Research Of Key Technologies Of MGWD Navigation System In Harsh Multilateral Sidetracks Horizontal Well Drilling Environments  Posted on:20171027  Degree:Doctor  Type:Dissertation  Country:China  Candidate:T Li  Full Text:PDF  GTID:1318330542972203  Subject:Navigation, guidance and control  Abstract/Summary:  PDF Full Text Request  Horizontal drilling(HD)technology arises wide attention from home and abroad and achieves tremendous economic value since its application in well drilling industry.The multilateral sidetracks horizontal well drilling(MSHWD)as an effective way to contact the marginal wells and revive the old wells for improving the production rates of the oil and gas.The MSHWD employs numerous lateral sidetracks radiated from the main wellbore to target the spaced reservoirs.The emerging of high pressure waterjet drilling(HPWD)technology promotes the development of MSHWD because conventional large rotary drill bit can be replaced by a small diameter(less than 24 millimeters)high pressure waterjet drill bit.The difficult problem of MSHWD is how to obtain navigation parameters of high pressure water jet drill bit information.Traditional MWD navigation system employs 3axis magnetometers and 3axis accelerometer to obtain the attitude of the BHA(BottomHoleAssembly),however,the magnetometers would increase the azimuth error and even loss the information of azimuth of the navigation system under strong magnetic interferences.The nonmagnetic drill collar can eliminate the external magnetic interference,however,the magnetometer still cannot meet the requirement of the highprecision MWD because of the ferromagnetic materials in the drill bit and the BHA.According to the above analysis,the gyroscopebased MWD(GWD)was proposed to continuously provide the attitude,velocity and position.The GWD can save considerable rig time and remove the blind during the kickoff.The GWD tools can evaluate the inclination of the drill bit in all directions compared with the traditional magnetometer based MWD tools without the problem of singular value.However,the large size of the fiber optic gyroscopes,laser optic gyroscopes and electrostatic gyroscopes limits their applications in the small diameter downhole applications.With the development of the microelectromechanical system(MEMS),the MEMS gyroscopes are brought into the field of high pressure waterjet MSHWD.Conventional MEMS inertial measurement unit is composed of threeaxis gyroscope and threeaxis accelerometer,but the diameter of the drill pipe of the sidetrack is less than 24 mm in the MSHWD.Itâ€™s hard to buy the small size of the gyroscope on the market.In addition,the harsh environments(high temperature,high pressure)increase the design difficulties of electronic components in the MGWD.Based on the above analysis,we use high temperature resistance,high voltage and miniature MEMS sensors to design the miniature gyroscope based measurement while drilling(MGWD)navigation system which mades up of twoaxis gyroscope and threeaxis accelerometer.The rotation angular rate MGWD navigation system in other axis is calculated by the measurements of accelerometer.The inertial navigation system is characteristiced by the high precision in short time and real time data transmission,however,due to gyroscope and accelerometer errors of inertial navigation solution makes integration algorithm causes the navigation error accumulation increased.The errors of inertial sensors are composed of two parts: deterministic errors and stochastic errors.The deterministic errors can partially be reduced by the calibration approach.Therefore,the residuals of deterministic errors and the stochastic errors are main error sources of inertial sensors.The stochastic errors can be modelled by firstorder Markov model,however,the result of the autocorrelation calculation for the stochastic errors of the MEMS sensors in MGWD does not meet the curve of the firstorder Markov model.Therefore,the fourthorder linear autoregressive(AR)model was proposed for the stochastic errors of the MEMS sensors,and then the fourthorder nonlinear AR model is proposed for the stochastic errors of MEMS sensors in harsh drilling environments.The coefficients of the nonlinear error model solved by linear least square(LMS)algorithm are not accurate because the errors of data matrix is neglected in LMS.This paper proposes the total least square algorithm to solve the coefficients of the nonlinear model of the stochastic errors and compares the estimated the navigation parameters between the linear AR model and the nonlinear AR model.The derivation of the linear error model of the strapdown inertial navigation system(SINS)such as the Psiangle error model and Phiangle error model is based on the assumption that the errors of the attitude so small that the direction cosine matrix can be approximated or simplified by the smallangle attitude errors.Therefore,the modified nonlinear error model of the MGWD system is proposed under the largeangle attitude errors conditions.The Kalman filter is suboptimal because the neglected error terms in the linear error model would introduce errors for the filter and the system noises and the measurements noises in MGWD are not Gaussian.The particle filter uses the random particles and the corresponding weights to represent the posterior probability density function of the system states.The errors caused by the external disturbance and the error residuals of the MGWD system accumulate with time and traditional indrilling alignment(IDA)and rotaryin drilling alignment(RIDA)cannot meet the design requirements in high pressure waterjet MSHWD.Terefore,the helixindrilling alignment and calibration technology is proposed to compensate the errors aforementioned.The symmetrical constant errors can be eliminated by the rotary movement and the reciprocating movement of the helixindrilling alignment and calibration technology can compensate the symmetrical constant errors and the system accumulated errors respectively.  Keywords/Search Tags:  MGWD, MSHWD, nonlinear AR model, nonlinear error model, helixindrilling alignment  PDF Full Text Request  Related items 
 
