Deepwater blowout hidden danger identification and risk analysis method based on historical data

Abstract

The invention discloses a deep water blowout hidden danger identification and risk analysis method based on historical data, and the analysis method comprises the steps: S1, extracting and artificially complementing key information of a blowout accident, and obtaining blowout accident data; s2, hidden danger keywords in the accident data are extracted and selected, classification is carried out according to hidden danger features, working condition stages and well control stages, and a hidden danger list is obtained; s3, establishing a fault tree model from three aspects of a well drilling stage, a well completion stage and a well repair stage; s4, establishing a Bayesian network model, and performing node probability allocation based on historical data; s5, target time period data are selected, the posterior probability of blowout risk factors is calculated through GeNIe, and a Bayesian network diagram is drawn; and S6, identifying main risk inducements and formulating a risk control scheme. And formulating a risk control scheme to reduce the well control risk.

Description

Technical Field

[0001] The invention relates to the technical field of petroleum engineering, and in particular to a method for identifying hidden dangers of deepwater blowout and analyzing risks based on historical data. Background Art

[0002] Due to the continuous increase in my country's demand for oil and gas resources and the limitation of total land oil and gas production, the development of deepwater oil and gas resources will be an important guarantee for the sustainable development of my country's oil industry in the future. The marine environment where deepwater oil and gas resources are located is often complex and uncertain. Therefore, uncontrolled blowout is one of the most catastrophic accidents in the process of deepwater oil and gas exploration and development. When a blowout occurs, the wellhead and offshore drilling platform may be full of oil and natural gas in a very short time. If there is an open flame and sparks generated by high-speed jets hitting the drilling platform equipment, it is very likely to cause a fire or even an explosion. The uncontrolled blowout of an offshore drilling platform has strong heat radiation, a short time, and few rescue means. Even if the blowout is uncontrolled and does not catch fire or the fire source is small, the ejected fluid will cause huge pollution to the ocean and atmospheric environment, affect the ecology, and cause damage to the health of platform workers.

[0003] At present, the development of deepwater oil and gas wells in my country is still in its early stages. Although my country's marine exploration has made rapid progress in recent years, oil and gas extraction technology has been continuously improved, and the initial application of mechanized, informationized and automated drilling equipment has improved the safety assurance capabilities of oil and gas extraction to a certain extent, deepwater blowout accidents are still the most serious, most worrying and least desirable accidents that may occur in offshore drilling operations.

[0004] According to Heinrich's Law: Behind every casualty accident, there are 29 times more minor injuries, 300 times more non-injury accidents, and thousands of potential accidents that may lead to accidents. Therefore, in order to control the final occurrence of accidents, it is necessary to start with the investigation and treatment of hidden accident hazards in deepwater oil and gas well projects. At present, the investigation of hidden dangers in deepwater drilling is mainly based on the monitoring and regular inspection of major blowout risk components by relevant drilling personnel. The hidden dangers that are focused on are subjective on the one hand, and lack comprehensiveness on the other hand. Therefore, a hidden danger investigation method based on facts is urgently needed in the deepwater drilling industry.

[0005] On the other hand, the current accident risk assessment strategy for deepwater oil and gas wells is often to statically evaluate a certain accident that has occurred through different mathematical models and algorithms. These evaluations lack both systematicity and universality in actual performance. Deepwater blowout accidents that occur in reality often have various causes. Relying solely on a single accident risk assessment experience and conclusion will not only fail to clearly and accurately determine the source of danger, but will also delay the best emergency rescue opportunity when the accident cannot be avoided. In-depth understanding of the structure and laws of hidden dangers of deepwater oil and gas well blowout accidents will help us collect and analyze data on accident hazards found in typical deepwater blowouts and production safety inspections. Establishing a scientific risk assessment model based on relevant data can obtain more universal laws, discover safety weaknesses in deepwater drilling, and avoid blowout accidents from the root. Summary of the invention

[0006] In view of the above problems, the present invention is proposed to provide a deepwater blowout hazard identification and risk analysis method based on historical data that overcomes the above problems or at least partially solves the above problems.

[0007] According to one aspect of the present invention, a method for identifying and analyzing the hidden dangers of deepwater blowouts based on historical data is provided, and the analysis method comprises:

[0008] Step S1: Based on the existing blowout accident database, extract and manually complete the key information of the blowout accident to obtain the blowout accident data;

[0009] Step S2: identifying hidden dangers one by one for the blowout accident data, extracting and selecting hidden danger keywords in the accident data, classifying them according to hidden danger characteristics, working condition stage and well control stage, and obtaining a hidden danger list;

[0010] Step S3: Using the hidden danger list as risk factors, a fault tree model is established from three perspectives: drilling stage, completion stage and workover stage;

[0011] Step S4: transforming the fault tree model, establishing a Bayesian network model, and performing node probability allocation based on historical data;

[0012] Step S5: Select the target time period data, calculate the prior probability of each risk factor according to the statistical probability method, calculate the posterior probability of the blowout risk factor using GeNIe and draw a Bayesian network diagram;

[0013] Step S6: Identify the main risk factors and develop risk control plans.

[0014] Optionally, the step S1: based on the existing blowout accident database, extracting and manually completing key information of the blowout accident, and obtaining the blowout accident data specifically includes:

[0015] The key information of the blowout accident is sorted out from six dimensions: location information, well conditions, operating conditions, blowout causes, blowout characteristics and remaining data.

[0016] Optionally, the step S2: identifying hidden dangers one by one for the blowout accident data, extracting and selecting hidden danger keywords in the accident data, classifying them according to hidden danger characteristics, working condition stage and well control stage, and obtaining a hidden danger list specifically includes:

[0017] When identifying hidden danger keywords, retain hidden danger keywords that appear twice or more, and selectively retain hidden danger keywords that appear once;

[0018] According to the characteristics of hidden dangers, hidden dangers are classified into drilling system hidden dangers and residual hidden dangers. According to the working stage, drilling system hidden dangers are divided into drilling stage hidden dangers, completion stage hidden dangers and workover stage hidden dangers. According to the well control stage, drilling system hidden dangers are divided into primary well control hidden dangers and secondary well control hidden dangers.

[0019] Optionally, the remaining hidden dangers include external risks such as typhoons, tsunamis, external impacts, and human operational errors and management defects.

[0020] Optionally, the step S4: transforming the fault tree model, establishing a Bayesian network model, and performing node probability allocation based on historical data specifically includes:

[0021] The fault tree model is converted into a Bayesian network model by a structural mapping method, wherein the top-level events of the fault tree model correspond to the sub-nodes of the Bayesian network model;

[0022] The bottom event of the fault tree model corresponds to the parent node of the Bayesian network model, and the logic gates such as AND gates and OR gates correspond to the conditional probabilities of the Bayesian network model.

[0023] Optionally, the calculating the prior probability of each risk factor according to the statistical probability method specifically includes:

[0024]

[0025] Where P(B) is the prior probability of the risk factor; N f is the number of accidents; N w is the total number of operations.

[0026] Optionally, the posterior probability of the risk factor is obtained by using the existing prior probability and conditional probability, and the calculation formula is as follows:

[0027]

[0028] In the formula, P(A i |B) is the prior probability of the risk factor; P(BAi ) is the conditional probability of the corresponding node.

[0029] Optionally, the identifying of major risk factors specifically includes:

[0030] The difference between the prior probability and the posterior probability is calculated. The larger the difference, the higher the contribution of the risk factor to the deepwater blowout.

[0031] The present invention provides a deepwater blowout hidden danger identification and risk analysis method based on historical data, and the analysis method comprises: step S1: based on the existing blowout accident database, extracting and artificially completing the key information of the blowout accident to obtain the blowout accident data; step S2: identifying the hidden dangers of the blowout accident data one by one, extracting and selecting the hidden danger keywords in the accident data, classifying them according to the hidden danger characteristics, working condition stage and well control stage, and obtaining a hidden danger list; step S3: taking the hidden danger list as a risk factor, establishing a fault tree model from three perspectives of drilling stage, completion stage and workover stage; step S4: transforming the fault tree model, establishing a Bayesian network model, and allocating node probabilities based on historical data; step S5: selecting the target time period data, calculating the prior probability of each risk factor according to the statistical probability method, using GeNIe to calculate the posterior probability of the blowout risk factor and drawing a Bayesian network diagram; step S6: identifying the main risk inducements and formulating risk control plans. It can effectively organize the hidden dangers of deepwater blowouts based on the database and analyze the main risk factors of deepwater blowouts through Bayesian networks, so as to formulate risk control plans to reduce well control risks.

[0032] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0034] Figure 1 A flowchart of a method for identifying and analyzing deepwater blowout hazards based on historical data provided by an embodiment of the present invention;

[0035] Figure 2 The FT model of uncontrolled blowout in the drilling stage provided by the embodiment of the present invention;

[0036] Figure 3 The blowout uncontrolled FT model in the completion stage provided by the embodiment of the present invention;

[0037] Figure 4 The FT model of uncontrolled blowout in the workover stage provided by the embodiment of the present invention;

[0038] Figure 5 FT-BN mapping model diagram provided by an embodiment of the present invention;

[0039] Figure 6 A flowchart of probability allocation and BN modeling provided by an embodiment of the present invention;

[0040] Figure 7 The BN model for the drilling stage provided by the embodiment of the present invention;

[0041] Figure 8 A BN model for the completion phase provided by an embodiment of the present invention;

[0042] Fig. 9 A BN model for the well repair phase provided by an embodiment of the present invention;

[0043] Fig.10 A probability analysis diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0045] The terms "comprises" and "having" and any variations thereof in the description embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0046] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0047] like Figure 1 As shown, a deepwater blowout hazard identification and risk analysis method based on historical data includes:

[0048] S1. Based on the existing blowout accident database, extract and manually complete the key information of the blowout accident and complete the blowout accident data sorting. The specific operations are as follows:

[0049] According to the database, key information from each accident data is extracted from six dimensions, and some missing information is supplemented artificially. A total of 323 deepwater blowout data from 1950 to 2020 are collected. The key information dimensions are shown in Table 1, and the data collection sample is shown in Table 2.

[0050] Table 1 Key information dimensions of blowout accidents

[0051]

[0052]

[0053] Table 2 Blowout accident data information sample

[0054]

[0055] S2. Identify hidden dangers one by one in the collected data, extract and select hidden danger keywords in the accident data, classify them according to hidden danger characteristics, working conditions and well control stages, and output a hidden danger list. The specific operations are as follows:

[0056] Based on the collected data, when identifying hidden danger keywords, hidden danger keywords that appear twice or more are retained, and hidden danger keywords that appear once are selectively retained; hidden dangers are classified into drilling system hidden dangers (i.e., internal hidden dangers of the drilling system) and other hidden dangers (i.e., external risks such as typhoons, tsunamis, external impacts, and human operational errors and management defects) according to hidden danger characteristics; hidden dangers of the drilling system are divided into drilling stage hidden dangers, completion stage hidden dangers, and workover stage hidden dangers according to the working stage; and hidden dangers of the drilling system are divided into primary well control hidden dangers and secondary well control hidden dangers according to the well control stage. Finally, the list of internal hidden dangers of deepwater blowout drilling system is shown in Table 3, and the list of other hidden dangers of deepwater blowout accidents is shown in Table 4:

[0057] Table 3 List of internal hidden dangers in deepwater blowout drilling system

[0058]

[0059]

[0060] Table 4 List of internal hidden dangers in deepwater blowout drilling system

[0061]

[0062]

[0063] S3. Taking the main hidden dangers of deepwater well blowout drilling identified as risk factors, explore its internal mechanism, and establish a fault tree model from the three perspectives of drilling stage, completion stage and workover stage. The specific operations are as follows: selectively convert the obtained drilling system hidden dangers into risk factors of drilling system blowout, selectively eliminate unreasonable risk factors, and establish a fault tree model from the three perspectives of drilling stage, completion stage and workover stage. Figure 2 As shown in Figure 2, the fault tree model of the completion stage is as follows: Figure 3 As shown in Figure 2, the fault tree model of the workover stage is as follows: Figure 4 shown.

[0064] S4. Transform the fault tree, establish a Bayesian network model, and perform node probability allocation based on historical data. The specific operations are as follows:

[0065] Firstly, by leveraging the excellent logical analysis capabilities and simple FT structure, the network method of mapping FT to BN is used to determine the basic correlations between nodes and construct the initial BN structure.

[0066] The specific mapping method is as follows Figure 5 As shown. Taking the simplest AND gate and OR gate structure as an example, the top-level event of FT corresponds to the child node of BN; the bottom event of FT corresponds to the parent node of BN, and the logic gates such as AND gate and OR gate correspond to the conditional probability of BN (different logic gates correspond to different conditional probabilities).

[0067] After completing the network structure, a value is assigned to each node in the network. Based on historical accident data and expert experience, a statistical probability method is used to calculate the conditional probability table of each node corresponding to the risk factor. Figure 6 As shown, the probability distribution table is mainly based on BSEE related data, combined with the blowout mechanism of the risk factor and expert experience.

[0068] S5. Select the data of the target time period, calculate the prior probability of each risk factor according to the statistical probability method, use GeNIe to calculate the posterior probability of the blowout risk factor and draw a Bayesian network diagram. The specific operations are as follows:

[0069] Based on the deepwater well operation data, the statistical probability method is used to calculate the prior probability of risk factors. The calculation formula is as follows:

[0070]

[0071] Where P(B) is the prior probability of the risk factor; N f is the number of accidents; N w is the total number of operations.

[0072] After calculating the prior probability, the posterior probability of the risk factor is obtained through the existing prior probability and conditional probability. The calculation formula is as follows:

[0073]

[0074] In the formula, P(A i |B) is the prior probability of the risk factor; P(BA i ) is the conditional probability of the corresponding node.

[0075] The posterior probability of the blowout risk factors was calculated using the GeNIe Bayesian calculation software and the Bayesian network diagram was drawn. The resulting BN diagram of the blowout at different drilling stages is shown in the figure below: Figure 7-9 The failure probability table of blowout risk factors at different drilling stages is shown in Table 5-7:

[0076] Table 5 Failure probability of blowout risk factors during drilling stage

[0077]

[0078]

[0079] Table 6 Failure probability of blowout risk factors during completion stage

[0080]

[0081]

[0082] Table 7 Failure probability of blowout risk factors during workover phase

[0083]

[0084]

[0085] S6. Compare the prior probability and the posterior probability to obtain the main risk factors of the blowout accident and formulate a risk control plan. The specific operations are as follows:

[0086] Draw a probability analysis diagram between prior probability and posterior probability, such as Fig.10 As shown in the figure, the serial number in the probability analysis diagram corresponds to the coefficient of the risk probability in the failure probability table. The two data points corresponding to each serial number are the prior probability and the posterior probability, and the line between the data points represents the change in the prior probability and the posterior probability of the risk factor before and after the inversion. Longer lines usually represent greater changes and greater contributions of the risk factor to the blowout, reflecting the importance of the risk factor in deepwater blowout accidents.

[0087] From the analysis results, shallow gas, abnormal high pressure and pumping are the main risk factors for primary well control in the drilling stage, poor cementing quality, guide device failure and blowout preventer failure are the main risk factors for secondary well control in the drilling stage; pumping and low design density completion fluid are the main risk factors for primary well control in the completion stage, poor cementing quality and blowout preventer failure are the main risk factors for secondary well control in the completion stage: oil and gas accumulation and pumping are the main risk factors for primary well control in the workover stage; failure to close the blowout preventer in time and failure of the downhole safety valve are the main risk factors for secondary well control failure in the workover stage. Controlling risk factors can effectively reduce the risk of deepwater blowouts, and the hidden danger investigation and risk analysis in the case of the present invention are now completed.

[0088] Beneficial effects: The present invention can provide a more reliable list of potential blowout hazards based on the historical blowout accident database, avoiding traditional subjective assumptions, and based on this, provide real and reasonable risk factors for the Bayesian network. By comparing the prior probability with the posterior probability, the main risk factors can be quickly identified, and ultimately the risk control of deepwater blowouts can be effectively achieved.

[0089] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying and analyzing the hidden dangers of deepwater blowout based on historical data, characterized in that: The analysis method comprises: Step S1: Based on the existing blowout accident database, extract and manually complete the key information of the blowout accident to obtain the blowout accident data; Step S2: identifying hidden dangers one by one for the blowout accident data, extracting and selecting hidden danger keywords in the accident data, classifying them according to hidden danger characteristics, working condition stage and well control stage, and obtaining a hidden danger list; Step S3: Using the hidden danger list as risk factors, a fault tree model is established from three perspectives: drilling stage, completion stage and workover stage; Step S4: transforming the fault tree model, establishing a Bayesian network model, and performing node probability allocation based on historical data; Step S5: Select the target time period data, calculate the prior probability of each risk factor according to the statistical probability method, calculate the posterior probability of the blowout risk factor using GeNIe and draw a Bayesian network diagram; Step S6: Identify the main risk factors and develop risk control plans.

2. The method for identifying and analyzing the hidden dangers of deepwater blowout based on historical data according to claim 1 is characterized in that: The step S1: based on the existing blowout accident database, extracting and manually completing the key information of the blowout accident, and obtaining the blowout accident data specifically includes: The key information of the blowout accident is sorted out from six dimensions: location information, well conditions, operating conditions, blowout causes, blowout characteristics and remaining data.

3. The method for identifying and analyzing the hidden dangers of deepwater blowout based on historical data according to claim 1 is characterized in that: The step S2: identifying hidden dangers one by one for the blowout accident data, extracting and selecting hidden danger keywords in the accident data, classifying them according to hidden danger characteristics, working condition stage and well control stage, and obtaining a hidden danger list specifically includes: When identifying hidden danger keywords, retain hidden danger keywords that appear twice or more, and selectively retain hidden danger keywords that appear once; According to the characteristics of hidden dangers, hidden dangers are classified into drilling system hidden dangers and residual hidden dangers. According to the working stage, drilling system hidden dangers are divided into drilling stage hidden dangers, completion stage hidden dangers and workover stage hidden dangers. According to the well control stage, drilling system hidden dangers are divided into primary well control hidden dangers and secondary well control hidden dangers.

4. The method for identifying and analyzing the hidden dangers of deepwater blowout based on historical data according to claim 3 is characterized in that: The remaining hidden dangers include external risks such as typhoons, tsunamis, external impacts, as well as human operational errors and management defects.

5. The method for identifying and analyzing the hidden dangers of deepwater blowout based on historical data according to claim 1 is characterized in that: The step S4: transforming the fault tree model, establishing a Bayesian network model, and performing node probability allocation based on historical data specifically includes: The fault tree model is converted into a Bayesian network model by a structural mapping method, wherein the top-level events of the fault tree model correspond to the sub-nodes of the Bayesian network model; The bottom event of the fault tree model corresponds to the parent node of the Bayesian network model, and the logic gates such as AND gates and OR gates correspond to the conditional probabilities of the Bayesian network model.

6. The method for identifying and analyzing deepwater blowout hazards based on historical data according to claim 1 is characterized in that: The calculation of the prior probability of each risk factor according to the statistical probability method specifically includes: Where P(B) is the prior probability of the risk factor; N f is the number of accidents; N w is the total number of operations.

7. The method for identifying and analyzing deepwater blowout hazards based on historical data according to claim 1 is characterized in that: The posterior probability of the risk factor is obtained through the existing prior probability and conditional probability, and the calculation formula is as follows: In the formula, P(A i |B) is the prior probability of the risk factor; P(BA i ) is the conditional probability of the corresponding node.

8. The method for identifying and analyzing the risks of deepwater blowout hazards based on historical data according to claim 1 is characterized in that: The identification of major risk factors specifically includes: The difference between the prior probability and the posterior probability is calculated. The larger the difference, the higher the contribution of the risk factor to the deepwater blowout.

Images