Reinforcement Learning with Probabilistic Boolean Network Models of Smart Grid Devices

Figure 10

Maximum reward obtained for the Fault 2 operation mode of the IPR in one year of operation.