High-Precision Automatic Crimping Technology for Cable Splice Joints based on Artificial Intelligence and Multi-Sensor Assistance

This study aims to develop a high-precision automatic crimping technology for cable middle joints, using AI-assisted multi-sensor data fusion. A multi-sensor system, consisting of temperature, pressure, and displacement sensors, was designed to collect key parameters in the crimping process. The data was processed through filtering, denoising, and normalization to provide reliable input for the AI model. YOLOv5, a deep learning-based object detection model, was optimized and trained to detect key features of the crimping process. Experimental results show that the system can monitor the crimping status in real-time, make intelligent adjustments based on detected anomalies, and ensure high crimping quality. The proposed method improves the automation and precision of the crimping process, making it more reliable and efficient than traditional manual methods. The novelty of this paper lies in integrating multi-sensor fusion and AI to achieve adaptive control and precise crimping quality monitoring.