Microgrid system allocation using a bi-level intelligent approach and demand-side management

Abstract

Demand-side management (DSM) segregates the elastic and inelastic loads and restructures the load demand model of a distribution system by minimizing the operational cost of the entire process. This is done by optimally transferring the flexible loads to hours when the per-unit cost of utility is lower. This paper performs a bi-level optimization strategy to lower the operating expense of a low-voltage microgrid (LV MG) system operating in grid-connected mode, comprising battery energy storage (BES), renewable energy sources (RES), and fossil fuel-powered generators. In the first level of optimization, the load model is restructured as per the DSM participation level. Thereafter, the restructured load demand model is considered, and optimal allocation for distributed generators (DGs) is percolated for minimizing the generation cost of the microgrid system in the second level. A recently developed hybrid swarm intelligence algorithm that has already been used in solving diverse power system optimization problems was used as the optimization tool for the study. The generation cost was minimized for different grid participation types and grid pricing strategies with and without consideration of DSM. The numerical results show a 55–75% reduction in generation cost when 20–30% DSM participation was considered.

Graphical abstract

Highlights

• i. The generation cost of an LV microgrid (MG) system was evaluated for diverse grid-dependent scenarios.

• ii. The impact of demand-side management on the performance of the MG system and generation costs was studied.

Discussion

The work described in this paper initially restructured the forecasted load demand for different DSM participation levels to reduce the peak demand and improve the load factor of the MG system. Thereafter, the generation costs were evaluated for diverse grid-dependent scenarios and compared for various load demand models obtained after DSM implementation.

Appendix

The optimization tool used for minimization of generation cost for the microgrid system is a hybrid algorithm combining the crow search algorithm and JAYA algorithm. The mathematical modelling details of the algorithm are discussed below (Fig. 11).

Figure 11

Flow chart of CSAJAYA for DED in grid-connected microgrid.

$$X^{u,iter + 1} = \left\{ \begin{gathered} X^{u,iter} + rand_{u} \times fl^{u} \times (m^{v,iter} - X^{u,iter} ),{\text{if}}\;\;rand_{v} \ge AP^{v} \hfill \\ {\text{a}}\;{\text{random}}\;{\text{position}},\,\,\,\,\,\,{\text{otherwise}} \hfill \\ \end{gathered} \right.$$

(17)

$$m^{u,iter + 1} = \left\{ \begin{gathered} X^{u,iter + 1} ,\;{\text{if}}\;f(X^{u,iter + 1} )\; < \;f(m^{u,iter} ) \hfill \\ m^{u,iter} ,\,\,\,\,\,\,\,\,\,\,{\text{otherwise}} \hfill \\ \end{gathered} \right.$$

(18)

$$X_{k,q,iter + 1}^{{}} = X_{k,q,iter} + c^{\prime}*(X_{k,best,iter} - \left| {X_{k,q,iter} } \right|) - c^{\prime\prime}*(X_{k,worst,iter} - \left| {X_{k,q,iter} } \right|)$$

(19)

$$X^{u,iter + 1} = \left\{ \begin{gathered} X^{u,iter} + rand_{1} \times fl^{u} \times (m^{v,iter} - X^{u,iter} )\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,,{\text{when}}\,\,rand_{j} \ge AP\,\,\,\,\,\& \,\,\,\,k = 1:n \hfill \\ X^{u,iter,k} + rand_{1} \times m_{best}^{k} \times \left| {X^{u,iter,k} } \right| - rand_{2} \times m_{worst}^{k} \times \left| {X^{u,iter,k} } \right|\,\,\,,{\text{when}}\,\,rand_{j} \le AP\,\,\,\,\,\& \,\,\,\,k = 1:n \hfill \\ {\text{end}} \hfill \\ \end{gathered} \right.$$

(20)

	CSA44	JAYA45	CSAJAYA37,38,39
Governing equation	(17)	(19)	(20)
Memory update	(18)	NA	(18)