24-09-2012, 10:37 AM
NEURAL NETWORK CONTROLLED ENERGY SAVER FOR INDUCTION
MOTOR DRIVE
Artificial neural network.pptx (Size: 2 MB / Downloads: 33)
Introduction
Single phase induction motors.
Most houses and offices are supplied by single phase supply.
Reliable, cheaper in cost and easy to repair.
Wide applications : fan, refrigerators, centrifugal pumps.
Controlling energy consumption of IM.
Use of ANN.
Why……………..?
To implement energy saving scheme of single phase IM under no load and partial load conditions.
In most of the applications induction motor woks on no load and partial load.
Presence of full rated voltage across the stator causes loss of energy in the form of heat.
Necessary to control the stator voltage.
ANN helps to apply the optimum value of voltage.
What is ANN……?
An information processing paradigm.
Inspired by the way biological nervous systems.
Composed of a large number of highly interconnected processing elements (neurones).
Adaptive learning.
Take a different approach to problem solving than that of conventional computers.
Training is required.
Description
V ,I,T,N are sensed to train the neural network
Obtain the optimum value of duty cycle.
Duty ratio is used to drive the PWM.
Change in duty ratio changes the average voltage.
Helps to obtain optimum voltage to stator.
PWM Ac chopper helps to attain better power factor.
It eliminates the harmonics to some extent and reduces the losses due to harmonics.
PWM AC CHOPPER
In differential topology the S2 and S1 switches, the voltage source and the load are serial connected.
Moving S2 switch between the voltage source and S1- non-differential topology.
If ua is positive, S1 and S1c switches are PWM controlled with a constant duty ratio (α), while S2 and S2c switches are fully turned on.
When the sign of the voltage source is changed, the switching pattern is reversed, S2 and S2c being complementary PWM controlled with a constant duty ratio and S1 and S1c are fully turned on.
us =α ⋅ua
Advantages
Helps to obtain an optimum voltage to drive induction motor.
Energy loss (Cu loss ) at no load and partial load conditions can be eliminated to some extent.
Energy saving of about 58% can be achieved at no load when about 20% rated voltage is applied instead of 100%.
Energy saving of 23% can be achieved in partial load conditions when about 70% of rated voltage is applied instead of 100%.
Losses due to harmonics can be eliminated.
Conclusion
Controlling energy consumption of an IM is an important task.
Combining ANN and hardware helps to obtain optimum value of duty cycle.
ANN can be trained using matlab simulation
PWM AC chopper supplies only required voltage
PWM AC chopper reduces harmonic content.
Powerfactor improvement can be achieved
About 58% of energy can be saved at no load
Difficulty is in interfacing hardware and software\
Energy saving is an important task