Real-time blockage risk prediction method of geothermal tail water recharge system and related device

Abstract

The invention discloses a real-time blockage risk prediction method of a geothermal tail water recharge system and a related device, and belongs to the technical field of geothermal recharge monitoring. The method comprises the steps that historical tail water recharge parameters of the geothermal tail water recharge system after a previous cleaning time node are obtained, and a tail water recharge parameter time sequence is generated; geological environment parameters, medium characteristic parameters and geothermal tail water recharge system design parameters are obtained, a geothermal tail water recharge system model is constructed, and tail water recharge twin bodies are obtained; blocking factors are preset, the tail water recharge parameter time sequence is input into a tail water recharge twinborn body, and blocking factor distribution state data are generated; real-time tail water recharge parameters are obtained, blockage risk analysis under blockage factor transfer is carried out based on the blockage factor distribution state data and the real-time tail water recharge parameters, and blockage risk judgment information is generated; and carrying out blocking risk early warning based on the blocking risk judgment information. According to the invention, the technical problems of lag and low accuracy of blockage risk prediction in the prior art are solved.

Description

Technical Field

[0001] This invention belongs to the field of geothermal reinjection monitoring technology, and relates to a real-time blockage risk prediction method and related device for geothermal tailwater reinjection systems. Background Technology

[0002] Geothermal resources, as a clean and stable renewable energy source, are increasingly widely used in heating, power generation, and other fields. Tailwater reinjection is a crucial link in achieving sustainable geothermal resource development, effectively maintaining underground reservoir pressure and preventing geological environmental damage. However, reinjection wells are prone to blockage during long-term operation, becoming a core technical bottleneck restricting the efficient and stable operation of geothermal systems. Reinjection well blockage is mainly caused by the combined effects of multiple factors: suspended particles in the tailwater deposit in the pores of the surrounding formation, forming physical blockage; minerals in the water reach a supersaturated state due to temperature and pressure changes, precipitating and forming chemical precipitation blockage; dissolved gases in the groundwater precipitate and accumulate with environmental changes, occupying seepage channels; simultaneously, suitable temperature and pressure environments promote the growth and reproduction of microorganisms, whose metabolic products combine with impurities to form biological blockage. These blockage problems lead to a continuous decline in reinjection capacity, and in severe cases, cause the reinjection well to be abandoned, significantly increasing geothermal development costs.

[0003] Existing geothermal tailwater reinjection systems typically assess blockage risk by real-time monitoring of reinjection parameters such as flow rate, pressure, temperature, and turbidity, combined with preset empirical thresholds. However, the blockage formation mechanism is significantly complex and coupled, dynamically influenced by multiple factors including formation lithology, pore structure, tailwater chemical composition, and operating conditions. Empirical threshold methods struggle to characterize this complex blockage evolution process. This approach cannot accurately predict the specific timing and severity of blockages, nor does it possess the capability for in-depth modeling of historical operating data. It also fails to quantify the cumulative effects of various factors on the blockage trend over long-term operation, resulting in significant lag and low accuracy in blockage risk prediction. Consequently, it cannot provide effective support for preventative maintenance, severely impacting the safe and stable operation of geothermal tailwater reinjection systems. Summary of the Invention

[0004] The purpose of this invention is to solve the technical problem that the existing technology cannot predict when the blockage will occur and the extent of the blockage, resulting in delayed and inaccurate blockage risk prediction. This invention provides a real-time blockage risk prediction method and related device for geothermal tailwater reinjection systems.

[0005] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, this invention discloses a method for real-time blockage risk prediction of a geothermal tailwater reinjection system, comprising: Obtain historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time point, and generate tailwater reinjection parameter time series; Geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters are obtained, a geothermal tailwater reinjection system model is constructed, and a tailwater reinjection twin is obtained. Preset the blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data; Real-time tailwater reinjection parameters are obtained, and based on the distribution data of the clogging factors and the real-time tailwater reinjection parameters, a clogging risk analysis is performed under the transfer of clogging factors to generate clogging risk judgment information. A congestion risk warning is issued based on the congestion risk assessment information.

[0006] Further improvements are made in the following aspects: The construction of the geothermal tailwater reinjection system model, resulting in the tailwater reinjection twin, specifically includes: Based on the geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters, a geothermal tailwater reinjection system model is constructed by integrating fluid dynamics and mineral precipitation simulation of the geothermal tailwater reinjection system, thus obtaining the tailwater reinjection twin.

[0007] The process of acquiring real-time tailwater reinjection parameters, performing a blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution data and real-time tailwater reinjection parameters, and generating blockage risk assessment information specifically includes: Read preset blocking factors; Based on the preset congestion factors, historical congestion state transition data are collected; Train a state transition analyzer based on the historical congestion state transition data; Using the state transition analyzer as a starting point, the state transition prediction is performed based on the real-time tailwater reinjection parameters to generate the predicted transition result; Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information.

[0008] Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information, specifically as follows: Construct threshold values ​​for the characteristics of congestion factors that cause congestion; Calculate the similarity between the predicted transfer result and the characteristic threshold of the congestion factor to generate a predicted congestion risk; Determine whether the predicted congestion risk is greater than a preset risk threshold, and generate the congestion risk assessment information.

[0009] The congestion risk warning based on the aforementioned congestion risk assessment information specifically includes: When a blockage risk warning signal is generated, a cleaning equipment signal is automatically sent to start cleaning the reinjection path of the geothermal tailwater reinjection system, and a backup reinjection path is activated during the cleaning period.

[0010] After issuing a congestion risk warning based on the congestion risk assessment information, the method further includes: Analyze the correlation between the historical tailwater reinjection parameters and preset blockage factors; Based on the aforementioned correlation and the design parameters of the geothermal tailwater reinjection system, an adaptive optimization model for the reinjection parameters is constructed. The real-time tailwater reinjection parameters are optimized based on the adaptive optimization model of the reinjection parameters.

[0011] The preset clogging factors include particulate deposition, chemical precipitation, gas evolution, and biological clogging, and each clogging factor corresponds to a specific tailwater reinjection parameter characterization item: Particulate sedimentation corresponds to turbidity and suspended particulate concentration in the tailwater. Chemical precipitation corresponds to the pH value and calcium and magnesium ion concentration of the effluent; The gas release corresponds to the reinjection pressure and the dissolved gas content in the tailwater. Biological blockage corresponds to effluent temperature and organic matter concentration.

[0012] Secondly, this invention discloses a real-time blockage risk prediction system for a geothermal tailwater reinjection system, comprising: The historical tailwater reinjection parameter acquisition module is used to acquire historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time node and generate tailwater reinjection parameter time sequence. The tailwater reinjection twin construction module is used to obtain geological environment parameters, medium characteristic parameters and geothermal tailwater reinjection system design parameters, construct a geothermal tailwater reinjection system model, and obtain a tailwater reinjection twin. The blockage factor distribution status data determination module is used to preset blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data. The real-time blockage risk assessment information determination module is used to acquire real-time tailwater reinjection parameters, perform blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution status data and real-time tailwater reinjection parameters, and generate blockage risk assessment information. The congestion risk warning module is used to issue congestion risk warnings based on the congestion risk assessment information.

[0013] Thirdly, the present invention discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned real-time blockage risk prediction method for the geothermal tailwater reinjection system.

[0014] Fourthly, the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for predicting the real-time blockage risk of the geothermal tailwater reinjection system.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a real-time blockage risk prediction method for geothermal tailwater reinjection systems. The "time node anchoring" design for historical parameter acquisition enhances data effectiveness and lays the foundation for accurate prediction. The method explicitly uses the "previous cleanup time node" as the starting point for historical parameter acquisition. This node corresponds to the "initial baseline state" of the reinjection system's blockage risk, effectively filtering outdated data that was invalid before cleanup. This ensures that the generated tailwater reinjection parameter time series only includes data directly related to the current blockage evolution process, avoiding invalid data interfering with modeling accuracy and solving the problem of prediction baseline deviation caused by mixed historical data in existing technologies. The construction of a tailwater reinjection twin achieves "virtual-real fusion" simulation, overcoming the limitations of empirical judgment. By integrating geological environment, medium characteristics, and system design parameters to construct a twin, the physical morphology, operational characteristics, and formation response of the reinjection system are accurately mapped in virtual space. Loading the historical parameter time series into the "blockage factor distribution status data" generated by the twin is essentially a digital reproduction of the blockage evolution process. Compared to existing technologies that rely solely on real-time parameters and empirical thresholds, this method achieves visualization and quantifiable characterization of the spatial distribution and cumulative effects of blockage factors. The "historical baseline + real-time dynamic" risk analysis model improves the timeliness and accuracy of predictions. Based on the distribution of blockage factors output by the twin, the method combines real-time reinjection parameters to conduct blockage factor transfer analysis. This preserves the historical patterns of blockage evolution while dynamically capturing the abrupt changes in blockage factors under real-time operating conditions. It effectively captures early, weak signals of blockage development, avoiding the lag problem of existing technologies that "can only passively identify blockages after they become apparent." Simultaneously, the linkage analysis of historical and real-time data significantly reduces misjudgments caused by fluctuations in single parameters, minimizing false alarms and missed alarms during the prediction process, and making the prediction results more closely match the actual operating status of the reinjection system. Precise early warning provides proactive protection for stable system operation. Based on the blockage risk assessment information and early warning signals generated by the above precise analysis, maintenance personnel can grasp the potential time and severity of blockages in advance, providing a scientific basis for developing preventative maintenance strategies. This avoids problems such as reinjection interruptions and equipment damage caused by sudden blockages, significantly improving the operational stability and efficiency of the geothermal tailwater reinjection system and reducing maintenance costs. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a real-time blockage risk prediction method for a geothermal tailwater reinjection system according to an embodiment of the present invention; Figure 2 This is a block diagram of a real-time blockage risk prediction system for a geothermal tailwater reinjection system according to an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0021] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 This invention discloses a method for real-time blockage risk prediction of a geothermal tailwater reinjection system, comprising: S1, obtain the historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the previous cleaning time node, and generate the tailwater reinjection parameter time sequence; S2, obtain geological environment parameters, medium characteristic parameters and geothermal tailwater reinjection system design parameters, construct geothermal tailwater reinjection system model, and obtain tailwater reinjection twin; S3, preset the blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data; S4, acquire real-time tailwater reinjection parameters, perform blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution data and real-time tailwater reinjection parameters, and generate blockage risk judgment information; S5, based on the blockage risk assessment information, issue a blockage risk warning.

[0022] This invention discloses a real-time blockage risk prediction method for geothermal tailwater reinjection systems. The "time node anchoring" design for historical parameter acquisition enhances data effectiveness and lays the foundation for accurate prediction. The method explicitly uses the "previous cleanup time node" as the starting point for historical parameter acquisition. This node corresponds to the "initial baseline state" of the reinjection system's blockage risk, effectively filtering outdated data that was invalid before cleanup. This ensures that the generated tailwater reinjection parameter time series only includes data directly related to the current blockage evolution process, avoiding invalid data interfering with modeling accuracy and solving the problem of prediction baseline deviation caused by mixed historical data in existing technologies. The construction of a tailwater reinjection twin achieves "virtual-real fusion" simulation, overcoming the limitations of empirical judgment. By integrating geological environment, medium characteristics, and system design parameters to construct a twin, the physical morphology, operational characteristics, and formation response of the reinjection system are accurately mapped in virtual space. Loading the historical parameter time series into the "blockage factor distribution status data" generated by the twin is essentially a digital reproduction of the blockage evolution process. Compared to existing technologies that rely solely on real-time parameters and empirical thresholds, this method achieves visualization and quantifiable characterization of the spatial distribution and cumulative effects of blockage factors. The "historical baseline + real-time dynamic" risk analysis model improves the timeliness and accuracy of predictions. Based on the distribution of blockage factors output by the twin, the method combines real-time reinjection parameters to conduct blockage factor transfer analysis. This preserves the historical patterns of blockage evolution while dynamically capturing the abrupt changes in blockage factors under real-time operating conditions. It effectively captures early, weak signals of blockage development, avoiding the lag problem of existing technologies that "can only passively identify blockages after they become apparent." Simultaneously, the linkage analysis of historical and real-time data significantly reduces misjudgments caused by fluctuations in single parameters, minimizing false alarms and missed alarms during the prediction process, and making the prediction results more closely match the actual operating status of the reinjection system. Precise early warning provides proactive protection for stable system operation. Based on the blockage risk assessment information and early warning signals generated by the above precise analysis, maintenance personnel can grasp the potential time and severity of blockages in advance, providing a scientific basis for developing preventative maintenance strategies. This avoids problems such as reinjection interruptions and equipment damage caused by sudden blockages, significantly improving the operational stability and efficiency of the geothermal tailwater reinjection system and reducing maintenance costs.

[0023] The present invention will be further described below with reference to specific embodiments: Step 1: Obtain historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time point, and generate tailwater reinjection parameter time series; Step 2: Obtain geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters; construct a geothermal tailwater reinjection system model; and obtain a tailwater reinjection twin. Based on the geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters, a geothermal tailwater reinjection system model is constructed by integrating fluid dynamics and mineral precipitation simulation of the geothermal tailwater reinjection system, thus obtaining the tailwater reinjection twin.

[0024] Step 3: Preset clogging factors and input the tailwater reinjection parameters into the tailwater reinjection twin to generate clogging factor distribution data. The preset clogging factors include particle deposition, chemical precipitation, gas evolution, and biological clogging, and each clogging factor corresponds to specific tailwater reinjection parameter characterization items: particle deposition corresponds to tailwater turbidity and suspended particle concentration; chemical precipitation corresponds to tailwater pH and calcium and magnesium ion concentration; gas evolution corresponds to reinjection pressure and tailwater dissolved gas content; and biological clogging corresponds to tailwater temperature and organic matter concentration.

[0025] Step 4: Obtain real-time tailwater reinjection parameters; perform a blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution data and real-time tailwater reinjection parameters; and generate blockage risk judgment information. Read preset blocking factors; Based on the preset congestion factors, historical congestion state transition data are collected; Train a state transition analyzer based on the historical congestion state transition data; Using the state transition analyzer as a starting point, the state transition prediction is performed based on the real-time tailwater reinjection parameters to generate the predicted transition result; Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information.

[0026] Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information, specifically as follows: Construct threshold values ​​for the characteristics of congestion factors that cause congestion; Calculate the similarity between the predicted transfer result and the characteristic threshold of the congestion factor to generate a predicted congestion risk; Determine whether the predicted congestion risk is greater than a preset risk threshold, and generate the congestion risk assessment information.

[0027] Step 5: Issue a congestion risk warning based on the congestion risk assessment information.

[0028] When a blockage risk warning signal is generated, a cleaning equipment signal is automatically sent to start cleaning the reinjection path of the geothermal tailwater reinjection system, and a backup reinjection path is activated during the cleaning period.

[0029] Step 6: Analyze the correlation between the historical tailwater reinjection parameters and the preset blockage factors; Based on the aforementioned correlation and the design parameters of the geothermal tailwater reinjection system, an adaptive optimization model for the reinjection parameters is constructed. The real-time tailwater reinjection parameters are optimized based on the adaptive optimization model of the reinjection parameters.

[0030] Example 1: This invention provides a real-time blockage risk prediction method for a geothermal tailwater reinjection system. The method is executed by a real-time blockage risk prediction device for the geothermal tailwater reinjection system, and specifically includes the following steps: S100: Determine the previous cleaning time node of the geothermal tailwater reinjection system, and start from the previous cleaning time node to collect historical tailwater reinjection parameters and generate tailwater reinjection parameter time sequence.

[0031] Specifically, a geothermal tailwater reinjection system refers to a system that reinjects used cooling hot water (tailwater) into the underground reservoir during geothermal heating or power generation. It includes injection wells, tailwater delivery pipelines, water quality conditioning equipment, and monitoring devices. The last cleaning time point is the date when chemical cleaning, airlift unblocking, high-pressure water flushing, or other maintenance operations were last performed on the reinjection well or system. Determining the specific date of the most recent well cleaning or system maintenance of the geothermal reinjection system typically indicates that the system was clean or unblocked at that time.

[0032] Starting from the previous cleaning point in time, operating parameters of the geothermal tailwater reinjection system are collected, recording the process from cleanliness to gradual blockage. In other words, from the previous cleaning point in time, during normal operation, various sensors monitor and record operational data in real time. By continuously collecting information such as injection pressure, flow rate, and water quality, and organizing it into a time series format, a complete historical dataset of operational status is constructed. Historical tailwater reinjection parameters are key data collected during the reinjection process, including but not limited to: injection pressure, injection flow rate, reinjection water temperature, and water quality parameters. These historical tailwater reinjection parameters are organized into a time series data according to time sequence (e.g., by minute, hour, or day), resulting in the tailwater reinjection parameter time series. By determining the previous cleaning point in time and constructing the tailwater reinjection parameter time series from that point, a baseline for the blockage evolution of the geothermal tailwater reinjection system is established, recording the dynamic process and trends of parameter changes over time.

[0033] S200: Perform physical simulation modeling of the geothermal tailwater reinjection system and construct a tailwater reinjection twin.

[0034] This includes: collecting modeling data for the geothermal tailwater reinjection system; performing an integrated simulation of fluid dynamics and mineral precipitation in the geothermal tailwater reinjection system based on the modeling data, and constructing the tailwater reinjection twin.

[0035] Specifically, comprehensive data collection should be conducted on the structure and operating environment of the geothermal tailwater reinjection system, including well structure parameters such as well depth, well diameter, and distribution of filter pipe sections, as well as the permeability coefficient, porosity, and pressure distribution of the underground reservoir. It should also include key physicochemical properties of the tailwater, such as temperature, pH value, conductivity, and concentrations of calcium ions and bicarbonate ions.

[0036] Based on the collected modeling data, a physical model of the tailwater reinjection system was established on a professional simulation platform. The fluid dynamics model was used to simulate the flow behavior of the tailwater in the injection well and formation, including the distribution of flow velocity, flow direction, and pressure drop. Combined with a mineral precipitation reaction model, the reaction trends and precipitation rates of inorganic ions in the tailwater under formation conditions were further simulated, such as the precipitation process of calcium carbonate and silicate minerals. Through the integrated coupling of flow simulation and chemical precipitation simulation, a dynamic and continuously evolving geothermal tailwater reinjection twin was constructed. The tailwater reinjection twin can not only replay historical reinjection processes but also load real-time parameters for forward prediction, thus possessing a certain degree of autonomous evolution and trend extrapolation capabilities.

[0037] Fluid mechanics is the study of the mechanical laws governing fluids (liquids and gases) in equilibrium and motion, and their applications. In geothermal tailwater reinjection systems, fluid mechanics focuses on the velocity, pressure distribution, and flow pattern (laminar or turbulent) of the tailwater flowing through pipes, wells, and other channels. These factors directly affect the transport and deposition of blockages. Mineral precipitation refers to the process by which dissolved minerals (such as calcium carbonate and silicates) in geothermal tailwater precipitate out of the solution under specific conditions (such as changes in temperature, pressure, and pH), forming solid precipitates. These precipitates are one of the important causes of blockages in reinjection systems.

[0038] Based on the collected modeling data, a comprehensive simulation of fluid dynamics and mineral precipitation was conducted on the geothermal tailwater reinjection system using professional computational fluid dynamics software and a chemical precipitation model. A three-dimensional geometric model was established, a computational grid was divided, and boundary conditions (such as inlet flow rate and outlet pressure) were set. The equations describing fluid flow and the chemical model describing the mineral dissolution-precipitation equilibrium were coupled to track the flow path and velocity distribution of the fluid in the system. At the same time, the saturation of minerals (such as CaCO3) under specific temperature, pressure, and chemical conditions was calculated to predict their precipitation rate and accumulation at pipe walls, filters, well shafts, etc.

[0039] After repeatedly verifying and correcting the model parameters to ensure that the simulation results are basically consistent with the actual observed trends in pressure drop and flow rate, this verified calculation model, which can simultaneously simulate fluid dynamics and mineral precipitation processes, constitutes the tailwater reinjection twin.

[0040] By creating a highly realistic virtual model—the tailwater reinjection twin—that can dynamically simulate the complex physicochemical processes inside the reinjection system, the analysis can be reproduced and even predicted through integrated simulation, including how fluids flow in the system and where and at what rate minerals precipitate to form blockages, thus improving the depth and accuracy of the analysis.

[0041] S300: Load the tailwater reinjection parameter timing with the tailwater reinjection twin to generate the distribution state of blockage factors.

[0042] Specifically, the tailwater reinjection parameters are loaded into the tailwater reinjection twin in time series. The fluid and hydrochemical parameters at each moment are correspondingly input into the twin model, driving its internal fluid dynamics and mineral precipitation modules to perform evolution calculations. Based on the tailwater reinjection parameter time series, the tailwater reinjection twin dynamically calculates indicators such as velocity distribution, pressure drop changes, supersaturation index, and sedimentation rate in the wellbore and formation pore space at each time point, generating the distribution state of clogging factors. The strength and location distribution of clogging-related factors are characterized using three-dimensional spatial coordinates and the time dimension. For example, in one simulation, the model output showed a significant decrease in flow velocity between 1450 and 1520 meters in the wellbore depth during days 15 to 25 of operation. The precipitation rate of calcium carbonate, formed by the coupling of calcium ions and bicarbonate concentrations, increased from an initial 0.006 g / cm² / day to 0.021 g / cm² / day, with a local porosity decrease of over 22%. The injection pressure also rose from 1.36 MPa to 1.54 MPa during this period, clearly indicating a sustained localized blockage evolution process in this region. The distribution of blockage factors not only includes the spatial accumulation of precipitates but also tracks the spatiotemporal superposition of various blockage inducing factors such as changes in particulate matter concentration and gas evolution trends.

[0043] By loading the tailwater reinjection parameters into the tailwater reinjection twin, the blockage evolution behavior of the geothermal tailwater reinjection system during its historical operation was simulated, and the distribution of blockage factors in the wellbore and reservoir was clearly output, revealing the pattern and spatial characteristics of blockage formation, as well as when and under what operating conditions the blockage material mainly accumulates rapidly.

[0044] S400: Collect real-time tailwater reinjection parameters, perform blockage risk analysis under the transfer of blockage factors based on the distribution status of the blockage factors, and generate blockage risk judgment information.

[0045] The process includes: reading preset blockage factors; collecting historical blockage state transition data for the preset blockage factors; training a state transition analyzer based on the historical blockage state transition data; using the state transition analyzer to predict state transitions based on the distribution state of the blockage factors and the real-time tailwater reinjection parameters, generating predicted transition results; and performing blockage risk analysis based on the predicted transition results to generate blockage risk judgment information.

[0046] Furthermore, the present invention also includes the following steps: constructing a congestion factor feature threshold that leads to congestion; calculating the similarity between the predicted transfer result and the congestion factor feature threshold to generate a predicted congestion risk; determining whether the predicted congestion risk is greater than a preset risk threshold to generate the congestion risk judgment information.

[0047] Specifically, the system retrieves or reads a set of preset clogging factors. These are key factors or indicators that may cause or exacerbate clogging, pre-defined based on an understanding of the characteristics of the geothermal tailwater reinjection system, historical experience, or preliminary analysis. Examples include sediment thickness, particulate matter concentration, percentage reduction in flow velocity in a specific area, or the clogging level of a component (such as a filter or elbow). Based on these preset clogging factors, historical clogging state transition data is collected. This data shows how the state of the preset clogging factors changes from one level to another at different points in time. For example, it records how the sediment thickness in a section of pipeline increased from 0.05m to 0.10m, and then to 0.15m over the past year, including various intermediate states and time points.

[0048] Using historical state transition sample data as input, a state transition analyzer is trained. This analyzer, based on an LSTM neural network time series classifier, learns the transition probabilities, transition rates, and key feature combinations of changes in various factors between states. The state transition analyzer is a machine learning model or rule-based analysis engine that learns the patterns inherent in historical congestion state transition data—that is, understanding under what conditions, and in what ways (speed, direction), congestion factors change. After training, it can predict possible future congestion paths based on the current congestion state and new input information (such as real-time operating parameters).

[0049] Using this historical data with time-series labels, a specialized state transition analyzer (such as a recurrent neural network (RNN) or a long short-term memory (LSTM) model) is trained. This training process teaches the analyzer how sediment at bends typically grows faster when flow rate decreases and temperature increases, and how filter clogging rates might change when pressure fluctuates significantly—these are the inherent relationships. After training, the analyzer functions like an experienced prediction expert. In practical applications, the current distribution of clogging factors, calculated using a twin model, is received first. Real-time tailwater reinjection parameters are then input into the state transition analyzer to predict state transitions, yielding the predicted transition results. These predicted transition results are calculated by the state transition analyzer based on the current distribution of clogging factors, combined with the input real-time tailwater reinjection parameters, to predict the possible changes in the state of clogging factors over a future period (e.g., the next 24 or 72 hours).

[0050] Based on practical experience, design standards, or parameter characteristics prior to severe congestion events, a set of characteristic thresholds for congestion factors is constructed as a standard for judging the severity of congestion. The predicted transfer results are then compared with these characteristic thresholds using multidimensional similarity calculations. Methods such as cosine similarity are employed to output a quantitative predicted congestion risk. This predicted congestion risk is a quantitative assessment of the likelihood of congestion occurring, based on the similarity calculation results. High similarity corresponds to high risk, and low similarity corresponds to low risk. This risk value is a numerical value between 0 and 1 (or 0% to 100%), representing the probability or severity of a congestion event requiring attention within the predicted timeframe.

[0051] A preset risk threshold is a pre-defined risk level limit used to determine whether the currently predicted congestion risk has reached a level requiring an alert. For example, the threshold can be set to 0.7 (or 70%). When the predicted risk exceeds this value, the congestion risk is considered high, and action is required. The system determines whether the predicted congestion risk exceeds the preset risk threshold and generates congestion risk assessment information. For example, the state transition analyzer, combined with the twin output, determines that the trend is highly similar to the historical stage 3 moderate congestion evolution path, calculating a similarity of 0.84, exceeding the set risk threshold of 0.75. It then outputs high congestion risk assessment information and recommends scheduling a cleaning operation or adjusting the injection flow within 24 hours.

[0052] By training a state transition analyzer, the potential evolution of future congestion states can be predicted proactively based on the current state and real-time operating conditions. Ultimately, the risk level of congestion is quantified, improving the accuracy and timeliness of congestion risk prediction.

[0053] S500: Issue a congestion risk warning based on the congestion risk assessment information.

[0054] This includes: when a blockage risk warning signal is issued, automatically activating the cleaning equipment to clean the backflow path, and activating the backup backflow path during the cleaning period.

[0055] Furthermore, this application also includes the following steps: collecting historical data to analyze the correlation between tailwater reinjection parameters and blockage factors; constructing an adaptive optimization model for reinjection parameters based on the correlation and the parameter control range of the geothermal tailwater reinjection system; and performing optimization analysis of the real-time tailwater reinjection parameters using the adaptive optimization model for reinjection parameters.

[0056] Specifically, based on blockage risk assessment information, a blockage risk warning is issued. That is, when the prediction module determines a high level of blockage risk, it not only issues a warning signal but also automatically controls the cleaning equipment to initiate unblocking and cleaning operations on the main reinjection path, while simultaneously switching to a backup injection path to ensure uninterrupted geothermal tailwater reinjection. Control signals activate cleaning equipment deployed on the main reinjection path, such as high-pressure airlift systems, acid injection systems, or mechanical flushing devices, simultaneously opening one or more backup reinjection paths. Backup paths must be pre-equipped with operational capabilities, including independent injection pumps, metering devices, and pressure control systems.

[0057] During cleaning periods, backup reinjection paths are activated. This means that during the actual working hours of the cleaning equipment, backup reinjection paths are initiated. Backup reinjection paths refer to one or more additional reinjection channels designed in addition to the main reinjection path when designing a geothermal tailwater reinjection system. These backup paths may be in standby or low-flow operation under normal circumstances. Their main function is to take over part or all of the reinjection tasks when the main path requires maintenance, cleaning, or malfunctions, ensuring that geothermal energy production is not significantly affected and maintaining the continuity and stability of the system.

[0058] Historical data was collected and analyzed to determine the correlation between tailwater reinjection parameters and clogging factors. This involved retrospectively analyzing a large amount of reinjection data collected during historical operating cycles, including injection pressure, injection flow rate, tailwater temperature, water quality parameters (such as turbidity, conductivity, and calcium ion concentration), operating time, and clogging status records (such as cleaning frequency and pressure rise rate). Statistical analysis, principal component analysis (PCA), or correlation thermography were used to identify the quantitative relationships between different parameters and clogging factors. The dominant factors influencing clogging may differ under different geothermal well or geological conditions. For example, in one project, a significant positive correlation was found between injection flow rate fluctuation and calcium carbonate precipitation rate, with a correlation coefficient of 0.81; while changes in conductivity had a weaker impact on clogging, with a coefficient of only 0.22.

[0059] Based on the control range of the operating parameters of the geothermal tailwater reinjection system, an adaptive optimization model for reinjection parameters is constructed. Multi-objective optimization methods (such as particle swarm optimization and multi-objective genetic algorithms) are employed. The objective functions typically include maximizing reinjection flow rate, minimizing the blockage risk score, and maintaining injection pressure within a safe range. The model's constraints are derived from equipment limits and formation tolerance, such as a maximum injection flow rate not exceeding 130 cubic meters per hour, a pressure not exceeding 1.8 MPa, and a minimum reinjection temperature not lower than 45 degrees Celsius. The adaptive optimization model receives tailwater reinjection parameter data in real time during operation and performs rapid analysis and dynamic optimization. It finds a set of operating parameter values ​​that, while meeting production requirements (such as ensuring a certain reinjection volume), can minimize or delay the occurrence and development of blockages. The adaptive optimization model can dynamically adjust the optimal parameter combination based on real-time changes in operating conditions (such as tailwater quality fluctuations and seasonal temperature changes), rather than remaining fixed.

[0060] The actual tailwater reinjection parameters (flow rate, pressure, temperature, etc.) collected at the current moment are input into the reinjection parameter adaptive optimization model. Based on the current specific operating conditions, combined with the previously established correlation relationships and parameter control ranges, the model calculates which adjustable parameters (such as pump speed and valve opening) should be set to what values ​​to achieve the best anti-clogging effect under these conditions. By establishing a clear correlation between parameters and clogging, and performing real-time optimization accordingly, the model automatically avoids operating conditions that easily lead to rapid clogging while meeting basic production needs. This significantly extends the interval between two cleaning cycles, reduces the frequency and severity of clogging, reduces unplanned downtime and maintenance costs caused by clogging, and ultimately improves the continuous and stable supply capacity and overall efficiency of geothermal energy.

[0061] In summary, the real-time blockage risk prediction method for geothermal tailwater reinjection systems provided by this invention determines the previous cleaning time node of the geothermal tailwater reinjection system, collects historical tailwater reinjection parameters from the previous cleaning time node, and generates a tailwater reinjection parameter time series; performs physical simulation modeling of the geothermal tailwater reinjection system to construct a tailwater reinjection twin; loads the tailwater reinjection parameter time series onto the tailwater reinjection twin to generate a blockage factor distribution state; collects real-time tailwater reinjection parameters, performs blockage risk analysis under the blockage factor transfer based on the blockage factor distribution state, and generates blockage risk judgment information; and provides a blockage risk warning based on the blockage risk judgment information. In other words, by using historical data (from the last cleaning) to construct the tailwater reinjection parameter time series and generating a blockage factor distribution state through physical simulation modeling, and performing blockage risk analysis, early signals of blockage development are captured, the prediction results are closer to the actual situation, false alarms and false negatives are reduced, and the accuracy and timeliness of blockage risk prediction are improved.

[0062] See Figure 2 This invention also provides a real-time blockage risk prediction system for a geothermal tailwater reinjection system, comprising: The historical tailwater reinjection parameter acquisition module is used to acquire historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time node and generate tailwater reinjection parameter time sequence. The tailwater reinjection twin construction module is used to obtain geological environment parameters, medium characteristic parameters and geothermal tailwater reinjection system design parameters, construct a geothermal tailwater reinjection system model, and obtain a tailwater reinjection twin. The blockage factor distribution status data determination module is used to preset blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data. The real-time blockage risk assessment information determination module is used to acquire real-time tailwater reinjection parameters, perform blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution status data and real-time tailwater reinjection parameters, and generate blockage risk assessment information. The congestion risk warning module is used to issue congestion risk warnings based on the congestion risk assessment information.

[0063] A third objective of this invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a real-time blockage risk prediction method for the geothermal tailwater reinjection system.

[0064] The real-time blockage risk prediction method for the geothermal tailwater reinjection system includes the following steps: Obtain historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time point, and generate tailwater reinjection parameter time series; Geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters are obtained, a geothermal tailwater reinjection system model is constructed, and a tailwater reinjection twin is obtained. Preset the blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data; Real-time tailwater reinjection parameters are obtained, and based on the distribution data of the clogging factors and the real-time tailwater reinjection parameters, a clogging risk analysis is performed under the transfer of clogging factors to generate clogging risk judgment information. A congestion risk warning is issued based on the congestion risk assessment information.

[0065] The fourth objective of this invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements a real-time blockage risk prediction method for the geothermal tailwater reinjection system.

[0066] The real-time blockage risk prediction method for the geothermal tailwater reinjection system includes the following steps: Obtain historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time point, and generate tailwater reinjection parameter time series; Geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters are obtained, a geothermal tailwater reinjection system model is constructed, and a tailwater reinjection twin is obtained. Preset the blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data; Real-time tailwater reinjection parameters are obtained, and based on the distribution data of the clogging factors and the real-time tailwater reinjection parameters, a clogging risk analysis is performed under the transfer of clogging factors to generate clogging risk judgment information. A congestion risk warning is issued based on the congestion risk assessment information.

[0067] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0068] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0069] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0071] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for real-time blockage risk prediction of a geothermal tailwater reinjection system, characterized in that, include: Obtain historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time point, and generate tailwater reinjection parameter time series; Geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters are obtained, a geothermal tailwater reinjection system model is constructed, and a tailwater reinjection twin is obtained. Preset the blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data; Real-time tailwater reinjection parameters are obtained, and based on the distribution data of the clogging factors and the real-time tailwater reinjection parameters, a clogging risk analysis is performed under the transfer of clogging factors to generate clogging risk judgment information. A congestion risk warning is issued based on the congestion risk assessment information.

2. The real-time blockage risk prediction method for a geothermal tailwater reinjection system according to claim 1, characterized in that, The construction of the geothermal tailwater reinjection system model, resulting in the tailwater reinjection twin, specifically includes: Based on the geological environment parameters, medium characteristic parameters, and geothermal tailwater reinjection system design parameters, a geothermal tailwater reinjection system model is constructed by integrating fluid dynamics and mineral precipitation simulation of the geothermal tailwater reinjection system, thus obtaining the tailwater reinjection twin.

3. The real-time blockage risk prediction method for geothermal tailwater reinjection systems according to claim 1, characterized in that, The process of acquiring real-time tailwater reinjection parameters, performing a blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution data and real-time tailwater reinjection parameters, and generating blockage risk assessment information specifically includes: Read preset blocking factors; Based on the preset congestion factors, historical congestion state transition data are collected; Train a state transition analyzer based on the historical congestion state transition data; Using the state transition analyzer as a starting point, the state transition prediction is performed based on the real-time tailwater reinjection parameters to generate the predicted transition result; Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information.

4. The real-time blockage risk prediction method for a geothermal tailwater reinjection system according to claim 3, characterized in that, Based on the predicted transfer results, a congestion risk analysis is performed to generate the congestion risk assessment information, specifically as follows: Construct threshold values ​​for the characteristics of congestion factors that cause congestion; Calculate the similarity between the predicted transfer result and the characteristic threshold of the congestion factor to generate a predicted congestion risk; Determine whether the predicted congestion risk is greater than a preset risk threshold, and generate the congestion risk assessment information.

5. The real-time blockage risk prediction method for a geothermal tailwater reinjection system according to claim 1, characterized in that, The congestion risk warning based on the aforementioned congestion risk assessment information specifically includes: When a blockage risk warning signal is generated, a cleaning equipment signal is automatically sent to start cleaning the reinjection path of the geothermal tailwater reinjection system, and a backup reinjection path is activated during the cleaning period.

6. The real-time blockage risk prediction method for a geothermal tailwater reinjection system according to claim 1, characterized in that, After issuing a congestion risk warning based on the congestion risk assessment information, the method further includes: Analyze the correlation between the historical tailwater reinjection parameters and preset blockage factors; Based on the aforementioned correlation and the design parameters of the geothermal tailwater reinjection system, an adaptive optimization model for the reinjection parameters is constructed. The real-time tailwater reinjection parameters are optimized based on the adaptive optimization model of the reinjection parameters.

7. The real-time blockage risk prediction method for a geothermal tailwater reinjection system according to claim 1, characterized in that, The preset clogging factors include particulate deposition, chemical precipitation, gas evolution, and biological clogging, and each clogging factor corresponds to a specific tailwater reinjection parameter characterization item: Particulate sedimentation corresponds to turbidity and suspended particulate concentration in the tailwater. Chemical precipitation corresponds to the pH value and calcium and magnesium ion concentration of the effluent; The gas release corresponds to the reinjection pressure and the dissolved gas content in the tailwater. Biological blockage corresponds to effluent temperature and organic matter concentration.

8. A real-time blockage risk prediction system for a geothermal tailwater reinjection system, characterized in that, include: The historical tailwater reinjection parameter acquisition module is used to acquire historical tailwater reinjection parameters of the geothermal tailwater reinjection system since the last cleaning time node and generate tailwater reinjection parameter time sequence. The tailwater reinjection twin construction module is used to obtain geological environment parameters, medium characteristic parameters and geothermal tailwater reinjection system design parameters, construct a geothermal tailwater reinjection system model, and obtain a tailwater reinjection twin. The blockage factor distribution status data determination module is used to preset blockage factors and input the tailwater reinjection parameters into the tailwater reinjection twin in sequence to generate blockage factor distribution status data. The real-time blockage risk assessment information determination module is used to acquire real-time tailwater reinjection parameters, perform blockage risk analysis under the transfer of blockage factors based on the blockage factor distribution status data and real-time tailwater reinjection parameters, and generate blockage risk assessment information. The congestion risk warning module is used to issue congestion risk warnings based on the congestion risk assessment information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the real-time blockage risk prediction method for the geothermal tailwater reinjection system according to any one of claims 1-7.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the real-time blockage risk prediction method for the geothermal tailwater reinjection system according to any one of claims 1-7.

Images