Application of Artificial Neural Network to Robust Speed Control of Servodrive

Particle swarm optimization of artificial-neural-network-based on-line trained speed controller for battery electric vehicle

“…determined during an off-line optimization procedure, or can be tuned continuously during normal operation of the drive in on-line mode. Examples of off-line and on-line trained controllers can be found in [6][7][8][9][10][11].…”

Robust speed and position control based on neural and fuzzy techniques

“…Each of the methods mentioned yields a different structure of the nonlinear module and requires a different method of synthesis. In the earlier papers [3,4] the concept of robust controller synthesis was described by the authors.…”

“…The assumed rule base and the membership functions create a control surface presented in Figure 3, which represents nonlinear controller characteristics. This concept of neural controller was described and analyzed in a previous work by the authors [4]. The proposed neural network has two inputs and three layers forming a 2×6×3×1 structure which is shown in Figure 4 .…”

“…Because of the nonlinear character of the controller, the Random Weight Change (RWC) procedure is selected [3,4]. The advantage of this procedure is its insensitivity to the local minima of the optimised criterion.…”

“…Next, the sum of both criterion values was assumed for optimisation. The flow chart of the RWC algorithm is given in [3,4].…”