[제어공학]Generalised Predictive Control for DC motor (예측 제어)

목차

1. Introduction
1) Background
2)Experiment method

2. Experiment result and discussion
1) system identification using leasr square method
2)GPC using CARIMA model
3)system idntification using step input
4)GPC using step response model

3. Conclusion

4. Bibliography

본문내용

Aim of this experiment is to control the DC motor in MATLab simulation using generalized predictive control
(GPC) [1] The DC motor is defined in form of systemXXX.p file to show that the system is unknown. The objective was
to ; (1) use least square method to identify the model of the system in CARIMA form. (2) Design GPC and change the
tuning parameter to inspect the effect of each parameter. (3) Identify the model of the system in step response
model. (4) implement GPC to the model and compare the difference with CARIMA control model.

2. Experiment result and discussion
2.1 System Identification using least square method
This section refers to the file GPC_task1.m [2]. In code, it executed the procedure of least square method to
find the Mean Square error for five times, then, averaged the data up. The result is shown in mesh grid provided
Figure 1.

Figure 1 shows the square error in different order of input and output. The data where order of input is lager
than order of output is zero.

참고 자료

없음