problems, and the analysis module i

problems, and the analysis module is used to calculate the responsevalue of the target variable during the optimization iteration. Thesystem passes the optimized design variable dij to the subsystem, and itsvalue becomes the target of the subsystem; On the premise of inequalityconstraint gsub≤0 and equality constraint hsub=0, the design module ofthe subsystem optimizes its own problem fsub, and the analysis moduleapproaches the optimized value pij of the design module to dij by introducing the penalty term π(dij − pij). π is a large positive number.As can be seen from the relationship between ADN and MGs inFig. 1, this hierarchical structure is consistent with the basic principle ofTCAM, and TCAM can be applied to the modeling and solution of dispatching problem for ADN with multi-MGs. Where, ADN corresponds tothe system layer, and MGs correspond to the subsystem layer. At thesame time, in order to realize the power decoupling of ADN-MG andMG-MG tie lines, taking ADN-MG as an example, this paper establishesthe decoupling mechanism as shown in Fig. 3. The power of the tie lineis decoupled into the equivalent generator power at the ADN side andthe equivalent load power at the MG side. The former is transferred tothe subsystem layer after optimization at the system layer, and thelatter to the system layer after optimization at the subsystem layer. Bysetting the lagrange penalty function, they are iterated alternately untilthe error requirements are met.