Pumping unit servo motor controller design
1. Working principle of pumping unit switched reluctance motor
At present, most of the new long-stroke pumping units use a switched reluctance motor (SRM) as their power component. The working principle is to drive the driving wheel through the reduction gear, and then drive the weight box and the sucker rod to lower and rise to pump the oil. This structure will effectively simplify the transmission structure, improve the overall efficiency, and save energy for the pumping unit. Provides the possibility.
The stator and rotor of the three-phase switched reluctance motor adopt a double salient pole structure, the stator pole has concentrated windings, and the corresponding pole windings are connected in series with each other. The rotor has no windings and no permanent magnets, thus forming A, B, C three-phase winding. When the winding of one of the three windings is energized, an electromagnetic torque is generated which causes the adjacent rotor pole to coincide with the winding axis, and energizing the three-phase windings in sequence, thereby driving the rotor to rotate continuously . By changing the energization order of the windings, the steering of the motor can be changed; by controlling the on-off time and current, the speed and torque of the switched reluctance motor can be changed.
The governing system of the long-stroke pumping unit of the reluctance motor is mainly composed of a controller, a power converter and an SRM motor. The controller designed in this paper adopts double closed-loop control (as shown in Fig. 2), and the inner ring is turned. Moment control, the outer ring is speed control, in which the main function of the speed loop is to control the motor speed and obtain the reference torque of the torque inner loop.
2, fuzzy speed regulator design
Fuzzy control and neural network composite control are added to the neural network based on fuzzy control. The fuzzy controller structure diagram shown in Figure 3, the error of given speed and detection speed e=V*-V and error rate of change As the input variable of the system, the output value of the system is the reference torque of the inner loop of the torque, thus forming a single output-two input fuzzy controller. When the input quantity is fuzzified, the fuzzy controller will approximate the error of some values to 0, so there will be a dead zone of control, it is difficult to achieve good control precision, and the system will always have static static difference. Therefore, the system designed in this paper switches to neural network control at a time when the speed error is relatively small. In order to effectively control the fuzzy control blind zone, the system also added the integral link. The switching conditions of the two control modes are determined by the magnitude of the velocity error.
