | 224.4 | 207.2 |
Are the conclusions supported by the results? ( ) (x) ( ) ( )
Comments and Suggestions for Authors
This paper presents an FPI control approach for the Cooling System Fan. The presented idea is interesting, and the simulations validate the effectiveness of the proposed scheme. However, there are some issues to be improved:
1- For the first time, the authors must define abbreviations such as FPI, VFPI-VL, VFPI and ....
Reply
à
We have added the abbreviation definition to the paper to reflect the comments of the reviewer.
ADD 1)
INDEX | |||
COP( | Coefficient of performance about thermoelectric device | Amount of heat absorbing [W] | |
P | Input Power [W] | Seebeck coefficient [V/K] | |
Seebeck coefficient of P type [V/K] | Seebeck coefficient of N type [V/K] | ||
Hot side temperature of thermoelectric [℃] | Coefficient of thermal coefficient [W/m∙K] | ||
Heat sink temperature [℃] | Cold side temperature of thermoelectric [℃] | ||
Object temperature [℃] | Ambient air temperature [℃] | ||
Integral gain of PI controller | Proportional gain of PI controller | ||
GC | Error change gain of Fuzzy control | GE | Error gain of Fuzzy control |
NL | Negative Large | NM | Negative Medium |
NS | Negative Small | ZE | Zero |
PL | Positive Large | PM | Positive Medium |
PS | Positive Small | Reference temperature [℃] | |
Compensation Temperature by Fuzzy control [℃] | Compensation reference temperature [℃] | ||
VFPI | Variable gain Fuzzy Proportional Integral controller | FPI | Fuzzy Proportional Integral Controller |
VFPI-VL | Variable gain Fuzzy Proportional Integral with Variable Limit controller | ||
2- As there are a large number of variables, parameters, and sets, the authors should provide nomenclatures in the first section of the paper. It helps the reader to follow the paper conveniently.
Reply
à
A description of the variables and parameters has been added to the paper as follows:
ADD 1)
INDEX | |||
COP( | Coefficient of performance about thermoelectric device | Amount of heat absorbing [W] | |
P | Input Power [W] | Seebeck coefficient [V/K] | |
Seebeck coefficient of P type [V/K] | Seebeck coefficient of N type [V/K] | ||
Hot side temperature of thermoelectric [℃] | Coefficient of thermal coefficient [W/m∙K] | ||
Heat sink temperature [℃] | Cold side temperature of thermoelectric [℃] | ||
Object temperature [℃] | Ambient air temperature [℃] | ||
Integral gain of PI controller | Proportional gain of PI controller | ||
GC | Error change gain of Fuzzy control | GE | Error gain of Fuzzy control |
NL | Negative Large | NM | Negative Medium |
NS | Negative Small | ZE | Zero |
PL | Positive Large | PM | Positive Medium |
PS | Positive Small | Reference temperature [℃] | |
Compensation Temperature by Fuzzy control [℃] | Compensation reference temperature [℃] | ||
VFPI | Variable gain Fuzzy Proportional Integral controller | FPI | Fuzzy Proportional Integral Controller |
VFPI-VL | Variable gain Fuzzy Proportional Integral with Variable Limit controller | ||
3- In the introduction section, the literature review should be expanded. Moreover, the references are old. Please review the below references in the introduction section:
-- Optimal intelligent control for HVAC systems. Journal of Power Technologies. 2012 Oct 1;92(3):192-200.
-- Optimal Type-2 Fuzzy Controller For HVAC Systems. Automatika: Journal for Control, Measurement, Electronics, Computing & Communications. 2014 Jan 1;55(1).
-- Intelligent robust PI adaptive control strategy for speed control of EV (s). IET Science, Measurement & Technology. 2016 Aug 1;10(5):433-41.
-- A new intelligent online fuzzy tuning approach for multi-area load frequency control: Self Adaptive Modified Bat Algorithm. International Journal of Electrical Power & Energy Systems. 2015 Oct 1;71:254-61.
Reply
à
We have added references to the introduction.
ADD 1)
Reference
Khooban, M. H.; Soltanpour, M. R.; Abadi, D. N. M.; Esfahani, Z. Optimal Intelligent Control for HVAC System. Journal of Power Technologies. 2012, 92, 192-200
Khooban, M. H.; Adadi, D. N. M.; Alfi, A.; Siahi, M. Optimal Type-2 Fuzzy Controller For HVAC System. Automatika. 2014, 55, 69-78.
Khooban, M. H.; Niknam, T. A new intelligent online fuzzy tuning approach for multi-area load frefuency control: Self Adatpvie Modified Bat Algorithm. International Journal Electrical Power and Energy Systems. 2015, 71, 254-261.
Khooban. M. H.; Naghash-Almasi, O.; Niknam, T.; Sha-Sadeghi, M. Intelligent robust Pi adaptive control strategy for speed control of EV(s). The Institution of Engineering and Technology(IET). 2016, 10, 433-441.
4- Please highlight the contributions of the paper in the last paragraph of the introduction section.
Reply
à
We added the contribution of the paper to the introduction.
ADD 1)
In this paper, we propose an FPI controller that controls the input value of the PI controller using fuzzy control, a VFPI controller that adjusts the fuzzy control gain of the FPI controller, and a VFPI_VL controller that adjusts the output limit of the VFPI controller. The limitation of the performance improvement by the fixed gain value of the conventional PI controller can be solved by adjusting the input value using the fuzzy control and the control performance can be improved by adjusting the gain value of the fuzzy control and the limit value of the controller output. The control method proposed in this paper is applied to the control system using FAN. The cooling system using the FAN changes the power consumption according to the operating time and speed of the FAN. The method proposed in this paper optimally controls the operation time and speed of the FAN according to the operating conditions, thereby reducing the power consumption of the FAN. The controller proposed in this paper can be applied to various variable speed drive systems, and it is expected to improve speed control performance and reduce power consumption.
5- Please check the Fig. 7. it has some problems.
Reply
à
Figure 7 has been modified as follows.
CHANGE 1)
Figure 7. The general structure of the fuzzy controlle
6- How the PI Controller 1 in Fig. 10 can update the membership functions of the fuzzy system!!!
Reply
PI Controller 1 in Figure 10 is not used to update the membership function of the fuzzy control. PI Controller 1 in Figure 10 performs the same operation as PI Controller 1 in Figure 9 and is used to adjust the output gain (GU) value of the fuzzy controller. The adjustment of the GU value is done by Eq. (18).
Figure 9. Structure of VFPI controller
PI controller 1 outputs the output gain (F_Gain) of the fuzzy controller with the error (e) and the error change value (ce) as inputs.
Section 5 shows the results for the experiment. We have added the experimental equipment configuration and experiment method as follows.
CHANGE 1)
Figure 12 shows the configuration of the experimental device to test the performance of the proposed method. Arduino mega2580 is used as the main controller, and DS18B20 temperature sensor is used for temperature of hot side and cold side of thermoelectric device. DHT22 temperature / humidity sensor was used for ambient air temperature and IRF520 Power MOSFET for PWM control of thermoelectric device and FAN. TEC1-12708 is used as the thermoelectric element and NF-S12 FXL is used as the cooling fan. Details of the parts used are shown in Table 2. The sampling period for the experiment is 1 [sec], and the switching frequency for PWM control is 980 [Hz].
The process for performance testing was as follows.
1. Keep ambient temperature and humidity constant for experiment.
2. Operate the experimental device until the temperature of the hold side and cold side of the thermoelectric device is stabilized.
3. When the temperature of the thermoelectric element stabilizes, data is acquired at the set time intervals.
4. Repeat steps 1 to 3 for all control methods.
8- How actually the switching control signal is generated under the approach? Try to describe from the flow chart.
Reply
->
We have added a flow chart and a description of it as follows.
ADD 1)
Figure 11 shows the flow chart for the PWM control that is the speed control signal of the FAN. Set the reference temperature (Set_Tem) and calculate the temperature difference ( by measuring the temperature of the hot side () and cold side () of the thermoelectric element. This temperature difference is used to maintain the PWM through the Main controller or to calculate a new PWM signal. The control ends when the set time is over or when the user enters the stop.
ADD 2)
Figure 11. Flow chart for control
9- The control signals for each simulation are NOT given. This makes it more difficult to understand the benefit of the proposed approach. The author must give the control signals obtained for simulations.
Reply
à
FAN control signal, PWM, has been added to the experimental results.
ADD 1)
Figure 15 (a) shows the temperature change and Figure 15 (b) shows the PWM signal for controlling the FAN according to temperature. In case of Arduino, PWM is used from 0 to 255, PWM 0 means duty ratio 0 [%], and PWM 255 means duty ratio 100 [%][54].
ADD 2)
PWM signal
Figure 15. Temperature control by PI controller
ADD 3)
PWM signal
Figure 17. Temperature control by FPI controller
ADD 4)
PWM signal
Figure 19. Temperature control by VFPI controller
ADD 5)
PWM signal
Figure 20. Temperature control by VFPI-VL controller
10- What's the computation effort of the proposed method compared to other references?
Reply
à
We added a comparison of the proposed methods to the conclusion of the paper.
ADD 1)
In this paper, we propose a FPI method to control the input value of the PI controller, the VFPI controller to control the fuzzy controller output gain, and the VFPI-VL controller to adjust the output limit value of the fuzzy control. Each method enables faster control by amplifying the output according to the input value in the transient state. In particular, when the overshoot occurs, the control direction is reversed so that it can be stabilized more quickly. As a result, the rise time of the transient state was accelerated, and the stabilization time decreased in the steady state.
11- what is the rate of sampling? Whether the inverter is operating with a fixed switching frequency or variable switching frequency under the proposed approach?
Reply
à
We added the sampling period used in the experiment and the switching frequency for PWM control to the paper.
ADD 1)
Figure 12 shows the configuration of the experimental device to test the performance of the proposed method. Arduino mega2580 is used as the main controller, and DS18B20 temperature sensor is used for temperature of hot side and cold side of thermoelectric device. DHT22 temperature / humidity sensor was used for ambient air temperature and IRF520 Power MOSFET for PWM control of thermoelectric device and FAN. TEC1-12708 is used as the thermoelectric element and NF-S12 FXL is used as the cooling fan. Details of the parts used are shown in Table 2. The sampling period for the experiment is 1 [sec], and the switching frequency for PWM control is 980 [Hz].
12- There is no mention of how the experimental results were carried out, the authors should clearly describe the experimental setup, what hardware was used and so on!!
Reply
à
Section 5 describes the parts, configuration and experimental procedures used in the experiment. In addition, we supplemented the contents by adding the order of control.
ADD 1)
Figure 12 shows the configuration of the experimental device to test the performance of the proposed method. Arduino mega2580 is used as the main controller, and DS18B20 temperature sensor is used for temperature of hot side and cold side of thermoelectric device. DHT22 temperature / humidity sensor was used for ambient air temperature and IRF520 Power MOSFET for PWM control of thermoelectric device and FAN. TEC1-12708 is used as the thermoelectric element and NF-S12 FXL is used as the cooling fan. Details of the parts used are shown in Table 2. The sampling period for the experiment is 1 [sec], and the switching frequency for PWM control is 980 [Hz].
The process for performance testing was as follows.
1. Keep ambient temperature and humidity constant for experiment.
2. Operate the experimental device until the temperature of the hold side and cold side of the thermoelectric device is stabilized.
3. When the temperature of the thermoelectric element stabilizes, data is acquired at the set time intervals.
4. Repeat steps 1 to 3 for all control methods.
ADD 2)
Figure 11 shows the flow chart for the PWM control that is the speed control signal of the FAN. Set the reference temperature (Set_Tem) and calculate the temperature difference ( by measuring the temperature of the hot side () and cold side () of the thermoelectric element. This temperature difference is used to maintain the PWM through the Main controller or to calculate a new PWM signal. The control ends when the set time is over or when the user enters the stop.
Figure 11. Flow chart for control
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have answered all the questions raised by the reviewer with great effort. The reviewer has no other question and recommends that the paper may be published in my perspective.
Reviewer 2 Report
The paper can be accepted in this form.
