*5.2. Multi-Period Simulation Analysis*

For further verifying the robustness of the proposed two-layer consensus algorithm, we further added a time period from 12:00 to 13:00 to illustrate the effectiveness. In the multi-period, the load is changed from 400 to 460 kW, and the AC/DC hybrid network is taken as the research object. Figures 9–11 illustrate the corresponding simulation results, respectively.

**Figure 9.** IC update of each node of the multi-period in the AC/DC hybrid microgrid.

**Figure 10.** Power output of each node of the multi-period in the AC/DC hybrid microgrid.

The IC update of each node of the multi-period is presented in Figure 9. Before the 20th iteration, the simulation results are the same as those of Figure 3, in the 20th and subsequent iterations, the nodes are abruptly changed due to the load, the IC also changes accordingly, and finally converges to 9.9196. Figure 10 shows the power output of each node of the multi-period. After the load is abrupt, then the output of each node changes accordingly, and finally converges to 151.21, 24.65, 62.78, 144.18, 14.40, and 62.78 kW, respectively. The sum is 460 kW, which just meets the load demand. Figure 11 is the power mismatch of the multi-period in the hybrid microgrid, following which we can find that at each time period, the total power output meets the total load demand.

**Figure 11.** Power mismatch of the multi-period in the AC/DC hybrid microgrid.

## *5.3. Full Time-Period Simulation Analysis*

The proposed two-layer consensus strategy was further verified in a full time period (24 h). By using the data from Tables 1 and 2, the simulation results obtained by the developed two-layer consensus method are provided in Figure 12. From this figure, we know that regardless of the time period, the power output can satisfy the total load demand, so we can conclude that the proposed algorithm has better robustness and adaptability.

**Figure 12.** Power mismatch of the full time period (24 h) in the AC/DC hybrid microgrid.

In the same conditions, by using the methods developed in References [30,31], they can all realize the power balance in a short time. However, due to the fact that no nodes were hierarchically controlled in the mentioned results, the interaction times were longer, and the fluctuations were larger, thus we can conclude that the proposed hierarchical consensus algorithm could be much more effective and satisfactory.

**Remark 5.** *In the circumstances of a multi-period simulation and a full time-period simulation, two or more time periods were considered to verify the proposed algorithm. Although there exists a time span, the power balance in the AC*/*DC hybrid microgrid can also be realized. It should be noted that at the moment of time varying between the two time spans, the total load power will mutate to another value, and as time goes by, the new power demand can be satisfied by the output power of the generator, PV, and WT. Finally, the total active power reaches the dynamic balance at each time-steady state.*
