Artificial neural networks for tuning magnetic field of colour cathode ray tube deflection Yoke

This paper presents artificial neural networks (ANN) using for tuning magnetic fields of deflection yokes (DY). The method designed to identify the number of ferroelastic correction shunts and their position and also metal shunts position for deflection yoke tuning to correct residual misconvergence of colours of cathode ray tube. The method consists of two phases: learning and operating. The learning phase is executed only once when the system is adapted to correct the misconvergence for deflection yokes of given type. In the operating phase, the trained neural networks are used to predict changes in misconvergence depending on correction shunt position. The deflection yoke is tuned correctly if 18 primary and 4 secondary parameters fall inside given intervals. During the experimental investigation, 98% of deflection yokes analyzed have been tuned correctly. The software developed is easy adapted for deflection yokes of different types by training neural networks used.