|
CSME 2024/12
Volume 45 No.6
:
605-618
|
|
Optimization of Control Parameters for Robot Integrated Joint Modules Based on an MA-BPNN Model
Qiu-Shi Hu a, Heng Li* b, Hao-Qiang Li b, Guo-Wei Wang b and Lei Li b
aSchool of Mechanical Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China.
bSchool of Mechanical Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China.
|
Abstract:
This study introduces a method to predict the
responsiveness performance and optimize the control
parameters of joint modules. An integrated joint
module was created to assess overshoot and response
time under various operating conditions and control
parameter values. The results reveal nonlinear impact
of conditions and parameters on the responsiveness
performance, indicating a complex multifactorial
issue. An MA-BPNN model was then established by
merging the BPNN model with the MA algorithm,
offering improved prediction accuracy and
computational efficiency compared to traditional
models. The application of Latin hypercube sampling
and the MA-BPNN model yielded the Pareto optimal
solutions for the control parameters of the joint
module under commonly known operating conditions.
Notably, higher load and joint speed values require a
suitably higher proportional gain coefficient in order
to meet the system demands. These findings hold
significant value for the design and optimization of
actual joint modules.
|
|
Keywords: joint modules, responsiveness performance, control parameters, back-propagation neural network.
|
|
|
*Corresponding author; e-mail:
|
©
2024
CSME , ISSN 0257-9731
|