Research On The Interpretation Method Of Complex Conditions In Managed Pressure Drilling

Oil and gas development in deep complicated formation has become a new pivot for increasing of oil and gas production in China.However,the geological features of deep complicated formation lead to the frequent occurance of complex conditions in the drilling process.Managed pressure drilling(MPD),based on information monitoring and interpretation of complex conditions,and aiming at wellbore pressure control,can effectively reduce downhole complexities and has a broad application prospect in deep complicated formation.The key points to realize pressure control in wellbore lies on rapid identification,classification and quantitative interpretation of complex conditions.Overflow and loss are the most common complexities in MPD.Meanwhile,drilling tool failures also occur from time to time in the drilling of deep wells.Inadequate ability to interprete downhole complexities is the main reason for the ineffective control of overflow and loss,and the improper WOB imposed on bit is an important factor to induce drilling tool failure.Therefore,this paper focuses on overflow,loss,drilling tool failure and WOB calibration to carry out the following research works.I.Research on the detection method of downhole complex conditions in MPD.To accurately interprete complex conditions,it is necessary to be able to detect complex conditions.Therefore,a new method for early detection of complex conditions in MPD has been proposed.Based on the drift flow model,a transient gas-liquid two-phase pressure calculation model is established considering the variation of drilling fluid parameters with temperature and pressure.The sensitive characteristic parameters of early gas kick detection are optimized by numerical simulation,and a gas kick detection method based on UKF parameter estimation is proposed.Then the detection method is extended to the detection of loss,drillstring washout and other complex conditions to form a new early detection method for complex conditions in MPD.Numerical simulation and experiment show that the new early detection method for complex conditions in MPD can detect downhole complex conditions in time and realize early warning.II.Research on the classification of different complex conditions.In order to ensure drilling safety,it is necessary to further realize the classification of complex conditions according to the detection results.Therefore,classification models of downhole complex conditions in MPD have been established.The UKF-BP classification model is established based on BP neural network theory,and the UKF-NNE and UKF-RF classification models are established based on integrated learning.The above models take pressure loss correction factor and flow rate correction factor as input features to quantitatively characterize the variation trends of feature parameters in the short-term evolution process of downhole complex conditions.Studies show that all the UKF-BP,UKF-NNE and UKF-RF models can well classify complex conditions of accident data at different well depth,while,the UKF-NNE and UKF-RF models have a higher recognition rate of complex conditions when considering the variation of drilling fluid parameters with temperature and pressure.III.Research on the quantitative interpretation of complex conditions.After determining the type of complex condition,it is also necessary to make a quantitative interpretation of the complex condition.Therefore,a inversion method of complex condition parameters in MPD has been proposed.Based on the UKF inversion theory and combined with the characteristics of MPD,a inversion model of complex condition parameters is established,which consideres the influence of temperature and pressure.Studies show that relying on bottomhole annular pressure,standpipe pressure and annular outlet flow rate can ensure the accurate inversion of loss position/washout position and loss rate/washout rate for one point loss or washout.For two point loss condition,the double loss position and double loss rate are used as inversion parameters firstly.Theoretical analysis and numerical simulation show that the double loss position and double loss rate can be inversed simultaneously by using the downhole three annular pressure and the annular outlet flow rate.IV.Research on WOB calibration.According to the mechanical state of BHA in hydraulic environment,the factors influencing WOB measurement are analyzed in detail and the calibration model of WOB measurement data based on strain principle has been established.This model not only considers the influence of internal pressure and external pressure,but also considers the influence of friction force of drilling fluid and BHA vibration.The axial internal force coefficient and the internal pressure coefficient in the correction model are determined by the axial load experiment and the internal pressure experiment respectively.The external pressure coefficient can be obtained by fitting the measured data of field experiment.Field experiments show that the bit WOB obtained from the WOB measurement data can accurately reflect the real working condition on site.