Academia.eduAcademia.edu

Figure 10 – uploaded by Waluyo Adi Siswanto

See full PDFdownloadDownload figure
Fig. 8. Step up speed response cmparison between conventional PID SVPWM and NN-SVPWM controller when the speed reference is varied from 80 to 110 rad/s.

Figure 8 Step up speed response cmparison between conventional PID SVPWM and NN-SVPWM controller when the speed reference is varied from 80 to 110 rad/s.

Related Figures (14)

Fig. 1. SVPWM-DTC induction motor drive. The SVPWM-DTC produces the stator reference voltage vectors by adopt of both torque and flux errors, then feed to the SVPWM algorithm to modulate. The constant switching frequency signal is then exported to the three phase inverter [8] as illustrated in Fig.1.
Fig. 1. SVPWM-DTC induction motor drive. The SVPWM-DTC produces the stator reference voltage vectors by adopt of both torque and flux errors, then feed to the SVPWM algorithm to modulate. The constant switching frequency signal is then exported to the three phase inverter [8] as illustrated in Fig.1.
The induction model in the stator-fixed d-q reference frame is described by [9]. The mechanical equations are given as below. The induction motor stator flux and torque are calculated in the flux and torque calculator as follows:
The induction model in the stator-fixed d-q reference frame is described by [9]. The mechanical equations are given as below. The induction motor stator flux and torque are calculated in the flux and torque calculator as follows:
Fig. 2. The block diagram of proposed NN-DTC.
Fig. 2. The block diagram of proposed NN-DTC.
Fig. 3. Diagram block of neural network speed control for DTC induction motor drive. The controller consists of input layer, hidden layer and output layer. Based on number of the neuron in the layers, the NN-DTC is defined as a 1-5-1 network structure. The first neuron of the output layer is used as a torque reference signal (a7,=m,; ). The connections weight parameter between j® and i® neuron at m mayer i is given by w"y, while bias parameter of this layer at i i” neuron is given by b". Transfer function of the network at i® neuron in m®™ layer is defined by:
Fig. 3. Diagram block of neural network speed control for DTC induction motor drive. The controller consists of input layer, hidden layer and output layer. Based on number of the neuron in the layers, the NN-DTC is defined as a 1-5-1 network structure. The first neuron of the output layer is used as a torque reference signal (a7,=m,; ). The connections weight parameter between j® and i® neuron at m mayer i is given by w"y, while bias parameter of this layer at i i” neuron is given by b". Transfer function of the network at i® neuron in m®™ layer is defined by:
TABLE I. Motor PARAMETERS To verify performance of the proposed NNDTC, the simulation results for a conventional DTC-SVPWM and the neural network DTC-SVPWM_ proposed controller are
TABLE I. Motor PARAMETERS To verify performance of the proposed NNDTC, the simulation results for a conventional DTC-SVPWM and the neural network DTC-SVPWM_ proposed controller are
Simulation was carried out to investigate the performance of the N -DTC. In this section the dynamic model of a three- phase induction motor, space vector PWM and neural network control model have been developed. The simulation is developed using Borland C++, and then embedded as S- function in Simulink-Matlab. The parameters for the motor are given in Table 1 while the simulation block diagram of the proposed M drive systems is shown in Fig. 4. Fig. 4. The simulink block diagram of the proposed NN-DTC.
Simulation was carried out to investigate the performance of the N -DTC. In this section the dynamic model of a three- phase induction motor, space vector PWM and neural network control model have been developed. The simulation is developed using Borland C++, and then embedded as S- function in Simulink-Matlab. The parameters for the motor are given in Table 1 while the simulation block diagram of the proposed M drive systems is shown in Fig. 4. Fig. 4. The simulink block diagram of the proposed NN-DTC.
Fig. 6. The motor stator current (Ja, Ib, Ic) for NN-DTC controller. Fig. 5. Start-up speed response comparison between conventional PID-DTC and NN-DTC controller.
Fig. 6. The motor stator current (Ja, Ib, Ic) for NN-DTC controller. Fig. 5. Start-up speed response comparison between conventional PID-DTC and NN-DTC controller.
Fig. 7. The motor stator current (Ia, Ib, Ic) for PID-DTC controller. Fig. 5 illustrated the start-up response of the system. From the figure, it shows that NN-controller have a start-up response improved from the conventional PID controller. It is clearly
Fig. 7. The motor stator current (Ia, Ib, Ic) for PID-DTC controller. Fig. 5 illustrated the start-up response of the system. From the figure, it shows that NN-controller have a start-up response improved from the conventional PID controller. It is clearly
compared. With the same speed and load torque reference, the simulations of both methods are run simultaneously. The simulation is start at the speed on 80 rad/s with a constant load applied. The start-up speed response of both system is shown in Fig.5. Meanwhile, the start-up motor stator current response is shown in Fig. 6 and Fig. 7.
compared. With the same speed and load torque reference, the simulations of both methods are run simultaneously. The simulation is start at the speed on 80 rad/s with a constant load applied. The start-up speed response of both system is shown in Fig.5. Meanwhile, the start-up motor stator current response is shown in Fig. 6 and Fig. 7.
Related topics:
Induction Motor DrivesNeural Network Based Controllers
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved