Realistic Home Energy Management System Considering the Life Cycle of Photovoltaic and Energy Storage Systems

Round 1

Reviewer 1 Report

This manuscript describes a computer simulation of a home energy management system which seeks to optimize the use of the public electricity grid, home PV systems, home storage batteries and home electrical appliances. It also simulates the generation of CO2 by fossil fuel power plants. The simulation was done using Matlab and a time-stepping algorithm with a one-minute time step.

It presents the equations used in the simulation, but gives little justification for them. Home appliance power usage and operating times were taken from a single home in Vigo, Spain. No consideration is given of how these parameters could vary from home to home. No consideration is given to large scale renewable energy power plants, except to say their operating limitations make home energy management systems necessary.

The diagrams are illegible and add nothing to the paper. The results are of little value as they are based on dubious cost benefit equations and unrepresentative data

English adequate

Author Response

Response to the comments of reviewer#1

 

We express our sincere gratitude and appreciation for the valuable feedback provided by the reviewer. With great consideration and attention, we have thoroughly revised our manuscript entitled "Realistic Home Energy Management System Considering the Life Cycle of Photovoltaic and Energy Storage Systems " according to your comments. We have prepared a comprehensive list of itemized responses addressing each comment.

 

Point 1: "The paper presents the equations used in the simulation, but gives little justification for them?".

 

Response 1: we acknowledge the reviewer's concern regarding mathematical model equations that would enhance the structure of the paper. To address this, we have revised many sections as follows:

Lines 154 to 180: we clarified the energy demand model in additional detail. Home appliances have classified into two categories (fixed & controllable). Changing the operation time of the fixed appliances is not possible. Where the controllable appliances can be shifted to the low energy-price periods. The energy consumption of the fixed and controllable appliance is calculated each minute.

 Lines 182 to 226: we expanded and rearranged the battery energy usage price section to be clearer to the readers. The proposed model of the energy usage cost was developed in three steps: firstly, we calculated the cost of storing energy in the battery during its lifespan (€/kWh). Secondly, we calculated the cost of charged battery energy from the PV/grid and stored this amount of charged energy in the battery. Thirdly, the battery energy price model is updated at each charging process. The proposed model takes into account the cost of storing energy, the cost of purchased energy from the grid/PV, the battery lifecycle, and the battery installation cost.

Lines 235 to 252: we explained the model of selling energy to the grid better. We attempted to manage the process of selling energy to the grid efficiently by considering the price of grid-selling and the cost of PV and battery energy usage. This model was introduced into the optimization problem.

Lines 253 to 278: we explained the proposed flow energy management algorithm in more detail and added uncompleted steps to the figure. This model takes into account the energy usage price from the grid, battery, and PV  at each minute to operate the PV and battery efficiently regarding the PV energy generation cost and battery energy usage cost.

Lines 279 to 294: the objective function model and user preferences constraints have been clarified better. In this work, the objective function is to minimize the energy usage cost from the grid, battery, and PV. This help in reducing real cost, saving energy, operating the battery efficiently, and reducing wasteful energy sold to the grid.

We cited the equations in (sections 3.1.1,3.1.2), and we developed the rest of equations and derived the objective function.

All equations throughout the manuscript were reviewed and all parameters and units were defined.

Point 2: "Home appliance power usage and operating times were taken from a single home in Vigo, Spain. No consideration is given of how these parameters could vary from home to home. No consideration is given to large scale renewable energy power plants, except to say their operating limitations make home energy management systems necessary".

 

Response 2: we appreciate the reviewer's comment regarding the mentioned limitations. We explained the scope of our study and the future work that covers these limitations as follows:

We carried out a real case study in order to show the real economic benefits of our proposed system to the Spanish household energy consumers, especially after the new electricity tariff. This would encourage them to install a PV-Battery system and manage their energy consumption. Moreover, the proposed system was not modeled based on the collected data, and it is able to deal with any other data and user preferences. The main objective of this work is to show the actual cost reduction regarding PV-battery energy generation costs.

 

As we focus on the home energy management system, large-scale power plants are out of our scope. The local PV system size is constrained by the usual home energy consumption; there is an upper limit for the capacity of the installed PV system. The PV size is 1.035 KWp to evaluate our proposed system.

 

Lines 542 to 547: we have mentioned some future work that covers the limitation of this study such as taking multiple houses in the study for applying peer-to-peer technology to exchange energy between them.

 

Point 3:  "The diagrams are illegible and add nothing to the paper".

 

Response 3: we appreciate the reviewer's attention to detail regarding the quality of the diagrams and agree that it is an important aspect to consider. To address this, we presented the results and the charts/graphs in a more visually understandable format (utilizing Origin Pro software and MATLAB with better resolution). Figure (4-15)

 

Point 4: "The results are of little value as they are based on dubious cost benefit equations and unrepresentative data"

 

Response 4: we understand your perspective regarding the cost benefits equations and data. Allow us to provide further details about our work.

Many studies adopted reducing the electricity cost based on the cost of energy imported from the grid. No consideration for the cost of using PV and battery energy, though they have investment costs and lifecycles. To address this:

Lines 181 to 252: we modeled an energy usage price for the PV-Battery system that expresses the cost of using the battery during its lifespan and the cost of purchased energy stored in the battery. The cost of using PV power in the Spanish residential sector is 0.092 euro/KWh, according to the International Renewable Energy Agency. For selling energy to the grid, the price of selling energy to the grid is usually low. For that, we modeled the real profit of selling energy to the grid to manage this process well and reduce wasting energy. All that are mentioned were introduced in the optimization problem to schedule the home appliances.

 

Lines: 378 to 527:the results show the effectiveness of the proposed system in terms of reducing real cost, maintaining the battery lifespan, reducing the energy losses, reducing the cost of energy losses, reducing the carbon emissions, and reaching the breakeven point of the system.

 

We cited the equations in (sections 3.1.1,3.1.2), and we developed the rest of equations and derived the objective function.

In the development section, all equations (parameters & units) were defined clearly.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper presents a well-designed and well-executed study that addresses an important issue related to home energy management systems. The authors' contribution of designing a novel energy flow management algorithm that considers the life cycle of the battery and PV and pricing models for energy usage of PV and battery is commendable. The real case study data collected from a home in Vigo, Spain, and various scenarios conducted in the study provide credible and convincing results. The authors have stated their methodology, algorithm, and the applied AGWO algorithm for solving the optimization problem.

General Suggestions

Abstract needs revision to provide more details regarding the study's objectives, methodology, and results. The authors should provide more details regarding the selection of the AGWO algorithm, including a comparison of its performance with other optimization algorithms. Additionally, the authors should discuss the limitations of the study, such as the generalizability of the results to other regions and households, and provide more details regarding the cost of implementing the proposed system, including the cost of the required hardware and software. To enhance the study, the authors may also consider providing more details on the user preferences that were taken into account while designing the scheduling algorithm, as this would provide additional insights into the study's methodology.

Some additional comments about the paper format and presentation:

The quality of Figures 1 and 5 should be increased.

Throughout the paper and especially in the equations starting from Eq. 13, there are some undescribed parameters. If authors add what they mean and how they are determined, readers will understand the methodology more easily.

It seems that some steps are missing/uncompleted in the algorithm given in Figure 2. For example, after the "check whether if charging grid is economic or not" step, the algorithm should make a decision and direct the operation to a certain step accordingly. However, this cannot be visually understood in the flowchart.

The objective function plays a crucial role in the optimization algorithm as it defines the ultimate aim that the algorithm is trying to achieve. Therefore, choosing the right objective function is critical, as it guides the direction and focus of the optimization process. For example, in a scheduling problem, the objective function may aim to minimize the total time taken to complete all the tasks, while in a resource allocation problem, the objective function may aim to maximize the utilization of resources. It would be helpful to provide more details on how Equation (23) was derived and its representation in the context of the study. It can be discussed in detail how the various variables and constraints relate to the objective function and whether there are any competing objectives or trade-offs considered in the optimization process. Additionally, please clarify how the objective function is related to the overall goals and objectives of the study.

The English is almost good and clear, however, there are some minor grammatical errors throughout the paper. Please check the paper grammatically one more time.

Author Response

Response to the comments of Reviewer#2

 

We express our sincere gratitude and appreciation for the valuable feedback provided by the reviewer. With great consideration and attention, we have thoroughly revised our manuscript entitled "Realistic Home Energy Management System Considering the Life Cycle of Photovoltaic and Energy Storage Systems " according to your comments. We have prepared a comprehensive list of itemized responses addressing each comment.

 

Point 1: Abstract needs revision to provide more details regarding the study's objectives, methodology, and results

 

Response 1: we acknowledge the reviewer's concern regarding the clarity of the abstract. To address this:

Lines 12 to 26: we rearranged the abstract and clarified the objectives, methodology, and results in more deitals.

1. Objectives: reducing the real electricity cost while maintaining a longer battery lifespan, reducing battery charging/discharging losses, and using PV power efficiently.
2. Methodology: we developed a pricing model of battery energy usage, in addition to modeling the PV energy usage cost based on the Levelized Cost of Energy (LCOE) for PV systems. The PV-battery energy usage cost models were introduced into the optimization problem solved using the Augmented Grey Wolf Optimization (AGWO) and Particle Swarm Optimization (PSO) algorithms in MATLAB. Besides, we developed an efficient energy flow management algorithm. We collected real data from a home in Vigo, Spain, and simulated four scenarios.
3. The results: we described the effictivenss of our work in reducing the cost , extending the battery lifespan, reducing energy losses using the AGWO and PSO algorithms.

 

 

Point 2: The authors should provide more details regarding the selection of the AGWO algorithm, including a comparison of its performance with other optimization algorithms

 

Response 2: we appreciate your observation regarding implementing another optimization algorithm and providing more details of the algorithms to enhance the manuscript. To do so:

Lines 310 to 340: we added a brief background about the AGWO and PSO algorithms with references and obtained new results using the PSO algorithm. The results and discussion sections show the results of AGOW and PSO.

 

Related to the selection of the algorithms, we implemented optimization algorithms to evaluate the proposed system with another energy usage pricing model. The main contributions are modeling PV and battery energy usage prices and introducing them in the scheduling optimization process. This help to operate these systems efficiently regarding their installation cost and lifecycle. We implemented AGWO and PSO to evaluate the effectiveness of the proposed system in terms of real cost reduction, energy saving, and battery lifespan. The results proved that implementing AGWO and PSO in the proposed system achieves better results compared to an existing energy usage pricing model.

 

Point 3: The authors should discuss the limitations of the study, such as the generalizability of the results to other regions and households, and provide more details regarding the cost of implementing the proposed system, including the cost of the required hardware and software

 

Response 3: - we appreciate the reviewer's comment regarding the mentioned limitations. We explained the scope of our study and the future work that covers these limitations as follows:

Related to the generalizability of the results to other regions and households, we carried out a real case study in order to show the real economic benefits of the proposed system to the Spanish household energy consumers, especially after the new electricity tariff in Spain. This would encourage them to install a PV-Battery system and manage their energy consumption. The proposed system was not modeled based on the collected data, and it is able to deal with any other data and user preferences.

Lines 542 to 547: we have mentioned some future work that covers the limitation of this study such as taking multiple houses for applying peer-to-peer technology to exchange energy between them and consider each user’s preferences.

 

Lines 360 to 377 and Lines 507 to 527: we modeled and discussed the expected breakeven point of the installed system based on its investment cost (PV, battery, inverter, converter and wires, and other components) and the system performance. However, the cost of a home automation controller is out of our scope. We focus on maintaining the system lifespan and reducing the electricity cost according to the energy usage cost from the grid, battery, and PV.

 

Point 4: The authors may also consider providing more details on the user preferences that were taken into account while designing the scheduling algorithm, as this would provide additional insights into the study's methodology

 

Response 4: thank you for bringing up the concern regarding enhancing the manuscript contents. Related to the user preferences:

Lines 289 to 294: we carried out a real case study, the user classifed the home appliances to fixed and controllable appliances, low and high priority appliances, and minimum/maximum starting operation time boundaries of the controllable appliances. The user preferences are provided in more detail.

 

 

Point 5: The quality of Figures 1 and 5 should be increased

 

Response 5: we appreciate the reviewer's attention to detail regarding the quality of the diagrams and agree that it is an important aspect to consider.

We modified Figure 1. Related to Figure 5 and other results figures, we substituted them with high-quality figures and presented the result in a more clear manner (utilizing Origin Pro software and MATLAB with better resolution).

 

Point 6: Throughout the paper and especially in the equations starting from Eq. 13, there are some undescribed parameters

 

Response 6: we acknowledge the reviewer's concern regarding mathematical model parameters. To address this:

All equations throughout the manuscript were reviewed and all parameters and units were defined. We cited the equations in (sections 3.1.1,3.1.2), and we developed the rest of equations and derived the objective function.

Related to Equation (13), the Energy Usage Pricing Model section has expanded and Equation (13) became  Equation (19).

                                                                                   (19)

 : the price of battery energy (€/kWh) for the next time slot (t+1) after adding the new charged energy price . (Lines : 222 to 223)

: The total price of power purchased and stored in the battery (€/kWh) (Line: 197 to 198)

 : the price of the battery energy usage in the time slot t (€/kWh). ( Lines: 223 to 224)

 : the battery net power at the time slot t (kW). ( Lines: 224 to 225)

, : are the amount of power sent from the battery to the home and grid at each time slot t (kW), respectively. ( Lines: 225 to 226)

,  are the amount of PV, and grid stored power in the battery at each time slot t (kW), respectively. ( Lines: 206 to 207)

Point 7: It seems that some steps are missing/uncompleted in the algorithm given in Figure 2

 

Response 7: we appreciate your insightful comment regarding the uncompleted steps in Figure 2 and we agree that it is important to consider.

Lines 253 to 278: we added the uncompleted steps that you mentioned in Figure 2. Moreover, the steps of the algorithm have been explained in more detail regarding processes of managing the flow of the PV excess power, meeting the load in the case of PV shortage, and charging/ discharging the battery to be more clear to the readers.

 

Point 8: It would be helpful to provide more details on how Equation (23) was derived and its representation in the context of the study. It can be discussed in detail how the various variables and constraints relate to the objective function and whether there are any competing objectives or trade-offs considered in the optimization process. Additionally, please clarify how the objective function is related to the overall goals and objectives of the study

 

Response 8: we acknowledge your point about expanding the objective function section. To address this:

Lines 279 to 295: we provided a good explanation of the objective function and its relation to the objective of the study. The main aim of this work is to schedule home appliances and manage the energy flow considering the installation cost and life cycle of the PV and battery. This work modeled the usage cost of using PV and battery and considered them in the objective function. Moreover, the profit from selling energy to the grid has been considered in the objective function.

In addition to more details about the user preferences and constraints of the appliances' operation time have been provided. The objective function is subjected to the constraints of user preferences based on the minimum/maximum start time of each controllable appliance as shown in Table 1. The priority of shifting the appliances have classified into low and high. The low-priority appliances have the possibility to change the starting operation time within a period that is longer than the high-priority appliances' period. The user selected the controllable appliances, their minimum/maximum start time, and the priority of each appliance

 

All equations throughout the manuscript were reviewed and all parameters and units were defined. We cited the equations in (sections 3.1.1,3.1.2), and we developed the rest of equations and derived the objective function.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors have designed and proposed an electricity management system for a house. Overall the manuscript broadly covers the current trend of installing the PVs and batteries. However, several confusions need to be addressed before the possible publication of the manuscript.

1. The authors have not mentioned the effect of harsh conditions (high and low temperatures) in the current model

2. Please elaborate whether the current model is focusing on summer or winter condition.

3. What are the battery sizing and power rating of the installed setup? How the authors conclude that the battery life is increased by 4.5 years. In general, there are several degradations inside the battery, which affect the battery. these degradations are not considered in the current model.

4. Please elaborate on how long it will take to breakeven point if someone installs the current setup.

5. The figures are too difficult to read. Please increase the font size and put high quality figures instead of taking the snapshots. The authors may use the OriginPro software instead of plottting in Excel.

Moderate English editing is required.

Author Response

Response to the comments of Reviewer#3

 

We express our sincere gratitude and appreciation for the valuable feedback provided by the reviewer. With great consideration and attention, we have thoroughly revised our manuscript entitled "Realistic Home Energy Management System Considering the Life Cycle of Photovoltaic and Energy Storage Systems " according to your comments. We have prepared a comprehensive list of itemized responses addressing each comment.

 

Point 1: The authors have not mentioned the effect of harsh conditions (high and low temperatures) in the current model.

 

Response 1: we appreciate the reviewer's comment regarding the mentioned limitations. We explained the scope of our study and the future work that covers these limitations as follows:

We focuses on managing energy consumption based on the power rating of the appliances and the user preferences, considering the installation cost and lifespan of the integrated systems. Moreover, we scheduled the operation time of each appliance based on the grid, battery, and PV prices. Thus, the temperate does not take a part in our study scope. For the PV model, there are many PV mathematical models, some of those models consider the temperature, and some consider other parameters. The chosen model in this work is based on the area and efficiency of the solar panel, solar radiation, and inverter efficiency.

Lines 542 to 547: We have mentioned some future work that covers the limitation of this study such as integrating thermostatically controlled loads into the energy demand model.

 

Point 2: Please elaborate whether the current model is focusing on summer or winter condition

 

Response 2: we appreciate the reviewer's attention to detail regarding the case study data. In response to your suggestion:

Lines 301 to 302: We described the months of the data. The collected data of the solar radiation and grid prices are for an average of two months (June, and July).

 

Point 3: What are the battery sizing and power rating of the installed setup? How the authors conclude that the battery life is increased by 4.5 years. In general, there are several degradations inside the battery, which affect the battery. these degradations are not considered in the current model.

 

Response 3: we appreciate your observation regarding the battery system. We clarified the mentioned queries as follows:

In the lines (303-305), we explained the battery system size ( 2.4 KWh, 80% depth of discharge, 6000 cycles), and we defined the PV system size (1.035 kWp) in the line 299.

Lines 342 to 351: we explained the expected battery lifespan calculation in more detail. As known, each battery lifecycle has a number of cycles during its lifespan. The battery lifespan is mainly based on the frequency-time of charging/discharging, and the amount of charging/discharging energy. The manufacturers normally indicate the relation between the depth of charge and the number of cycles. In this work, the depth of charge is 80% and the number of cycles is 6000. So, we calculate the number of cycles during the day. Then, we divide the number of daily cycles by the( 6000* number of days in a year) to expected the battery lifespan.

 

 

 

Related to the battery degradation, it can be caused by a number of factors, such as depth of discharge, charging rate, overcharging, frequency of use, and temperature. In this work, we have considered the depth of discharge (line 147), charging rate (line 148), overcharging (line 149). Additionally, we considered the frequesncy of use as follow:

Lines 182 to 226: we proposed a battery energy usage price and introduced it into the optimization problem, aiming to use the battery efficiently at the right time with a suitable amount of charging/ discharging energy to sustain its lifespan.

 

Lines 253 to 278: we designed an algorithm to ensure efficient use of the battery (time of use, amount of charging/discharging) to sustain its lifespan.

 

Point 4: Please elaborate on how long it will take to breakeven point if someone installs the current setup.

 

Response 4: we appreciate your suggestion. We calculated the expected breakeven point of the system.

Lines 360 to 377: we proposed a mathematical model that expected the breakeven point of the system.

Lines 507 to 527: we showed the simulation results of the expected breakeven point for each scenario using AGWO and PSO.

We used PV and battery systems. Firstly, we calculate the cost reduction of using a PV-Battery in the proposed scheduling strategy. However, the battery lifespan usually is less than the PV lifespan. For that, after the battery stops working, the cost reduction will result from the PV power generated only, it is equal to the amount of PV power supplied to the home multiplied by the grid price at each time step. In addition to the profit of selling the excess PV power generation to the grid. Secondly, we subtract the cost saved during the battery lifespan from the investment cost and then calculate how many times we will need to reach the breakeven point after the battery stops working by using PV only. Finally, we add this time to the battery lifespan which gives the breakeven point.

 

 

Point 5: The figures are too difficult to read.

 

Response 5: we appreciate the reviewer's attention to detail regarding the quality of the diagrams and agree that it is an important aspect to consider. To address this, we presented the results and the charts/graphs in a more visually understandable format (utilizing Origin Pro software and MATLAB with better resolution). Figure (4-15)

Author Response File: Author Response.pdf

Reviewer 4 Report

This paper discusses about an energy management approach considering photovoltaic and energy storage systems. Although the topic seems interesting, there are several comments and concerns as follows that should be addressed before publication:

 

1. The contribution of the paper is not clear. Please discuss more about the aim and benefits of this research. What are its contributions? What are its differences with other proposed techniques in the literature?

2. This paper used AGWO algorithm to solve the optimization problem but there is no information about this algorithm or any appropriate references for future readers.

3. It is important that the authors compare their obtained results using AGWO algorithm with the results achieved by some other heuristic algorithms (i.e., GA, PSO, TLBO, etc.)

4. The author solved the optimization problem with 100 populations and 200 iterations. How did you reach to those values? It is necessary to have a proper judgment because the performance of heuristic algorithms highly depends on the initial population size and number of iterations.

5. Please provide the running time of the algorithm as well as the system specification that the MATLAB code was run on.

6. Please provide a figure showing the convergence characteristic of the AGWO algorithm.

7. There are some grammatical issues and typos throughout the paper. It should be carefully re-read by the authors or a native English-speaking person to solve those issues.

 

There are some grammatical issues and typos throughout the paper. It should be carefully re-read by the authors or a native English-speaking person to solve those issues.

Author Response

Response to the comments of Reviewer#4

 

We express our sincere gratitude and appreciation for the valuable feedback provided by the reviewer. With great consideration and attention, we have thoroughly revised our manuscript entitled "Realistic Home Energy Management System Considering the Life Cycle of Photovoltaic and Energy Storage Systems " according to your comments. We have prepared a comprehensive list of itemized responses addressing each comment.

 

Point 1: The contribution of the paper is not clear. Please discuss more about the aim and benefits of this research. What are its contributions? What are its differences with other proposed techniques in the literature.

 

Response 1: we appreciate the reviewer's attention to detail regarding the contributions In response to your suggestion:

Lines 83 to 96: we provided the objectives and contributions of this work, and the differences with other techniques in more detail. The previous studies focused on optimizing the operation time of home appliances based on minimizing the cost of energy used from the grid. We focuses on managing home energy consumption using PV and ESS by scheduling the operation time of home appliances considering the installation cost and lifecycle of the PV-ESS system. The objectives are realistic cost reduction of using energy, energy savings through efficient energy utilization, and maintaining the integrated system lifespan. The main contributions are developing battery and PV energy usage cost models, a model of selling energy to the grid, and introducing them to the optimization problem. Additionally, creating an efficient energy flow management algorithm and a model of a one-minute time slot is considered for the optimization problem, which makes the results more accurate. These contributions consider the installation cost and lifecycle of the integrated system, which help in efficient energy utilization and reduce energy losses and wasted energy. Moreover, we carried out a real case study to obtain realistic results to encourage household energy consumers to manage their energy consumption.

 

Point 2: This paper used AGWO algorithm to solve the optimization problem but there is no information about this algorithm or any appropriate references for future readers

 

Response 2: thank you for raising this concern, to address this:

Lines 310 to 340: we added brief backgrounds about AGWO and PSO algorithms.

Regarding the AGWO, we have provided a more detailed explanation of its inspiration from the social structure and hunting behavior of wolf packs. We have highlighted the four levels within the pack, namely alpha (?), beta (?), delta (?), and omega (?) wolves, along with their respective roles and interactions. This addition will enhance the reader's understanding of the foundation upon which the AGWO algorithm is built.

For PSO, we have provided a summary of PSO, highlighting its inspiration from the social behavior of bird flocking or fish schooling. We have emphasized that PSO involves a population of particles that navigate a search space to find the optimal solution for an optimization problem. Each particle represents a potential solution and adjusts its position and velocity based on personal experience and the collective knowledge of the swarm. By continuously updating their positions and velocities, particles explore and exploit the search space, converging towards the global optimum.

 

Point 3: It is important that the authors compare their obtained results using the AGWO algorithm with the results achieved by some other heuristic algorithms (i.e., GA, PSO, TLBO, etc.).

 

Response 3: we appreciate your observation regarding implementing another optimization algorithm and providing more details of the algorithms to enhance the manuscript. To do so:

 We obtained new results using the PSO algorithm, in addition to the AGWO algorithm. The results of the AGWO and PSO are described in the results and discussion sections. We implemented AGWO and PSO to evaluate the effectiveness of the proposed system in terms of real cost reduction, energy saving, and battery lifespan. The results proved that implementing AGWO and PSO in the proposed system achieves better results compared to an existing energy usage pricing model.

 

Point 4: The author solved the optimization problem with 100 populations and 200 iterations. How did you reach to those values? It is necessary to have a proper judgment because the performance of heuristic algorithms highly depends on the initial population size and number of iterations.

 

Response 4:  we acknowledge your point about optimization algorithms parameters.

In the beginning, we started with less population and iteration to see how the results goes. By changing these parameters, we reached that 100 population and 200 iterations are appropriate to evaluate our proposed model. It is clear from the convergence curve in the results section that the proposed system using AGWO and PSO obtains the best solution early. Moreover, our selected parameters are close to the average of the parameters used in previous studies. We only have a lower and upper bound of constraints.

 

Point 5: Please provide the running time of the algorithm as well as the system specification that the MATLAB code was run on.

 

Response 5:  we appreciate the reviewer's attention to detail regarding the optimization process detail.In response to your suggestion:

Lines 380 to 383: we added the system specification and the running time to the manuscript. For simulations, we used MATLAB R installed on Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz   2.90 GHz and 8 GB RAM with Windows 10 Pro. Each simulation run took around 7 hours to finish the 200 iterations and obtain the best solutions.

 

Point 6: Please provide a figure showing the convergence characteristic of the AGWO algorithm.

 

Response 6: we acknowledge your point that would enhance the manuscript’s results.

We add the convergence characteristic of the AGWO and PSO to the third and fourth sections of the results.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The authors have thoroughly revised the manuscript.

Extensive editing of English language required

Reviewer 4 Report

No further comments.