Drilling machine disc brake control test system and control test method based on cloud platform

Abstract

The invention relates to the technical field of drilling machine disc brake control testing, in particular to a drilling machine disc brake control testing system and method based on a cloud platform. The system collects apparent electric control signals and hidden physical layer data; the method comprises the following steps: calling a degradation mechanism model to calculate a physical degradation state based on hidden working condition data; the digital twin body updates and simulates a virtual limit working condition according to the virtual limit working condition; the evaluation module calculates a recessive degradation weight factor and a real reliability index in combination with a virtual simulation result and an apparent index; and the adaptive adjustment module dynamically adjusts simulation and sampling frequencies according to the reliability change rate. According to the invention, limitation of conventional electric control signals is penetrated, reduction of the real reliability of the disc brake system caused by high-frequency vibration, hydraulic oil pollution and other hidden physical degradation is accurately evaluated, and the false safety problem that the appearance of the system is normal but may be located at a failure edge is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drilling rig disc brake control testing, in particular to a drilling rig disc brake control testing system and method based on a cloud platform. BACKGROUND

[0002] The drilling rig disc brake system is the core equipment to ensure the safety of drilling operations, and its operation reliability is crucial. At present, the industry generally relies on conventional apparent self-checking methods, mainly monitoring apparent electric control signals such as electromagnetic valve driving current and voltage. However, such methods are seriously divorced from real physical conditions such as high-frequency vibration and hydraulic oil pollution at the well site, and cannot quantify the hidden physical damage caused by these factors. This leads to a serious technical defect: even if the system passes self-checking and appears normal, its internal key components may have reached the failure edge due to physical degradation, and there is a false safety problem. Therefore, how to penetrate the limitations of apparent electric control signals, accurately assess the real reliability decline caused by hidden physical degradation, and predict sudden safety risks is a technical problem that needs to be solved in the field. SUMMARY

[0003] To solve the above technical problems, the present application provides a drilling rig disc brake control testing system and method based on a cloud platform. Specifically, the technical solution of the present application is as follows: A drilling rig disc brake control testing system based on a cloud platform, comprising: a well site data acquisition module that acquires apparent electric control signals of a drilling rig disc brake system and hidden physical layer data reflecting real working conditions, the hidden physical layer data including high-frequency vibration data and hydraulic oil data, to obtain apparent index data and hidden working condition data; a degradation mechanism modeling module that, based on the hidden working condition data, calls a preset degradation mechanism model of a key component, quantitatively calculates a physical degradation state parameter of the key component, and obtains dynamic degradation parameter information; a digital twin simulation module that, based on the dynamic degradation parameter information, dynamically updates key parameters of a preset disc brake system digital twin, and applies a preset virtual extreme working condition braking simulation to the updated digital twin at a high frequency to solve a virtual performance index, the virtual performance index including a virtual braking time and a virtual braking torque, and obtain a virtual simulation result; a real reliability evaluation module that, based on the virtual simulation result and the apparent index data, calculates a hidden degradation weight factor and a disc brake system real reliability index, and generates a real-time reliability score; An adaptive adjustment module calculates a change rate of the real reliability index contained in the real-time reliability score, and reversely adjusts a simulation frequency of the digital twin simulation module and a data sampling rate of the well site data acquisition module to obtain a dynamic frequency adjustment instruction in response to the change rate.

[0004] Preferably, the hidden working condition data includes a high-frequency vibration root mean square value and a vibration main frequency; the hidden working condition data further includes a hydraulic oil granularity, a dynamic viscosity, and a water content; the dynamic degradation parameter information includes an electromagnetic valve effective friction, a hysteresis time, and an elastic modulus; and the real-time reliability score includes an implicit degradation weight factor and a real reliability index.

[0005] Preferably, the well site data acquisition module includes: A hidden layer data acquisition submodule acquires high-frequency vibration data through an acceleration sensor and acquires hydraulic oil product data through an online oil quality sensor; An apparent layer data acquisition submodule acquires conventional electric control signals, including control instructions and driving currents and voltages; An edge processing submodule performs time domain and frequency domain analysis on the high-frequency vibration data in a local edge layer, extracts a root mean square value and a main frequency of a key frequency band to reduce an uploaded data amount, and generates the hidden working condition data.

[0006] Preferably, the degradation mechanism modeling module includes: An offline training submodule uses a high-precision physical model to simulate different combinations of oil pollution, vibration, and viscosity, and establishes a high-dimensional database; A proxy model training submodule trains a lightweight proxy model using the high-dimensional database; An online inference submodule runs the lightweight proxy model on a cloud side, instantaneously outputs the physical degradation state parameter based on the hidden working condition data received in real time, and solves a contradiction between a high-precision model and real-time cloud computing.

[0007] Preferably, the online inference submodule in the degradation mechanism modeling module is specifically used for: Calculating a dynamic basic friction force, which is a function of fluid viscosity variation; Calculating an additional friction force, which is a result of cumulative abrasion caused by solid particles and corrosion caused by water; Calculating a vibration amplification factor, which is a function of vibration amplitude and vibration main frequency; Coupling a calculated effective friction, summing the dynamic base friction and the additional friction, multiplying by the vibration amplification factor, resulting in an electromagnetic valve effective friction as the dynamic degradation parameter information.

[0008] Preferably, the real reliability evaluation module comprises: An implicit degradation weight calculation submodule, based on the virtual braking time and virtual braking torque in the virtual simulation result, comparing the preset safety specification maximum braking time and safety specification minimum braking torque, calculating the time margin and torque margin; and multiplying the time margin and the torque margin to obtain an implicit degradation weight factor; A reliability fusion calculation submodule, multiplying the apparent self-inspection pass rate in the apparent index data and the implicit degradation weight factor to obtain the disc brake system real reliability index; A score generation submodule, combining the implicit degradation weight factor and the real reliability index to generate the real-time reliability score.

[0009] Preferably, the adaptive adjustment module is specifically used for: Calculating the first derivative of the disc brake system real reliability index to obtain a degradation change rate; Calling a preset stable threshold and a sudden degradation threshold; Judging whether the degradation change rate is less than the stable threshold, if yes, setting the simulation frequency to low frequency; Judging whether the degradation change rate is greater than or equal to the stable threshold and less than the sudden degradation threshold, if yes, increasing the simulation frequency to medium frequency; Judging whether the degradation change rate is greater than or equal to the sudden degradation threshold, if yes, increasing the simulation frequency to high frequency, and synchronously instructing the well site data acquisition module to increase the data sampling rate.

[0010] Preferably, the system further comprises: A self-calibration closed loop module, for when a physical entity occurs a real braking event: Capturing real working condition data and real braking results of the real braking event; Instructing the digital twin simulation module to use the real working condition data for reproduction simulation to obtain a virtual braking result; Calculating the error between the real braking result and the virtual braking result; And based on the error, through an optimization algorithm, reversely fine-tuning the model parameters in the degradation mechanism modeling module to solve the model mismatch problem.

[0011] A drilling rig disc brake control test method based on a cloud platform, comprising the following steps: Step S1: Obtain the apparent electric control signal of the drilling rig disc brake system and hidden physical layer data reflecting the real working condition, the hidden physical layer data including high-frequency vibration data and hydraulic oil data, to obtain apparent index data and hidden working condition data; Step S2: Based on the hidden working condition data, call the preset degradation mechanism model of the key component, quantitatively calculate the physical degradation state parameter of the key component, and obtain dynamic degradation parameter information; Step S3: Based on the dynamic degradation parameter information, dynamically update the key parameters of the preset disc brake system digital twin, and apply a preset virtual limit working condition braking simulation to the updated digital twin at a high frequency, to solve the virtual performance index including virtual braking time and virtual braking torque, and obtain a virtual simulation result; Step S4: Based on the virtual simulation result and the apparent index data, calculate the implicit degradation weight factor and the real reliability index of the disc brake system, and generate a real-time reliability score; Step S5: Based on the real reliability index contained in the real-time reliability score, calculate the change rate, and in response to the change rate, adjust the simulation frequency of step S3 and the data sampling rate of step S1 in reverse.

[0012] Compared with the prior art, the present application has the following advantages: 1. The present application can penetrate the limitations of conventional electric control signals, accurately evaluate the real reliability decline of the disc brake system caused by high-frequency vibration, hydraulic oil pollution and other implicit physical degradation factors, and solve the false safety problem that the system appears normal but may be on the edge of failure; 2. The present application can predict the limit performance index of the system when it encounters the maximum impact in real time, such as virtual braking time and virtual braking torque, by dynamically updating the digital twin and applying virtual limit working condition braking simulation, so as to predict the sudden safety risk in advance; 3. The present application constructs a closed-loop feedback mechanism based on the change rate of the real reliability index, which can dynamically adjust the digital twin simulation frequency and well site data acquisition sampling rate according to the accumulation speed of the risk, and realizes the optimization of the allocation of computing resources and the instantaneous high-precision capture of sudden degradation; 4. The present application adopts the strategy of offline training agent model and online inference, solves the contradiction between the large amount of calculation of high-precision physical model and the requirement of millisecond-level response of real-time cloud computing, and guarantees the real-time and accuracy of the degradation parameter calculation. BRIEF DESCRIPTION OF DRAWINGS

[0013] The present application will be further explained in conjunction with the drawings and examples: Figure 1 is a structural diagram of the system of the present application.

[0014] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0016] Example 1: Please see Figure 1 A cloud-based drilling rig disc brake control test system includes: The well site data acquisition module acquires the apparent electrical control signals of the drilling rig disc brake system, as well as the hidden physical layer data that reflects the actual working conditions. The hidden physical layer data includes high-frequency vibration data and hydraulic oil data, thus obtaining apparent index data and hidden working condition data. The degradation mechanism modeling module, based on hidden operating condition data, calls the preset degradation mechanism model of key components, quantifies and calculates the physical degradation state parameters of key components, and obtains dynamic degradation parameter information. The digital twin simulation module dynamically updates the key parameters of the preset disc brake system digital twin based on dynamic degradation parameter information, and applies preset virtual extreme working condition braking simulation to the updated digital twin at high frequency to calculate virtual performance indicators, including virtual braking time and virtual braking torque, and obtains virtual simulation results. The real reliability assessment module calculates the implicit degradation weight factor and the real reliability index of the disc brake system based on virtual simulation results and apparent index data, and generates a real-time reliability score. The adaptive adjustment module calculates the rate of change of the disc brake system's true reliability index included in the real-time reliability score, and in response to the rate of change, adjusts the simulation frequency of the digital twin simulation module and the data sampling rate of the well site data acquisition module in reverse to obtain a dynamic frequency adjustment command.

[0017] This embodiment provides a cloud-based drilling rig disc brake control test system. Its overall architecture is based on a physical-digital fusion health assessment framework, which aims to solve the technical problem that existing technologies rely only on apparent indicators and are detached from real working conditions, thus failing to predict the safety risks caused by hidden damage. The system specifically includes: The well site data acquisition module is responsible for acquiring two types of key data during the operation of the drilling rig's disc brake system: one is apparent data reflecting the status of the electrical control system, and the other is hidden data reflecting the actual physical operating conditions and sources of degradation. In this embodiment, the apparent electrical control signals acquired by this module include, but are not limited to, control commands. Solenoid valve drive current and voltage These signals, once processed, can generate apparent indicator data, such as traditional self-test pass rates. This value is almost always 100% false normal in existing technology; Meanwhile, the hidden physical layer data acquired by this module is a key input that distinguishes this invention from existing technologies, as it quantifies the hidden physical factors of damage that are ignored by the G05B23 standard test. Specifically, this module collects high-frequency vibration data through acceleration sensors installed on the disc brake base and hydraulic station, and collects hydraulic oil data such as particle size, dynamic viscosity, and water content through online oil quality sensors. These raw data are then processed locally to form characteristic hidden operating condition data such as the root mean square value of vibration. and clock speed wait; The degradation mechanism modeling module acts as a translator, converting collected, scattered hidden operating condition data, such as vibration and oil contamination, into centralized, quantifiable changes in physical parameters that the digital twin can understand. Deployed in the cloud, this module receives hidden operating condition data in real time and invokes pre-defined degradation mechanism models for key components. (Computational model) Physical degradation parameters refer to the implicit changes in the internal physical properties of key components such as solenoid valve cores and hydraulic cylinder seals caused by factors such as vibration, oil contamination, and wear. These changes may include nonlinear increases in effective friction force, decreases in elastic modulus, or extensions in response hysteresis time. This module uses a degradation mechanism model to quantitatively calculate these physical degradation parameters in real time and outputs dynamic degradation parameter information, such as the effective friction force of the solenoid valve. And pass it to the simulation module; The core purpose of the digital twin simulation module is no longer to perform low-load self-testing in a greenhouse environment, but to answer a core safety question in real time: whether the physical system that carries all the current hidden degradation can still work reliably when it encounters an extreme impact in the next second; This module is deployed in the cloud computing layer, and it receives dynamic degradation parameter information from the modeling module, such as... Based on dynamic degradation parameter information, this module dynamically updates the key parameters of the preset digital twin of the disc brake system; for example, it updates the friction parameters of the solenoid valve model in the twin from the initial design value to the real-time degradation value. value; This module applies preset virtual extreme condition braking simulations to the degraded twin at high frequencies, such as simulating the maximum braking command during a full-load drop; by calculating the dynamic response of the degraded twin, it outputs virtual performance indicators. In this embodiment, these indicators specifically include virtual braking time. and virtual braking torque These metrics constitute the results of the virtual simulation. The true reliability assessment module integrates the apparent electronic control status. Simulation results of implicit physical degradation This will generate a completely new comprehensive health indicator that can eliminate false feelings of security. This module is deployed at the cloud application layer. It receives virtual simulation results from the simulation module and apparent index data from the acquisition module; this module calculates the implicit degradation weighting factor. This factor is used to quantify the performance margin decline caused by physical degradation; it takes apparent indicator data. and Multiply by each product to calculate the true reliability index of the disc brake system. ; This The fusion method is one of the core innovations of this invention; it solves the problems in the existing technology. The problem of seemingly normal operation but the system is actually on the verge of failure; module generation includes... and Real-time reliability scores are used for predictive alerts and maintenance decisions; The adaptive adjustment module aims to build an intelligent closed-loop feedback system that matches the system's attention level (i.e., computing resource consumption) with the risk level, thereby achieving optimal allocation of computing resources and instantaneous response to sudden risks. This module receives the disc brake system's true reliability index, which is included in the real-time reliability score. And calculate its rate of change, i.e., its first derivative, in real time. ; The magnitude of represents the acceleration of the system's latent degradation; This module responds to the rate of change and performs dynamic adjustments: when When stable Low, it outputs instructions to reduce the simulation frequency of the digital twin simulation module and the data sampling rate of the well site data acquisition module to save resources; when detected During accelerated descent It immediately outputs dynamic frequency modulation commands to increase the simulation frequency, for example, to 10 times / second and the data sampling rate, in order to capture sudden degradation with high precision.

[0018] Example 2: Hidden operating condition data includes the root mean square value of high-frequency vibration and the dominant vibration frequency; hidden operating condition data also includes hydraulic oil particle size, dynamic viscosity and water content; dynamic degradation parameter information includes the effective friction force, hysteresis time and elastic modulus of the solenoid valve; virtual simulation results include virtual full-load braking time and virtual peak braking torque; real-time reliability score includes hidden degradation weighting factor and real reliability index.

[0019] Based on Example 1, this embodiment further defines the data and parameters processed by the system to ensure the clarity and feasibility of the technical solution; this specific definition of parameters is the basis for the accurate calculation of subsequent degradation and reliability models. Hidden operating condition data is not merely the raw signal, but rather feature-enhanced data extracted through edge computing; specifically, it includes: Root mean square value of high frequency vibration With the dominant vibration frequency Both of these are derived from the original vibration signal. The data were extracted through time-frequency domain analysis, such as FFT, and used to quantify the impact of vibration energy and key vibration modes on components. Hydraulic oil particle size Dynamic viscosity With water content These three parameters are directly measured by online oil quality sensors to quantify the impact of oil quality deterioration on components such as valves, including wear, corrosion, and basic friction. Dynamic degradation parameter information is the output of the degradation mechanism modeling module and is a core parameter injected into the digital twin; in this embodiment, it preferably includes, but is not limited to: Effective friction force of solenoid valve This is a coupled parameter that integrates the effects of vibration, oil contamination, and viscosity, and is a key variable affecting valve response time; Hysteresis time and elastic modulus: These parameters are also affected by hidden operating conditions and together characterize the physical degradation state of the component; The virtual simulation results are the output of the digital twin simulation module, used to evaluate the limiting performance of the degraded system; specifically, they include: Virtual full-load braking time This refers to the time required for the load on the twin to come to a stop under virtual extreme conditions. Virtual peak braking torque This refers to the maximum torque that the twin can apply during virtual braking. The real-time reliability score is the final output of the evaluation module, and it must include two core metrics to provide comprehensive decision-making information: Latent degradation weighting factor This factor, ranging from 0 to 100%, individually reflects the degree of decrease in physical performance margin. True Reliability Index This factor (0-100%) incorporates apparent indicators. and latent degradation .

[0020] Example 3: The well site data acquisition module includes: The hidden layer data acquisition submodule collects high-frequency vibration data through an acceleration sensor and hydraulic oil data through an online oil quality sensor. The surface layer data acquisition submodule acquires conventional electronic control signals, including control commands and drive current and voltage. The edge processing submodule performs time-domain and frequency-domain analysis on high-frequency vibration data at the local edge layer, extracts the root mean square value and dominant frequency of key frequency bands to reduce the amount of data uploaded, and generates hidden operating condition data.

[0021] Based on Example 1, this embodiment specifically defines the internal structure and working method of the well site data acquisition module; this edge-cloud collaborative architecture aims to resolve the engineering contradiction between the acquisition of massive amounts of raw data from the well site and the limited transmission bandwidth in the cloud. This module specifically includes: The hidden layer data acquisition submodule is responsible for acquiring the raw data of the physical degradation source; specifically, it continuously acquires the raw time-domain signal of high-frequency vibration by installing an accelerometer on the disc brake base or hydraulic station. And by using online oil quality sensors to monitor the pipeline in real time, hydraulic oil data is collected, such as... , , ; The surface layer data acquisition submodule is responsible for acquiring the G05B23 conventional electronic control signals; specifically, it acquires control commands. Drive current With voltage These data are used to calculate apparent indicators such as And, during self-calibration, it serves as the input condition for reproducing the simulation; Edge processing submodule: This is the key optimization of this module, which is to perform preprocessing locally at the data source to reduce the amount of data uploaded, while extracting high-value feature information; this submodule is deployed at the local edge layer, such as the edge gateway or PLC at the well site; Time-domain and frequency-domain analysis refers to the analysis of the raw high-frequency vibration data acquired by the hidden layer data acquisition submodule. Local computations are performed, such as transforming the time domain to the frequency domain using the Fast Fourier Transform (FFT). After performing this analysis, this submodule extracts the root mean square values ​​of key frequency bands from the results. Reflects vibrational energy and dominant frequency Reflects vibration modes; This submodule will only and These two high-value characteristics, along with oil product data that changes less frequently, are used to generate hidden operating condition data and uploaded to the cloud.

[0022] Example 4: The degradation mechanism modeling module includes: The offline training submodule uses a high-precision physical model to simulate different combinations of oil contamination, vibration, and viscosity, and establishes a high-dimensional database. The agent model training submodule uses a high-dimensional database to train a lightweight agent model. The online inference submodule runs a lightweight agent model in the cloud and instantly outputs physical degradation state parameters based on real-time received hidden operating condition data, in order to resolve the contradiction between high-precision models and real-time cloud computing.

[0023] Based on Example 1, this embodiment specifically defines the implementation strategy of the degradation mechanism modeling module. This strategy aims to solve the technical difficulty 1, namely the sharp contradiction between the high precision physical model being accurate but computationally intensive and unable to run in real time, and the requirement for millisecond-level response in real-time cloud computing. This solution innovatively adopts an offline training-online inference proxy model strategy. This module specifically includes: Offline training submodule: This step is executed asynchronously in the background of the cloud before system deployment; it is used to solidify the complex knowledge of high-precision physical models. High-precision physical models refer to simulation models that are built based on complete fluid dynamics and structural dynamics, such as CFD or finite element analysis, and can accurately reflect the degradation mechanism of components, such as the vibration-oil-friction coupling effect, but with huge computational requirements. This submodule uses this high-precision model to address different types of oil contamination. ,vibration Viscosity By combining these methods to perform massive simulation calculations, a high-dimensional database of input conditions and output degradation parameters is established. The surrogate model training submodule is responsible for learning and fitting the complex nonlinear mapping relationships contained in the high-dimensional database; Lightweight surrogate models refer to computationally extremely efficient mathematical models, such as deep neural networks (DNNs) or radial basis function (RBF) networks; they are trained to reproduce the computational results of high-precision models at extremely high speeds. This submodule uses a high-dimensional database as a training set to train the lightweight surrogate model to achieve the preset fitting accuracy. Online Inference Submodule: This is the actual execution unit of the degradation mechanism modeling module running in real time in the cloud; its purpose is to realize the instantaneous calculation of degradation parameters. This submodule runs a lightweight agent model on the cloud platform; when it receives hidden operating condition data from the well site data acquisition module in real time, such as... , At this time, it no longer needs to run a complex physical model, but instead performs a fast forward propagation inference through a surrogate model, instantaneously outputting the physical degradation state parameters such as .

[0024] Example 5: The online inference submodule within the degradation mechanism modeling module is specifically used for: Calculate the dynamic foundation friction force, which is a function of the fluid viscosity change; Calculate the additional friction force, which is the cumulative result of abrasive wear caused by solid particles and corrosive wear caused by moisture. Calculate the vibration amplification factor, which is a function of the vibration amplitude and the dominant frequency. The effective friction force is calculated by coupling the dynamic basic friction force and the additional friction force, and then multiplying it by the vibration amplification factor to obtain the effective friction force of the solenoid valve as a dynamic degradation parameter.

[0025] Based on Examples 1 and 4, this embodiment further specifies the core computational logic executed by the online inference submodule, i.e., the lightweight proxy model, in the degradation mechanism modeling module. This logic corresponds to the innovative model 1 of this invention, which is to construct an effective friction force model of a solenoid valve coupled with vibration, oil stains and viscosity, breaking through the limitation of traditional G05B23 test that only focuses on electrical signals. The online inference submodule calculates the effective frictional force of the solenoid valve. At the same time, it innovatively couples three effects across physical domains; its core computational logic is as follows: Core Model Definition Specific definition: Define the effective frictional force of the solenoid valve spool. It consists of dynamic basic friction influenced by viscosity and additional friction influenced by contaminant accumulation, and this sum will also be nonlinearly amplified by well site vibration: Model Analysis: in, The calculated effective frictional force of the solenoid valve, in Newtons (N), is derived from the final calculation result of this step and will be output to Module 3 as dynamic degradation parameter information. The dynamic basic friction force is calculated by the following steps; Additional friction force, calculated using the following steps; Vibration amplification factor, dimensionless, is calculated using the following steps; Logic and Effect Integration: This Class B improvement creatively incorporates hydraulic fluid characteristics. Cumulative wear and instantaneous vibration Three factors across physical domains are multiplicatively coupled; this design can precisely quantify the hysteresis amplification caused by vibration and the friction accumulation caused by oil contamination, which is not possible with existing technologies. The formula The dimensions on both sides of the equation are the same, both being Newtons (N); Supporting model implementation Calculate dynamic foundation friction Implementation: This value is a function of the fluid viscosity change; for example, through... Calculation, where It is the basic frictional force calibrated through experiments, for example, the static frictional force of the valve core at startup measured under the standard viscosity of a clean oil. It is the dynamic viscosity measured in real time by the well site data acquisition module. It is a B-order improved function used to describe the change of frictional force with viscosity, for example, a simplified polynomial obtained by fitting experimental data to characterize the Stribeck effect; Calculate additional friction force Implementation: This value represents solid particulate matter. Abrasive wear and moisture caused The result of the combined accumulation of corrosion and wear; for example, through the improved Arrhenius cumulative damage model. calculate; in, This indicates only the degree of oil contamination. or The data obtained by the well site data acquisition module exceeded the safety threshold. , Damage only begins to accumulate when set according to industry standards; and These are the key offline calibration coefficients; To ensure that implementability complies with Article 26.3 of the Patent Law, the calibration process is clearly defined as follows: Calibration Prepare a group of particles with different constant particulate matter pollution levels. The rate of increase in valve core friction was measured in a controlled experiment using oil samples. ; By using this calibration dataset Perform least squares regression analysis and obtain the fit. dimension: ; Calibration Use and Similar logic, based on a set of data with different constant water contents calibration dataset The fitting yielded dimension: ; After dimensional verification, the integral formula is dimensionally consistent. ; Calculate the vibration amplification factor Implementation: This value is the vibration amplitude. and vibration dominant frequency Functions; for example, through A-level original models. calculate; in, and Provided by the well site data acquisition module. It is the vibration influence threshold determined based on the vibration resistance performance test of the component; Dimensionless basic magnification factor and Dimensionless resonant amplification function, such as one with a natural frequency The parameters in the Gaussian function centered at the center are as follows Offline calibration is also required. The calibration process is as follows: the key natural frequencies of the component are determined through modal analysis, such as frequency sweep excitation. Apply a set of constant amplitudes. and constant frequency The excitation was measured to determine the corresponding frictional amplification effect. ; By using this calibration dataset Perform nonlinear surface fitting to identify and Other unknown parameters in the function; Coupled calculation of effective friction force Implementation: The online inference submodule sums the results of calculating the dynamic basic friction force and the additional friction force, and then multiplies them by the result of calculating the vibration amplification factor to obtain the final effective friction force of the solenoid valve. And output it.

[0026] Example 6: The true reliability assessment module includes: The implicit degradation weight calculation submodule calculates the time margin and torque margin based on the virtual braking time and virtual braking torque in the virtual simulation results, compares them with the preset maximum braking time and minimum braking torque of the safety specifications, and then multiplies the time margin and torque margin to obtain the implicit degradation weight factor. The reliability fusion calculation submodule multiplies the apparent self-inspection pass rate in the apparent index data with the implicit degradation weight factor to obtain the true reliability index of the disc brake system. The scoring generation submodule combines implicit degradation weighting factors with the true reliability index to generate a real-time reliability score.

[0027] This embodiment, based on embodiment 1, specifically defines the internal structure and calculation logic of the true reliability assessment module; this logic corresponds to the innovative point 2A-level original parameters of this invention. and Its core objective is to integrate the multi-dimensional virtual simulation results (time and torque) output from Module 3 with the apparent index data from Module 1 to calculate a single, reliable index that can guide decision-making. ; This module specifically includes: Implicit degradation weight calculation submodule: The question is how much margin the performance indicators of the quantized degraded twin are still close to the failure boundary in the virtual limit test. Implementation: This submodule receives the virtual braking time from the virtual simulation results. and virtual braking torque ; Safety regulations for maximum braking time and safety specifications minimum braking torque This refers to the safety red line required by industry standards or design specifications, such as Second, These values ​​are stored in the module as preset design parameters. This submodule compares the simulation results with safety specifications and calculates the time margins respectively. and torque margin ; Definition: The margin calculation preferably adopts the original Class A model with an exponential decay function to quantify the risks caused by exceeding or falling short of the target: Model: in, Time margin and torque margin, dimensionless (0-1), calculated; Output of the digital twin simulation module; Preset industry safety standard parameters; : The ramp function indicates that the slope only applies when the margin is negative. or When the value inside the parentheses is positive, the factor will begin to decrease from a full margin of 1. Risk attenuation coefficient, dimensionless, derived from offline calibration; to ensure feasibility, its value logic is explicitly defined: this parameter is set based on a risk tolerance strategy, for example, This means that when the virtual braking time exceeds the limit by 20%, At that time, the margin decreased to This reflects a high sensitivity to penalties for exceeding time limits; Fusion: This submodule multiplies the time margin and torque margin to obtain the implicit degradation weighting factor. ; This multiplicative logic conforms to the series failure model, meaning that either exceeding the time limit or insufficient torque will lead to... The system flagged the price as declining and deemed it risky. Reliability fusion calculation submodule: The goal is to achieve a quantitative integration of apparent electrical reliability and implicit physical reliability; Implementation: This submodule will calculate the apparent self-inspection pass rate from the apparent indicator data. The G05B23 system measured, typically 100% and Multiply by the product to obtain the true reliability index of the disc brake system. ; Definition: True Reliability Index The formula for calculating a Level A original model is: Model: in, : Reliability index, dimension 0-100%; The apparent self-inspection pass rate was measured by the G05B23 system of the well site data acquisition module. : Implicit degradation weight factor, obtained from the implicit degradation weight calculation submodule; Fusion: This original model is based on the concept of a series model in system reliability engineering; it ensures that even if Displaying 100%, Also because The decline, for example Increased Extend, make It dropped to 90% and then declined; This submodule ultimately packages the implicit degradation weight factor. With the true reliability index This generates a real-time reliability score, which is used for final decision output and alerts, for example: .

[0028] Example 7: The adaptive adjustment module is specifically used for: Calculate the first derivative of the true reliability index of the disc brake system to obtain the degradation rate. Invoke the preset stability threshold and sudden degradation threshold; Determine if the rate of degradation is less than the stability threshold; if so, set the simulation frequency to low frequency. Determine whether the degradation rate is greater than or equal to the stability threshold and less than the sudden degradation threshold. If so, increase the simulation frequency to the intermediate frequency. Determine whether the degradation rate is greater than or equal to the sudden degradation threshold. If so, increase the simulation frequency to a high frequency and simultaneously instruct the well site data acquisition module to increase the data sampling rate.

[0029] Based on Example 1, this embodiment specifically defines the internal working logic of the adaptive adjustment module. This logic corresponds to the Class 1A innovation mechanism of this invention, which is to construct a closed-loop feedback based on the risk change rate and dynamically adjust the computational resource input of the system to achieve the best balance between cost and risk sensitivity. The adjustment logic of the adaptive adjustment module can be specifically divided into: Calculate the rate of change The module receives the disc brake system real-time reliability index output by the real reliability assessment module in real time. ; Calculate using methods such as finite difference. The first derivative yields the rate of degradation. Right now ; The magnitude reflects the rate at which the system degrades, i.e., the risk accumulates; Call threshold Module calls preset stability threshold and sudden degradation threshold ; To ensure feasibility, the logic for setting these thresholds is explicitly defined: and ( This is determined through statistical analysis of a large dataset of historical degradation data; for example, It can be set to the P90 quantile value that covers 90% of the normal operating volatility, for example... ,and The P99 quantile value can be set to characterize sudden failures such as oil emulsification, for example... This ensures that the regulation logic has clear statistical significance and engineering robustness; Execute three-stage adjustment logic Determine the rate of degradation Is it less than the stability threshold? : If so, the system is determined to be in a stable state; the module sets the simulation frequency. For low frequencies, such as once per minute, the well site data acquisition module can be synchronously instructed to use a low sampling rate to save computing and communication resources; Determine the rate of degradation Is it greater than or equal to? And less than the sudden degradation threshold : If so, the system is determined to have entered an observable, slow degradation phase; the module increases the simulation frequency. To improve monitoring accuracy, the frequency can be reduced to medium frequency, for example, once per 10 seconds. Determine the rate of degradation Is it greater than or equal to the sudden degradation threshold? : If so, the system is determined to be experiencing a sudden damage, such as component resonance, which poses an extremely high risk; the module immediately increases the simulation frequency. Up to a high frequency, such as 10 times / second, and synchronously instruct the well site data acquisition module to increase the data sampling rate, so as to achieve instantaneous high-precision capture of sudden risks.

[0030] Example 8: The system also includes: A self-calibrating closed-loop module is used when a real braking event occurs in a physical entity: Capture real-world operating condition data and real-world braking results for real braking events; The instruction digital twin simulation module uses real-world operating condition data to reproduce the simulation and obtain virtual braking results. Calculate the error between the actual braking results and the virtual braking results; Based on the error, the model parameters in the degradation mechanism modeling module are fine-tuned in reverse through optimization algorithms to solve the model mismatch problem.

[0031] This embodiment adds a self-calibration closed-loop module to the system in Embodiment 1. This module corresponds to the solution of the innovation point 4A-level innovation architecture and technical difficulty 2 of this invention. Its core purpose is to solve the degradation mechanism modeling module, such as the wear coefficient, in the model parameters of Embodiments 4 and 5. Resonance amplification factor Natural frequency Model mismatch issues caused by time drift or individual differences; The self-calibration closed-loop module's workflow is triggered when a real braking event occurs on the physical drilling rig, as follows: Capturing real events When the well site data acquisition module detects a real braking event that is typically a routine, non-extreme condition, module one immediately captures the entire process of the event's real-world data, such as actual braking conditions. Sequence and actual braking results, such as actual braking time ; Twin recurrence The self-calibration closed-loop module instructs the digital twin simulation module Module 3 to reproduce the simulation: The digital twin simulation module pauses the regular virtual limit simulation and instead executes a reproduction simulation; it uses the exact same real-world operating condition data captured in the previous step as input, and runs it on the digital twin in the current degradation state, which uses the current parameters of Module 2, to obtain the virtual braking result. ; Calculation error The self-calibrating closed-loop module calculates the actual braking results. With virtual braking results Error between ;this It accurately reflects the deviation between the current digital twin model and the actual state of the physical entity; Reverse calibration The self-calibration closed-loop module is based on error. As an objective function, such as minimization The square of the value is used to fine-tune the model parameters in Module 2 of the degradation mechanism modeling module through optimization algorithms such as Particle Swarm Optimization (PSO) or gradient descent. The offline calibration coefficients defined in Example 5 are used until... It converges to the minimum.

[0032] Example 9: Please see Figure 2 A cloud-based drilling rig disc brake control test method includes the following steps: Step S1: Obtain the apparent electrical control signals of the drilling rig disc brake system and the hidden physical layer data reflecting the actual working conditions. The hidden physical layer data includes high-frequency vibration data and hydraulic oil data, and obtain the apparent index data and hidden working condition data. Step S2: Based on the hidden operating condition data, call the preset degradation mechanism model of the key components, quantify and calculate the physical degradation state parameters of the key components, and obtain dynamic degradation parameter information; Step S3: Based on the dynamic degradation parameter information, dynamically update the key parameters of the preset disc brake system digital twin, and apply the preset virtual extreme working condition braking simulation to the updated digital twin at high frequency to calculate the virtual performance index including virtual braking time and virtual braking torque, and obtain the virtual simulation results. Step S4: Based on the virtual simulation results and apparent index data, calculate the implicit degradation weight factor and the true reliability index of the disc brake system, and generate a real-time reliability score; Step S5: Based on the disc brake system real reliability index included in the real-time reliability score, calculate its rate of change, and in response to the rate of change, adjust the simulation frequency of step S3 and the data sampling rate of step S1 in reverse.

[0033] This embodiment provides a cloud platform-based drilling rig disc brake control test method. This method achieves the functions of the system in Embodiment 1 through a series of steps. This method is preferably executed by the system modules of Embodiments 1-8. This method specifically includes the following steps: Step S1: Obtain data This step is preferably performed by the well site data acquisition module; The apparent electronic control signals of the drilling rig disc brake system are acquired through sensors, such as... And hidden physical layer data that reflects the actual working conditions; the hidden physical layer data specifically includes high-frequency vibration data and hydraulic oil data such as particle size, viscosity, and water content; To further optimize, as defined in Example 3, the high-frequency vibration data is processed at the local edge layer to extract... and The apparent index data obtained are as follows With hidden operating condition data; Step S2: Calculate degradation parameters This step is preferably performed in the cloud by the degradation mechanism modeling module; Hidden operating condition data obtained based on S1 Wait, invoke the preset degradation mechanism model of key components; To ensure the real-time performance and accuracy of the calculations, as defined in Examples 4 and 5, this step involves running a lightweight surrogate model and performing vibration-oil contamination-viscosity coupling calculations. To quantify and calculate the physical degradation state parameters of key components, such as This yields dynamic degradation parameter information; Step S3: Perform simulation This step is preferably performed in the cloud by a digital twin simulation module; Dynamic degradation parameter information based on S2 output Dynamically update the key parameters of the preset disc brake system digital twin, i.e., inject degradation; For the updated, i.e. degraded, digital twin, a preset virtual extreme condition braking simulation is applied by S5 at a high frequency. The simulation results were calculated, including the virtual braking time. and virtual braking torque The virtual performance metrics are used to obtain virtual simulation results; Step S4: Assess reliability This step is preferably performed in the cloud by the real reliability assessment module; Virtual simulation results based on S3 Apparent index data of S1 ; To achieve accurate quantification of reliability, as defined in Example 6, this step involves margin calculation. Obtain the implicit degradation weighting factor ; Calculate the true reliability index of the disc brake system ; Generate includes and Real-time reliability score; Step S5: Adaptive Adjustment This step is preferably performed in the cloud by the adaptive adjustment module; Real-time reliability score generated based on S4 Calculate its rate of change ; In response to the rate of change, to construct a closed-loop feedback, as defined in Example 7, this step is based on the rate of change and , Based on the comparison results, the simulation frequency in step S3 is adjusted in reverse. and the data sampling rate in step S1.

[0034] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A cloud-based drilling rig disc brake control test system, characterized in that, The system includes: The well site data acquisition module acquires the apparent electrical control signals of the drilling rig disc brake system, as well as the hidden physical layer data reflecting the actual working conditions. The hidden physical layer data includes high-frequency vibration data and hydraulic oil data, thus obtaining apparent index data and hidden working condition data. The degradation mechanism modeling module, based on the hidden operating condition data, calls the preset degradation mechanism model of the key components, quantifies and calculates the physical degradation state parameters of the key components, and obtains dynamic degradation parameter information; The digital twin simulation module dynamically updates the key parameters of the preset disc brake system digital twin based on the dynamic degradation parameter information, and applies preset virtual extreme working condition braking simulation to the updated digital twin at high frequency to calculate virtual performance indicators, including virtual braking time and virtual braking torque, and obtains virtual simulation results. The real reliability assessment module calculates the implicit degradation weight factor and the real reliability index of the disc brake system based on the virtual simulation results and the apparent index data, and generates a real-time reliability score. The adaptive adjustment module calculates the rate of change of the disc brake system's true reliability index included in the real-time reliability score, and in response to the rate of change, adjusts the simulation frequency of the digital twin simulation module and the data sampling rate of the well site data acquisition module in reverse to obtain a dynamic frequency adjustment command.

2. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The hidden operating condition data includes the root mean square value of high-frequency vibration and the dominant vibration frequency; the hidden operating condition data also includes hydraulic oil particle size, dynamic viscosity and water content; the dynamic degradation parameter information includes the effective friction force, hysteresis time and elastic modulus of the solenoid valve; the virtual simulation results include virtual full-load braking time and virtual peak braking torque; the real-time reliability score includes the hidden degradation weighting factor and the true reliability index.

3. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The well site data acquisition module includes: The hidden layer data acquisition submodule collects high-frequency vibration data through an acceleration sensor and hydraulic oil data through an online oil quality sensor. The surface layer data acquisition submodule acquires conventional electronic control signals, including control commands and drive current and voltage. The edge processing submodule performs time-domain and frequency-domain analysis on the high-frequency vibration data at the local edge layer, extracts the root mean square value and dominant frequency of key frequency bands to reduce the amount of uploaded data, and generates the hidden operating condition data.

4. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The degradation mechanism modeling module includes: The offline training submodule uses a high-precision physical model to simulate different combinations of oil contamination, vibration, and viscosity, and establishes a high-dimensional database. The agent model training submodule uses the high-dimensional database to train a lightweight agent model; The online inference submodule runs the lightweight agent model in the cloud and instantly outputs the physical degradation state parameters based on the hidden working condition data received in real time, so as to resolve the contradiction between the high-precision model and real-time cloud computing.

5. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The online inference submodule within the degradation mechanism modeling module is specifically used for: Calculate the dynamic base friction force, which is a function of the fluid viscosity change; Calculate the additional frictional force, which is the cumulative result of abrasive wear caused by solid particles and corrosive wear caused by moisture; Calculate the vibration amplification factor, which is a function of the vibration amplitude and the dominant vibration frequency; The effective friction force is calculated by coupling the dynamic basic friction force and the additional friction force, and then multiplying the sum by the vibration amplification factor to obtain the effective friction force of the solenoid valve, which serves as the dynamic degradation parameter information.

6. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The true reliability assessment module includes: The implicit degradation weight calculation submodule calculates the time margin and torque margin based on the virtual braking time and virtual braking torque in the virtual simulation results, compares them with the preset maximum braking time and minimum braking torque of the safety specifications, and multiplies the time margin and torque margin to obtain the implicit degradation weight factor. The reliability fusion calculation submodule multiplies the apparent self-inspection pass rate in the apparent index data with the implicit degradation weight factor to obtain the true reliability index of the disc brake system. The scoring generation submodule combines the implicit degradation weight factor with the true reliability index to generate the real-time reliability score.

7. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The adaptive adjustment module is specifically used for: Calculate the first derivative of the true reliability index of the disc brake system to obtain the degradation rate. Invoke the preset stability threshold and sudden degradation threshold; Determine whether the degradation rate is less than the stability threshold; if so, set the simulation frequency to low frequency. Determine whether the degradation rate is greater than or equal to the stability threshold and less than the sudden degradation threshold; if so, increase the simulation frequency to the intermediate frequency. Determine whether the degradation rate is greater than or equal to the sudden degradation threshold. If so, increase the simulation frequency to a high frequency and simultaneously instruct the well site data acquisition module to increase the data sampling rate.

8. The drilling rig disc brake control test system based on a cloud platform according to claim 1, characterized in that, The system also includes: A self-calibrating closed-loop module is used when a real braking event occurs in a physical entity: Capture the actual operating condition data and actual braking results of the actual braking events; The digital twin simulation module described in the instruction uses the real working condition data to perform a simulation to obtain virtual braking results; Calculate the error between the actual braking result and the virtual braking result; Based on the error, the model parameters in the degradation mechanism modeling module are fine-tuned in reverse using an optimization algorithm to solve the model mismatch problem.

9. A cloud-based drilling rig disc brake control test method, based on the cloud-based drilling rig disc brake control test system according to any one of claims 1 to 8, characterized in that, The method includes the following steps: Step S1: Obtain the apparent electronic control signals of the drilling rig disc brake system and the hidden physical layer data reflecting the actual working conditions. The hidden physical layer data includes high-frequency vibration data and hydraulic oil data to obtain apparent index data and hidden working condition data. Step S2: Based on the hidden operating condition data, call the preset degradation mechanism model of the key components, quantify and calculate the physical degradation state parameters of the key components, and obtain dynamic degradation parameter information; Step S3: Based on the dynamic degradation parameter information, dynamically update the key parameters of the preset disc brake system digital twin, and apply preset virtual extreme working condition braking simulation to the updated digital twin at high frequency to calculate the virtual performance index including virtual braking time and virtual braking torque, and obtain the virtual simulation results. Step S4: Based on the virtual simulation results and the apparent index data, calculate the implicit degradation weight factor and the true reliability index of the disc brake system, and generate a real-time reliability score; Step S5: Based on the disc brake system real reliability index included in the real-time reliability score, calculate its rate of change, and in response to the rate of change, adjust the simulation frequency of step S3 and the data sampling rate of step S1 in reverse.

Images