Application of brain emotional learning‐based intelligent controller to power flow control with thyristor‐controlled series capacitance

Semi-active vibration control of building structure by Self Tuned Brain Emotional Learning Based Intelligent Controller

“…(27) (28) for each input variable (e n and ) 7 membership functions (MFs) are assigned as displayed in Fig. 8.…”

“…The online learning capacity, minimal computational complexity, and, most importantly, no need for prior knowledge of system dynamics makes BELBIC a distinctive controller over formerly debated intelligent controllers, such as ANNs and FLC. Furthermore, it is simple, with fewer tuning parameters in emotional controllers, and, unlike traditional ANNS, it doesn't require an additional iterative process for learning or updating parameters [28]. Contrasting, in neural network control, the network topology, like that of the number of layers, nodes, and parameters in the activation functions, are critical considerations that must be considered appropriately [29].…”

Artificial emotional reinforcement learning for automatic generation control of large‐scale interconnected power grids

“…A promising approach is to introduce emotional learning to RL, which could deal with continuous variables and generate the control policies by not only a logical but also a humanistic intelligent part. In recent years, emotional learning-based intelligent (ELI) controller has been presented to handle the above issue, which showed the superior control performance in various real-world applications, including non-linear control of an interconnected power system [19], doubly-fed induction generator [20], power-flow control [21], sensorless speed control of switched reluctance motor [22], dynamic voltage regulator [23], interline power-flow controller [24], asymmetrical six-phase induction motor [25], real-time position control of a servo-hydraulic rotary actuator [26], unmanned ground vehicle navigation [27] and so on. Moreover, emotional learning was developed with combining Qlearning for traffic flow forecasting of multi-agent systems [28,29].…”