Transmission line operation environment is complex, prone to tilt collapse accident, so a transmission tower tilt real-time early warning model based on multi-sensor data is established to judge whether the tower is stable operation by detecting the tilt state of the early warning tower. The pressure and inclination sensors are deployed at different positions of the transmission tower to collect the inclination and stress of the tower in real time, and transmit them to the remote monitoring terminal through the wireless network to send out an alarm. The operators can timely adjust according to the alarm situation to maintain the safe operation of the transmission line. The experimental results show that the model can realize the real-time warning of transmission tower tilt, the measurement accuracy can meet the needs of comprehensive detection of tower state, and the application can effectively ensure the safety of staff and reduce the work intensity.