Design Of The Attitude Monitoring System Of Hydraulic Support Based On Multi-sensor Fusion Technology

The working posture of the hydraulic support has an important influence on the efficiency and safety of underground mining,remote monitoring of the working posture of the hydraulic support is a key link in the realization of intelligent and unmanned fully mechanized mining face.This article addresses the problems of the current attitude monitoring system of hydraulic supports,such as low accuracy of collected attitude parameters and low real-time data transmission,the design of the attitude monitoring system of hydraulic support based on multi-sensor fusion technology has theoretical research and practical application significance.This article is based on the theory of multi-sensor fusion,the overall architecture of the hydraulic support monitoring system is built,and the software and hardware design of the data acquisition,data transmission,upper computer and other modules in the architecture are carried out.The support height and pitch angle of the hydraulic support are fused and solved through the attitude calculation model.In the solution of the support height,in order to find a high-precision solution algorithm,simulation and comparison experiments were carried out on batch estimation fusion,optimal weighted fusion and BP neural network fusion algorithms.The results showed that the data processing accuracy of BP neural network with higher fusion level high,but it has problems such as unstable training results and easy to fall into local extreme values.For this reason,this thesis introduces particle swarm algorithm and genetic algorithm to improve it,which improves the stability of the system and reduces the data fusion error by 52.1%.Further analysis shows that the individual sorting speed of particle swarms is too slow.This thesis introduces a radix sorting algorithm suitable for binary sorting to optimize the sorting speed.In the fusion solution of the pitch angle,a complementary filter fusion algorithm is used to fuse the pitch angle data collected by the gyroscope and accelerometer.However,the complementary filter fusion algorithm has the problem that the parameters of the PI controller cannot be adjusted with the change of the carrier motion state.For this reason,this thesis proposes an adaptive complementary filtering fusion algorithm to adjust the proportional parameters according to the motion state of the hydraulic support.Simulation experiments show that the pitch angle data processed by the adaptive complementary filtering algorithm has a 41.4% increase in accuracy compared with the complementary filtering algorithm.In summary,the hydraulic support attitude monitoring system designed in this thesis meets the design requirements in terms of accuracy and real-time performance.