Oil-water well acidification continuous construction electric control system based on multi-parameter monitoring

Abstract

The invention discloses an oil-water well acidification continuous construction electric control system based on multi-parameter monitoring, and relates to the technical field related to acidification control in oil-gas field development, and the system comprises a multi-source data fusion monitoring module which integrates wellhead pressure, temperature, flow, pH value, H2S concentration and acid liquid conductivity, and an underground microwave imaging sensor carries out multi-physical quantity collection; a multi-objective optimization controller is designed, and the acid etching depth, the crack flow conductivity and the construction cost are balanced; the self-adaptive construction control module is based on a variable-frequency and variable-speed acid pump system controlled by a fuzzy sliding mode and is used for realizing stepless regulation of displacement; an intelligent shunting system is arranged, and acid liquor distribution is dynamically adjusted according to the suction volume of each layer section. According to the variable-frequency acid pump system based on fuzzy sliding mode control, stepless adjustment of the displacement is achieved; the intelligent flow distribution system dynamically prepares liquid according to the suction volume of each layer section; and through a three-stage safety interlocking mechanism, power can be cut off and emergency acid discharge can be started within 0.5 second when the temperature rise is greater than or the pressure suddenly changes, so that the well control risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field related to acidizing control in oil and gas field development, and in particular to an electric control system for continuous acidizing construction of oil and water wells based on multi-parameter monitoring. Background Art

[0002] In oil and gas field development, acidizing of oil and water wells is a key technology for removing contamination near the wellbore and restoring or increasing formation permeability. Traditional acidizing processes typically employ an intermittent operation mode, relying on manual control of acid injection parameters (such as displacement, pressure, and acid concentration). This has the following significant drawbacks:

[0003] Traditional methods mostly use offline detection of a few parameters such as pressure and flow, lacking real-time perception of key parameters such as downhole microwave imaging, acid concentration, and H2S concentration. In addition, chemical sampling has a delay (usually >5 minutes), which cannot reflect the dynamic changes of the formation in a timely manner, and can easily lead to excessive acid erosion or uneven injection.

[0004] Acid pump displacement adjustment relies on manual operation, has low resolution (usually graded adjustment), and lacks a closed-loop feedback mechanism; when faced with abnormal operating conditions (such as sudden pressure changes), emergency response relies on manual judgment, with delays of several seconds to tens of seconds, posing a well control risk.

[0005] Single-well construction data is stored in a decentralized manner, lacking integrated analysis and sharing of multi-well site data. This results in construction plans relying on local experience and making it difficult to optimize acid compatibility and diversion strategies for complex formations (such as multi-layer heterogeneous reservoirs). Traditional systems only use threshold alarms to indicate risks, without implementing automated interlocking protection.

[0006] Therefore, it is necessary to propose an electric control system for continuous acidizing construction of oil and water wells based on multi-parameter monitoring to solve the above problems. Summary of the Invention

[0007] (1) Technical problems solved

[0008] In response to the shortcomings of the existing technology, the present invention provides an electronic control system for continuous construction of oil and water well acidizing based on multi-parameter monitoring, which solves the problems of traditional oil and water well acidizing technology such as insufficient real-time monitoring, delayed manual control, isolated data and passive risk warning, which lead to low efficiency and high risk of construction.

[0009] (2) Technical solution

[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0011] An electric control system for continuous acidizing construction of oil and water wells based on multi-parameter monitoring, comprising:

[0012] Multi-source data fusion monitoring module: Integrates wellhead pressure, temperature, flow, pH value, H2S concentration, acid conductivity, and downhole microwave imaging sensors to collect multiple physical quantities and pre-process the data;

[0013] Intelligent Decision-Making Engine: Build a bidirectional digital twin of the underground and surface systems, using deep learning models to predict the attenuation curve of the formation's acid absorption capacity; develop a dynamic acid compatibility optimization algorithm to automatically adjust the HF and HCl ratio based on real-time rock mineral composition analysis; and design a multi-objective optimization controller to balance acid etching depth, fracture conductivity, and construction costs.

[0014] Adaptive construction control module: A variable frequency speed regulation acid pump system based on fuzzy sliding mode control achieves stepless displacement adjustment; an intelligent diversion system dynamically adjusts acid distribution based on the intake volume of each layer; and a three-level safety interlock mechanism is built to cut off the power source and initiate the emergency acid discharge procedure when abnormal temperature rise or pressure mutation is detected.

[0015] Cloud-edge collaborative operation and maintenance module: deploy edge computing gateways to achieve 20ms-level real-time control; build a digital twin cloud platform to support multi-well site construction experience sharing and algorithm iteration; innovate blockchain data storage technology to ensure the traceability of the construction process.

[0016] Optionally, the multi-source data fusion monitoring module adopts a four-layer heterogeneous fusion architecture, including:

[0017] Perception layer: performs signal conditioning unit, analog-to-digital conversion unit and feature extraction unit;

[0018] Transport layer: includes wired transmission unit, wireless transmission unit and data encapsulation unit;

[0019] Edge computing layer: includes time alignment unit, spatial alignment unit and anomaly detection unit; the time alignment unit uses interpolation to compensate for the difference in sampling rates of different sensors; the spatial alignment unit uses weighted average based on well depth; the anomaly detection unit identifies sudden changes based on the standard deviation of the sliding window;

[0020] Cloud fusion layer: includes data storage unit and fusion algorithm; the data storage unit is a time series database that stores multidimensional data tables; the fusion algorithm is Kalman filtering, which is used for linear system state estimation.

[0021] Optionally, the multi-source data fusion monitoring module includes a wellhead pressure sensor, a wellhead temperature sensor, a flow sensor, a pH sensor, an H2S concentration sensor, an acid conductivity sensor, a downhole microwave imager, an acoustic ranging module, an optical fiber distributed acid concentration sensor, a rock mineral spectrometer, a casing damage monitoring sensor, and an acid flow rate sensor;

[0022] The wellhead pressure sensor is installed in the high-pressure area of the wellhead; the wellhead temperature sensor is installed on the outside of the wellhead oil pipe; the flow sensor is installed in the injection pipeline; the pH sensor is installed in the wellhead return pipeline and inserted into the return fluid pipeline; the H2S concentration sensor is installed in the wellhead return pipeline, including an electrochemical sensor and an infrared sensor; the acid conductivity sensor is installed in the injection pipeline or the downhole tool short section;

[0023] The downhole microwave imager is installed in the downhole tool assembly and transported to the target depth through the coiled tubing. It transmits microwaves in a directionally oriented manner, covers 360 degrees around the well, and generates real-time images of the well wall to monitor the morphology of acid-etched cracks.

[0024] The acoustic ranging module is installed in the pipe string; the optical fiber distributed acid concentration sensor is installed on the inner wall of the downhole oil pipe, and the optical fiber is wrapped around the outer wall of the oil pipe; the laser tuning technology is used to collect the acid absorption spectrum in sections;

[0025] The rock and mineral spectrometer is installed on the downhole drill bit; the casing damage monitoring sensor is installed on the inner wall of the casing; and the acid flow rate sensor is installed on the injection pipeline or the inlet of the downhole tool.

[0026] Optionally, the hierarchical architecture of the intelligent decision engine is as follows:

[0027] Perception layer: Obtains full-dimensional data covering both underground and surface areas from the multi-source data fusion monitoring module;

[0028] Data fusion layer: data fusion based on Kalman filtering;

[0029] Model layer: includes digital twin models and AI models; the digital twin model integrates geomechanics, fluid dynamics, and chemical reaction models; the AI model includes the Transformer+LSTM acid absorption capacity prediction model and the XGBoost regression acid concentration model;

[0030] Decision-making layer: dynamic acid compatibility algorithm, multi-objective optimization controller, output control instructions;

[0031] Execution layer: The acid injection pump and throttle valve are adjusted through the electronic control system to achieve parameter closed-loop control.

[0032] Optionally, the model of the underground and surface bidirectional digital twin is constructed as follows:

[0033] Mapping physical entities with virtual models: Downhole data is collected through a sensor network and combined with surface construction parameters to construct the following mapping relationships in the digital twin model:

[0034] Formation parameters: porosity, permeability, mineral composition;

[0035] Fluid parameters: acid concentration, flow rate, viscosity;

[0036] Equipment status: pipe string vibration frequency and stress distribution;

[0037] Bidirectional data closed loop: Real-time data is transmitted from the underground to the surface via optical fiber to the ground control center to update the digital twin model. From the surface to the underground, construction parameters are dynamically adjusted based on the model's prediction results, and instructions are sent to the downhole tools through the electronic control system.

[0038] The deep learning model architecture is as follows:

[0039] Input features: historical construction data, real-time monitoring data, and geological static parameters;

[0040] LSTM layer: extracts time-dependent features;

[0041] Transformer layer: captures long-range dependencies.

[0042] Optionally, the dynamic acid compatibility optimization algorithm is as follows:

[0043] Real-time analysis of mineral composition: Rapidly identify rock mineral components through downhole XRD spectrometer and input into chemical reaction kinetics model;

[0044] Dynamic ratio optimization: Determine the optimal HF and HCl ratio based on the target reaction rate and constraints;

[0045] Dynamically adjust the ratio through Lagrange multiplier method or genetic algorithm;

[0046] The architecture of the multi-objective optimization controller is as follows:

[0047] Optimization objectives: maximize acid etching depth, crack conductivity, and minimize construction costs;

[0048] Dynamic weight allocation strategy: adjust target priorities according to the construction stage.

[0049] Optionally, the architecture of the variable frequency speed regulation acid pump system based on fuzzy sliding mode control is as follows:

[0050] Fuzzy control: takes displacement error and error change rate as input, outputs initial control quantity through fuzzy rules, generates based on error and error change rate, and is dynamically adjusted by the digital twin model;

[0051] Sliding mode control: Design a sliding surface. When the system state deviates from the sliding surface, it is forced to return by switching the control quantity.

[0052] Frequency conversion speed regulation: convert the control quantity into motor frequency, drive the acid pump through the frequency converter, and realize continuous adjustment of displacement;

[0053] The role of digital twins: Use the twin model to predict changes in the acid absorption capacity of the formation, dynamically adjust the target, and modify the control parameters and fuzzy rule base in real time.

[0054] The architecture of the intelligent diversion system is as follows:

[0055] Layered section absorption monitoring unit: The downhole flow sensor array is used to monitor the acid absorption of each layer in real time, and the permeability and mineral composition differences of each layer are analyzed in combination with the digital twin model;

[0056] Objective function: maximize global flow conductivity;

[0057] Allocation algorithm: Using an improved particle swarm optimization algorithm, the acid distribution ratio of each layer is dynamically adjusted according to real-time data;

[0058] Actuator: The flow rate of each layer branch is adjusted through the electronically controlled throttle valve to achieve real-time adjustment of acid distribution.

[0059] Optionally, the cloud-edge collaborative operation and maintenance module includes an edge computing layer, a cloud platform and a network layer; the edge computing layer is deployed at the well site and is responsible for real-time data acquisition, local closed-loop control, preprocessing and caching; the cloud platform is based on digital twins and blockchain to achieve global optimization, cross-well site experience sharing, and trusted data storage; the network layer uses industrial Ethernet to achieve low-latency communication between the edge and the cloud, and supports real-time command issuance and batch data synchronization.

[0060] Optionally, the architecture of the digital twin cloud platform is:

[0061] Dynamic parameter calibration: Based on real-time data from the well site, Kalman filtering is used to online correct formation permeability and acid-rock reaction rate;

[0062] Federated learning algorithm iteration: Each well site edge device trains the model locally, and the cloud aggregates the global model;

[0063] Case library construction: Historical construction data is stored in the cloud, and stratum types are divided through clustering algorithms to form typical construction templates.

[0064] (3) Beneficial effects

[0065] The present invention provides an electric control system for continuous acidizing construction of oil and water wells based on multi-parameter monitoring, which has the following beneficial effects:

[0066] 1. The present invention integrates 12 types of sensors (pressure, temperature, flow, pH, H2S, conductivity, acoustic ranging module, etc.) to achieve full parameter coverage of the wellhead and downhole, accurately capture the vibration state of the pipe string, and effectively prevent accidents such as sand sticking; by setting up acid concentration optical fiber distributed sensing technology, it replaces traditional chemical sampling to achieve real-time online monitoring of acid concentration and improve data refresh rate.

[0067] 2. The present invention uses a bidirectional digital twin system for both underground and surface areas and a deep learning model to dynamically predict the attenuation curve of the formation's acid absorption capacity, improving prediction accuracy. It combines XRD spectroscopy with real-time analysis of rock mineral composition and automatically adjusts the ratio of HF and HCl through a dynamic compatibility optimization algorithm to match the acid etching effect with the formation characteristics, thereby improving fracture conductivity. Furthermore, it balances acid etching depth, conductivity, and construction costs, generating an optimal construction plan through a multi-objective controller to reduce overall costs.

[0068] 3. The variable frequency acid pump system based on fuzzy sliding mode control of the present invention can achieve 0.1-10m 3 / min displacement is steplessly regulated; the intelligent diversion system dynamically distributes fluid according to the suction volume of each layer, improving the acid absorption balance between layers; through the three-level safety interlock mechanism, when the temperature rise is greater than or the pressure suddenly changes, the power can be cut off within 0.5 seconds and emergency acid discharge can be started, reducing well control risks.

[0069] 4. The present invention deploys an edge computing gateway with a 20ms response level to achieve local closed-loop control, reduce cloud dependence, and ensure system reliability under extreme working conditions; the digital twin cloud platform aggregates data from multiple well sites and forms standardized construction templates through algorithm iteration, shortening the preparation time for new well site operations; blockchain evidence storage technology encrypts and stores construction data throughout the entire process, supporting full-process backtracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a module schematic diagram of the electric control system for continuous acidizing construction of oil and water wells of the present invention. DETAILED DESCRIPTION

[0071] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0072] Example 1

[0073] See also Figure 1 , an electric control system for continuous acidizing construction of oil and water wells based on multi-parameter monitoring, including:

[0074] Multi-source data fusion monitoring module: Integrates wellhead pressure, temperature, flow, pH value, H2S concentration, acid conductivity, and downhole microwave imaging sensors to collect multiple physical quantities. An acoustic ranging module is introduced to monitor string vibration in real time to prevent sand sticking accidents. Fiber-optic distributed sensing technology for acid concentration is deployed to overcome the bottleneck of traditional sampling delays. Data is also pre-processed.

[0075] Intelligent Decision-Making Engine: Build a bidirectional digital twin of underground and surface systems, using a deep learning model (Transformer + LSTM) to predict the attenuation curve of the formation's acid absorption capacity. Develop a dynamic acid compatibility optimization algorithm to automatically adjust the HF and HCl ratio based on real-time rock mineral composition analysis (XRD spectral data). Design a multi-objective optimization controller to balance acid etching depth, fracture conductivity, and construction costs.

[0076] Adaptive construction control module: variable frequency speed regulation acid pump system based on fuzzy sliding mode control, achieving displacement of 0.1-10m 3 / min stepless adjustment; an intelligent diversion system is set up to dynamically adjust the acid distribution according to the suction volume of each layer; a three-level safety interlock mechanism is built: when an abnormal temperature rise (greater than 15℃ / min) or a sudden pressure change (greater than 3MPa / min) is monitored, the power source is cut off within 0.5 seconds and the emergency acid discharge program is initiated;

[0077] Cloud-edge collaborative operation and maintenance module: deploy edge computing gateways to achieve 20ms-level real-time control; build a digital twin cloud platform to support multi-well site construction experience sharing and algorithm iteration; innovate blockchain data storage technology to ensure the traceability of the construction process.

[0078] The multi-source data fusion monitoring module adopts a four-layer heterogeneous fusion architecture to achieve the transformation from raw data to decision-making information, including:

[0079] Perception layer (for multi-physical quantity acquisition and preprocessing): signal conditioning unit, analog-to-digital conversion unit and feature extraction unit; the signal conditioning unit is used to amplify and filter analog signals (such as pressure and temperature); the analog-to-digital conversion unit is used for 24-bit AD sampling with a sampling rate of f s ≥2kHz; the feature extraction unit is used to reduce noise of spectral data by wavelet transform;

[0080] Transport layer (high-reliability data transmission): includes wired transmission unit, wireless transmission unit, and data encapsulation unit; the wired transmission unit is Modbus RS485 bus (wellhead) and armored optical cable (underground, tensile strength ≥10kN); the wireless transmission unit is LoRaWAN (ground base station and wellhead), with transmission delay <50ms; the data encapsulation unit uses the OPC UA protocol to unify data formats and define multi-source data tags;

[0081] Edge computing layer (real-time data processing and preliminary fusion): includes a time alignment unit, a spatial alignment unit, and anomaly detection unit. The time alignment unit uses interpolation to compensate for differences in sensor sampling rates. The spatial alignment unit uses a weighted average based on the well depth L. The anomaly detection unit identifies sudden changes based on the sliding window standard deviation.

[0082] Cloud fusion layer (global optimization and intelligent decision-making): includes a data storage unit and a fusion algorithm; the data storage unit is a time series database that stores multidimensional data tables (fields: time, well depth, parameter 1...parameter 12); the fusion algorithm is a Kalman filter, which is used for linear system state estimation.

[0083] The multi-source data fusion monitoring module includes wellhead pressure sensor, wellhead temperature sensor, flow sensor, pH sensor, H2S concentration sensor, acid conductivity sensor, downhole microwave imager, acoustic ranging module, fiber optic distributed acid concentration sensor, rock and mineral spectrometer, casing damage monitoring sensor and acid flow rate sensor;

[0084] Wellhead pressure sensors are installed in the high-pressure area of the wellhead (tubing head, Christmas tree). There are three sensors: one is installed in the main output pipeline to monitor the system back pressure, one is installed at the injection pump outlet to monitor the injection pressure in real time, and a spare sensor is used for abnormality verification to monitor bottomhole pressure changes to prevent overpressure from causing casing damage or acid backflow.

[0085] Two wellhead temperature sensors are installed on the outside of the wellhead tubing (near the Christmas tree). One monitors the tubing's outer wall temperature, reflecting the wellbore fluid temperature; the other monitors the ambient temperature and is used to correct data. This allows real-time monitoring of acid temperature changes to prevent high-temperature decomposition or increased low-temperature viscosity.

[0086] The flow sensor is installed in the injection pipeline (between the surface pump and the wellhead) and uses an electromagnetic flow meter (for conductive acid) or an ultrasonic flow meter (for non-conductive liquid). It is installed full of the pipe to accurately control the acid injection volume and achieve continuous operation displacement adjustment.

[0087] The pH sensor is installed in the wellhead flowback pipeline (the acid outflow end), inserted into the flowback fluid pipeline, in contact with the fluid core area, and combined with the temperature compensation algorithm to eliminate the influence of temperature on pH measurement. The pH value of the flowback fluid is monitored in real time to determine the progress of the acid-rock reaction.

[0088] The H2S concentration sensor is installed on the wellhead flowback pipeline (high up, away from direct impact of the liquid). It includes an electrochemical sensor and an infrared sensor. The electrochemical sensor is close to the fluid surface and responds quickly; the infrared sensor is installed in a safe area for remote monitoring. Double verification avoids false alarms and detects hydrogen sulfide spills to ensure the safety of wellsite personnel.

[0089] Acid conductivity sensors are installed in two sections of the injection pipeline (surface) or downhole tool nipples. When installed on the surface, they are integrated into the same nipple as the pH sensor to reduce delays. When installed downhole, a straight pipe section with a stable flow pattern is selected to avoid turbulent interference. The sensors indirectly reflect changes in acid concentration and assist in determining the efficiency of the acid-rock reaction.

[0090] The downhole microwave imager is installed in the downhole tool assembly (e.g., above the packer or sliding sleeve) and transported to the target depth (e.g., near the perforation section) via coiled tubing. It transmits microwaves in a directionally controlled manner, covering 360° around the wellbore and generating real-time images of the wellbore wall. It monitors the morphology of acid-etched cracks and identifies natural cracks or collapse risks.

[0091] The acoustic ranging module is installed in the tubing (such as pumps, packers, and reducers). The ultrasonic transmitter and receiver are installed in pairs, with a spacing of 0.5-1m. They are fixed to the outer wall of the tubing, pointing toward the inner wall of the casing, and measuring the vibration displacement of the tubing. This prevents sand sticking accidents and monitors tubing fatigue damage.

[0092] The acoustic ranging module uses the ultrasonic pulse reflection method to send sound waves through the transmitting transducer, and the receiving transducer detects the echo time to calculate the vibration displacement of the pipe string. The acoustic wave data processing uses short-time Fourier transform to analyze the vibration spectrum and identify the characteristic frequency of sand impact.

[0093] The fiber-optic distributed acid concentration sensor (TDLAS) is installed on the inner wall of the downhole tubing (along the entire injection path), with the optical fiber wrapped around the outer wall. Laser tuning technology is used to collect the acid absorption spectrum in segments (with a resolution of 0.5 m). This provides real-time data on downhole acid concentration distribution and optimizes acid squeezing strategies. Fiber-optic distributed acid concentration monitoring is based on the Lambert-Beer law of gas absorption spectroscopy, with the laser wavelength tuned to the characteristic absorption peak of the acid.

[0094] The rock and mineral spectrometer is installed on the downhole drill bit or drilling tool, integrated behind the drill bit and close to the wellbore rock. It uses near-infrared spectroscopy (NIR) to analyze rock mineral composition (such as quartz and calcite content) and dynamically adjust the acid formula to match the formation characteristics.

[0095] Casing damage monitoring sensors are installed on the inner wall of the casing (high stress areas or corrosion-sensitive sections). Fiber Bragg gratings (FBGs) are attached to the inner wall of the casing to monitor strain changes, and wireless sensors (such as MEMS accelerometers) are embedded in the casing coupling to transmit data remotely. This provides early warning of casing deformation, cracks, or corrosion perforation risks.

[0096] The acid flow rate sensor is installed on the injection pipeline (surface) or at the inlet of the downhole tool. When installed on the surface, a Doppler effect flow meter (non-invasive) is used; when installed downhole, it is integrated upstream of the flow control valve to measure the fluid Reynolds number; the injection displacement is optimized to avoid turbulence-induced pipe erosion.

[0097] The hierarchical architecture of the intelligent decision-making engine is as follows:

[0098] Perception layer: Acquires full-dimensional data covering both the underground and surface areas from multi-source data fusion monitoring modules (such as fiber optic pressure measurement, spectrometer mineralogy, and pH meter acid concentration measurement);

[0099] Data fusion layer: Data fusion is performed based on Kalman filtering to resolve conflicts in multi-source data (such as mismatch between pressure mutations and flow fluctuations);

[0100] Model layer: includes digital twin models and AI models; the digital twin model integrates geomechanics (such as the poroelasticity equation), fluid dynamics (Navier-Stokes equations), and chemical reaction models; the AI model includes the Transformer+LSTM acid absorption capacity prediction model and the XGBoost regression acid concentration model;

[0101] Decision-making layer: dynamic acid mixing algorithm, multi-objective optimization controller, output control instructions (such as pump speed N, HF and HCl ratio r);

[0102] Execution layer: The acid injection pump and throttle valve are adjusted through the electronic control system to achieve parameter closed-loop control.

[0103] In this embodiment, an integrated fiber-optic distributed acid concentration sensor is used to monitor acid concentration directly in situ downhole, avoiding the delay of traditional sampling methods (usually tens of minutes to several hours). This achieves a second-level response to acid concentration, providing real-time data support for acidizing reaction process control. The acoustic ranging module monitors the vibration state of the tubing in real time and, by analyzing the vibration spectrum (such as high-frequency sand impact signals), can provide early warning (within minutes) before sand sticking accidents occur. Compared with traditional manual observation or post-processing, this significantly reduces operational risks.

[0104] The system integrates 12 types of sensors (pressure, temperature, flow, pH, H2S concentration, etc.), combined with downhole microwave imaging and rock and mineral spectral analysis, to achieve multi-dimensional monitoring of the wellbore environment, acid performance, and formation characteristics. For example, through pressure-flow coupling analysis, the formation absorption capacity can be determined and the acid displacement can be dynamically adjusted. The pH-conductivity joint inversion can be used to accurately calculate the acid-rock reaction rate to avoid over-acidification. Based on the results of multi-source data fusion, the electronic control system can automatically adjust the acid injection pump frequency and flow rate, and even trigger an emergency shutdown (for example, when H2S exceeds the standard), realizing a fully closed-loop intelligent control of "monitoring-analysis-execution" and reducing human intervention errors.

[0105] Traditional acidizing construction requires multiple sampling and testing (such as acid concentration and rock cuttings composition). However, the present invention replaces most manual sampling processes through fiber optic distributed sensing and spectral online analysis, saving about 30% of testing time for a single well and reducing operating costs. Casing damage monitoring sensors (such as FBG fiber optic strain sensors) monitor casing stress changes in real time to avoid casing damage caused by acid corrosion or high pressure, reducing subsequent well repair costs. Rock mineral spectrometers analyze well wall mineral composition (such as carbonate rock content) in real time and, combined with acid concentration data, dynamically adjust the acid formulation (such as the ratio of HF and HCl) to achieve precise acidizing on a "one well, one policy" basis. In high-temperature and high-pressure wells, wellhead pressure-temperature combined monitoring can provide real-time early warning of abnormal temperature and pressure changes to prevent well control risks. In sulfur-containing formations, H2S concentration sensors and the electronic control system quickly initiate emergency procedures to ensure operational safety.

[0106] Integrate multi-source downhole data (such as pressure, flow, and spectral analysis of rock composition) with ground construction parameters to overcome the limitations of traditional single parameter control (such as relying solely on pressure or flow) and adapt to complex formation conditions such as high temperature, high pressure, and high permeability.

[0107] Example 2

[0108] This embodiment is a further optimization based on the first embodiment. Specifically, the model construction method of the underground and surface bidirectional digital twin is as follows:

[0109] Mapping physical entities with virtual models: Downhole data (such as pressure, temperature, acid concentration, and rock mineral composition) is collected through a sensor network (fiber optics, acoustic waves, and spectrometers). Combined with surface construction parameters (such as pump speed and displacement), the following mapping relationships are constructed in the digital twin model:

[0110] Formation parameters: porosity, permeability, mineral composition (such as carbonate content M carbonate );

[0111] Fluid parameters: acid concentration C acid (t), flow velocity v, viscosity μ;

[0112] Equipment status: pipe string vibration frequency and stress distribution;

[0113] Bidirectional data closed loop: Real-time data is transmitted from the underground to the surface via 5G / fiber optics to the ground control center to update the digital twin model. From the surface to the underground, operation parameters (such as acid ratio and displacement) are dynamically adjusted based on model predictions (such as sand sticking risk and acid absorption capacity degradation). Instructions are then sent to the downhole tools via the electronic control system.

[0114] The deep learning model (Transformer+LSTM) architecture is as follows:

[0115] Input features: historical construction data (pressure P(t), flow Q(t)), real-time monitoring data (wellhead pressure, temperature), geological static parameters (formation permeability k, mineral composition M i LSTM layer: extracts time-dependent features (such as the decay pattern of acid concentration over time);

[0116] Transformer layer: captures long-range dependencies (such as the impact of historical construction on the current acid absorption capacity of the formation);

[0117] Output result: Acid absorption capacity attenuation curve S(t), defined as the rate of acid absorption per unit time, the formula is: S(t) = f model (P(t),Q(t),C acid (t),k,Mi); where f model is the model fitting function;

[0118] Dynamic correction mechanism: If the deviation between real-time monitoring data and predicted values exceeds a threshold (e.g., a 10% pressure surge), the model is triggered to update online:

[0119] S corrected (t) = S model (t)+α·(S measured (t)-S model (t));

[0120] Among them, α is the correction coefficient (take 0.1 to 0.3), S model (t) is the original prediction value of the model, S measured (t) is the real-time monitoring value.

[0121] The dynamic acid compatibility optimization algorithm is as follows:

[0122] Real-time analysis of mineral composition: Rapid identification of rock mineral components (such as carbonate rock M carbonate , Quartz M quartz ), input to the chemical reaction kinetics model:

[0123] Among them, R reaction is the total reaction rate, which characterizes the acid's ability to dissolve rocks, k HF 、k HCl is the reaction rate constant, n and m are the reaction orders;

[0124] Dynamic ratio optimization: according to the target reaction rate R target and constraints (including total acid concentration C total ), solve the optimal HF and HCl ratio: min|R reaction -R target |subject to CHF +C HCl =C total Among them, C HF is the concentration of hydrofluoric acid, C HCl is the concentration of hydrochloric acid;

[0125] The ratio is dynamically adjusted by the Lagrange multiplier method or genetic algorithm, which can be expressed mathematically as follows:

[0126] Where, ∈ is a small value to prevent division by zero, is the optimal hydrofluoric acid HF concentration;

[0127] The architecture of the multi-objective optimization controller is as follows:

[0128] Optimization goal: maximize the acid etching depth D etching , fracture conductivity F conductivity , minimize the construction cost C cost ;

[0129] Constraint: Bottom hole pressure P bottom ≤P max 、Displacement Q≤Q max , acid concentration C acid ∈[C min ,C max ];

[0130] Dynamic weight allocation strategy: Adjust target priority according to the construction stage: Fitness = w1·D etching +w2·F conductivity -w3·C cost Among them, w1, w2, and w3 are dynamic weight coefficients, and their priorities are adjusted according to the construction stage.

[0131] In this embodiment, the digital twin model is used to map downhole working conditions (such as acid concentration, pressure, and rock reaction) in real time. Combined with AI, it predicts the attenuation trend of the formation's acid absorption capacity, dynamically adjusts the acid ratio (such as the ratio of HF and HCl) and operation parameters (such as displacement and pump speed), and avoids the blindness of traditional empirical construction. It also provides early warning of the risk of the formation "not absorbing acid" (such as a sudden increase in pressure or a sudden drop in concentration), and adjusts the strategy in time to avoid ineffective acid injection or equipment jamming, shortening the construction period. The digital twin model accumulates construction data under different formations and working conditions to form a reusable knowledge base, replacing the traditional model that relies on manual experience and improving the level of industry standardization.

[0132] By precisely matching the acid formula (such as dynamically adjusting the ratio of HF and HCl according to the rock mineral composition), acid waste is reduced and material costs are lowered; at the same time, the subsequent environmental treatment pressure caused by excessive acid injection is avoided; by precisely controlling the acid concentration and displacement, the conductivity of the acid-etched fractures is ensured to match the formation, delaying the decline of oil well production and improving long-term recovery rates.

[0133] Example 3

[0134] This embodiment is a further optimization based on the second embodiment. Specifically, the architecture of the variable frequency speed regulation acid pump system based on fuzzy sliding mode control is as follows (to achieve stepless regulation of acid liquid discharge and adapt to dynamic changes in formation acid absorption capacity):

[0135] Fuzzy control: displacement error e(t) = Q 目标 -Q 实际 and the error change rate Δe(t) as input, and the initial control quantity u is output through fuzzy rules (such as "the error is large and changes rapidly, so the frequency is adjusted significantly"). fuzzy (t), generated based on the error and error rate of change, where Q 目标 is the target acid discharge rate, which is dynamically adjusted by the digital twin model, Q 实际 The actual acid liquid discharge is achieved through variable frequency speed regulation;

[0136] Sliding Mode Control: Designing the Sliding Surface Among them, c is the integral coefficient. When the system state deviates from the sliding surface, the control quantity u is switched. sliding (t) forced regression, the formula is: u(t) = u fuzzy (t)+k·sing(s(t)); where u(t) is the final control variable, k is the sliding mode gain, which is used to overcome uncertainty interference, u sliding (t) is the switching control variable generated by the sliding mode control, which is used to force the system to return to the sliding surface. sing(s(t)) is a sign function, which takes the value of +1 when s(t)>0 and -1 otherwise.

[0137] Frequency conversion speed regulation: convert the control quantity u(t) into motor frequency f(t), drive the acid pump through the frequency converter, and realize continuous adjustment of displacement Q(t);

[0138] Function of digital twin: Use the twin model to predict changes in the acid absorption capacity of the formation, dynamically adjust the Q target, and correct the control parameter k and fuzzy rule base in real time.

[0139] The architecture of the intelligent diversion system is:

[0140] Layered section suction monitoring unit: Real-time monitoring of the acid absorption Q of each layer through the downhole flow sensor array i(t)(i=1,2,...,m), analyze the differences in permeability and mineral composition of each layer by combining the digital twin model;

[0141] Dynamic allocation strategy:

[0142] Objective function: maximize global flow capacity Among them, f(Q i (t)) represents a monotonically increasing function, which indicates the contribution of the acid absorption of the i-th layer to the conductivity; while satisfying the constraint Among them, Q 总 (t) is the total acid absorption of each layer;

[0143] Allocation algorithm: Using the improved particle swarm optimization (PSO) algorithm, the acid distribution ratio of each layer is dynamically adjusted according to real-time data. i (t), the formula is: Q i (t) = α i (t)·Q 总 (t); where α i (t) PSO is iteratively updated to preferentially allocate more acid to the layers with high acid absorption efficiency;

[0144] Actuator: The flow rate of each layer branch is adjusted through the electronically controlled throttle valve to achieve real-time adjustment of acid distribution;

[0145] The monitoring and triggering conditions of the three-level safety interlock mechanism are:

[0146] First level interlock: monitor pressure P(t) and temperature T(t), if P(t)>P 阈值 or T(t)>T 阈值 , triggering an early warning;

[0147] Secondary interlock: When the pressure change rate ΔP(t)>ΔP 阈值 Or temperature rises suddenly ΔT(t)>ΔT 阈值 , determined to be an abnormal operating condition;

[0148] Level 3 interlock: If the level 2 interlock exceeds the limit for 0.5 seconds, the power source will be cut off and the acid will be discharged immediately;

[0149] The emergency response process of the three-level safety interlock mechanism is as follows:

[0150] Cut off power: stop the acid pump motor and close the main valve. Time t 响应 ≤0.5 seconds;

[0151] Emergency acid drainage: Start the standby centrifugal pump to direct the residual acid in the well into the emergency storage tank to avoid secondary damage;

[0152] Digital twin assistance: Use the twin model to simulate the accident diffusion path and optimize the acid discharge path and flow rate.

[0153] In this embodiment, a fuzzy sliding mode control algorithm is used to achieve real-time and continuous adjustment of the acid pump discharge rate, with a response speed of milliseconds. For example, when the acid absorption capacity of the formation decreases, the system can reduce the discharge rate within 0.5 seconds to avoid ineffective acid injection. Conversely, the discharge rate can be quickly increased in high-permeability layers, shortening the construction period.

[0154] The intelligent diversion device dynamically adjusts the acid distribution ratio based on the real-time intake volume of each layer, allowing the low-permeability layer to receive more acid and improving the uniformity of reservoir reconstruction.

[0155] The three-level safety interlock mechanism independently monitors pressure and temperature. When the pressure suddenly changes or the temperature rises beyond the standard, it cuts off the power source within 0.5 seconds and starts emergency acid drainage, avoiding blowouts or equipment damage and reducing the risk of shutdowns.

[0156] Example 4

[0157] This embodiment is a further optimization based on Example 3. Specifically, the cloud-edge collaborative operation and maintenance module includes an edge computing layer, a cloud platform, and a network layer. The edge computing layer is deployed at the well site and is responsible for real-time data acquisition, local closed-loop control, preprocessing, and caching. The cloud platform is based on digital twins and blockchain to achieve global optimization, cross-well site experience sharing, and trusted data storage. The network layer uses industrial Ethernet to achieve low-latency communication between the edge and the cloud, supporting real-time command issuance and batch data synchronization.

[0158] The 20ms-level real-time control architecture of the edge computing gateway is:

[0159] Adopt edge computing gateway, integrated FPGA acceleration module, support hardware offloading of real-time control algorithm; sensor data acquisition cycle T sample ≤10ms, deterministic transmission is guaranteed by time-triggered bus;

[0160] Edge-cloud collaboration:

[0161] Local closed loop: The edge gateway directly handles real-time control (such as displacement adjustment and pressure response) without waiting for cloud instructions;

[0162] Cloud intervention: The cloud simulates the long-term optimization results (such as the acid absorption attenuation trend of the formation) through digital twin simulation, and periodically updates the set values of the edge controller with an update frequency of f update ≤1Hz.

[0163] The architecture of the digital twin cloud platform is:

[0164] Dynamic parameter calibration: Based on real-time well site data (such as pressure P(t), flow rate Q(t)), Kalman filtering is used to online correct formation permeability and acid-rock reaction rate;

[0165] Federated learning algorithm iteration: Each well site edge device trains models locally (such as acid distribution strategy), and the cloud aggregates the global model: Among them, w i is the weight of the ith well site, ablaF i is the local gradient of the ith well site, ablaG represents the current global gradient, and η is the learning rate, which controls the model update step size;

[0166] The cross-well site experience sharing is as follows:

[0167] Case library construction: Cloud storage of historical construction data (such as {P(t), Q(t), C acid (t)}), using clustering algorithms (such as DBSCAN) to divide the stratum types and form a typical construction template;

[0168] Knowledge transfer: When constructing a new well site, match similar formation templates (such as cosine similarity cosθ>0.85) and load the optimal control strategy to reduce trial and error costs.

[0169] In this embodiment, by deploying an edge computing gateway, local real-time processing of key parameters such as wellhead pressure, flow, and temperature is achieved, and the control command response time is shortened to 20 milliseconds, significantly improving the ability to quickly respond to sudden anomalies (such as sudden pressure changes and excessive temperature rise) during acidizing construction, and effectively reducing the risk of downhole equipment damage or construction accidents; the edge end can independently complete real-time control logic (such as emergency acid discharge and power cut-off) without relying on cloud communication, ensuring system reliability under extreme working conditions; the edge end focuses on real-time control (such as variable frequency speed regulation and shunt regulation), while the cloud is responsible for complex computing tasks (such as digital twin modeling and big data analysis), reducing the computing power requirements of edge devices while realizing dynamic allocation of resources across well sites (such as sharing acid pump equipment and optimizing construction queue sequence);

[0170] The digital twin cloud platform aggregates construction data from different well sites (such as formation acid absorption capacity, acid distribution patterns, and pressure change curves). Using machine learning algorithms, it discovers the optimal combination of construction parameters and forms a global optimization strategy. For example, the stratified acid distribution experience of one well site can be directly reused in other similar well sites, reducing trial and error costs. The high-precision cloud-based digital twin model can simulate the entire acidizing construction process, predict formation reactions and pressure change trends, provide early warning of potential risks (such as wellbore blockage and excessive acid erosion), and provide real-time optimization targets (such as adjusting displacement and diversion ratio) to the edge.

[0171] Blockchain technology is used to distribute and encrypt construction data (such as pressure, flow, and temperature curves) and operation records (such as valve opening and equipment start and stop) to ensure the authenticity and integrity of the data and avoid human tampering or data loss. The entire process data from acid mixing and equipment commissioning to construction execution is stored on the chain, which can accurately trace the operational details of each link and provide a credible basis for accident analysis and responsibility determination.

[0172] In summary: Through technological innovations such as multi-parameter monitoring, digital twin decision-making, adaptive control, and cloud collaboration, this invention achieves "safe, efficient, intelligent, and economical" acidizing operations, significantly improves the production of oil and water wells, reduces operational risks and costs, and has important technological advancements and industrial application value.

[0173] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring, characterized in that: include: Multi-source data fusion monitoring module: Integrates wellhead pressure, temperature, flow, pH value, H2S concentration, acid conductivity, and downhole microwave imaging sensors to collect multiple physical quantities and pre-process the data; Intelligent Decision-Making Engine: Build a bidirectional digital twin of the underground and surface systems, using deep learning models to predict the attenuation curve of the formation's acid absorption capacity; develop a dynamic acid compatibility optimization algorithm to automatically adjust the HF and HCl ratio based on real-time rock mineral composition analysis; and design a multi-objective optimization controller to balance acid etching depth, fracture conductivity, and construction costs. Adaptive construction control module: A variable frequency speed regulation acid pump system based on fuzzy sliding mode control achieves stepless displacement adjustment; an intelligent diversion system is set up to dynamically adjust acid distribution according to the suction volume of each layer; a three-level safety interlock mechanism is established: when abnormal temperature rise or pressure mutation is detected, the power source is cut off and the emergency acid discharge procedure is initiated; Cloud-edge collaborative operation and maintenance module: deploy edge computing gateways to achieve 20ms-level real-time control; build a digital twin cloud platform to support multi-well site construction experience sharing and algorithm iteration; innovate blockchain data storage technology to ensure the traceability of the construction process.

2. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The multi-source data fusion monitoring module adopts a four-layer heterogeneous fusion architecture, including: Perception layer: performs signal conditioning unit, analog-to-digital conversion unit and feature extraction unit; Transport layer: includes wired transmission unit, wireless transmission unit and data encapsulation unit; Edge computing layer: includes time alignment unit, space alignment unit and anomaly detection unit; the time alignment unit uses interpolation to compensate for the differences in sampling rates of different sensors; the space alignment unit is based on weighted average of well depth; the anomaly detection unit identifies mutation values based on the standard deviation of the sliding window; cloud fusion layer: includes data storage unit and fusion algorithm; the data storage unit stores multidimensional data tables in a time series database; the fusion algorithm is Kalman filtering, which is used for linear system state estimation.

3. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The multi-source data fusion monitoring module includes a wellhead pressure sensor, a wellhead temperature sensor, a flow sensor, a pH sensor, an H2S concentration sensor, an acid conductivity sensor, a downhole microwave imager, an acoustic ranging module, an optical fiber distributed acid concentration sensor, a rock and mineral spectrometer, a casing damage monitoring sensor, and an acid flow rate sensor; The wellhead pressure sensor is installed in the high-pressure area of the wellhead; the wellhead temperature sensor is installed on the outside of the wellhead oil pipe; the flow sensor is installed in the injection pipeline; the pH sensor is installed in the wellhead return pipeline and inserted into the return fluid pipeline; the H2S concentration sensor is installed in the wellhead return pipeline, including an electrochemical sensor and an infrared sensor; the acid conductivity sensor is installed in the injection pipeline or the downhole tool short section; The downhole microwave imager is installed in the downhole tool assembly and transported to the target depth through the coiled tubing; Directed microwave emission, covering 360° around the well, generates real-time images of the well wall and monitors the morphology of acid-etched cracks; The acoustic ranging module is installed in the pipe string; the optical fiber distributed acid concentration sensor is installed on the inner wall of the downhole oil pipe, and the optical fiber is wrapped around the outer wall of the oil pipe; the laser tuning technology is used to collect the acid absorption spectrum in sections; The rock and mineral spectrometer is installed on the downhole drill bit; the casing damage monitoring sensor is installed on the inner wall of the casing; and the acid flow rate sensor is installed on the injection pipeline or the inlet of the downhole tool.

4. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The hierarchical architecture of the intelligent decision engine is as follows: Perception layer: Obtains full-dimensional data covering both underground and surface areas from the multi-source data fusion monitoring module; Data fusion layer: data fusion based on Kalman filtering; Model layer: includes digital twin models and AI models; the digital twin model integrates geomechanics, fluid dynamics, and chemical reaction models; the AI model includes the Transformer+LSTM acid absorption capacity prediction model and the XGBoost regression acid concentration model; Decision-making layer: dynamic acid compatibility algorithm, multi-objective optimization controller, output control instructions; Execution layer: The acid injection pump and throttle valve are adjusted through the electronic control system to achieve parameter closed-loop control.

5. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The model construction method of the underground and surface bidirectional digital twin is as follows: Mapping physical entities with virtual models: Downhole data is collected through a sensor network and combined with surface construction parameters to construct the following mapping relationships in the digital twin model: Formation parameters: porosity, permeability, mineral composition; Fluid parameters: acid concentration, flow rate, viscosity; Equipment status: pipe string vibration frequency and stress distribution; Bidirectional data closed loop: Real-time data is transmitted from the underground to the surface via optical fiber to the ground control center to update the digital twin model. From the surface to the underground, construction parameters are dynamically adjusted based on the model's prediction results, and instructions are sent to the downhole tools through the electronic control system. The deep learning model architecture is as follows: Input features: historical construction data, real-time monitoring data, and geological static parameters; LSTM layer: extracts time-dependent features; Transformer layer: captures long-range dependencies.

6. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The dynamic acid compatibility optimization algorithm is as follows: Real-time analysis of mineral composition: Rapidly identify rock mineral components through downhole XRD spectrometer and input into chemical reaction kinetics model; Dynamic ratio optimization: Determine the optimal HF and HCl ratio based on the target reaction rate and constraints; Dynamically adjust the ratio through Lagrange multiplier method or genetic algorithm; The architecture of the multi-objective optimization controller is as follows: Optimization objectives: maximize acid etching depth, crack conductivity, and minimize construction costs; Dynamic weight allocation strategy: adjust target priorities according to the construction stage.

7. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The architecture of the variable frequency speed regulation acid pump system based on fuzzy sliding mode control is as follows: Fuzzy control: takes displacement error and error change rate as input, outputs initial control quantity through fuzzy rules, generates based on error and error change rate, and is dynamically adjusted by the digital twin model; Sliding mode control: Design a sliding surface. When the system state deviates from the sliding surface, it is forced to return by switching the control quantity. Frequency conversion speed regulation: convert the control quantity into motor frequency, drive the acid pump through the frequency converter, and realize continuous adjustment of displacement; The role of digital twins: Use the twin model to predict changes in the acid absorption capacity of the formation, dynamically adjust the target, and modify the control parameters and fuzzy rule base in real time.

8. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The architecture of the intelligent diversion system is as follows: Layered section absorption monitoring unit: The downhole flow sensor array is used to monitor the acid absorption of each layer in real time, and the permeability and mineral composition differences of each layer are analyzed in combination with the digital twin model; Objective function: maximize global flow conductivity; Allocation algorithm: Using an improved particle swarm optimization algorithm, the acid distribution ratio of each layer is dynamically adjusted according to real-time data; Actuator: The flow rate of each layer branch is adjusted through the electronically controlled throttle valve to achieve real-time adjustment of acid distribution.

9. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The cloud-edge collaborative operation and maintenance module includes an edge computing layer, a cloud platform, and a network layer. The edge computing layer is deployed at the well site and is responsible for real-time data collection, local closed-loop control, preprocessing, and caching. The cloud platform is based on digital twins and blockchain to achieve global optimization, cross-well site experience sharing, and trusted data storage; the network layer uses industrial Ethernet to achieve low-latency communication between the edge and the cloud, supporting real-time command issuance and batch data synchronization.

10. The electric control system for continuous acidizing of oil and water wells based on multi-parameter monitoring according to claim 1, characterized in that: The architecture of the digital twin cloud platform is: Dynamic parameter calibration: Based on real-time data from the well site, Kalman filtering is used to online correct formation permeability and acid-rock reaction rate; Federated learning algorithm iteration: Each well site edge device trains the model locally, and the cloud aggregates the global model; Case library construction: Historical construction data is stored in the cloud, and stratum types are divided through clustering algorithms to form typical construction templates.

Images