Optimal Energy Management in a Smart Micro Grid with Demand Side Participation

– The energy management in energy systems is the main solution for energy companies in order to provide minimization of the energy generation costs and emission polluting. In this work, a multi-criteria optimization model is implemented for minimizing the generation cost and emission in a smart micro grid (SMG) at day-ahead planning. In this modelling, the demand side participates in optimal energy management through two strategies such as demand shifting and onsite generation by the energy storage system (ESS). The optimal participation of demand side is modelled based on energy price in energy market. Implementation of the proposed approach in GAMS software is done, and weight sum method (WSM) is employed for solving multi-criteria optimization. The desired optimal solution of multi-criteria objectives is found via the max-min fuzzy procedure. Finally, confirmation of the proposed approach is analysed by numerical simulation in two case studies.


INTRODUCTION
The employment of smart micro grids (SMGs) technologies alongside different generator units (GUs) such as electrical, thermal and storage systems are developing in energy system infrastructures in many countries [1]- [5].In SMGs communication links have been provided between the generation side and the demand side, in order to achieve the optimal energy consumption via economic signals to demand side.These signals lead to a decrease in energy consumption at high energy price in the energy market [6]- [10].The participation of the demand side in energy management with the name demand response (DR) programmes is defined [11]- [15].The DR can be implemented in various types by demand side with attention to energy prices in the day-ahead market [16]- [20].On the other side, DR programmes are not limited to economic indices, and optimal energy consumption has notable effects on generation emission from the generation side [20]- [25].Using DR, optimal energy balance based-economic and environmental indices are provided in between generation and demand sides [26]- [30].
The investigations on energy management in different systems are studied in diverse types of modelling and approach.In [31], the economic modelling of the SMG is presented without consideration of the environment index and DR programs.The energy management in the residential section is studied in [32] to minimise the demand side bill.In [33], the optimal installation of the GUs in a SMG is implemented based on load demand uncertainty.The energy-saving in smart buildings with optimal charging and discharging of electric vehicles (EVs) is proposed in [34].The self-scheduling of the load demand based on the energy output of renewable energy is implemented in [35].In [36] demand interrupting approach is used for objectives optimization such as economic and technical indices.In [37] minimizing the generation costs and maximizing the reliability indices of the SMG are studied.The demand reduction in [38] is done to increase energy efficiency by optimal scheduling of appliances in the residential section.In [39] energy management of an SMG is proposed, to improve technical and economic issues by reconfiguration approach and installation of the energy storage system (ESS).In [40] and [41] scheduling modelling of the SMG is studied for energy management using peak shaving modelling.
In this article, a multi-criteria optimization problem based-economic and environment model is proposed for optimal energy management in an SMG at the day-ahead.The objectives include minimization of the 1) generation costs and 2) emission pollution on the generation side.As well, load scheduling of DR programs on the demand-side is considered as demand shifting and onsite generation by ESS with attention to electrical price in the energy market.The objective functions are optimized using the weight sum method (WSM), and the desired optimal solution in Pareto frontier solutions of objectives is determined via Environmental and Climate Technologies ____________________________________________________________________________ 2022 / 26 the max-min fuzzy method.Accordingly, the novelty and contributions in this article can be listed by the following items: 1.A multi-criteria optimization is proposed to provide minimization of the generation cost and emission; 2. Two strategies are provided to the demand side participation in optimal energy management; 3. The WSM is used to obtain Pareto frontier solutions of multi-criteria optimization; 4. Using max-min fuzzy method, desired optimal solution is found.The outline of the article is arranged as follows: The topology of the SMG is presented in section 2. In section 3, an optimization method is provided.The simulation is done in section 4. Finally, the conclusion is in section 5.

SMG TOPOLOGY
In Fig. 1. the topology of proposed SMG is shown.The SMG comprised of four sections as follows: 1. Demand side.This section has required energies, such as electrical and thermal for consumption.On the other hand, the demand side plays an optimal role in energy consumption, the modelling of which will be analysed in the next subsection; 2. Energy markets.The electrical market is considered as the main energy supplier, which has diverse price in each time [42], [43]; 3. GUs.The GUs are diesel generator (DG), boiler and combined heat and power (CHP).
These units are feed by fossil fuels to energy generation [44]; 4. Energy management operator.In this section, optimal coordination via communication links between demand side and generation side is done based on pre-defined objective functions and existing constraints to optimal energy management at day-ahead. .

Energy Markets
Electrical Market

Modelling Demand Shifting
Using demand shifting strategy, load demand at high price electrical market can be moved to low price.This strategy modelling is as follow [35]: ' ' where Eq. ( 1) and Eq. ( 2) are demand shifted from t to t', and limit of the demand shifted, respectively.

Modelling Onsite Generation
The onsite generation strategy using ESS is done.The demand side can be charged self-ESS at low electrical price and discharged at high price to meet self-demand.The modelling onsite generation strategy based-ESS is as follow [36]: The Eq. (3) and Eq. ( 4) are limit of discharge power and charge power of ESS, respectively.The Eq. ( 5) is an onsite generation model by ESS.

Objective Functions Modelling
In this subsection, objective functions including generation cost and emission polluting are modelled by Eq. ( 6) and Eq. ( 7), respectively.
Environmental and Climate Technologies ____________________________________________________________________________ 2022 / 26 The generation costs of GUs are modelled by the first and second terms of Eq. ( 6), respectively.The third, fourth and fifth terms of the Eq. ( 6) are respectively generation costs of the electrical market, discharge and charge of ESS.And emission generation by GUs and the electrical market is modelled in the first and second terms of Eq. ( 7), respectively.

Constraints
The proposed approach is optimized according to several constraints: min max ( , ) The constraints Eqs. ( 8)-( 11) are electrical energy balance, thermal energy balance, electrical and thermal generation bound of GUs, respectively.

OPTIMIZATION METHOD
Since, the multi-criteria model including generation cost and emission are optimized, simultaneous.The WSM is employed to optimization of the objective functions.In this method, objectives are turned to one objective and Pareto frontier solutions are obtained by diverse values of weights.The WSM is modelled as follows [38]: subject to: Here, w2 and w2 are weights for objective functions such as generation cost and emission, respectively.

Max-Min Fuzzy Procedure
In this subsection, the max-min fuzzy procedure is used to determine the desired optimal solution in between Pareto frontier solutions.The operator as a decision maker should implement the maximum trade-off for objectives.The process of the max-min fuzzy procedure is modelled as follows [45]- [49]: Environmental and Climate Technologies ____________________________________________________________________________ 2022 / 26 Using Eq. ( 14), all objectives are normalized, and by Eq. ( 15) maximum trade-off can be calculated.Here fn max , fn min and fn(k) are maximum, minimum value of nth objective and value of the nth objective in kth solution, respectively.

NUMERICAL SIMULATION
In this section, two case studies are considered for confirmation of the multi-criteria optimization problem and the proposed approach.The case studies are as follows: − Case 1) Multi-criteria optimization problem in SMG without demand side participation; − Case 2) Multi-criteria optimization problem in SMG with demand side participation.In this study, GUs include two CHPs, two boilers and two DGs.The data of GUs are provided in Table 1 and Table 2 [50]- [52].The data of ESS is given in Table 3.In Fig. 2 and Fig. 3, electricity prices in day-ahead energy market and energy demand are depicted, respectively [38], [39].The emission factor of the electrical market is taken by 980 g/kWh.The value of the demand participation in demand shifting is equal to 40 %.The simulation is implemented using GAMS software in core i7 PC system with CPU 2.5 GHz by DICOPT solver and mixed integer nonlinear program (MINLP).Fig. 2. Electrical price in day-ahead [38], [39].

Results Evaluation
As mentioned before, two case studies are considered to validation of the proposed approach subject to demand side participation in SMG.In this subsection, obtained results of the simulation are analysed as a comparative study.
In Table 4. Pareto frontier solutions of objectives in case studies are listed with attention to different values of weights by WSM.The weights are changed with a value of 0.2 for each objective.As well in this table, the desired solution of each case with maximum trade-off is bolded using the max-min fuzzy procedure.The maximum trade-off of objective functions for Cases 1 and 2 in, the desired solution is equal to 0.761 and 0.763, respectively.In Case 1, the third solution is selected as the desired solution.The generation cost and emission in Case 1, have values of $ 426 132.2 and 5330.1 kg, respectively.However, with the participation of the demand side in Case 2, emission and generation costs are minimized by 10.43 % and 21.32 % than Case 1, respectively.In Case 1, generation cost and emission of the electrical market are equal to $ 281 802.3 and 3573.51kg, respectively.This while, emission and generation costs of the electrical market in Case 2 are decreased by 25.4 % and 20.3 % according to Case 1, respectively.The optimal reshaping of electrical demand by demand shifting and utilization of the ESS with low cost leads to more participation of GUs and a reduction of the energy market participation in meeting demand.
In Fig. 4, the electrical demand with consideration of the demand shifting strategy is shown.It is visible that the demand for the high electrical price is shifted to low prices.This strategy

Electrical demand
Thermal demand leads to reduce generation costs and emissions of the electrical market, due to high electrical prices at peak demand and high emission factors of the electrical market.
In Tables 5 and 6, electrical and thermal generation in Cases 1 and 2 are listed, respectively.In Table 5, electrical purchased from the electrical market is done at hours 5-19 in high electrical price and peak demand, due to low electrical generation by GUs.On the other hand, in Table 6, with the participation of the demand side in optimal energy consumption and onsite generation by ESS, electrical generation from the electrical market is reduced by 32.3 % than Case 1.In Table 6, the power charge of ESS is scheduled at hours 1-3, 19 and 24.The ESS at mentioned hours is fed by DGs and CHPs with low cost and emission; and discharged at a high electrical price in the market.This strategy has an optimal effect on electrical purchases from the energy market and leads to decrease costs and emissions.The amount of the generation cost and emission production by GUs to meet thermal demand in both cases is similar.

CONCLUSION
In this article, the optimal energy management of the SMG is analysed with consideration of the demand side participation in the day-ahead electrical market.The demand side has two strategies of the DR onsite generation by ESS and shifting their demand.The multi-criteria problem such as generation cost and emissions are minimized subject to demand side participation and constraints.Hence, the main conclusions of the results can be expressed as follows: 1.The optimal shifting of the electrical demand leads to modification of the consumption pattern with attention to electrical price in the energy market.The demand shifting and ESS are provided reshape of the peak electrical demand at operation time to optimal energy dispatch of generation side in meet demand with minimum cost and emission; 2. The participation of the demand side leads to a decrease in electrical generation by the electrical market with a high emission factor.The charging of ESS is done by GUs and low electrical price, and feed demand by ESS occurs at a high price.It can be concluded that the demand side participation in energy management has been provided with the optimal level of the energy balance among the generation side and demand side to enhance the efficiency of the proposed SMG, and improve environmental and economic indices.

TABLE 1 .
ECONOMIC AND EMISSION DATA OF GUS

TABLE 2 .
ENERGY GENERATION LIMIT OF GUS

TABLE 4 .
PARETO FRONTIER SOLUTIONS OF CASE STUDIES

TABLE 5 .
OPTIMAL ENERGY GENERATION IN CASE 1

TABLE 6 .
OPTIMAL ENERGY GENERATION IN CASE 2