Intelligent Lead Fish Design Based On Kalman Filter

At present in the field of hydrology,the cableway measurement still plays an indispensable role because of its low cost and strong reliability.In order to measure the velocity and depth of the rivers,it is usually used to measure the flow velocity of lead fish with rotor current meter,and to measure the depth of water by calculat ing the length of water entering the cable.However,it is impossible to judge whether the pose of lead fish is stable under the impact of water flow.When the velocity is large,the angle of the cable is constantly changing,and the depth of the water entry can not be measured accurately,which affects the accuracy of the flow and flow measurem ent.In order to realize the accurate measurement of flow velocity and water depth,a method of improving suspension of lead fish and the suspension angle adjustment of lead fish is commonly used.The lead fish in parallel with the direction of water flow,while a certain extent improve the rotor velocity measurement precision.But the measurement personnel on-site operation experience requirements is too high,the accuracy improvement is limited.For this problem,installing sensors on lead fish is a good treatment plan at present.It not only improves the accuracy of depth measurement,but also timely feedback the real-time posture of underwater lead fish and it has the advantages of small volume,low cost and convenient installation.Using the Kalman filtering algorithm and combining the characteristics of MEMS sensor,the original data are filtered and fused,and the optimal attitude angle is obtained.The specific research works are as follows:First of all,aiming at the shortcomings of traditional Hydrometric measurement such as low accuracy and complicated operation process,a new intelligent signal monitoring bucket for intelligent plumb hydrology information is designed accordi ng to actual needs.The monitoring bucket is installed on the traditional fish gauge and the attitude data of the underwater lead fish are collected in real time by the built-in nine axis posture sensor so as to judge whether the current lead fish posture is stable and record the direction of the water flow.The high precision absolute pressure sensor is used to collect water pressure data and the real-time water entering depth of lead fish is calculated according to the pressure value.At the same time,the pressure sensor is equipped with temperature sensor so that we can accurately measure the informat ion of velocity,depth,direction and temperature of the rivers.Aiming at the existing problems of large random drift component data which hydrological information monitoring sensor gets,I propose to use the Kalman filt er algorithm to filter the sensor measurement noise.The basic Kalman filter(KF)and the extended Kalman filter(EKF)is extended by filtering and data fusion of attitude sensor measured data,through the experimental comparison of two filtering effect filtering methods,it is concluded that the method effectively improves the accuracy of attitude estimation of underwater lead.At the same time,aiming at the problem of noise measurement error of pressure sensor,by using Kalman filtering to deal with the pressure data and using the principle of surge filter algorithm and the Bernoull i equation,based on the attitude angle to correct the error of the pressure value messurement under the dynamic conditions caused by the installation posit ion deviation and the flow rate of water to obtain more accurate water depth data.