1.3 System of Energy Management
Extensive research has been going on upon the application of an energy management system in micro-grid operated in either connected with grid network mode or isolated from grid mode. For optimal MSE the objective function has to be defined following the system constraints (mechanism and functioning mode). The MSE puts a significant impact on the environment, the life span of the generation unit and system performance. The energy management modeling is the uncertainty modeling techniques with specific objective functions and constraints [18–24].
An MSE will be able to implement in two manners, such as integrated and disintegrated. In integrated MSE there is a central controller that controls the exchange of power in MG and optimizes that with respect to market prices and security constriction. In the MG having decentralized MSE, distributed supplies and loads are having a higher degree of freedom, and it maximizes the revenue by communicating the components of MG with each other. The main aim of any MSE is to maintain the load–supply balance [25, 26].
1.3.1 Classification of MSE
According to literature, the MG energy management system is categorized into four groups:
1 a. MSE based on conventional sources
2 b. MSE based on SSE
3 c. MSE based on DSM
4 d. MSE based on Hybrid system.
1.3.1.1 MSE Based on Conventional Sources
According to this MSE during the failure of the energy storage system, the renewable energy sources are used with a backup source of energy like gas engines, diesel generators and microturbines in the MG [21].
1.3.1.2 MSE Based on SSE
The MGs face problems in managing the non-conventional sources like solar energy and wind energy because of its fluctuating nature. Often there is a variation in forecasted and real-time production of energy. SSE (storage system of energy) is the best solution to this problem. The SSE maintains a balance between energy production and consumption as load by preserving power during low-peak times and releasing power during high-peak time. Different optimization techniques are focused on improving the utilization in MG [27–29].
1.3.1.3 MSE Based on DSM
An additional way to deal with the energy unbalance in supply and load in MG is the utilization of DSM (demand-side management). The goal of this MSE technique is to match the generated power with power consumption by modifying customers’ behavior or load profile [30, 31]. The DSM is divided into to categories:
1 i) Energy efficiency: This minimizes the consumption of power with increasing the commodities effectiveness at the demand side.
2 ii) Demand reaction: This (DR) changes the amount of power handling by the consumer in the response of changes in the price of electricity and encouragement payments as an aim to minimize the power expenditure during high-price hours or peak time or when the system is facing any threat in reliability.
The DR skims categorized in two ways. Such as: based on price DR and based on incentive DR [32].
1.3.1.4 MSE Based on Hybrid System
In this type, more than one of the above-mentioned types is practised together to solve MSE problems in the micro-grid [33].
1.3.2 Steps of MSE During Problem Solving
To solve a problem in micro-grid, the MSE follows some steps:
1 a. Prediction of uncertain parameters
2 b. Uncertainty modeling
3 c. Mathematical formulation
4 d. Optimization
1.3.2.1 Prediction of Uncertain Parameters
Uncertainty is the possibility of deference in real and forecasted values due to a lack of information [34]. This uncertainty in MG may be in operational parameters or in economic parameters. The operational parameters include the quantity of power generated and the quantity of power consumption. On the other hand, the economic parameter includes the effects on economic aspects such as production cost, financial growth and rate of interest, etc. [35].
The uncertain parameters can be predicted in various time ranges that are of the very small time period, a few minutes to a couple of days, which is called prediction of short term. In the prediction of mid-term, the time range is from several weeks to a few months. And in the prediction of the long term, it ranges from few months to several years [36]. The MG problems are considered for hourly intervals, so short term prediction is the best method for this.
1.3.2.2 Uncertainty Modeling
The MSE faces difficulty during decision making because of the uncertainty. That is why a verity of ways have been implemented to manage the uncertainty. Those methods are called uncertainty modeling [35]. There are some methods like stochastic method, ANN method, fuzzy method, robust optimization method and information gap decision theory [37–39].
1.3.2.3 Mathematical Formulation
The management of energy in a microgrid can be formulated mathematically as a problem of optimization with the primary aim to schedule the functionality of DGs, SSEs and loads for the short term with specific objective functions and the constraints of the components of MG. The capital cost and the operational cost are taken as the objective functions. Fuel cost, maintenance cost, start-up and short-down losses and degradation are considered as the operational cost. There are different constraints which can affect an MSE. Some cases maximum and minimum limit of the generation unit affects their safety and economic performance. The source and load balance are also a constraint that has to be taken. The rate of charging and rate of discharging of energy storage body can also be taken as a constraint. The bus voltage, feeder current, frequency, etc. are some technical constraints [40, 41].
1.3.2.4 Optimization
In the literature, there are many optimization techniques used for MSE in MGs [26]. Some examples are like (i) Heuristic approach (ii) MAS (iii) CPLEX solver (iv) SNOPT solver, etc.
1.3.3 Micro-Grid in Islanded Mode (Figure 1.6)
1.3.3.1 Objective Functions and Constraints of System
In Ref. [42] the authors have studied the micro-grid consisting of several renewable energy sources with EMS. They have taken the goal function as the cost minimization of energy, which is the net cost of RES per annual cost of the net energy produced. Now the optimization objective is subjected to various constraints, such as reliability, economic conditions and environmental conditions. The criteria of reliability can be considered as the amount of energy cannot be supplied when the amount of power demand is more than the power generation. In economic condition, all of the cost includes installation cost, running cost. The environmental constraint is about the amount of CO2 emitted related