Kick Risk Forecasting and Evaluating During Drilling Based on Autoregressive Integrated Moving Average Model

Yin, Hu; Si, Menghan; Li, Qian; Zhang, Jinke; Dai, Liming

doi:10.3390/en12183540

Cited by 10 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) Obtaining intrusive gas transportation velocity within the wellbore under shut-in conditions Bubble cluster transport velocity experiments were conducted using various solutions under shut-in conditions, following the same experimental procedure as described in the experiment (1). Subsequently, the microporous aerator was replaced with a 0.5 mm diameter nozzle, and experiments were conducted to measure the transport velocity of individual bubbles under various solution properties.…”

Section: Methodsmentioning

confidence: 99%

“…In the process of offshore drilling, if gas invasion and overflow are not detected in time, a blowout can rapidly occur. Influx fluids tend to have a high pressure; a gas kick is much more detrimental than a liquid kick due to gas expansion and, as a result, has a higher variation in pressure [1]. To ensure the safety of drilling operations and facilitate the rapid progress of drilling, the development of a precise model for gas intrusion rates and bubble transportation is of significant importance.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Study on the Mechanism of Gas Intrusion and Its Transportation in Wellbore Under Shut-In Conditions

Zhu,

Xiang,

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper presents a comprehensive study based on multiphase seepage and wellbore multiphase flow theories. It establishes a gas intrusion rate calculation model that considers various factors including formation pore permeability, bottomhole pressure difference, drilling fluid rheology, and surface tension. Experiments were conducted to investigate the gas intrusion mechanism under shut-in conditions, and the experimental results were employed to validate the reliability of the proposed gas intrusion rate calculation method. Furthermore, the research explores the transportation rates of single bubbles and bubble clusters in drilling fluid under shut-in conditions. Meanwhile, empirical expressions were derived for the drag coefficient for single bubbles and bubble clusters in the wellbore. These expressions can be used to calculate gas transportation rates for various equivalent radii of single bubbles and bubble clusters. Additionally, a method was developed for calculating the rising velocity of bubble clusters in water based on experimental results. The study reveals that the average bubble size in the bubble cluster is significantly smaller than the size of the single bubble generated from the orifice. When the viscosity of the drilling fluid is low, the bubble cluster transportation velocity exhibits a positive correlation with the average bubble diameter. When the average bubble diameter exceeds 1 mm, the bubble velocity no longer varies with the change in the bubble cluster diameter. The initial bubble size of intrusive gas, the transportation speed of intrusive gas in the wellbore, the gas intrusion rate, and variations of the wellbore pressure after gas intrusion were analyzed. The research results provide theoretical support for wellbore pressure prediction and pressure control under shutdown conditions.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Study on the Mechanism of Gas Intrusion and Its Transportation in Wellbore Under Shut-In Conditions

Zhu,

Xiang,

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Different input parameters have been tested which include majorly prior highlighted primary and secondary kick indicators. Different models have been tested such as Bayesian classifier, decision tree, k-nearest neighbor, random forest, support vector machine, different neural networks, and autoregressive models [31,32]. Extensive gas-kick datasets were generated autonomously via 108 tests from a pilot-scale test well experimental setup equipped with a complete drilling system and a comprehensive mud logging system for surface monitoring of relevant drilling and geological parameters complimented with Doppler wave sensors just above the BOP for riser monitoring of gas migration and downhole pressure monitoring via pressure gauges.…”

Section: The Use Of Numerical Modeling and Machine Learning To Aid Ea...mentioning

confidence: 99%

Advances in Well Control: Early Kick Detection and Automated Control Systems

Olamigoke¹,

James²

2022

Drilling Engineering and Technology - Recent Advances New Perspectives and Applications

View full text Add to dashboard Cite

The devastating impact of the Macondo blowout incident has underscored the need for effective well barriers, early kick detection systems, and increased automation of well-control operations toward successful drilling and completion operations particularly in deep offshore environments. Early kick detection systems should be capable of detecting a gas influx both during drilling and tripping operations regardless of the drilling fluid system with minimal false-negative alarms, while automated control systems regain well-control eliminating delays or omissions due to human error. In this chapter, developments in the deployment of early kick detection and automated control systems in conventional and managed pressure drilling operations are reviewed. We discuss the use and placement of surface sensors such as the Coriolis flowmeter, smart flowback fingerprinting when the rig pumps are off, real-time gas monitoring along the marine riser and downhole measurements complimented with machine learning algorithms for early kick detection. We then focus on the application of automated well-control systems for managed pressure drilling operations for which gas kicks are circulated without stopping the pumps or shutting in the well and in conventional well operations requiring intelligent tool joint space-out prior to well shut in especially for deep offshore operations.

show abstract

“…As A, B, C, and D exponential logarithmic yield sequence have the feature of thick tail, it indicates that the T distribution has an obvious advantage in describing the yield sequence with peak and thick tail distribution feature. In order to eliminate autocorrelation, autoregressive moving average (ARMA) model [26] is used for data processing in this study. Compared with the GARCH model, the generalised autoregressive score (GAS) model can make full use of the density function, so it is adopted to overcome the heteroscedasticity in the data [27].…”

Section: Data Preprocessing and Edge Distribution Determinationmentioning

confidence: 99%

Cognitive Computational Model Using Machine Learning Algorithm in Artificial Intelligence Environment

Liu

Khadidos

Abdulrazzqa

2021

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

In order to accurately describe the risk dependence structure and correlation between financial variables, carry out scientific financial risk assessment, and provide the basis for accurate financial decision-making, first the basic theory of Copula function is established and the mixed Copula model is constructed. Then the hybrid Copula model is nested in a hidden Markov model (HMM), the risk dependences among banking, insurance, securities and trust industries are analysed, and the Copula–Garch model is constructed for empirical analysis of investment portfolio. Finally, the deep learning Markov model is adopted to predict the financial index. The results show that the mixed Copula model based on HMM is more effective than the single Copula and the mixed Copula models. The empirical structure shows that among the four major financial industries in China, the banking and insurance industries have strong interdependence and high probability of risk contagion. The investment failure rate under 95%, 97.5% and 99% confidence intervals calculated by Copula–Garch model are 4.53%, 2.17% and 1.08%, respectively. Moreover, the errors of deep learning Markov model in stock price prediction of Shanghai Pudong Development Bank (sh600000), Guizhou Moutai (sh600519) and China Ping An Insurance (sh601318) are 2.56%, 2.98% and 3.56% respectively, which indicates that the four major financial industries in China have strong interdependence and risk contagion, so that the macro or systemic risks may arise, and the deep-learning Markov model can be adopted to predict the stock prices.

show abstract

Kick Risk Forecasting and Evaluating During Drilling Based on Autoregressive Integrated Moving Average Model

Cited by 10 publications

References 25 publications

Study on the Mechanism of Gas Intrusion and Its Transportation in Wellbore Under Shut-In Conditions

Study on the Mechanism of Gas Intrusion and Its Transportation in Wellbore Under Shut-In Conditions

Advances in Well Control: Early Kick Detection and Automated Control Systems

Cognitive Computational Model Using Machine Learning Algorithm in Artificial Intelligence Environment

Contact Info

Product

Resources

About