Distributed neurodynamic optimization for multi-energy management with time-varying external disturbances considering time-varying emission limitations and load demand in multi-microgrid.

An emerging time-varying distributed multi-energy management problem (MEMP) considering time-varying load and emission limitations for resisting time-varying external disturbances and communication time delays in the multi-microgrid (MMG) system is investigated. Each microgrid (MG) contains some smaller microgrids (SMGs), which are connected by energy routers (ERs) of the system and can monitor energy in real-time with each other. In addition, a time-varying multi-energy management optimization model (MEMOM) is proposed in this paper in order to minimize the total cost of the MEMP which considers environmental cost, renewable energy cost and fuel cost. Furthermore, time-varying distributed neurodynamic optimization algorithms are proposed for solving the above MEMP based on consensus theory and sliding mode control technique. Compared with the optimization algorithms which consist of symbolic functions proposed in traditional energy management problems, algorithms consisting of hyperbolic tangent functions proposed in this paper can effectively reduce the oscillation of the algorithms and improve the stability of algorithms. Furthermore, the algorithm can converge the optimal trajectory of optimization problems with time-varying external disturbances and communication time delays. Meanwhile, the stability and convergence of the algorithms are proved theoretically by constructing appropriate Lyapunov functions. Finally, the performance evaluation results of numerical simulations show that the proposed algorithms can efficiently handle energy trading under time-varying load and maintain excellent stability with time-varying external disturbances and communication time delays.