Energy Management Strategy for Dc Micro-grid System With an Important Penetration of Renewable Energy

This paper presents an energy management strategy using stateflow controller related to DC micro-grids with an important penetration of renewable energy. The increase in world electricity demand is one of the principal drivers of the exhaustion of fossil fuels and expanded greenhouse gas emissions. To solve these problems, several countries have adopted actions for large renewable energy deployment, which includes wind energy, solar power, biomass power, tidal, and hydropower. These sources are considered as significant in delivering clean energy and reducing greenhouse gas emissions for sustainable improvement. In such a case, these are referred to as distributed generation systems. Distributed generation can impact negatively on the performance of the distribution network as the distribution network will no longer operate with a unidirectional power flow pattern. To address these issues, micro-grids are being used as a platform to integrate distributed generation systems, as they provide significant benefits to end-users and to the distribution network. The objective of this paper was to implement an energy management system to ensure the proper operation of the developed DC microgrid and this was developed using Simulink blocks available in the MATLAB/Simulink software. The simulation results shown that this control strategy is unconditionally reliable to ensure proper operation of the microgrid systems. Moreover, the developed algorithm model presents another advantage, which enables the users to access and to change any control parameters within the DC microgrid. By comparing these results with the literature, the developed energy management algorithm provides safety and automatic control of the microgrid.

P wind : Power produced by wind (Watt)

INTRODUCTION
The augmentation of world electricity requirements is one of the principal drives to the exhaustion of fossil fuels and expanded greenhouse gas emissions.To solve these problems, several countries have adopted actions for large deployment renewable energy sources, which include wind energy, solar power, biomass energy, tidal and hydropower [1].These sources are considered as significant in delivering clean energy and reducing greenhouse gas emissions for sustainable development [2].Topić et al. reported that in 2016, a total of 921 GW was achieved from the deployment of renewable energy resources, excluding hydropower, which results in the increased awareness of climate change.These renewable energy sources are often connected to the conventional power system through the distribution network near the loads, thus no transmission system is needed.In such a case, these are referred to as distributed generation systems [3].Distributed generation can impact negatively on the performance of the distribution network as the distribution network will no longer operate with a unidirectional power flow pattern [4].The issues related to bidirectional power flow pattern are intensified by increasing the levels of the distributed generation systems in the distribution network.Some of the known issues effect power quality, desensitized relays, increased fault currents, increased maintenance of equipment utilized, and even a large portion of the distribution grid.To address these issues, micro-grids are used as a platform to integrate distributed generation systems, as they provide significant benefits to end-users and to the distribution network.A microgrid can minimize disruption, lower costs, and optimize the size of the system components, thereby reducing operation costs and ensuring access to affordable, reliable, and sustainable forms of energy [6].Besides distributed generation systems, a typical microgrid consists of a controllable load and an energy storage system.An energy storage system refers to a device that converts energy from one form (usually electrical energy) to a storable form, and then the stored energy can be converted back into electricity when required.The utilization of energy storage systems, especially in renewable microgrids, has a significant impact on the reliability of the electric power as it can smooth the power fluctuation, reduce power quality problems, control the microgrid frequency and voltage, deliver initial energy when there is a transition between grid connection and islanded mode operation of micro-grids and provides ride-through capability in case of dynamic variations in intermittent energy sources and enables distributed generations to operate as dispatchable units [7] [8].Three different types of micro-grids exist, namely, the AC microgrid, the DC microgrid and the hybrid microgrid [9].This paper considers a DC microgrid.The rapid increase of renewable energy resources requires a sturdy energy management system to connect these renewable energy sources, including energy storage systems.Based on the inexact models, deterministic estimations, and balance conditions, different researchers have reported, which focus on enhancing the efficient action and operation of EMS that require much more improvement.Thus, new researchers need to be performed to improve the reliability of microgrid systems.The aim of the paper is to develop an energy management strategy for DC microgrids to provide power to a remote town.The principal objective of this paper is to present the design of an energy management strategy that attempts to ensure a proper operation of the developed DC microgrid.The developed DC microgrid simulation model is performed using MATLAB / Simulink software.This paper is organized as follows.The concept of microgrid systems is presented in Section 2. Section 3 discusses the mathematical modelling of the system components.In Section 4, the developed energy management system is presented.Section 5 explores and evaluates the simulation results of the developed method and finally Section 6 presents the conclusion of the paper.

Concept of Micro-Grid Systems
A typical microgrid refers to a set of distributed generation (DG) systems based on renewable and/or nonrenewable sources, incorporating an energy storage system (ESS) and local controllable loads, generally connected to the distribution system [10].It may operate in both grid connection as well as isolated mode depending on the load condition.Micro-grids can be grouped into diverse categories depending on the location (such as campus, military, residential, commercial, and industrial), size (such as small, medium, and large scale), application (such as premium power, resilience-oriented, and loss mitigation) [11].A microgrid implicates the integration of several distributed energy sources; the electricity from these sources is collected, converted, and distributed according to the load requirements.A control system is needed to ensure a proper operation of the microgrid when the power electronics interface with it to form a single unit.The control system is very essential, besides providing flexibility, it also preserves the specific energy production and the power quality [12].Each of these three sorts of microgrid presents advantages and disadvantages.The following Figure 1 presents the different types of micro-grids.DC microgrids have recently received much attention, especially for commercial and residential small-scale applications, as they provide increased efficiency and controllability with additional power conversion stages being eliminated, synchronization and compensation of reactive power no longer needed [13].In a typical microgrid, the common bus is DC, hence, AC generators are connected to the DC bus via rectifiers, while inverters are used to supply AC loads as shown in Figure 2. The DC microgrid can operate connected to the grid or isolated from it.It gives various operational advantages [14].Many of the devices, which are connected to a DC microgrid are generally electronic devices.As there are directly connected to a DC microgrid, no power systems like AC-to-DC, DC-to-AC, or AC-to-DC-to-DC are needed, which could be needed for an AC microgrid.No transformer is used in a DC microgrid; this characteristic makes it to be more efficient, reduced size, and more reliable in a DC power system [15].However, a DC microgrid still presents various problems that must be surmounted.There are no good practices implemented to manage fault situations and a basic protective element such as circuit breakers, fuses, and protection relays is lacking, as in the case of AC micro-grids [16].The hybrid micro-grids result from the combination of AC and DC micro-grids, they include both AC and DC types of buses [17].The type of connection to each bus depends on the proximity of the distributed generation unit and the load on the bus.The following steps in the electricity improvement distribution system are the AC microgrid systems.They can operate in a grid-connected or island mode.Based on the tasks that they have been designed to accomplish, their grid-connected or island modes can possess many suboperational states and or topological configurations [18].DC-AC converters are used to connect DC generators and energy storage systems to AC buses, while rectifiers are used to supply DC loads.To optimize the components of a microgrid system, an energy management system is used, which resolves decision-making strategies.These strategies consider increased system energy efficiency, increased reliability, reduced power consumption, reduced operating cost of distributed energy resources, reducing system losses, and mitigating greenhouse gas emissions for sustainable improvement [19].

Distributed Generation Sources
A distributed generation system is composed of different sources, both renewable and nonrenewable.Renewable sources are generators that utilize renewable energy technologies such as photovoltaic, solar thermal power, wind power, biomass, tidal power, and geothermal power as the primary energy to generate power [20].Whereas, nonrenewable sources utilize natural resources that are not naturally replenished to produce power [21].
A) Solar Power: Solar power can produce power through two different forms, which include solar thermal generation and photovoltaic conversion).The first method which is the solar thermal process, it is a process of converting solar power into heat and that produces heat runs a steam turbine which produces power and photovoltaic conversion converts sunlight into electricity using photoelectric phenomena [22].The solar cell is a basic component of a PV system.Solar cells can be grouped in series and parallel to constitute modules, which are also connected in series to form strings, and the strings are finally connected in parallel to form arrays. Different from the photovoltaic system, the solar thermal generation utilizes solar collectors to collect heat and the most widely used include parabolic, concentrated solar collectors, etc. [23].B) Wind Power: The wind turbine operating mode is focused on two methods.Firstly, the kinetic energy of the moving air is converted into mechanical energy.This conversion is obtained by using aerodynamic rotor blades and mechanical power control.Secondly, the electromechanical energy is converted into electric power via a generator [24].According to the wind turbine topology, they can be classified into two categories (horizontal axis and vertical axis) [25] , however, the most used topology is the horizontal axis wind turbine [26].Various generator topologies are utilized to convert the electro-mechanical energy into electricity.C) Geothermal Power: A geothermal power system converts the energy contained in the hot rock into electricity by injecting water into the rock to absorb heat from the rock and transport that heat to the surface of the earth, where it can be converted into electrical power through a turbine generator.High-temperature water (> 240 ° C) is vaporized into steam and that heat is transferred to a low-pressure steam turbine, which converts the steam into mechanical energy [27].Usually, the exploitable geothermal reservoirs are mostly located close to young volcanic rock areas [28].D) Hydropower: Hydropower is a process that produces electricity from water flowing in the river or the oceans.The purpose of hydropower production is to generate clean energy [29].A hydropower plant is generally composed of a generator, a turbine, a penstock, and a wicket gate.Mainly, two types of turbines are utilized, which include a pulse turbine (a Pelton Wheel turbine), and a reaction turbine like Francis and Kaplan turbines.Generally, the generator and turbine are directly linked to the same vertical shaft [30].E) Biomass Power: A biomass power system utilizes biomass to produce biogas, which is utilized as fuel for a biomass generator [31].The utilized biomass includes both energy crops and wastes, such as forest residues and a range of other agricultural and industrial compounds.Actually, the use of biomass is a fundamental part of global renewable energy and contributes to an ever-increasing share of global electricity capacity [32].F) Micro-Turbines: Micro-turbines are energy generators, which have a range of capacity between 15 and 500 kW.They operate under the open cycle gas turbine principle and they have shown many different characteristics and they operate easily and at high speed.Some of their benefits include variable speed, easy to install, compact size, low maintenance, air bearings, low NOx emissions, and mostly a recuperator [33].Micro-turbine generators have demonstrated strong improvements for distributed small-scale power production.However, the results have shown that these generators cannot be able to supply power during peak periods, however they may meet peak demand and improve the power generation reliability as they can provide standby capabilities when the power grid fails [34].G) Diesel Generators: A diesel generator (also known as genset) is composed of an internal combustion engine and a synchronous generator connected to the same shaft.Diesel generators are generally utilized as backup or emergency power system in commercial and industrial installations.They are also widely utilized in new areas where connection to the distribution grid is unfeasible or expensive to perform [35].Their principal disadvantage is the negative impact that they can cause on the environment due to their high NOx.

Power Electronics Converters Control
The hierarchical control architecture of a microgrid system includes the primary, secondary, and tertiary controls.The primary control consists of the preliminary power sharing and regulation of current/voltage, whereas the secondary control is higher than the primary control and helps to compensate for the voltage and improve the performance sharing.Finally, tertiary control focuses on power management, energy management and economic dispatch [36].

Primary Control
The primary control consists of inner loop (current / voltage regulation) and droop control (preliminary power-sharing) based of control in the power converters.The response time of the primary control is very short, and there is no need of communication when it is utilized in the decentralized system [37].But when it is used in centralized control, distributed control, and master-slave control, there is a need of communication.This control is mostly used for regulating the frequency and voltage of the inner loop of control.

Secondary Control
A secondary control has a longer operating time than the primary control because it minimizes the frequency and voltage deflections that persevere even after primary control is performed and it can also be utilized for centralized and decentralized controls [38].The secondary control is performed by a central controller of the microgrid in centralized control and in decentralized control, it is implemented by local controllers.

Tertiary Control
The principal goal of the tertiary control is to regulate power flow and load sharing between converters and between several hybrid microgrid systems if they are connected in a group and finally synchronize individual incremental costs.Additionally, a tertiary controller is utilized to update the set point for the secondary controller depending on the production cost preferences by updating the loading ratio of each converter [39].Comparatively to the secondary controller, the tertiary controller is able to suppress voltage fluctuations and change voltage operating points to attain the optimal dispatch of power production.

Description of the Developed System
The developed DC microgrid architecture is for a remote town.Different loads such as commercial load, industrial load, and residential load can be found in a town.The capacity of the power produced by all DG sources are 150kWp for solar, 150kW for wind energy, and 345kWh for battery bank.The power generation from the PV and battery is DC, but for wind and biomass is AC, therefore, the power generation from the AC sources are converted to DC by using a rectifier.To reduce the voltage fluctuation, each DG source is equipped with a converter.Battery is used to assure power supply to consumers without any interruption.The DC microgrid feeds the load directly by 380Vdc and three types of loads are considered, commercial load, industrial load, and residential load.The total produced power, which includes PV and wind, are connected in parallel.The battery bank will be used as back-up power during peak demand and to decrease variations in the power produced by renewable sources.Thus, the total power required to supply the load is 280 kW. Figure 3 illustrates the block diagram of the developed DC microgrid system.

Mathematical Modelling of the System Components
The concern in this point is to design a dynamic model of the developed DC microgrid, which helps to verify the efficacy of the designed controller.The MATLAB/Simulink mathematical modules have been used to build the DC microgrid model including the battery system, which uses the component equivalent circuits.Four different power supplies are considered in this model, which include PV, wind, biomass, and battery power.

Photovoltaic Module Modelling
The output characteristic of a photovoltaic array is determined by cell temperature, solar irradiation, and output voltage.The operation mode of a PV either solar cell refers to the PN junction diode operation, which converts light energy into electricity by the photovoltaic effect [40].
With, N p and N s are numbers of cell connected in parallel and series,  is Boltzmann's constant, 1.380658e-23 J/K, A ideal diode factor between 1 and 5, IR s : inverse cell current saturation at T, I RS represents the solar cell reverses saturation current and I rr represents the reverse saturation current at T r .
Where,   represents the short circuit current at the base temperature of the cell, K i the coefficient of the short circuit temperature, T r refers to cell temperature and  represents the solar radiation in (w/m 2 ) [42].According to this configuration, the shunt resistance in parallel with the ideal shunt diode and the I-V characteristics are described by the following equation: The PV output power is a function of the solar irradiation and the area of PV module as shown in the following equation: Where, ƞ g is the production efficiency, i r radiation of solar in w/m 2 and A the area.

Modelling of Wind Power
This supply power model is characterized by the variation of wind speed with gust and wind speed.
• V w is the base wind velocity; • V g the gust wind velocity; • V wr the ramp wind component.
Wind power is given by the following expression: The energy drawn by the wind turbine is expressed by the formula below: With W w the energy drawn by wind turbine and  the air density.Based on Betz, the maximum output power wind turbine is given by: The substitution of the value for V 1 , and V 3 enable to obtain the Equation: The modelling of a wind turbine helps to describe the captured output power of the turbine.The wind power according to a given area is calculated as follows: Where, β is the pitch angle of the blade (℃); δ is the tip speed ratio of the turbine and C p is the power coefficient.The maximum produced power by a wind turbine is calculated by the following equation:

Modelling of the battery
The principal focusing parameters of the battery mathematical model are the voltage and current.The current can be determined by the changing of the terminal voltage of the battery.The current production is driven by the transmission of electrons from one electrode to another [43].The potential difference between the positive and negative electrodes determines the open circuit voltage of the battery [44].The following equations present the charging and discharging voltages of battery.
The composed relation model allows to modify the Equations ( 17) and ( 18).E 0 = represents the open circuit voltage of the battery expressed in (Volt); K is the polarization coefficient (Ω); Q represents the battery capacity (A/h) and R the internal resistance.Equations ( 17) and ( 18) present some limitations such as (i) battery aging and self-discharge, (ii) the current amplitude does not have impact on the battery capacity, and (iii) no consideration of the temperature coefficient.To surmount these limitations, it is crucial to consider the aspects that have impact on the life of the battery [45].The analyze of  condition is done at each instant and is calculated with the threshold capacity by the help of the following equation: The following equation determines the net power produced by a DC microgrid structure: P net = P PV + P wind + P bio (23)

Developed Management System
Generally, the energy management system is used to control power production and schedule programs for a group of power grid applications.However, it may be considered as another way to control the electrical loads in micro-grids.The present system is designed to meet the loads.As renewable resources are intermittent sources, the battery is used as a back-up system and this is also designed to meet the load as well.The developed system has three types of loads (Industrial, residential and commercial Loads), in the proposed model system, the industrial load is assumed to be a priority load.The system comprises two renewable sources with the storage system.The net generated power   and the load power   are calculated as follows: P G = P pv + P wind (24) P L = P Ind + P Res + P Com (25) Where, P G the generated power; P pv the power produced by PV; P wind the power produced by wind;   the load;   the Industrial load;   the Residential load and   the Commercial load.
The battery bank will charge when there is excess production and will discharge when the generated power is not able to handle the load demands.Based on this architecture, the power generation will supply the load through two conditions and by the support of the battery bank.✓ Firstly, when the power generation is equal to the overall load demand, the loads are supplied by the power generation from PV and wind without any disruption.✓ Secondly, when the power generation exceeds the total load.For this case, the power generation is more than the load demand, thus the total load demands are provided by the generation and the surplus of the production goes to charge the batteries.✓ When the power generation becomes less than the total load demand, then the load will be provided with the help of the battery bank.The battery is discharged until its  will reach its minimum value.
SoC min < SoC batt < SoC max = 20% < SoC batt < 100% ✓ When the SoC batt < 20%, the battery will be disconnected from the system and at this time the priority load, which is the industrial load, will be supplied by the generation.In the case where the production exceeds the industrial load, the excess of the production will be used to charge the battery.The system will permanently verify the productive power until the power generation will be enough to provide the load.

SIMULATION RESULTS AND DISCUSSION OF THE DEVELOPED MODEL
The present section gives the description of the developed energy management model structure and diverse specified simulation procedures and simulation results.The developed EMS is for a DC microgrid utilizing Simulink blocks available in the MATLAB / SIMULINK software.All circuit elements and machines are represented by their respective model blocks available in the software.Different parameters of the simulation are presented in Table 1.The developed EMS strategy will control the PV, wind, battery, and load demands.The load is subdivided in three categories: industrial load (which is the high priority load), residential load, and commercial load [46].The output voltage remains constant at 380   with very small ripples; the designed model allows a ripple with 0.9% variation from the steady state.The power generation stays constant at 149.68kW.
Figure 6 presents the universal bridge rectifier output voltage and the output power of the wind power model as a function of the simulation time.The steady state of the output voltage is attained at approximately 2.2s and after remains constant at 380 V dc with very small ripples.

Fig. 6: Wind output power and voltage from the developed DC micro-grid model
The output power generation from the wind model is also presented in Figure 6 and the steady state of this power starts to be reached at 2.2s and remains constant with small ripples.The average value of the power generation is 148.78kW.
The developed EMS algorithm is designed on Stateflow logical programming environment.In this section, different cases are considered.Table 2 summarizes the different operating modes of these cases.In this study, Load 1 is considered the priority load, which is the industrial load.The operating mode of the Stateflow environment corresponds to that of a logical system, either it is 0 or 1.When the system is in operation mode, the output indicates 1, and when the output indicates 0, it means that the system is not in operation.
Case 1: In this first case, the power supply is equal to the load demand as shown in Figure 7.At this stage, the battery as well as the auxiliary load are disconnected from the system.The total generation is localized to the load and the battery SoC is assumed to be less than 100%.In this case, the battery SoC fixed to be 50%.Here, the power generation becomes higher compared to the load demand as shown in Figure 9. Thus, the EMS checks if the battery is fully charged or not.If the SoC is less than 100% as shown in Figure 10, the load demand will be provided by the supplied power and at the same time, the battery is connected to the system where it is charged by the extra production until it reaches its charge limit which is 100%.When the battery reaches its maximum charge value (100%), it is disconnected from the system to wait for the discharge phase.The surplus of the power production will be transferred to the auxiliary load as shown in Figure 11, where it will dump until the power generation reaches the load demand or decreased.In Figure 11, the load and battery outputs are 1 while the damping load output is 0. This means that the loads are supplied and at the same time the battery is also charging.Figure 12 shows the transition from the Stateflow chart.The supplied power from the distributed generation is 295kW and the load demands are 280kW.The excess of the production (15kW) is stored in the battery bank as illustrated in Figures 11 and 12. Figure 13 shows that the power generation is still higher than the load demand and the battery is fully charged (100%), in this case the surplus is supplied to the auxiliary load as it can be seen in Figure 13.The power generation becomes lower than the load demand as it can be seen in Figure 14.Based on the EMS order, the load will be supplied by the help of the battery.Firstly, the SoC of the battery will be measured.If the SoC of the battery is higher than 20% as illustrated in Figure 15, then the battery will be connected to the system where it will be discharged.The condition is that, when the SoC becomes less than 20%, automatically the battery bank must be disconnected from the system.Once the SoC of the battery drops below 20%, the system disconnects the battery.The EMS allows the priority load, which is the industrial load (Load1) to be supplied.Before that, the EMS will measure the power generation to check if it will be able to provide power to the priority load and charge the battery bank at the same time, if the condition is approved, then the battery will also be charged at the same time.Figure 17 presents the simulation results showing that the priority load is supplied by the power generation and Figure 18 illustrates the transition state.When the supplied power becomes lower than the priority load, the EMS will check if the generated power is able to supply the residential load (Load2) or supply both the residential load and charge the battery at the same time.If the condition is approved, the residential load and the battery will be supplied by the power generation and if not, only the residential load will be supplied.When the power generation falls below the residential load, the proposed EMS will automatically disconnect all loads (Shut-down loads).In this case, no load can be supplied by the available power generation from the renewable energy sources.The loads are momently disconnected from the system and the available power generation is used to charge the battery until its SoC becomes more than 20% or the power generation becomes active.The result of this scenario is shown in Figure 19.

Figure 19: Output results from the developed energy management system chart
The simulation results of the developed DC microgrid and energy management algorithm model were presented in this section.
The model was finally designed and developed to ensure the energy management in a DC microgrid.The model was designed and developed using MATLAB/Simulink software.In this paper, it was a question of focusing on the energy management in all DC microgrid system to finally maintain its proper operation.The method used to design and develop this energy management system algorithm was the Stateflow logical programming environment in MATLAB/Simulink software.Based on all these scenarios, the developed energy management system algorithm has shown the ability to ensure the reliability, resiliency, robustness and proper operation of a microgrid system.

CONCLUSION
This paper treats the energy management strategy using Stateflow controller related to DC micro-grids with an important penetration of renewable energy.The developed energy management system model was implemented on MATLAB/Simulink.This developed DC microgrid model considered different loads such as commercial, industrial, and residential loads.The simulation was conducted to show the interaction between different components of the system and to demonstrate the operation of the developed energy management strategy model.The simulation result scenarios of the developed energy management system algorithm model have successfully shown that this control strategy is unconditionally reliable to ensure a proper operation of the microgrid systems and their durability.The main advantage of this developed algorithm is that it will supply power to the consumers with the deep penetration of renewable energy sources.

ACKNOWLEDGMENT
This journal paper is based on the thesis work and results of author C.Ndeke Bipongo.This paper is a summary of the most important aspects of his thesis entitled "Energy management of a battery energy storage system for renewable energy DC microgrid".

Ethical Approval
The data collection for this project did not require the consent of participants since humans or animals are not the subject of research.

Figure 3 :
Figure 3: Block diagram of the developed system.

Figure 4 :
Figure 4: A single equivalent circuit model of a PV cell [41].

Figure 4
Figure 4 represents the configuration of a single PV cell.According to this circuit, the solar irradiance is described through a current source  ℎ , and (diode current I d , output current   , series resistance R s , parallel resistance R p , and output voltage   ) represent the remaining circuit parameters of the configuration.The output current is determined by the following formula:

Figure 5
Figure 5 presents the PV output voltage and power as a function of the simulation time.

Fig. 5 :
Fig. 5: PV output power and voltage from the developed DC microgrid model

Fig. 7 :
Fig. 7: Power and voltage generated from the DC microgrid model

Figure 8 :
Figure 8: Transition state from the Stateflow chart

Figure 11 :Figure 12 :
Figure 11: Output results from the developed EMS chart

Figure 13 :
Figure 13: Output results from the developed energy management system chart

Figure 14 :Figure 15 :
Figure 14: Power generation from the proposed DC Micro-Grid Model

Figure 16 :
Figure 16: Transition state from Stateflow chart

Figure 17 :Figure 18 :
Figure 17: Output results from the developed energy management system chart