Abstract: With the improvement of intelligence and automation in power equipment, the safety issues of power equipment are also receiving increasing attention. In order to improve the safety performance of power equipment, an active sound and light safety classification warning device based on behavior prediction analysis is studied and designed. This device can predict potential safety issues by analyzing the predicted characteristics and safety risks of the surrounding environment of power equipment and the behavior of nearby personnel. And timely issue sound and light warnings to remind operators to take corresponding measures to prevent accidents from occurring.