Performance Optimization Analysis of Partial Discharge Detection Manipulator Based on STPSO-BP and CM-SA Algorithms

Area of Science:

• Robotics and Automation
• Electrical Engineering
• Artificial Intelligence

Background:

• Partial discharge (PD) detection in high-voltage switchgear is critical for grid reliability.
• Six-degree-of-freedom (6-DOF) manipulators face challenges in PD detection due to inverse kinematics (IK) redundancy and lack of synergistic optimization.
• Existing methods struggle to balance end-effector positioning accuracy with energy efficiency.

Purpose of the Study:

• To develop an adaptive dual-layer optimization model for 6-DOF manipulators in PD detection.
• To address the challenges of IK solution redundancy and synergistic optimization of accuracy and energy consumption.
• To enhance the performance of robotic systems in critical electrical infrastructure inspection.

Main Methods:

• A dual-layer adaptive optimization model integrating spatio-temporal correlation particle memory-based particle swarm optimization BP neural network (STPSO-BP) and chaotic mapping-based simulated annealing (CM-SA).
• The first layer uses STPSO-BP with long short-term memory (LSTM) for enhanced IK, improving positioning accuracy and adaptability.
• The second layer employs CM-SA with chaotic joint angle constraints and dynamic adjustments for collaborative optimization of energy consumption and positioning error, using cubic spline interpolation for trajectory smoothing.

Main Results:

• Positioning error reduced by 68.9% compared to traditional BP neural network algorithms.
• Energy consumption decreased by 60.18% relative to the pre-optimization state.
• The proposed model demonstrates significant synergistic optimization of accuracy and energy efficiency.

Conclusions:

• The developed dual-layer model offers an innovative solution for synergistic accuracy-energy control in 6-DOF manipulators for PD detection.
• This approach significantly enhances the efficiency and effectiveness of robotic inspection in high-voltage switchgear.
• The findings contribute to advancing robotic applications in critical electrical infrastructure maintenance and safety.