Table of Contents
Particularly, the Distinction block computes the motor's change in position (in counts) and the first Gain block divides by the sample time. Subsequent Gain obstructs transform the units from counts/sec to revolutions/sec, and after that from revolutions/sec to revolutions/min. The continuous representing the equipment ratio requires to be defined in the MATLAB office prior to the design can be run.
Decreasing the length of the simulation then running the design produces the list below output for motor speed in RPM. Taking a look at the above, we can see that the estimate for motor speed is quite loud. This develops for a number of factors: the speed of the motor is really differing, encoder counts are being occasionally missed, the timing at which the board is polled does not precisely match the recommended sampling time, and there is quantization associated with checking out the encoder.
Consider the following design with a basic first-order filter added to the motor speed estimate. This model can be downloaded here. Running this model with the sample time increased to 0. 05 seconds and a filter time constant of 0. 15 seconds produces the following time trace for the motor speed.
05; filter_constant = 0. 15;. By increasing the sampling duration and adding the filter, the speed quote undoubtedly is much less noisy. This is particularly helpful for enhancing the estimate of the motor's speed when it is performing at a consistent speed. A disadvantage of the filtering, nevertheless, is that it adds hold-up.
In essence we have lost information about the motor's real action. In this case, this makes identifying a model for the motor more difficult. In the case of feedback control, this lag can break down the efficiency of the closed-loop system. Decreasing the time continuous of the filter will lower this lag, however the tradeoff is that the sound will not be filtered too.
Thinking about that our input is a 6-Volt step, the observed response appears to have the kind of a first-order step reaction. Taking a look at the filtered speed, the DC gain for the system is then around 170 RPM/ 6 Volts 28 RPM/V. In order to approximate the time continuous, however, we require lower the filtering in order to much better see the real speed of the motor.
01 seconds, we get the following speed response. Remembering that a time constant defines the time it takes a procedure to accomplish 63. 2% of its total modification, we can estimate the time constant from the above chart. We will attempt to "eye-ball" a fitted line to the motor's response chart.
Presuming the very same steady-state performance observed in the more greatly filtered data, we can estimate the time consistent based upon the time it takes the motor speed to reach RPM. Since this appears to take place at 1. 06 seconds and the input appears to step at 1. 02 seconds, we can approximate the motor's time consistent to be roughly 0.
For that reason, our blackbox design for the motor is the following. (2) Remembering the model of the motor we obtained from very first principles, duplicated listed below. We can see that we anticipated a second-order design, however the response looks more like a first-order model. The description is that the motor is overdamped (poles are genuine) and that one of the poles dominates the response.
( 3) In addition to the fact that our design is reduced-order, the design is a more approximation of the real life in that it ignores nonlinear elements of the real physical motor. Based upon our direct design, the motor's output need to scale with inputs of various magnitudes. For example, the action of the motor to a 6-Volt step should have the very same shape as its action to a 1-V action, simply scaled by an element of 6.
This is because of the stiction in the motor. If the motor torque isn't large enough, the motor can not "break free" of the stiction. ייצור מכונות לתעשייה. This nonlinear habits is not recorded in our model. Typically, we use a viscous friction design that is linearly proportional to speed, instead of a Coulomb friction design that captures this stiction.
You could then compare the predictive capability of the physics-based design to the blackbox model. Another workout would be to produce a blackbox design for the motor based upon its frequency response, comparable to what was finished with the increase converter in Activity 5b. A benefit of using a frequency response technique to recognition is that it enables recognition of the non-dominant dynamics.
In Part (b) of this activity, we create a PI controller for the motor.
Table of Contents
Find Out More About Propulsion Systems