Dynamic environmental‐economic load dispatch in grid‐connected microgrids with demand response programs considering the uncertainties of demand, renewable generation and market price

Abstract

Microgrids (MGs) as a key building block of smart grids have been emerged to address the proliferation of distributed energy resources. In grid-connected MGs, dynamic economic load dispatch (DELD) module determines optimal schedule of distributed energy resources and adjustable loads and power to be exchanged with upstream grid, while all operational constraints of the MG are respected. DELD in MGs represents a constrained optimization problem with uncertain input data, as the forecasts of demand, renewable generation and market price are uncertain. In this research, particle swarm optimization (PSO) as a bio-inspired optimization algorithm is used to solve DELD in gridconnected MGs, while demand response program is integrated into MG and the uncertainties of demand, renewable power generation and market price are dealt with two-point estimate method (TPEM). Load curtailment as a demand response program is used for reducing operation cost of microgrids. The performance of PSO is compared with two optimization algorithms including grey wolf optimization and backtracking search algorithm. As per the results, at times with low grid power price, microgrid imports power from upstream grid and at times with high power price it exports power to the upstream grid. The results show that the integration of demand response has significantly reduced the operation cost of the microgrid. The effect of change in maximum curtailable power on the operation cost of the MG has been investigated.

Incentive-based demand response in grid-connected microgrid using quasi-opposed grey wolf optimizer

“…A detailed review of various DR models for optimization related to energy management system isreported in various literature as presented in (Fallahi and Smith, 2017;Hussain and Gao, 2018;Jordehi, 2019;Khalili et al, 2019;Mizutani et al, 2018;Mohammed and Albadi, 2020;Robert et al, 2018;Sedighizadeh et al, 2018;Shariatzadeh et al, 2015). Researchers have proposed various energy management systems and models (EMS) to solve the generation scheduling problems either by using the traditional approach (Kiptoo et al, 2020;Mehdizadeh et al, 2018;Nwulu and Xia, 2017;SoltaniNejad Farsangi et al, 2018) or by using metaheuristic algorithms (Aghajani et al, 2015;da Silva et al, 2020;Esther et al, 2016;Jordehi, 2020;Kakran and Chanana, 2019;Niharika and Mukherjee, 2018;Nosratabadi and Hooshmand, 2020;Roy et al, 2019;Sarker et al,2020;Shahryari et al, 2019) with the objective to either minimize operational cost or minimize both cost and emission.Grid-tied MiG with DR is seen as an excellent option by independent system operators (ISO) to achieve solutions to a large spectrum of modern-day electrical distribution problems.Futuristic MiG control systems are encouraged to include demand-side management (DSM) as an indispensable component (Imani et al, 2020;Ajoulabadi et al, 2020).…”

“…The multi-objective scheduling of MiG coupled with incentive-based DR has been investigated in (Aghajani et al, 2015;da Silva et al, 2020;Esther et al, 2016;Jordehi, 2020;Kakran and Chanana, 2019;Niharika and Mukherjee, 2018;Nosratabadi and Hooshmand, 2020;Olorunfemi and Nwulu, 2020;Roy et al, 2019;Shahryari et al, 2019)for minimization of operational cost and emissions released to the atmosphere. The augmented ɛ-constraint approach has been used to solve the multi-objective problem to minimize operation costs and emissions (Sedighizadeh et al, 2018).…”

“…In (Aghajani et al, 2015), multi-objective particle swarm optimization (MOPSO) has been utilized for minimizing operational cost and emission for MiG involving DR. Here, it is observed that the DR approach helps to mitigate power shortage caused due to the uncertainty of RES and is also useful in reducing operational cost and pollutant emission effectively.The benefit of using DR in grid-connected MiG after considering the uncertainties of demand, renewable power generation, and the market price is explored in (Jordehi, 2020). Here, particle swarm optimization (PSO) is used as the optimization algorithm.A multi-objective model is formulated in (Olorunfemi and Nwulu, 2020) to simultaneously minimize the annual cost of electricity production, minimize the carbon dioxide emission and maximize the customers' participation in the IBDRP to increase the customer benefit that comes with the program.An improved IBDR program for optimum multiobjective MiG scheduling is given in (Shahryari et al, 2019).…”

An improved particle swarm optimisation for unit commitment in microgrids with battery energy storage systems considering battery degradation and uncertainties

“…EMS includes two main modules; unit commitment (UC) and economic load dispatch. [10][11][12] In UC module, data of demand and DERs are sent into MG master controller and the set of committed dispatchable DGs along with their power, power exported to/imported from macrogrid and status and power of ESS units are determined, while all related constraints of MG are satisfied. 13 UC in ESSintegrated MGs represents a constrained, mixed binarycontinuous optimisation problem with uncertain input data as the forecast of demand, renewable power and market price include uncertainties.…”

A new power system active rescheduling method considering the dispatchable plug-in electric vehicles and intermittent renewable energies

“…Conventional MA algorithms include simulated annealing algorithm (SA) [15], genetic algorithm (GA) [16], differential evolution algorithm (DE) [17,18], ant colony optimization algorithm (ACO) [19], particle swarm algorithm (PSO) [20], social learning particle swarm algorithm (SLPSO) [21], firefly algorithm [22], reinforcement learning [23] and so on. In addition, due to that the switch state of the unit is a binary variable, improved optimization algorithms based on binary coding or mixed binary and real number coding schemes have also used to solve the UC problem, such as binary differential evolution (DBDE) [24], binary competitive group optimization algorithm (BCSO) [25], binary gravitational search algorithm (BGSA) [26], binary grey wolf optimiser (BGWO) [27], mixed binary-continuous particle swarm optimization [28] and etc. Though MA methods are shown to have better capability than traditional mathematical methods in solving UC problems, large-scale and multi-dimensional optimization problems may cause MAs falling into the local optimum.…”