Reservoir helium resource dynamic evaluation method based on real-time data

Abstract

The invention relates to the technical field of helium resource evaluation, and particularly discloses a reservoir helium resource dynamic evaluation method based on real-time data, and the method comprises the steps: obtaining precise helium concentration data; performing three-dimensional modeling and finite element stress field coupling analysis on the precise helium concentration data to generate helium distribution migration data; based on the helium distribution migration data, adopting a deep learning model for processing to obtain helium resource quantity evolution data; performing dynamic recovery rate evaluation and path optimization processing on the helium resource quantity evolution data under multiphase flow constraint to generate a final mining scheme; performing strategy self-correction processing on the final mining scheme, and finally outputting a dynamic evaluation and intelligent regulation and control scheme; potential exploitation risks can be predicted and avoided through reservoir stress simulation and multiphase flow optimization, safe exploitation of the reservoir is ensured, and the system is beneficial to prolonging the service life of the reservoir through efficient resource recovery and accurate resource evaluation.

Description

Technical Field

[0001] The present invention belongs to the technical field of helium resource evaluation, and particularly relates to a dynamic evaluation method for reservoir helium resources based on real-time data. Background Art

[0002] In the existing reservoir helium resource evaluation methods, they mainly rely on static geological exploration data and traditional resource evaluation models. These methods usually include steps such as geological survey, core analysis, reservoir simulation, etc. to obtain the structure, physical properties and fluid characteristics of the reservoir. However, these methods have obvious limitations, especially when facing complex and changeable reservoir environments and dynamically changing helium resources.

[0003] Traditional resource evaluation models are usually based on simplified assumptions and idealized conditions, and it is difficult to accurately simulate the actual flow and distribution of helium in the reservoir. These models often ignore the influence of reservoir stress and multiphase flow on the distribution and migration of helium, resulting in large errors in the evaluation results, and thus lacking the support of real-time data and the ability of dynamic adjustment. During the actual exploitation process, the state of the reservoir and the distribution of helium will change, while the traditional evaluation methods cannot timely obtain this change information, nor can they perform dynamic adjustment and optimization based on real-time data. This leads to exploitation plans often being based on outdated data and information and unable to effectively cope with the uncertain factors in the reservoir.

[0004] In view of this, the inventor proposes a dynamic evaluation method for reservoir helium resources based on real-time data to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a dynamic evaluation method for reservoir helium resources based on real-time data to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A dynamic evaluation method for reservoir helium resources based on real-time data, comprising: Obtaining helium concentration data in the reservoir, performing real-time monitoring and adaptive filtering denoising processing on the helium concentration in the reservoir to obtain accurate helium concentration data; Performing three-dimensional modeling and finite element stress field coupling analysis on the accurate helium concentration data, and generating helium distribution and migration data in combination with real-time formation pressure and temperature data; Based on the helium distribution and migration data, using a deep learning model to perform spatio-temporal feature extraction and dynamic prediction processing to obtain helium resource quantity evolution data; Performing dynamic recovery rate evaluation and path optimization processing under the constraint of multiphase flow on the helium resource quantity evolution data to obtain the final exploitation plan; Perform strategy self-correction processing on the final exploitation plan to obtain a dynamic evaluation and intelligent regulation plan.

[0007] Preferably, the adaptive filtering denoising algorithm is used to remove noise and errors from gas concentration monitoring devices; Kalman filtering is used as the denoising algorithm to predict and correct noise components in real-time data streams. The Kalman filtering formula is: ; where is the state estimate at the current time; is the current observation value; is the observation matrix, which describes the observation process of the system and converts the system state into an observation value; is the Kalman gain, which is used to determine the fusion weight of the system for the observation value and the predicted value; Through filtering, the random noise of the measuring device can be removed, and more accurate helium concentration data can be obtained; the reliability of the real-time monitoring system can be improved, providing accurate data input for subsequent dynamic evaluation.

[0008] Preferably, the finite element stress field coupling analysis is used for reservoir stress and gas distribution simulation. Based on the geological structure of the reservoir, the changes of stress, pressure, and temperature factors are simulated, and the spatial distribution and migration trend of helium are inferred. The formula for the finite element stress field coupling analysis is: ; where is the formation stress tensor, which describes the stress state of each point in the reservoir; J is the gas flow, which describes the flow behavior of helium in the reservoir and is affected by temperature and pressure factors; Q is the source term (such as heat source, gas flow, etc.), which describes the release and adsorption process of helium; represents the change of reservoir stress with time, reflecting the stress change caused by the exploitation process, gas flow, or other external factors; is used to describe the flow trend of helium in the reservoir. The divergence operator calculates the outflow or inflow of gas, helping to understand the gas distribution and the dynamic state of the reservoir; The finite element stress field coupling analysis can simulate the stress changes of the reservoir at different exploitation stages, predict the distribution and flow of helium, and optimize the recovery strategy.

[0009] Through the dynamic simulation of stress and gas distribution, potential exploitation risks can be identified in advance, and resource waste or wellhead damage caused by unstable gas distribution can be reduced; Simulate the stress state of the reservoir through finite element analysis to infer the distribution of helium. Combine the 3D modeling of the reservoir with real-time data, and use factors such as formation pressure, stress changes, and temperature to conduct dynamic simulation of the helium distribution in the reservoir.

[0010] Preferably, the deep learning model is a convolutional neural network (CNN) fused with a long short-term memory network (LSTM). The deep learning model is used to process time series data (such as helium concentration changes, temperature, pressure) and predict the dynamic changes of helium resources in the reservoir. Among them, the convolutional neural network is used to extract the spatial features in the time series data of helium concentration. The formula of the convolutional neural network is: ; Among them, ht is the output feature of the convolutional layer. W is the convolutional kernel weight, which extracts features from the input data and learns the spatial features of the helium concentration in the reservoir. b is the bias. xt is the input data (helium concentration data or other parameters). The long short-term memory network is used to capture the trend of helium concentration changing over time. The formula of the long short-term memory network is: ; Wf, Wi are the gating weights of the long short-term memory network: used to control the magnitude of information flow and capture important dynamic information in the time series.

[0011] is the prediction output: based on the combination of the convolutional neural network and the long short-term memory network, predict the future changes in the amount of helium resources in the reservoir. Combining CNN and LSTM can not only make short-term predictions of helium resources, but also model long-term change trends and dynamically adjust the mining strategy.

[0012] The real-time dynamic adjustment ability of this model significantly improves the accuracy and response speed of resource assessment.

[0013] Preferably, the multiphase flow constraint is used to optimize the flow path of helium in the reservoir. The particle swarm optimization algorithm (PSO) is used to update the particle velocity at time t + 1, and the best mining path and recovery plan are found through iteration to improve the recovery rate of resources. The formula of the particle swarm optimization algorithm is: ; Among them, vi(t) is the velocity of the particle at the t-th moment, which is used to adjust the velocity of the helium flow path and is updated based on the current state and the historical best position. $x_i$ is the position of the particle, representing the path of helium flow or the recovery strategy; $pbesti$ is the historical best position of the particle; $gbest$ is the global best position, representing the optimal helium recovery path; $w$ is the inertia weight; $c_1$, $c_2$ are acceleration constants; $r_1$, $r_2$ are random numbers; For the new position of each particle, i.e., the new scheme, a fitness evaluation is carried out; the fitness function combines the helium flow characteristics, reservoir state and recovery efficiency under multiphase flow constraints, reflecting the advantages and disadvantages of the scheme for helium recovery efficiency and risk control.

[0014] If the fitness of the new scheme is better than the best scheme in the history of the particle, then update the historical optimal solution of the particle; if its fitness is better than the global best solution, then update the global best scheme; Repeat the above iterative process until the preset maximum number of iterations is reached or the fitness converges; Finally, the particle position corresponding to the global best solution $gbest$ is the best helium extraction path and recovery scheme under multiphase flow constraints; Through the particle swarm optimization algorithm, it is possible to quickly find the optimized helium recovery path, reduce resource waste, and improve the recovery rate; the algorithm adaptively adjusts the recovery path, can dynamically optimize the extraction scheme according to real-time data, and improves the economic efficiency and sustainability of the entire extraction process.

[0015] Preferably, the formula for the self-correction process of the strategy is: ; where $u(t)$ represents the control strategy at time $t$, such as extraction parameters; $y_{target}(t)$ represents the expected target, such as the target recovery rate or resource volume; $y_{actual}(t)$ represents the actual data obtained through real-time monitoring; $\alpha$ is the adjustment factor; The above formula describes the process of continuously correcting the control strategy through real-time error feedback, i.e., the difference between the target and the actual, so as to achieve dynamic evaluation and intelligent regulation.

[0016] Preferably, the helium concentration data is obtained by real-time monitoring of the helium concentration in the reservoir using a high-sensitivity spectroscopic analyzer; The accurate helium concentration data is obtained by using an adaptive filtering and denoising algorithm to remove noise and correct errors from the helium concentration data.

[0017] Preferably, the dynamic recovery rate evaluation and path optimization process are realized by means of real-time reservoir data and a simulated gas-liquid phase conversion model.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: In the present invention, reservoir stress simulation and multiphase flow optimization can predict and avoid potential mining risks, ensuring the safe mining of the reservoir; at the same time, by improving the resource recovery efficiency, unnecessary mining costs and environmental impacts are reduced; real-time data visualization and automated report generation can improve management efficiency, reduce manual intervention, ensure that decisions are based on accurate real-time data, and at the same time provide real-time feedback and warning mechanisms for managers, improving work efficiency and accuracy; through efficient resource recovery and accurate resource assessment, the system helps to extend the service life of the reservoir, reduce the risk of over-mining, and promote environmental protection and sustainable mining. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a dynamic evaluation method for helium resources in a reservoir based on real-time data according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Embodiment 1: Please refer to Figure 1 As shown, the embodiment of the present application provides a dynamic evaluation method for helium resources in a reservoir based on real-time data, including: For step S1, helium concentration data in the reservoir is obtained, and the helium concentration in the reservoir is monitored in real time and processed by adaptive filtering and denoising to obtain accurate helium concentration data; For the above steps, the helium concentration data is obtained by real-time monitoring of the helium concentration in the reservoir by a high-sensitivity spectroscopic analyzer; The accurate helium concentration data is obtained by removing noise and correcting errors from the helium concentration data by an adaptive filtering and denoising algorithm; By using a high-sensitivity spectroscopic analyzer, such as an infrared spectrum, the helium concentration in the reservoir is monitored in real time to generate accurate helium concentration data. The real-time monitoring of the helium concentration is the basis for the evaluation of helium resources in the reservoir.

[0022] Gas analysis is carried out using an infrared spectrometer. The instrument and equipment can be installed in the reservoir monitoring well to ensure that the sampling point can represent the distribution of helium in the reservoir; The data is transmitted to the data processing platform through wireless communication (such as LoRa); The Kalman Filter is applied to smooth the collected real-time data and make adaptive adjustments to cope with changes in the external environment and equipment noise; The filtered data is passed to the subsequent stage for further analysis; The adaptive filtering denoising algorithm is used to remove the noise and errors from the gas concentration monitoring device; Kalman filtering is used as the denoising algorithm to predict and correct the noise components in the real-time data stream. The Kalman filtering formula is: ; Where, is the state estimate at the current moment; is the current observation value; is the observation matrix, which describes the observation process of the system and converts the system state into the observation value; is the Kalman gain, which is used to determine the fusion weight of the system for the observation value and the predicted value; Through filtering, the random noise of the measuring device can be removed, and more accurate helium concentration data can be obtained; the reliability of the real-time monitoring system can be improved, providing accurate data input for subsequent dynamic evaluation.

[0023] For step S2, three-dimensional modeling and finite element stress field coupling analysis are performed on the precise helium concentration data, and helium distribution migration data is generated by combining real-time formation pressure / temperature data; Finite Element Analysis (FEM): The finite element method is used to perform three-dimensional modeling of the reservoir, simulating the stress distribution and dynamic changes of helium in the reservoir.

[0024] Formation pressure and temperature monitoring: Combining real-time pressure and temperature data, calculate the stress changes in the reservoir and infer the migration path of helium in the reservoir.

[0025] Use seismic data and drilling data to generate a three-dimensional model of the reservoir, and apply finite element analysis software such as ANSYS to simulate the stress state of the reservoir at different production stages.

[0026] Combining the real-time monitored data, make real-time adjustments to the simulation results to predict the flow and distribution of helium; The finite element stress field coupling analysis is used for reservoir stress and gas distribution simulation. Based on the geological structure of the reservoir, simulate the changes of stress, pressure, and temperature factors, and infer the spatial distribution and migration trend of helium. The formula for the finite element stress field coupling analysis is: ; wherein, is the formation stress tensor, describing the stress state at each point in the reservoir; J is the gas flow, describing the flow behavior of helium in the reservoir, affected by temperature and pressure factors; Q is the source term (such as heat source, gas flow, etc.), describing the release and adsorption processes of helium; represents the change of reservoir stress with time, reflecting the stress change caused by the exploitation process, gas flow or other external factors; is used to describe the flow trend of helium in the reservoir. The divergence operator calculates the outflow or inflow of gas, helping to understand the gas distribution and the dynamic state of the reservoir; The finite element stress field coupling analysis can simulate the stress change of the reservoir at different exploitation stages, predict the distribution and flow of helium, and optimize the recovery strategy.

[0027] Through the dynamic simulation of stress and gas distribution, potential exploitation risks can be identified in advance, reducing resource waste or wellhead damage caused by unstable gas distribution; By simulating the stress state of the reservoir through finite element analysis, the distribution of helium can be inferred. Combining the three-dimensional modeling of the reservoir with real-time data, the dynamic simulation of the helium distribution in the reservoir is carried out using factors such as formation pressure, stress change and temperature.

[0028] Based on the helium distribution migration data, a deep learning model is used for spatio-temporal feature extraction and dynamic prediction processing to obtain the evolution data of helium resource volume; Deep learning prediction uses deep learning technology, combines real-time monitoring data and historical data, predicts the dynamic change of helium resource volume in the reservoir, and adjusts the recovery strategy according to the prediction results.

[0029] Deep learning training: Use historical data to train the model and perform dynamic prediction through real-time data input.

[0030] Construct a deep learning model integrating CNN and LSTM, and use frameworks such as TensorFlow and Keras for model training and application.

[0031] Combining real-time monitoring data and historical exploitation data, use a deep learning model to predict the helium resource volume and generate a dynamic change prediction result.

[0032] The deep learning model is a convolutional neural network (CNN) fused with a long short-term memory network (LSTM). The deep learning model is used to process time series data (such as helium concentration change, temperature, pressure) and predict the dynamic change of helium resources in the reservoir; Among them, the convolutional neural network is used to extract the spatial features in the time series data of helium concentration; The formula of the convolutional neural network is: ; Among them, ht is the output feature of the convolutional layer; W is the convolutional kernel weight, which extracts features from the input data and learns the spatial features of the helium concentration in the reservoir; b is the bias; xt is the input data (helium concentration data or other parameters); The long short-term memory network is used to capture the trend of the change of helium concentration over time. The formula of the long short-term memory network is: ; Wf, Wi are the gating weights of the long short-term memory network: used to control the magnitude of information flow and capture important dynamic information in the time series.

[0033] is the prediction output: based on the combination of the convolutional neural network and the long short-term memory network, predict the future changes of the helium resource volume in the reservoir; Combining CNN and LSTM can not only perform short-term prediction of helium resources, but also model the long-term change trend and dynamically adjust the mining strategy.

[0034] The real-time dynamic adjustment ability of this model significantly improves the accuracy and response speed of resource evaluation For step S3, perform dynamic recovery rate evaluation and path optimization processing on the helium resource volume evolution data under multiphase flow constraints to generate the final mining plan; the dynamic recovery rate evaluation and path optimization processing are realized by using real-time reservoir data and a simulated gas-liquid phase conversion model; According to the change results of the helium resource volume output by the deep learning model, combined with real-time reservoir data, perform dynamic recovery rate evaluation and automatically generate a dynamic evaluation report of helium resources; Dynamic recovery rate evaluation: Evaluate the recovery rate of the current reservoir according to the change of helium concentration and the predicted resource volume.

[0035] Dynamic report generation: Automatically generate a report on the current resource reserves, mining progress and predicted recovery rate to help decision-makers understand the current reservoir conditions.

[0036] Use real-time data input, such as pressure, concentration, etc., and combine the output of the deep learning model to perform recovery rate evaluation.

[0037] Generate a report through a data visualization tool with the evaluation results and provide real-time feedback to the operators; Optimize the extraction path and recovery plan of reservoir helium based on real-time reservoir data and gas-liquid phase transformation model to improve the recovery efficiency and reduce risks.

[0038] Multiphase flow simulation: Use the gas-liquid phase transformation model to simulate the interaction between helium and other gases or liquids in the reservoir and optimize the extraction plan.

[0039] Optimization of extraction path: Based on real-time data and simulation results, optimize the extraction path of helium to ensure efficient recovery.

[0040] Use multiphase flow simulation software (such as CMG, ECLIPSE), combined with real-time data to adjust the extraction path.

[0041] Optimization algorithms (such as particle swarm optimization, genetic algorithm) can be used to adjust the extraction path of helium to improve the recovery rate and efficiency; The multiphase flow constraint is used to optimize the flow path of helium in the reservoir. The particle swarm optimization algorithm (PSO) is used to update the particle velocity at time t+1, and the best extraction path and recovery plan are found through iteration to improve the recovery rate of resources; The formula of the particle swarm optimization algorithm is: ; where vi(t) is the velocity of the particle at the t-th moment, which is used to adjust the velocity of the helium flow path and is updated based on the current state and the historical best position; xi is the position of the particle, representing the flow path of helium or the recovery strategy; pbesti is the historical best position of the particle; gbest is the global best position, representing the optimal helium recovery path; w is the inertia weight; c1, c2 are acceleration constants; r1, r2 are random numbers; Evaluate the fitness of the new position of each particle, that is, the new plan; the fitness function combines the helium flow characteristics, reservoir state and recovery efficiency under multiphase flow constraints, reflecting the advantages and disadvantages of the plan for helium recovery efficiency and risk control.

[0042] If the fitness of the new plan is better than the best plan in the history of the particle, update the historical optimal solution of the particle; if its fitness is better than the global best solution, update the global best plan; Repeat the above iterative process until the preset maximum number of iterations is reached or the fitness converges; Finally, the particle position corresponding to the global best solution gbest is the best helium extraction path and recovery plan under multiphase flow constraints; Through the particle swarm optimization algorithm, it is possible to quickly find an optimized helium recovery path, reduce resource waste, and improve the recovery rate. This algorithm can adaptively adjust the recovery path, dynamically optimize the mining plan according to real-time data, and improve the economic efficiency and sustainability of the entire mining process.

[0043] Preferably, the formula for the strategy self-correction process is: ; where u(t) represents the control strategy at time t, such as mining parameters; ytarget(t) represents the expected target, such as the target recovery rate or resource volume; yactual(t) represents the actual data obtained from real-time monitoring; α is the adjustment factor; The above formula describes the process of continuously correcting the control strategy through real-time error feedback, that is, the difference between the target and the actual value, so as to achieve dynamic evaluation and intelligent control.

[0044] Perform strategy self-correction processing on the final mining plan, and finally output a dynamic evaluation and intelligent control plan.

[0045] As can be seen from the above, reservoir stress simulation and multiphase flow optimization can predict and avoid potential mining risks and ensure the safe mining of the reservoir. At the same time, by improving the resource recovery efficiency, unnecessary mining costs and environmental impacts are reduced. Real-time data visualization and automated report generation can improve management efficiency, reduce manual intervention, ensure that decisions are based on accurate real-time data, and provide real-time feedback and warning mechanisms for management personnel, improving work efficiency and accuracy. Through efficient resource recovery and accurate resource assessment, the system helps to extend the service life of the reservoir, reduce the risk of over-mining, and promote environmental protection and sustainable mining. Example Two:

[0046] Dynamic Evaluation and Prediction of Helium Concentration in the Reservoir Based on Real-Time Data: Application Scenario: In a medium-sized helium reservoir, through real-time monitoring of helium concentration, formation pressure, and temperature data, combined with a deep learning model (CNN + LSTM), dynamic evaluation and prediction of helium resource volume are carried out.

[0047] Specifically, the helium concentration data (monitoring frequency: once per hour): Initial concentration: 2.1% (initial concentration of the reservoir); Concentration change range: 2.1% - 4.8%; Monitoring frequency: once per hour, acquisition period: 24 hours; Formation pressure data (unit: MPa): Initial pressure: 12.0 MPa; Pressure change range: 12.0 MPa - 15.5 MPa; Monitoring frequency: once per hour, acquisition period: 24 hours; Reservoir temperature data (unit: °C): Initial temperature: 35°C; Temperature change range: 35°C - 38°C; Monitoring frequency: once per hour, acquisition period: 24 hours; Training dataset: Historical data: real-time monitoring data for the past six months, including helium concentration, formation pressure, temperature data, and recovery rate.

[0048] Calculation process: Data preprocessing: Use Kalman filtering to denoise the collected helium concentration, formation pressure, and temperature data, and eliminate the interference caused by external environmental factors or equipment noise.

[0049] The original helium concentration data is: 2.1%, 2.3%, 2.5%,..., 4.8% (collected once per hour).

[0050] The helium concentration data after using Kalman filtering may be smoothed to: 2.3%, 2.5%, 2.6%,..., 4.6%.

[0051] Feature extraction (Convolutional Neural Network CNN): Input data: helium concentration, formation pressure, and temperature data for 24 hours.

[0052] The features extracted by the CNN layer include information such as local temporal changes and fluctuation ranges. For example, the maximum, minimum, and mean values of concentration changes.

[0053] Deep learning model (CNN + LSTM) training and prediction: Input the features processed by the CNN into the LSTM layer, and the LSTM model learns the dynamic change law of helium concentration over time. Assume the predicted value for the next 24 hours output by the LSTM is: 4.3%.

[0054] Helium resource assessment: Assume the reservoir volume is 100,000 cubic meters and the current concentration is 4.3%, then the current helium resource volume in the reservoir can be calculated as: Helium resource volume = reservoir volume × helium concentration Helium resource volume = 100,000 m 3 × 0.043 = 4300 m 3 Therefore, the current helium resource volume in the reservoir is approximately 4300 cubic meters.

[0055] Effects and significance: Improved prediction accuracy: The prediction error of the deep learning model is less than 5%, which is about 15% more accurate than traditional statistical methods (such as linear regression).

[0056] Real-time dynamic assessment: By using real-time data and dynamic prediction, it is possible to timely understand the changes in helium resources in the reservoir, avoiding the static assessment method based on historical data in traditional methods.

[0057] Optimization of mining decisions: Through real-time prediction of helium concentration and estimation of resource volume, it is possible to optimize the mining plan, improve the recovery rate, and reduce the mining cost.

[0058] As can be seen from the above, by comprehensively applying real-time data monitoring, deep learning, optimization algorithms, and automated processing technologies, the system not only improves the efficiency of helium resource mining, reduces costs and risks, but also enhances the intelligence and automation level of the decision-making process, promoting the modernization and sustainable development of helium resource management.

[0059] Example 3: Optimization of helium mining path based on multiphase flow optimization: Application scenario: In a large-scale helium reservoir, by real-time monitoring of helium concentration, pressure, and temperature, combined with the multiphase flow optimization algorithm (PSO), the recovery path of helium is optimized to improve the recovery rate and extraction efficiency.

[0060] Furthermore, basic reservoir data: Reservoir volume: 500,000 cubic meters; Initial helium concentration: 2.5%; Initial pressure: 10.0 MPa; Initial temperature: 30°C; Real-time monitoring data: Helium concentration: 2.5% - 3.8% (collected once per hour); Pressure: 10.0 MPa - 12.5 MPa (collected once per hour); Temperature: 30°C - 32°C (collected once per hour); Particle swarm optimization (PSO) algorithm parameters: Particle swarm size: 50; Maximum number of iterations: 100; Inertia weight w = 0.9; Acceleration constants c1 = 1.5, c2 = 1.5; Random numbers r1, r2 ∈ [0, 1]; Calculation process: Particle swarm initialization: Initialize 50 particles, where the state of each particle represents a possible mining path. The position of the particle represents the selection of the production well, and the velocity represents the amplitude of path adjustment.

[0061] Objective function design: The goal is to maximize the recovery rate. Considering the real-time data of helium concentration, pressure, and temperature, the objective function is designed as follows: ; Where: Pi is the pressure of the i-th production well, Ci is the helium concentration of the i-th production well, Ti is the temperature of the i-th production well.

[0062] Suppose there are 3 production wells, and their pressure, helium concentration, and temperature data are as follows: Production well 1: Pressure P1 = 10.5 MPa, helium concentration C1 = 2.8%, temperature T1 = 31°C Production well 2: Pressure P2 = 11.0 MPa, helium concentration C2 = 3.2%, temperature T2 = 30°C Production well 3: Pressure P3 = 12.0 MPa, helium concentration C3 = 3.5%, temperature T3 = 32°C Substitute the data to calculate the objective function value: ; Particle swarm optimization process: Particle swarm optimization updates the particle position and velocity iteratively to maximize the objective function.

[0063] Furthermore, after 100 iterations, the particle swarm finally finds the optimal path, and its objective function value is increased to 1.25.

[0064] Mining path optimization: The result of the optimal path is as follows: Priority is given to mining areas with high pressure and high helium concentration, and the mining strategy is adjusted according to real-time monitoring data. This optimized path increases the recovery rate by about 12%.

[0065] As can be seen from the above, by optimizing the mining path through the particle swarm optimization algorithm, the recovery rate is increased by 12%, which can make helium recovery more efficient and reduce energy consumption and resource waste.

[0066] Through the dynamic feedback of real-time pressure, concentration, and temperature data, the mining path is optimized, reducing the risk of reservoir fracture or wellhead instability caused by excessive or too low local pressure.

[0067] The particle swarm optimization algorithm can dynamically adjust the mining strategy according to real-time data to adapt to reservoir changes, and has strong flexibility and adaptability.

[0068] As can be seen from the above, through the combination of real-time monitoring, data preprocessing, deep learning, and simulation optimization technologies, accurate and real-time helium resource assessment results can be provided, providing strong support for resource management and mining decisions. The application of dynamic monitoring and prediction models can significantly improve the scientific nature and efficiency of the mining plan. By adjusting strategies in real time, optimizing the mining path, increasing the helium recovery rate, and reducing waste.

[0069] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic evaluation method for reservoir helium resources based on real-time data, characterized in that, Including: Obtain the helium concentration data in the reservoir, perform real-time monitoring and adaptive filtering denoising processing on the helium concentration in the reservoir to obtain accurate helium concentration data; Perform three-dimensional modeling and finite element stress field coupling analysis on the accurate helium concentration data, and generate helium distribution migration data by combining real-time formation pressure and temperature data; Based on the helium distribution migration data, use a deep learning model to perform spatio-temporal feature extraction and dynamic prediction processing to obtain helium resource quantity evolution data; Perform dynamic recovery rate evaluation and path optimization processing under multiphase flow constraints on the helium resource quantity evolution data to obtain the final mining plan; Perform strategy self-correction processing on the final mining plan to obtain a dynamic evaluation and intelligent control plan.

2. The dynamic evaluation method of reservoir helium resources based on real-time data according to claim 1, wherein, The adaptive filtering denoising algorithm is used to remove the noise from the gas concentration monitoring device; Among them, Kalman filtering is used as the denoising algorithm to predict and correct the noise components in the real-time data stream, and the Kalman filtering formula is: ； Among them, is the state estimate at the current moment; is the current observed value; is the observation matrix, which is used to describe the observation process of the system and convert the system state into the observed value; is the Kalman gain, which is used to determine the fusion weight of the observed value and the predicted value of the system.

3. A dynamic evaluation method for reservoir helium resources based on real-time data according to claim 1, characterized in that, The finite element stress field coupling analysis is used for reservoir stress and gas distribution simulation. Based on the geological structure of the reservoir, simulate the changes of stress, pressure, and temperature factors, and speculate on the spatial distribution and migration trend of helium. The formula for the finite element stress field coupling analysis is: ； Among them, is the formation stress tensor, describing the stress state at each point in the reservoir; J is the gas flow, which describes the flow behavior of helium in the reservoir and is affected by temperature and pressure factors; Q is the source term, which describes the release and adsorption processes of helium; Indicates the change of reservoir stress with time, reflecting the stress change caused by the production process, gas flow or other external factors; To describe the flow trend of helium in the reservoir, the divergence operator calculates the amount of gas flowing out or in, helping to understand the gas distribution and the dynamic state of the reservoir.

4. A dynamic evaluation method for reservoir helium resources based on real-time data according to claim 1, characterized in that The deep learning model is a convolutional neural network fused with a long short-term memory network, which is used to process time series data and predict the dynamic changes of helium resources in the reservoir; Among them, the convolutional neural network is used to extract the spatial features in the helium concentration time series data; The formula for the convolutional neural network is: ； Among them, ht is the output feature of the convolutional layer; W is the convolutional kernel weight, which extracts features from the input data and learns the spatial features of the helium concentration in the reservoir; b is the bias; xt is the input data; The long short-term memory network is used to capture the trend of helium concentration changing over time. The formula for the long short-term memory network is: ； Wf, Wi are the gating weights of the long short-term memory network: used to control the magnitude of information flow and capture important dynamic information in the time series; For the prediction output: Based on the combination of a convolutional neural network and a long short-term memory network, predict the future changes in the helium resource volume of the reservoir.

5. A dynamic evaluation method for reservoir helium resources based on real-time data according to claim 1, characterized in that, The multiphase flow constraint is used to optimize the flow path of helium in the reservoir. The particle swarm optimization algorithm is used to update the particle velocity at time t + 1, and the best mining path and recovery plan are found through iteration; The formula for the particle swarm optimization algorithm is: ； Among them, vi(t) is the velocity of the particle at the t-th moment, which is used to adjust the velocity of the helium flow path and is updated according to the current state and the historical best position; xi is the position of the particle, representing the flow path of helium or the recovery strategy; pbesti is the historical best position of the particle; gbest is the global best position, representing the optimal helium recovery path; w is the inertia weight; c1, c2 are acceleration constants; r1, r2 are random numbers.

6. The dynamic evaluation method of reservoir helium resources based on real-time data according to claim 1, characterized in that The formula for the strategy self-correction processing is: ； Among them, u(t) represents the control strategy at time t; ytarget(t) represents the expected target; yactual(t) represents the actual data obtained from real-time monitoring; α is the adjustment factor.

7. A dynamic evaluation method for reservoir helium resources based on real-time data according to claim 1, characterized in that The helium concentration data is obtained by using a high-sensitivity spectral analyzer to monitor the helium concentration in the reservoir in real time; The precise helium concentration data is obtained by removing noise and correcting errors from the helium concentration data using an adaptive filtering and denoising algorithm.

8. A dynamic evaluation method for reservoir helium resources based on real-time data according to claim 1, characterized in that The dynamic recovery rate evaluation and path optimization process are realized based on real-time reservoir data and a simulated gas-liquid phase transformation model.

Images