Operational Cost Minimization of Grid Connected Microgrid System Using Fire Fly Technique

Present time, green energy sources interfacing to the utility grid by utilizing microgrid system is very vital to satisfy the ever increasing energy demand. Optimal operation of the microgrid system improved the generation from the distributed renewable energy sources at the lowest operational cost. Large amount of constraints and variables are associated with the microgrid economic operation problem. Thus, this problem is very complex and required efficient technique for handing the problem adequately. Therefore, this research utilized the efficient fire fly optimization technique for solving the formulated microgrid operation control problem. Fire fly algorithm is based on the behaviour and nature of the fire flies. A microgrid system modelling which incorporated various distributed energy sources such as solar photo voltaic, wind turbine, micro turbine, fuel cell, diesel generator, electric vehicle technology, battery energy storage system and demands. Energy storage system is utilized in this research for supporting renewable energy sources’ integration in more reliable and qualitative way. Further, the electric vehicle technology i.e. battery electric vehicle, plug-in hybrid electric vehicle and fuel cell electric vehicle are utilized to support the microgrid and utility grid systems with respect to variable demands. Optimal operational cost minimization problem of the developed microgrid system is solved by fire fly algorithm and compared with the grey wolf optimization and particle swarm optimization techniques. By comparative analysis it is clear that the fire fly algorithm provides the minimum operational cost of microgrid system as compared to the GWO and PSO. MATLAB software is utilized to model the microgrid system and implementation of the optimization techniques.


Introduction
The energy demand around the world is continuously increasing. Along with that the green house gas emission, energy efficiency reductions and appropriate renewable energy generation became main issues in the power system. The most suitable remedy for all of these above discussed problems is the construction of the microgrid system with green energy generators like solar photovoltaic, wind turbine, fuel cells, micro turbine, electric vehicle, and energy storage systems. There are two modes of operation in which the microgrid can be operated i.e. standalone micro grid system and grid integrated microgrid system. In the standalone operation mode microgrid can be operated as self sufficient energy grid. Further in the grid tied mode, energy is exported and imported from the grid. Other than t provide the green energy, micro grid system reduced the green house gases from the environment, generated from the thermal power plant with reduction in the power price [1,2]. Though there are numerous advantages of the micro grid system, but due to intermittent nature of the renewable energy sources such as solar photo voltaic and wind turbine, reliability of the micro grid system is affected. With the utilization of the energy storage systems, reliability issues of the micro grid system can be managed [3]. The main functions of the energy storage system are the supplying energy at the time of energy shortage in the micro grid. Further energy storage system stored surplus energy generated from renewable energy sources of micro grid system energy at the time of off peak load time. Integration of the energy storage devices' in the microgrid system is important for balancing the power in the system. Determination of the energy storage system capacity is very important for the economical operation of the microgrid system. Too much capacity of energy storage system will raise the total cost of the microgrid system while less capacity of energy storage system minimize the reliability of the system and increase the cost of energy generation from the traditional sources of generation.
Hence the determination of the optimal generation of various renewable energy generation sources with the optimal capacity of the energy storage system are the important issues for the economical operation and minimization of the operational cost of the microgrid system [4][5][6].
The optimized energy management techniques are utilized to find the optimal generation capacity of the various generation components of the microgrid system as well as the capacity of the energy storage system. Further, these techniques minimized the complete operational cost of the system.
The main issue with the some of the Meta heuristic algorithm is that these techniques search the best solution in their local space without incorporating the global solution space. These techniques may mislead the process of searching and due to that the optimization technique stuck into the local optimum value only. On the other hand some of the Meta heuristic techniques provide adequate global search capabilities but their local searching capacity is limited. Both the limitation affected the performance of the Meta heuristic techniques. Therefore, more efficient Meta heuristic techniques are required for good convergence and enhancement of the exploration process. For that purpose, an advanced Meta heuristic technique namely fire fly optimization approach provides the good equilibrium between the global and local search solution spaces, is applied in this research work. The main objective of this research work is the minimization of the operational cost of the grid connected microgrid system. Further optimal output of various generation components of the microgrid system is also computed. The applied technique is based on the food searching capabilities of Fire Flies with the high equilibrium between exploitation and exploration capabilities [7]. At last for verifying the stability and the performance of the applied technique it is executed on the representative low voltage microgrid network. Various case studies have been performed and a comparative analysis with the two renowned techniques i.e. particle swam optimization (PSO) [8] and grey wolf optimization (GWO) [9] is utilized. By comparative analysis it is concluded that the applied fire fly technique provides better results as compare to the other techniques. The fire fly results proved that the ability of the applied technique to obtain the global best solution in the optimization difficulty by means of computation efficacy and solution quality is better as compare to the PSO and GWO. In [10], Bridier (2016), utilized a meta-heuristic technique to describe the comparison of economical and technical sizing of energy storage system with microgrid generation components such as PV, wind and wave. In [11], Aghamohammadi and Abdolahinia (2014), depending on the primary frequency control of micro grid system, optimal capacity of battery energy storage system is computed. In [12][13][14][15][16][17], for addressing the optimal dispatch energy flow in the microgrid system, mixed integer linear programming technique is utilized. Further, the same technique is utilized to find the optimal capacity of the energy storage system. In [18][19][20][21][22][23], for finding the optimum capacity of the energy storage device in the microgrid system, various meta-heuristic techniques are implemented in the hybrid microgrid system. In [24,25], for computing the optimum scheduling issue of the standalone and grid connected microgrid system, dynamic programming is utilized. For the problem formulation, efficiency and the operation characterises of the energy storage system in the microgrid is considered. In [26,27], for calculating the optimal capacity of the different generation components such as PV, full cell and energy storage devices along with the distributed generation under the electricity hybrid market structure of the microgrid system, a GA based technique is utilized. This method is useful for enhancing the life cycle cost of microgrid system as well as minimizing the green house gases. In [28][29][30], researchers are utilized the PSO technique to compute the optimal capacity of the battery energy storage system at minimum cost. In [31], Sukumar et al. (2018) utilized the Grey Wolf optimization technique is utilized for finding the optimal capacity of battery energy storage system by solving the economic operation problem of microgrid system. This work utilized the batteries as the energy storage system but not included the electric vehicles technology for the energy storage purpose. In [32], Nimma et al. (2018) utilized the grey wolf optimization-technique dependent optimal energy-management system is developed for microgrid system. Further, this work computed the optimal capacity of the battery for grid connected microgrid system. This work utilized the batteries as the energy storage system but not included the electric vehicles technology for the energy storage purpose. The main contributions of the manuscript are as follows: 1. Applied a advanced optimization technique i.e. fire fly algorithm for the intelligent energy management which enhances the integration of the renewable energy generators and minimize the reliance on the conventional energy sources such as diesel generator in microgrid system.

2.
As compare to the existing system in [32], different types of electric vehicles and diesel generator are incorporated and then optimized the system. Three different types of electric vehicles are modelled in the system those are battery electric vehicles, plug in electric vehicles and fuel cell electric vehicles.
3. Minimized the overall operational cost of the grid connected microgrid system 4. Optimal generation outputs of various microgrid generators components is calculated in this work.
5. Optimal capacities of the batteries are computed in this work by fire fly algorithm. 6. A comparative analysis with GWO and PSO techniques presented the superiority of the fire fly algorithm. As compare to [32], in this work the operation cost of the system calculated with the fire fly algorithm and compared with the grey wolf optimization technique utilized in the base paper. By comparative analysis it is clear that fire fly algorithm has more improved performance as compare to the gray wolf optimization. Further, fire fly algorithm also provided better results as compare to the particle swam optimization [33].

Problem Formulation
Based on the above discussed literature review, the proposed microgrid optimal operational problem can be formulated as follows: The objective of optimal operation problem of micro grid system is to minimize the total operational cost of the system [34]. This problem is solved by utilizing the Firefly optimization method for solving the cost minimization problem of microgrid system,

Development of objective function
The developed problem is formulated as follows: The formulation of the optimal operation problem of microgrid is illustrated as follows: Minimization of the total cost of microgrid system is given by: where, collective total cost per day for all types of batteries utilized in this research is the summation of total cost per day of: The cost of start up of FC, FCEV and MT are provided in the following equations, respectively.
The regular operational and repair cost of DRG are presented by = ( In the micro grid system, operational charges of the utility grid, FCEV, PHEV, BEV and BES, operation and maintenance The developed operational cost reduction problem is optimized over the following constraint conditions

Constraint conditions 2.2.1 Power Balance Condition
The generated power must always be equal to the demand and losses in the system. In this work, the system losses ignore, therefore, the power balance equation can be modified as follows

Dispatchable DRGs constraints
The energy output limits of the various distributed renewable energy sources must be satisfied by the microgrid generation:

BES constraints
The minimum and maximum, charging and discharging rates of BES are presented as follows Discharging mode [35]: Charging mode: Where

Grid Constraints
Utility grid should provides within the mentioned generation limits represented by the following equation ≤ ≤ (23)

Diesel Generator Constraints
Diesel generator should generated energy within the mentioned limits represented by the following equation Journal of Informatics Electrical and Electronics Engineering (JIEEE) A2Z Journals

BEV Constraints
The minimum and maximum, charging and discharging rates of BEV are presented as follows Discharging mode: Charging mode: Where

PHEV Constraints
The minimum and maximum, charging and discharging rates of PHEV are presented as follows Discharging mode: Charging mode: where

Operating Reserve Constraints
Integration of the energy storage systems such as BES, EVTs and operational reserve, increased the reliability of the microgrid system. Operational reserve (ORE) capacity is the summation of the reserve generation capacity of the active BES, electric vehicle technologies, FC, MT, DEG and utility grid, in the each time duration. The ORE can supply to the microgrid system within 10 minutes and presented by the following equations: where, ORE(t) is the 10-min ORE requirement at time t.

Methodology
For obtaining the above mentioned objectives, this work utilized the methodology presented in figure 1. The basic concepts related to the microgrid system and detailed review of various literatures related to the optimal operation of grid

Fire Fly Algorithm
This technique is a Meta heuristic optimization algorithm, which is based on the natural inspired phenomenon. It is based on the nature and behavior of the fire files. There are three basic rules on which this algorithm is dependent [7]. 3. The objective function is directly proportional to the brightness.
The movement of a firefly i is attracted to another more attractive (brighter) firefly j is determined by where, β 0 represents the attraction at distance 0, r ij = ||x i − x j || is the space between any two fireflies i and j at x i and x j , respectively, ϵ i is a random numbers vector computed from a Gaussian or uniform distribution function and α represents the randomization parameter. Table 1 presented the various parameters utilized to model the fire fly algorithm in the developed micro grid optimization problem.

Start
Step 1. Initialize max number of iteration, , 0 , Step 2. Generate initial population Step 3. Formulate the Objective function ( ), Step 4. Determine Intensity (I) at cost (x) of each individual determined by ( ) Step

Introduction
In this paper, first the developed system is described. Further, the Fire Fly optimization technique is applied on the developed micro grid system and validated. For validation the purpose, the performance of the Fire fly optimization technique is compared with PSO and GWO methods. Two different operational scenarios are considered to show the effective-Journal of Informatics Electrical and Electronics Engineering (JIEEE) A2Z Journals ness of the Fire fly algorithm. In the first scenario, it is considers that, all the batteries are integrated to the system with no charge or minimum charging condition. The second scenario considered all connected batteries in charged situation. Figure 2 presented the proposed grid connected microgrid system.  Table 2 presents the description of the coefficients and generation limits utilized in this research.   For verifying the performance of the Fire fly algorithm, the optimal operation problem of microgrid system is solved To find the effectiveness of the firefly algorithm, optimal operation problem of microgrid system is solved under two operational scenarios. In the first scenario, it is considered that all types of batteries utilized in this work are utilized with no charge or minimum charging situation. In the second scenario, all batteries are integrated into the system with full charging condition and considered as generation sources.

Scenario 1: Batteries Charging Mode
In this scenario, various Li-ion batteries are integrated into In this particular scenario the optimal operational problem of microgrid system is minimized the operational cost

Scenario 2: Batteries Discharging Mode
This scenario considered that all the batteries integrated with the micro grid system are fully charged. The outputs of the utility grid, DEG, Solar PV, WT, PGEV, FCEV, BES, BEV, MT and FC are computed optimally with the help of fire fly algorithm. In this case it is beneficent to import power form the battery energy storage system and electric vehicles. This case, also considered the optimum battery size of 50 kWh.

Results Analysis
Fire fly optimization technique is utilized to solve the optimal operation problem of microgrid system. MATLAB software is utilized to model the microgrid system and implementation of the fire fly optimization technique. Further, a detailed comparative analysis is presented to show the effectiveness of the fire fly technique over GWO and PSO technique.

Scenario 1: Batteries Charging Mode
In this operational scenario, all interfaced batteries are considered at no charging or at the minimum charging condition.

Optimal Operation Cost Minimization using Fire Fly Algorithm
For scenario one, the results obtained from the fire fly algorithm are presented in the table 4.    Figure 3 presented the optimal outputs computed from the firefly algorithm obtained from the various generation elements of the microgrid system. Table 5 presented the optimal power output with the status of the various microgrid generation elements.     Table 6 presented the comparative analysis of the different operational costs of microgrid system computed form the fire fly, GWO and PSO techniques. From the table 6, it is clear that the operation cost of the microgrid system calculated with the firefly algorithm is lowest as compared to the grey wolf optimization and particle swarm optimization techniques.  18.54157% worst cost reduction is obtained with the help of firefly algorithm as compare to the grey wolf optimization algorithm in the case of scenario 1. Figure 5 presented the comparative best cost reduction percentage of the PSO and GWO.  Figure 6 presented the percentage Reduction in the average operational cost as compare to the GWO and PSO. Figure 7 presented the percentage reduction in the microgrid worst operational cost computed from fire fly algorithm as compare to the GWO and PSO techniques.

Comparative analysis of the output of microgrid components
In this research work, fire fly algorithm is utilized to compute the output of various microgrid generation components in scenario 1. A comparative analysis of the power output in (kW) of various microgrid generation components with GWO and PSO techniques is presented by the table 8.

Scenario 2: Batteries Discharging Mode
In this operational scenario, all interfaced batteries are considered at fully charging or at the maximum charging condition.

Optimal Operation Cost Minimization using Fire Fly Algorithm
For scenario two, the results obtained from the fire fly algorithm are presented in the table 9.   Figure 8 presented the optimal outputs in kW computed from the firefly algorithm obtained from the various generation elements of the microgrid system. Table 10 presented the optimal power output with the status of the various microgrid generation elements.   From the table 11, it is clear that the operation cost of the microgrid system calculated with the firefly algorithm is the lowest as compared with the grey wolf optimization and particle swam optimization techniques.  Figure 10 presented the percentage reduction in the microgrid best operational cost computed from fire fly algorithm as compare to the GWO and PSO techniques.

Comparative analysis of the output of microgrid components
Journal of Informatics Electrical and Electronics Engineering (JIEEE) A2Z Journals In this research work, fire fly algorithm is utilized to compute the output of various microgrid generation components in scenario 2. A comparative analysis of the power output in (kW) of various microgrid generation components with GWO and PSO techniques is presented by the table 13.

Conclusion
The energy demand around the world is continuously increasing. Thus, distributed renewable energy sources should be integrated to the utility grid along with traditional energy sources in the form of microgrid for satisfying the energy de-Journal of Informatics Electrical and Electronics Engineering (JIEEE) A2Z Journals mand. Microgrid system integrated various renewable and conventional sources of energy in a single platform. In this work, microgrid system is integrated with MT, FC, PV, WT, BES, EVTs, Diesel Generator and demand. For the efficient working of the microgrid system, operation of microgrid system should be effacingly optimized. Therefore, in this research work efficient fire fly optimization algorithm is utilized and the operation of the microgrid system is optimized. The operational cost of the microgrid system is minimized under two different operational scenarios. First scenario considered that all the integrated batteries are in charging mode while in scenario second discharging modes of all the batteries utilized in this work. Operational cost for scenario one is minimized to 109157.02 Rs/Day and operational cost for scenario two is minimized to 137734.30 Rs/Day. A comparative analysis with the GWO and PSO techniques is also presented. By the comparative analysis it is clear that the fire fly algorithm provided the most optimal solutions i.e. average, best and worst solutions, in both the operational scenarios.
The results of this research work will help the scientists and researchers to optimize the performance of the microgrid system by minimizing the operational cost of the various microgrid generation components under different constraint conditions. This work may also be helpful for the economic operation of microgrid. The futuristic enrichment of this research work can be developed a technique that will generate more minimized operation cost results for the practical micro grid system. Further, hardware implementation of the developed microgrid system with applied fire fly techniques can also do as futuristic enhancement.