Abstract: In view of the problems of response lag and error deviation of downhole gas turbine flowmeter under the dynamic working condition in Sihe Coal Mine,a dynamic error compensation algorithm based on edge computing is proposed.This algorithm combines nonlinear system modeling and edge intelligent theory,integrates Kalman filter and LSTM neural network model,and introduces adaptive weight adjustment mechanism to solve the dynamic error problems in complex working conditions such as sudden increase of flow velocity,vibration disturbance,high temperature and humidity and so on.This algorithm is deployed on the intrinsically safe industrial edge computing node to achieve real-time computing and dynamic correction with low delay and high reliability.The experimental results show that,compared with the traditional linear compensation method,the proposed algorithm decreases the average relative error by over 65% and shortens the response time to within 350 ms under the typical working condition,significantly improving the real-time and robustness of the monitoring system,providing a new technical path for error control of mine gas flow monitoring,and exploring a feasible implementation scheme for the application of edge computing in the field of industrial measurement and control.