Safety production monitoring method and device based on industrial Internet of Things

Abstract

The invention relates to the technical field of industrial internet of things, in particular to a safety production monitoring method and device based on the industrial internet of things, and the method comprises the steps: judging the well kick risk based on the size of a mud pit and the pressure of a casing in an analysis period, and determining an oscillation mark; the fluctuation characteristics of the stand pipe pressure in the analysis cycle are extracted to determine underground working condition stability coefficients, and then underground abnormal working conditions are divided; extracting intrusion features based on the slurry inlet and outlet flow and inlet and outlet density; integrating and analyzing the oscillation mark in the period, the underground working condition stability coefficient and the intrusion characteristic to construct a comprehensive risk index, and dividing risk grades; a hydraulic health metric is determined based on the hydraulic control oil pressure, leakage flow, and BOP instructions within the management cycle, and a closing speed reference value is updated in fusion with the comprehensive risk index within the management cycle. According to the method, accurate early warning of well control risks and dynamic optimization of the closing speed of the blowout preventer are achieved through multi-source data fusion analysis and a self-adaptive control strategy.

Description

Technical Field

[0001] This invention relates to the field of industrial Internet of Things (IoT) technology, and in particular to a method and device for safety production monitoring based on industrial IoT. Background Technology

[0002] Offshore oil drilling platforms operate in harsh environments with extremely high well control risks. As a critical piece of equipment ensuring drilling safety, the timeliness and reliability of the emergency interlock control of the blowout preventer (BOP) system directly affect the safety of the platform and personnel. Traditional BOP controls often employ fixed parameters or simple threshold alarms, resulting in problems such as delayed response, high false alarm rates, and inability to adapt to complex operating conditions.

[0003] Furthermore, existing control strategies struggle to effectively integrate multi-source monitoring data, resulting in insufficient early identification capabilities for abnormal conditions such as well kicks and leaks. They also fail to dynamically optimize shutdown strategies based on the health status of the hydraulic system, potentially leading to excessive equipment wear or delayed emergency response. Therefore, there is an urgent need for an intelligent emergency interlocking control method that can integrate multi-source information and possess adaptive capabilities. Summary of the Invention

[0004] The purpose of this invention is to provide a safety production monitoring method and device based on the Industrial Internet of Things (IIoT) to solve at least one of the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A safety production monitoring method based on the Industrial Internet of Things includes:

[0007] Based on the mud pit volume and casing pressure within the analysis period, well kick risk is determined, and oscillation indicators are identified;

[0008] Extract and analyze the fluctuation characteristics of riser pressure during the cycle to determine the downhole operating condition stability coefficient, and then classify abnormal downhole operating conditions;

[0009] Intrusion characteristics were extracted based on mud inlet and outlet flow rate and density.

[0010] A comprehensive risk index is constructed by integrating oscillation indicators, downhole operating condition stability coefficients, and invasion characteristics within the analysis period, and risk levels are classified accordingly;

[0011] Hydraulic health metrics are determined based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and the shut-off speed baseline value is updated by integrating the comprehensive risk index within the management cycle.

[0012] Furthermore, with a sampling period of 1 second, the volume change ΔV of the mud pit and the pressure change ΔP of the casing are calculated per second.

[0013] Set the volume change rate threshold Vth and the pressure change rate threshold Pth, and statistically analyze the number of times ΔV is greater than Vth (NV) and the number of times ΔP is greater than Pth (NP) within the period.

[0014] If NV is greater than or equal to the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be high risk, and the oscillation flag is set to 2; if NV is greater than or equal to the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be volume anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be pressure anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be normal, and the oscillation flag Fg is set to 0.

[0015] Furthermore, the average value Pta and standard deviation Pts of the riser pressure within the analysis period are calculated, and then the pressure fluctuation intensity coefficient Sf=Pts / Pta is determined.

[0016] Perform a fast Fourier transform on the riser pressure sequence within the analysis period to calculate its power spectral density. Find the frequency point with the largest power spectral amplitude in the target frequency band, denoted as fd, and record its corresponding power spectral amplitude PSD. Then determine the pulsation significance coefficient Sp=min(1,PSD / (PSDn×3)).

[0017] By integrating the pressure fluctuation intensity coefficient and the pulsation significance coefficient, a downhole working condition stability coefficient C is constructed, C=1-(c1×min(1,Sf / sf0)+c2×Sp);

[0018] Where c1 is the pressure fluctuation intensity weight, c2 is the pulsation significance weight, c1+c2=1, PSDn is the preset background noise baseline, and sf0 is the preset fluctuation intensity threshold.

[0019] Furthermore, based on the working condition stability coefficient C, abnormal working conditions in the well are classified. If C is greater than or equal to j1, the abnormal working condition in the well in the current analysis period is determined to be a normal working condition. If C is greater than or equal to j2 and less than or equal to j1, the abnormal working condition in the well in the current analysis period is determined to be a low-risk working condition. If C is less than j2, the abnormal working condition in the well in the current analysis period is determined to be a high-risk working condition.

[0020] Where j1 is the first preset stability threshold and j2 is the second preset stability threshold.

[0021] Furthermore, the inlet and outlet flow difference ΔF of each sampling period within the analysis period is calculated, and the number of sampling periods Nf in which the inlet and outlet flow difference is greater than the preset overflow threshold Fth within the analysis period is statistically analyzed, thereby determining the overflow characteristic coefficient Kf=min(1,Nf / n1).

[0022] Calculate the inlet and outlet density difference Δρ for each sampling period within the analysis period, and statistically analyze the number Nd of sampling periods where Δρ is less than the preset density decrease threshold ρth within the analysis period, thereby determining the density decrease characteristic coefficient Kd=min(1,Nd / n1).

[0023] The overflow characteristic coefficient and the density decrease characteristic coefficient are fused to obtain the intrusion characteristic Ci, Ci = β1×Kf + β2×Kd;

[0024] Where β1 is the overflow feature weight, β2 is the density decrease feature weight, β1+β2=1, and n1 is the preset quantity threshold.

[0025] Furthermore, a comprehensive risk index Rt is constructed by combining the oscillation indicator Fg, the downhole working condition stability coefficient C, and the invasion characteristic Ci within the analysis period, where Rt = w1 × Fg + w2 × C + w3 × Ci;

[0026] Risk levels are determined based on the comprehensive risk index Rt: if Rt is less than r1, the current analysis period is considered low risk; if Rt is greater than or equal to r1 and less than or equal to r2, the current analysis period is considered medium risk; if Rt is greater than r2, the current analysis period is considered high risk.

[0027] Where r1 is the first preset risk threshold, r2 is the second preset risk threshold, and w1, w2 and w3 are weighting coefficients.

[0028] Furthermore, the analysis cycle conditions are divided according to the BOP instruction: if there is a BOP instruction in the current analysis cycle, the current analysis cycle condition is divided into dynamic conditions; otherwise, the current analysis cycle is divided into steady-state conditions.

[0029] For analysis cycles with stable operating conditions within the management period, calculate the average hydraulic control oil pressure Pavg and the standard deviation of the hydraulic control oil pressure Pstd within the analysis cycle. Use (1-Pstd / Pavg) as the pressure stability coefficient for that analysis cycle. Calculate the average pressure stability coefficient within the management cycle as the pressure stability coefficient for the current management cycle, denoted as ηP. Simultaneously, calculate the average leakage flow rate Qavg for each analysis cycle with stable operating conditions within the management period, and calculate the leakage severity coefficient ηq for the current management cycle, ηq=min(1,Qavg / Qwarn).

[0030] The steady-state health index Hs is determined based on the pressure stability coefficient ηP and the leakage severity coefficient ηq, where Hs = a1 × ηP + a2 × (1 - ηq).

[0031] Where Qwarn is the preset early warning leakage flow rate, a1 is the pressure stability weight, a2 is the first leakage weight, and a1+a2=1.

[0032] Furthermore, for each BOP command within the management cycle, a dynamic evaluation window is formed, starting from the time the BOP command is issued and ending at the time the BOP command is completed. Within the dynamic evaluation window, the time Tr from the issuance of the BOP command to the pressure reaching the target value is calculated, and the pressure response anomaly coefficient Rey is calculated, Rey=min(1,max(0,(Tr-Tt)) / (α×Tt)). At the same time, the average leakage flow rate Qavgj within the dynamic evaluation window is calculated, and the leakage severity index Lj within the dynamic evaluation window is calculated, Lj=min(1,Qavgj / Qwarn).

[0033] The dynamic health index Ds is determined based on the pressure response anomaly coefficient Rey and the leakage severity index Lj, where Ds = b1 × Rey + b2 × (1 - Lj).

[0034] The hydraulic health measure D is determined based on the steady-state health index Hs and the dynamic health index Ds, where D = γ1 × Hs + γ2 × Ds.

[0035] Where Tt is the rated rise time, b1 is the pressure response weight, b2 is the second leakage weight, b1+b2=1, γ1 is the steady-state weight, γ2 is the dynamic weight, γ1+γ2=1, and α is the preset adjustment coefficient.

[0036] Furthermore, the average value of the comprehensive risk index Ravg for each analysis cycle within the management week is calculated, and the hydraulic health metric D is integrated to update the shut-off speed baseline value Vbn for the next management cycle.

[0037] If Ravg is less than r1, then Vbn = v1 × (0.6 + 0.4 × D);

[0038] If Ravg is greater than or equal to r1 and less than or equal to r2, then Vbn = v2 × (0.6 + 0.4 × D);

[0039] If Ravg is greater than r2, then Vbn = v3 × (0.6 + 0.4 × D);

[0040] Wherein, v1 is the first preset closing speed reference value, v2 is the second preset closing speed reference value, and v3 is the third preset closing speed reference value.

[0041] According to another aspect of this application, a safety production monitoring device based on the Industrial Internet of Things is provided, comprising:

[0042] The well kick analysis unit is used to determine the well kick risk based on the mud pit volume and casing pressure within the analysis period, and to identify oscillation indicators;

[0043] The working condition division unit is used to extract and analyze the fluctuation characteristics of riser pressure during the analysis period to determine the downhole working condition stability coefficient, and then classify abnormal downhole working conditions.

[0044] An intrusion feature determination unit is used to extract intrusion features based on mud inlet / outlet flow rate and inlet / outlet density.

[0045] The risk classification unit is used to integrate oscillation indicators, downhole operating condition stability coefficients, and invasion characteristics within the analysis period to construct a comprehensive risk index and classify risk levels.

[0046] The baseline update unit is used to determine hydraulic health metrics based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and to update the shut-off speed baseline value by integrating the comprehensive risk index within the management cycle.

[0047] The beneficial effects of this invention are as follows: This invention provides an intelligent emergency interlock control method for blowout preventer (BOP) systems based on the Industrial Internet of Things (IIoT), which has the following beneficial effects: First, through real-time fusion analysis of multi-source heterogeneous data, it achieves early and accurate identification of well kick risk, abnormal downhole conditions, and formation fluid intrusion, significantly improving the accuracy and timeliness of well control early warning. Second, it constructs a dual-dimensional assessment system of comprehensive risk index and hydraulic health measurement, enabling emergency strategies to simultaneously consider external risk situations and internal equipment status. Finally, based on the risk assessment results, it adaptively adjusts the BOP closing speed benchmark value, closing smoothly in low-risk situations to protect the equipment and responding quickly in high-risk situations to ensure safety, achieving a dynamic optimization balance between well control safety and equipment lifespan, and significantly improving the intelligence level and emergency response capability of the drilling platform BOP system. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used 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 those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the safety production monitoring method based on the Industrial Internet of Things in this embodiment.

[0050] Figure 2 This is a flowchart illustrating the method for classifying abnormal downhole working conditions in this embodiment.

[0051] Figure 3This is a flowchart illustrating the method for disabling the update of the speed reference value in this embodiment.

[0052] Figure 4 This is a schematic diagram of the safety production monitoring device based on the Industrial Internet of Things in this embodiment. Detailed Implementation

[0053] To more clearly illustrate the present invention, the following description, in conjunction with preferred embodiments and accompanying drawings, further explains the invention. Similar components in the drawings are indicated by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than restrictive and should not be construed as limiting the scope of protection of the present invention.

[0054] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0055] Specifically, this embodiment is applied to the intelligent emergency interlock control of the blowout preventer system on offshore oil drilling platforms.

[0056] Please see Figure 1 As shown, this is a flowchart illustrating the safety production monitoring method based on the Industrial Internet of Things in this embodiment. Before the method is executed, the system synchronously collects data through monitoring terminals deployed at key locations on the drilling platform, including:

[0057] Wellbore pressure and flow data:

[0058] Riser pressure, collected by a pressure sensor installed on the riser, characterizes the drilling fluid circulation pump pressure;

[0059] The casing pressure is collected by a pressure sensor installed at the casing head, which characterizes the annular pressure.

[0060] The inlet and outlet flow rates are collected by electromagnetic flow meters installed on the mud inlet and outlet pipelines, respectively.

[0061] The inlet and outlet densities were collected using densitometers installed on the mud inlet and outlet pipelines, respectively.

[0062] Mud pit and leak monitoring data:

[0063] The volume of the mud pit is collected using a mud pit level gauge.

[0064] Leakage flow is collected by a flow sensor on the return oil line;

[0065] BOP (Blowout Preventer) status data:

[0066] The hydraulic control oil pressure is acquired by the hydraulic sensor of the BOP control unit;

[0067] The gate position and response time are obtained in real time by magnetostrictive displacement sensors installed on each gate (full seal, half seal, shear), which acquire the gate opening percentage (%) and record the delay time from issuing the closing command to the gate starting to move and the time required for the gate to be completely closed. In this embodiment, the method of collecting the above data is not specifically limited, and those skilled in the art can set it freely according to their needs.

[0068] The method includes:

[0069] Step S1: Determine the risk of well kick based on the mud pit volume and casing pressure within the analysis period, and identify the oscillation indicator.

[0070] Specifically, with a sampling period of 1 second, the volume change of the mud pit per second is calculated as ΔV=(Vp-Vpi) / t, and the pressure change of the casing is calculated as ΔP=(Pc-Pca) / t.

[0071] Where Vp is the mud pit volume of the current sampling period, Vpi is the mud pit volume of the previous sampling period adjacent to the current sampling period, Pc is the casing pressure of the current sampling period, Pca is the casing pressure of the previous sampling period adjacent to the current sampling period, and t is the sampling period duration.

[0072] Set the volume change rate threshold Vth and the pressure change rate threshold Pth, and statistically analyze the number of times ΔV is greater than Vth (NV) and the number of times ΔP is greater than Pth (NP) within the period.

[0073] If NV is greater than or equal to the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be high risk, and the oscillation flag is set to 2; if NV is greater than or equal to the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be volume anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be pressure anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be normal, and the oscillation flag Fg is set to 0.

[0074] Specifically, in this embodiment, the volume change rate threshold is 0.05 m³ / s, the pressure change rate threshold is 0.02 MPa / s, the analysis cycle is one minute, and the preset abnormal threshold is 20.

[0075] Specifically, during normal drilling without any abnormal downhole conditions (such as well kick, well leakage, drill string leakage, etc.), the system selects a historical data sequence of riser pressure lasting no less than 60 seconds; performs a fast Fourier transform on the data sequence to calculate its power spectral density; then, within the preset target frequency band, calculates the arithmetic mean of the power spectral density corresponding to all frequency points, and uses this mean as the preset background noise baseline PSDn. This baseline can be recalibrated after each platform maintenance or at regular intervals to reflect the current noise background level of the downhole and sensor systems.

[0076] Specifically, this step achieves rapid preliminary screening of well kick risk by real-time monitoring of the change rate of mud pit volume and casing pressure, and statistically analyzing the frequency of abnormal changes per unit time. This method can effectively distinguish risk types and generate quantitative oscillation indicators, providing crucial early warning information for subsequent comprehensive risk assessment. It avoids the shortcomings of single-threshold alarms, which are prone to false alarms, and improves the sensitivity and accuracy of risk identification.

[0077] Please continue reading. Figure 1 As shown, the safety production monitoring method based on the Industrial Internet of Things also includes:

[0078] Step S2: Extract and analyze the fluctuation characteristics of riser pressure during the analysis period to determine the downhole working condition stability coefficient, and then classify abnormal downhole working conditions.

[0079] Please see Figure 2 As shown, the method for classifying abnormal downhole operating conditions includes:

[0080] Step S21: Extract and analyze the fluctuation characteristics of the riser pressure during the analysis period to determine the downhole operating condition stability coefficient.

[0081] Specifically, the average value Pta and standard deviation Pts of the riser pressure within the analysis period are calculated, and then the pressure fluctuation intensity coefficient Sf=Pts / Pta is determined.

[0082] Perform a fast Fourier transform on the riser pressure sequence within the analysis period to calculate its power spectral density. Find the frequency point with the largest power spectral amplitude in the target frequency band, denoted as fd, and record its corresponding power spectral amplitude PSD. Then determine the pulsation significance coefficient Sp=min(1,PSD / (PSDn×3)).

[0083] By integrating the pressure fluctuation intensity coefficient and the pulsation significance coefficient, a downhole working condition stability coefficient C is constructed, C=1-(c1×min(1,Sf / sf0)+c2×Sp);

[0084] Where c1 is the pressure fluctuation intensity weight, c2 is the pulsation significance weight, c1+c2=1, PSDn is the preset background noise baseline, and sf0 is the preset fluctuation intensity threshold.

[0085] Specifically, in this embodiment, the target frequency band is 0.5 Hz to 5 Hz, the preset fluctuation intensity threshold sf0 is 0.2, the pressure fluctuation intensity weight is 0.5, and the pulsation significance weight is 0.5.

[0086] Specifically, this step analyzes the time-domain fluctuation intensity and frequency-domain pulsation characteristics of riser pressure to construct a quantitative index that can comprehensively reflect the stability of downhole operating conditions. This index integrates the statistical characteristics and spectral energy distribution of pressure data, and can keenly capture pressure fluctuations caused by abnormal drill string movement, mud pulse interference, and early well kick, providing a reliable numerical basis for the identification of complex downhole operating conditions.

[0087] Please continue reading. Figure 2 As shown, the method for classifying abnormal downhole conditions also includes:

[0088] Step S22: Classify downhole abnormal working conditions based on the working condition stability coefficient.

[0089] Specifically, abnormal downhole conditions are classified based on the working condition stability coefficient C. If C is greater than or equal to j1, the abnormal downhole conditions in the current analysis period are determined to be normal conditions. If C is greater than or equal to j2 and less than or equal to j1, the abnormal downhole conditions in the current analysis period are determined to be low-risk conditions. If C is less than j2, the abnormal downhole conditions in the current analysis period are determined to be high-risk conditions.

[0090] Output the current operating status of the analysis cycle to the user;

[0091] Where j1 is the first preset stability threshold and j2 is the second preset stability threshold.

[0092] Specifically, in this embodiment, the first preset stability threshold is 0.8 and the second preset stability threshold is 0.5.

[0093] Specifically, this step, based on the working condition stability coefficient, divides the downhole condition into three levels: normal, low risk, and high risk, and outputs this information to the operators in an intuitive manner. This classification mechanism not only makes the degree of abnormality of the downhole working condition clear at a glance, but also provides key downhole environmental information for subsequent comprehensive risk assessment, which helps to distinguish between surface equipment failures and downhole geological anomalies, and improves the intelligence level of the monitoring system.

[0094] Please continue reading. Figure 1 As shown, the safety production monitoring method based on the Industrial Internet of Things also includes:

[0095] Step S3: Extract intrusion features based on mud inlet / outlet flow rate and density.

[0096] Specifically, the inlet and outlet flow difference ΔF = Fout - Fin is calculated for each sampling period within the analysis period, and the number of sampling periods Nf in which the inlet and outlet flow difference is greater than the preset overflow threshold Fth is statistically analyzed, thereby determining the overflow characteristic coefficient Kf = min(1, Nf / n1).

[0097] Calculate the inlet and outlet density difference Δρ=ρout-ρin for each sampling period within the analysis period, and statistically analyze the number of sampling periods Nd in which Δρ is less than the preset density decrease threshold ρth, and then determine the density decrease characteristic coefficient Kd=min(1,Nd / n1).

[0098] The overflow characteristic coefficient and the density decrease characteristic coefficient are fused to obtain the intrusion characteristic Ci, Ci = β1×Kf + β2×Kd;

[0099] Where β1 is the overflow feature weight, β2 is the density decrease feature weight, β1+β2=1, n1 is the preset quantity threshold, Fout is the outlet flow rate, Fin is the inlet flow rate, ρout is the outlet density, and ρin is the inlet density.

[0100] Specifically, in this embodiment, the preset overflow threshold is 2L / s, the preset density decrease threshold is -0.02 g / cm³, the preset quantity threshold is 30, the overflow feature weight is 0.6, and the density decrease feature weight is 0.4.

[0101] Specifically, this step extracts the quantitative characteristics of formation fluid invasion by real-time statistics of the flow rate difference and density difference between the mud inlet and outlet. This method can effectively identify the flow imbalance and density drop caused by low-density fluid entering the annulus in the early stage of the overflow. By statistically analyzing the frequency of anomalies and fusing them to form an invasion characteristic coefficient, the system's ability to detect well kick precursors is significantly enhanced, buying valuable time for early well control.

[0102] Please continue reading. Figure 1 As shown, the safety production monitoring method based on the Industrial Internet of Things also includes:

[0103] Step S4: Combine the oscillation indicators, downhole working condition stability coefficient and invasion characteristics within the analysis period to construct a comprehensive risk index and classify the risk levels.

[0104] Specifically, the oscillation indicator Fg, the downhole working condition stability coefficient C, and the invasion characteristic Ci within the analysis period are used to construct a comprehensive risk index Rt, Rt=w1×Fg+w2×C+w3×Ci;

[0105] Risk levels are determined based on the comprehensive risk index Rt: if Rt is less than r1, the current analysis period is considered low risk; if Rt is greater than or equal to r1 and less than or equal to r2, the current analysis period is considered medium risk; if Rt is greater than r2, the current analysis period is considered high risk.

[0106] The risk level is displayed to the user;

[0107] Where r1 is the first preset risk threshold, r2 is the second preset risk threshold, and w1, w2 and w3 are weighting coefficients.

[0108] Specifically, in this embodiment, w1 is 0.4, w2 is 0.2, w3 is 0.4, the first preset risk threshold is 0.3, and the second preset risk threshold is 0.7.

[0109] Specifically, this step weights and fuses risk characteristics from different dimensions to generate a comprehensive risk index that fully reflects the platform's security status. Based on this, risks are divided into three levels—low, medium, and high—by setting preset thresholds, achieving quantitative assessment and graded early warning of well control risks. This multi-source information fusion strategy greatly improves the robustness and reliability of risk criteria and effectively reduces the risk of false alarms or missed alarms caused by anomalies in a single parameter.

[0110] Please continue reading. Figure 1 As shown, the safety production monitoring method based on the Industrial Internet of Things also includes:

[0111] Step S5: Determine hydraulic health metrics based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and update the shut-off speed benchmark value by integrating the comprehensive risk index within the management cycle.

[0112] Please see Figure 3 As shown, the method for updating the closed speed reference value includes:

[0113] Step S51: Determine hydraulic health metrics based on hydraulic control oil pressure, leakage flow, and BOP commands during the management cycle.

[0114] Specifically, the analysis cycle conditions are divided according to the BOP instruction: if there is a BOP instruction in the current analysis cycle, the current analysis cycle condition is divided into dynamic conditions; otherwise, the current analysis cycle is divided into steady-state conditions.

[0115] For analysis cycles with stable operating conditions within the management period, calculate the average hydraulic control oil pressure Pavg and the standard deviation of the hydraulic control oil pressure Pstd within the analysis cycle. Use (1-Pstd / Pavg) as the pressure stability coefficient for that analysis cycle. Calculate the average pressure stability coefficient within the management cycle as the pressure stability coefficient for the current management cycle, denoted as ηP. Simultaneously, calculate the average leakage flow rate Qavg for each analysis cycle with stable operating conditions within the management period, and calculate the leakage severity coefficient ηq for the current management cycle, ηq=min(1,Qavg / Qwarn).

[0116] The steady-state health index Hs is determined based on the pressure stability coefficient ηP and the leakage severity coefficient ηq, where Hs = a1 × ηP + a2 × (1 - ηq).

[0117] Where Qwarn is the preset early warning leakage flow rate, a1 is the pressure stability weight, a2 is the first leakage weight, and a1+a2=1;

[0118] For each BOP command within the management cycle, a dynamic evaluation window is formed, starting from the time the BOP command is issued and ending at the time the BOP command is completed. Within the dynamic evaluation window, the time Tr from the issuance of the BOP command to the pressure reaching the target value is calculated, and the pressure response anomaly coefficient Rey is calculated, Rey=min(1,max(0,(Tr-Tt)) / (α×Tt)). At the same time, the average leakage flow rate Qavgj within the dynamic evaluation window is calculated, and the leakage severity index Lj within the dynamic evaluation window is calculated, Lj=min(1,Qavgj / Qwarn).

[0119] The dynamic health index Ds is determined based on the pressure response anomaly coefficient Rey and the leakage severity index Lj, where Ds = b1 × Rey + b2 × (1 - Lj);

[0120] The hydraulic health measure D is determined based on the steady-state health index Hs and the dynamic health index Ds, where D = γ1 × Hs + γ2 × Ds.

[0121] Where Tt is the rated rise time, b1 is the pressure response weight, b2 is the second leakage weight, b1+b2=1, γ1 is the steady-state weight, γ2 is the dynamic weight, γ1+γ2=1, and α is the preset adjustment coefficient.

[0122] Specifically, in this embodiment, the preset early warning leakage flow rate is 120 L / min, the pressure stability weight is 0.7, the first leakage weight is 0.3, the pressure response weight is 0.6, the second leakage weight is 0.4, the rated rise time is 15 seconds, the preset adjustment coefficient is 0.2, the steady-state weight is 0.4, and the dynamic weight is 0.6.

[0123] Specifically, in this embodiment, the management cycle is 30 minutes.

[0124] Specifically, this step distinguishes between steady-state and dynamic operating conditions, assesses the pressure stability, leakage severity, and dynamic response capability of the hydraulic system, and constructs a quantitative metric that comprehensively reflects the health status of the BOP hydraulic system. This metric not only focuses on leakage and pressure fluctuations during daily operation, but also specifically examines the response performance when BOP commands are executed, providing key equipment status information for subsequent shutdown strategy optimization.

[0125] Please continue reading. Figure 3 As shown, the method for updating the closed speed reference value includes:

[0126] Step S52: Update the shut-off speed baseline value based on hydraulic health metrics and comprehensive risk index during the management cycle.

[0127] Specifically, the average value of the comprehensive risk index Ravg for each analysis cycle within the management week is calculated, and the hydraulic health metric D is integrated to update the shutdown speed baseline value Vbn for the next management cycle.

[0128] If Ravg is less than r1, then Vbn = v1 × (0.6 + 0.4 × D);

[0129] If Ravg is greater than or equal to r1 and less than or equal to r2, then Vbn = v2 × (0.6 + 0.4 × D);

[0130] If Ravg is greater than r2, then Vbn = v3 × (0.6 + 0.4 × D);

[0131] Wherein, v1 is the first preset closing speed reference value, v2 is the second preset closing speed reference value, and v3 is the third preset closing speed reference value.

[0132] Specifically, in this embodiment, the first preset closing speed reference value v1, the second preset closing speed reference value v2, and the third preset closing speed reference value v3 are summarized as the target closing linear speed of the blowout preventer gate at low, medium, and high risk levels, respectively. In this embodiment, the rated closing linear speed of the blowout preventer gate is set to Vrate, then v1 = 0.8 × Vrate, v2 = Vrate, and v3 = 1.2 × Vrate. This rated speed V_rate is provided by the blowout preventer manufacturer. For example, when the rated closing time is 30 seconds and the gate travel is 300 mm, Vrate = 10 mm / s.

[0133] Specifically, this step integrates the overall risk average level within the management cycle with hydraulic system health metrics to dynamically adjust the baseline value for the blowout preventer (BOP) shut-off speed in the next cycle. When the risk is low and the hydraulic system is healthy, a gentler shut-off speed is used to protect the equipment; when the risk increases or the health level decreases, the shut-off speed is appropriately increased to cope with emergencies; when the risk is extremely high, the fastest emergency shut-off speed is activated. This adaptive adjustment mechanism achieves a dynamic balance between ensuring well control safety and extending equipment life, significantly improving the intelligent emergency response capability of the BOP system.

[0134] Please see Figure 4 As shown, the safety production monitoring device based on the Industrial Internet of Things includes:

[0135] The well kick analysis unit is used to determine the well kick risk based on the mud pit volume and casing pressure within the analysis period, and to identify oscillation indicators;

[0136] The working condition division unit is used to extract and analyze the fluctuation characteristics of riser pressure during the analysis period to determine the downhole working condition stability coefficient, and then classify abnormal downhole working conditions.

[0137] An intrusion feature determination unit is used to extract intrusion features based on mud inlet / outlet flow rate and inlet / outlet density.

[0138] The risk classification unit is used to integrate oscillation indicators, downhole operating condition stability coefficients, and invasion characteristics within the analysis period to construct a comprehensive risk index and classify risk levels.

[0139] The baseline update unit is used to determine hydraulic health metrics based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and to update the shut-off speed baseline value by integrating the comprehensive risk index within the management cycle.

[0140] The safety production monitoring device based on the Industrial Internet of Things provided in this application can execute the safety production monitoring method based on the Industrial Internet of Things provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of the execution method.

[0141] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A safety production monitoring method based on the Industrial Internet of Things, characterized in that, include: Based on the mud pit volume and casing pressure within the analysis period, well kick risk is determined, and oscillation indicators are identified; Extract and analyze the fluctuation characteristics of riser pressure during the cycle to determine the downhole operating condition stability coefficient, and then classify abnormal downhole operating conditions; Intrusion characteristics were extracted based on mud inlet and outlet flow rate and density. A comprehensive risk index is constructed by integrating oscillation indicators, downhole operating condition stability coefficients, and invasion characteristics within the analysis period, and risk levels are classified accordingly; Hydraulic health metrics are determined based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and the shut-off speed baseline value is updated by integrating the comprehensive risk index within the management cycle.

2. The safety production monitoring method based on the Industrial Internet of Things according to claim 1, characterized in that, Using a sampling period of 1 second, calculate the change in mud pit volume ΔV and the change in casing pressure ΔP per second. Set the volume change rate threshold Vth and the pressure change rate threshold Pth, and statistically analyze the number of times ΔV is greater than Vth (NV) and the number of times ΔP is greater than Pth (NP) within the period. If NV is greater than or equal to the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be high risk, and the oscillation flag is set to 2; if NV is greater than or equal to the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be volume anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is greater than or equal to the preset anomaly threshold, the well kick risk for the current analysis period is determined to be pressure anomaly-dominated risk, and the oscillation flag Fg is set to 1; if NV is less than the preset anomaly threshold and NP is less than the preset anomaly threshold, the well kick risk for the current analysis period is determined to be normal, and the oscillation flag Fg is set to 0.

3. The safety production monitoring method based on the Industrial Internet of Things according to claim 2, characterized in that, The average value Pta and standard deviation Pts of the riser pressure within the analysis period are calculated, and then the pressure fluctuation intensity coefficient Sf=Pts / Pta is determined. Perform a fast Fourier transform on the riser pressure sequence within the analysis period to calculate its power spectral density. Find the frequency point with the largest power spectral amplitude in the target frequency band, denoted as fd, and record its corresponding power spectral amplitude PSD. Then determine the pulsation significance coefficient Sp=min(1,PSD / (PSDn×3)). By integrating the pressure fluctuation intensity coefficient and the pulsation significance coefficient, a downhole working condition stability coefficient C is constructed, C=1-(c1×min(1,Sf / sf0)+c2×Sp); Where c1 is the pressure fluctuation intensity weight, c2 is the pulsation significance weight, c1+c2=1, PSDn is the preset background noise baseline, and sf0 is the preset fluctuation intensity threshold.

4. The safety production monitoring method based on the Industrial Internet of Things according to claim 3, characterized in that, The abnormal working conditions in the well are classified based on the working condition stability coefficient C. If C is greater than or equal to j1, the abnormal working conditions in the current analysis period are determined to be normal working conditions. If C is greater than or equal to j2 and less than or equal to j1, the abnormal working conditions in the current analysis period are determined to be low-risk working conditions. If C is less than j2, the abnormal working conditions in the current analysis period are determined to be high-risk working conditions. Where j1 is the first preset stability threshold and j2 is the second preset stability threshold.

5. The safety production monitoring method based on the Industrial Internet of Things according to claim 4, characterized in that, Calculate the inlet and outlet flow difference ΔF for each sampling period within the analysis period, count the number Nf of sampling periods where the inlet and outlet flow difference is greater than the preset overflow threshold Fth within the analysis period, and then determine the overflow characteristic coefficient Kf=min(1,Nf / n1). Calculate the inlet and outlet density difference Δρ for each sampling period within the analysis period, and statistically analyze the number Nd of sampling periods where Δρ is less than the preset density decrease threshold ρth within the analysis period, thereby determining the density decrease characteristic coefficient Kd=min(1,Nd / n1). The overflow characteristic coefficient and the density decrease characteristic coefficient are fused to obtain the intrusion characteristic Ci, Ci = β1×Kf + β2×Kd; Where β1 is the overflow feature weight, β2 is the density decrease feature weight, β1+β2=1, and n1 is the preset quantity threshold.

6. The safety production monitoring method based on the Industrial Internet of Things according to claim 5, characterized in that, The comprehensive risk index Rt is constructed by combining the oscillation indicator Fg, the downhole working condition stability coefficient C, and the invasion characteristic Ci within the analysis period. Rt = w1×Fg + w2×C + w3×Ci. Risk levels are determined based on the comprehensive risk index Rt: if Rt is less than r1, the current analysis period is considered low risk; if Rt is greater than or equal to r1 and less than or equal to r2, the current analysis period is considered medium risk; if Rt is greater than r2, the current analysis period is considered high risk. Where r1 is the first preset risk threshold, r2 is the second preset risk threshold, and w1, w2 and w3 are weighting coefficients.

7. The safety production monitoring method based on the Industrial Internet of Things according to claim 6, characterized in that, The analysis cycle is divided into operating conditions based on the BOP instruction: if there is a BOP instruction in the current analysis cycle, the current analysis cycle is divided into dynamic operating conditions; otherwise, the current analysis cycle is divided into steady-state operating conditions. For analysis cycles with stable operating conditions within the management period, calculate the average hydraulic control oil pressure Pavg and the standard deviation of the hydraulic control oil pressure Pstd within the analysis cycle. Use (1-Pstd / Pavg) as the pressure stability coefficient for that analysis cycle. Calculate the average pressure stability coefficient within the management cycle as the pressure stability coefficient for the current management cycle, denoted as ηP. Simultaneously, calculate the average leakage flow rate Qavg for each analysis cycle with stable operating conditions within the management period, and calculate the leakage severity coefficient ηq for the current management cycle, ηq=min(1,Qavg / Qwarn). The steady-state health index Hs is determined based on the pressure stability coefficient ηP and the leakage severity coefficient ηq, where Hs = a1 × ηP + a2 × (1 - ηq). Where Qwarn is the preset early warning leakage flow rate, a1 is the pressure stability weight, a2 is the first leakage weight, and a1+a2=1.

8. The safety production monitoring method based on the Industrial Internet of Things according to claim 7, characterized in that, For each BOP command within the management cycle, a dynamic evaluation window is formed, starting from the time the BOP command is issued and ending at the time the BOP command is completed. Within the dynamic evaluation window, the time Tr from the issuance of the BOP command to the pressure reaching the target value is calculated, and the pressure response anomaly coefficient Rey is calculated, Rey=min(1,max(0,(Tr-Tt)) / (α×Tt)). At the same time, the average leakage flow rate Qavgj within the dynamic evaluation window is calculated, and the leakage severity index Lj within the dynamic evaluation window is calculated, Lj=min(1,Qavgj / Qwarn). The dynamic health index Ds is determined based on the pressure response anomaly coefficient Rey and the leakage severity index Lj, where Ds = b1 × Rey + b2 × (1 - Lj); The hydraulic health measure D is determined based on the steady-state health index Hs and the dynamic health index Ds, where D = γ1 × Hs + γ2 × Ds. Where Tt is the rated rise time, b1 is the pressure response weight, b2 is the second leakage weight, b1+b2=1, γ1 is the steady-state weight, γ2 is the dynamic weight, γ1+γ2=1, and α is the preset adjustment coefficient.

9. The safety production monitoring method based on the Industrial Internet of Things according to claim 8, characterized in that, Calculate the average risk index Ravg for each analysis cycle within the management week, and integrate it with the hydraulic health metric D to update the shutdown speed baseline value Vbn for the next management cycle: If Ravg is less than r1, then Vbn = v1 × (0.6 + 0.4 × D); If Ravg is greater than or equal to r1 and less than or equal to r2, then Vbn = v2 × (0.6 + 0.4 × D); If Ravg is greater than r2, then Vbn = v3 × (0.6 + 0.4 × D); Wherein, v1 is the first preset closing speed reference value, v2 is the second preset closing speed reference value, and v3 is the third preset closing speed reference value.

10. A safety production monitoring device based on the Industrial Internet of Things (IIoT), applied to the safety production monitoring method based on the Industrial Internet of Things as described in any one of claims 1-9, characterized in that, include: The well kick analysis unit is used to determine the well kick risk based on the mud pit volume and casing pressure within the analysis period, and to identify oscillation indicators; The working condition division unit is used to extract and analyze the fluctuation characteristics of riser pressure during the analysis period to determine the downhole working condition stability coefficient, and then classify abnormal downhole working conditions. An intrusion feature determination unit is used to extract intrusion features based on mud inlet / outlet flow rate and inlet / outlet density. The risk classification unit is used to integrate oscillation indicators, downhole operating condition stability coefficients, and invasion characteristics within the analysis period to construct a comprehensive risk index and classify risk levels. The baseline update unit is used to determine hydraulic health metrics based on hydraulic control oil pressure, leakage flow, and BOP commands within the management cycle, and to update the shut-off speed baseline value by integrating the comprehensive risk index within the management cycle.

Images