Reinforcement Learning with Probabilistic Boolean Network Models of Smart Grid Devices

Figure 6

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