A membrane gas meter based on self-healing nanocomposite and system

Abstract

This invention relates to the field of gas metering equipment technology, specifically disclosing a diaphragm gas meter and system based on self-healing nanocomposite materials. Based on the self-healing nanocomposite material diaphragm, it integrates a sensor network, intelligent algorithms, and a control execution unit. First, embedded sensors collect diaphragm strain data in real time, and a crack propagation model is used to predict the evolution trend and safety cycle of microcracks. Then, based on the prediction results and environmental parameters, a reinforcement learning algorithm determines the optimal repair strategy and drives a micro-actuator to perform targeted and quantitative active repair stimulation on the risk area. Finally, the system continuously evaluates the repair efficiency and dynamically generates material and structural optimization suggestions for specific operating environments through machine learning analysis of historical data. This achieves intelligent management of the entire process from damage prediction and intelligent repair to material self-adaptation, significantly improving the metering accuracy, safety, reliability, and service life of the gas meter under harsh operating conditions.

Description

Technical Field

[0001] This invention relates to the field of gas metering equipment technology, specifically a diaphragm gas meter and system based on self-healing nanocomposite materials. Background Technology

[0002] As a commonly used gas metering device, the diaphragm of a diaphragm gas meter's core component can develop cracks and damage over time due to factors such as impurities in the gas, chemical corrosion, and frequent expansion and contraction. Once the diaphragm is damaged, it not only affects the metering accuracy but may also lead to safety hazards such as gas leaks, requiring frequent diaphragm replacements and increasing maintenance costs and manpower.

[0003] For example, application number "CN202411908814.5" discloses a gas overcurrent detection method and a diaphragm gas meter. During gas flow, a first timer with lower timing accuracy measures the pulse time interval between each two adjacent pulses to roughly determine if an overcurrent has occurred. If so, a second timer with higher accuracy is activated to precisely measure subsequent pulse time intervals, accurately determining whether an overcurrent has indeed occurred within each subsequent pulse time interval. When an overcurrent occurs within a preset number of consecutive pulse time intervals during the second timer's measurement process, an overcurrent can be accurately determined. Furthermore, by setting a preset number less than the number of complete pulse time intervals that must pass within the detection time limit when gas flows at the overcurrent threshold, overcurrent detection can be completed within the specified detection time limit, facilitating timely anomaly control. However, in the market... The membrane gas membranes commonly used in gas systems are mostly made of traditional rubber or polymer materials. These materials do not have the ability to self-repair after being damaged, and the problem can only be solved by manual replacement. Although some studies have attempted to improve the membrane materials, there is still no effective solution that can achieve rapid self-repair of micron-level cracks and significantly extend the service life of the membrane. At the same time, the existing technology lacks the ability to monitor and predict the membrane damage process in real time, and cannot provide early warning before the cracks expand to a dangerous size. The repair process depends entirely on the random stress triggered by the damage, and the timing and effect of the repair are uncontrollable. Moreover, the repair efficiency may decrease in complex and variable environments. In addition, the current material formula and structure are fixed once they are made, and they do not have the ability to self-optimize and adjust based on long-term use data and specific environments, making it difficult to cope with diverse and harsh application scenarios. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] This invention provides a membrane gas meter and system based on self-healing nanocomposite materials, which solves the problems mentioned in the background art.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: a diaphragm gas meter based on a self-healing nanocomposite material, comprising a metering module, wherein the metering module is the diaphragm gas meter body, and the core metering component of the metering module is a diaphragm, wherein the diaphragm is made of a self-healing nanocomposite material, the material comprising a polymer matrix and self-healing microcapsules dispersed therein, and further comprising:

[0008] A sensing unit is disposed on the membrane and is used to collect membrane deformation, strain or acoustic emission signals in real time.

[0009] A control and computing unit, connected to the sensing unit, is used to receive and process sensing data;

[0010] A repair execution unit, connected to the control and computing unit, is controlled to apply physical or chemical stimulation to a specific area of ​​the membrane to trigger a response from the self-repairing microcapsule;

[0011] The matrix material is a polyurethane elastomer or silicone rubber polymer with good flexibility and gas barrier properties. The self-healing microcapsule has a spherical structure, and its outer shell is made of a ruptureable polyurea formaldehyde or gelatin-gum arabic polymer. The inside is encapsulated with a repair fluid that can repair membrane cracks. The composition of the repair fluid is selected according to the characteristics of the matrix material. For example, when the matrix material is a polyurethane elastomer, the repair fluid may contain isocyanate prepolymer and polyol. The two can undergo a polymerization reaction after contact with air or moisture to fill and repair the cracks.

[0012] During the use of the membrane, when micron-level cracks appear on the surface, the stress generated at the crack will cause the shell of the nearby self-healing microcapsules to rupture, releasing the repair fluid. The repair fluid quickly fills the crack under surface tension and capillary action, and solidifies through a chemical reaction within 5 minutes, thus completing the repair of the crack and restoring the integrity and function of the membrane.

[0013] Furthermore, the self-healing microcapsules have a particle size of 1-10 micrometers and a mass fraction of 5%-15% in the matrix material. By controlling the particle size and content of the microcapsules, the self-healing effect can be guaranteed without significantly affecting the flexibility and gas barrier properties of the membrane.

[0014] A diaphragm gas meter system based on self-healing nanocomposite materials, comprising:

[0015] The data acquisition module is responsible for driving the sensing unit to acquire data in real time, and performing preliminary filtering, amplification and analog-to-digital conversion on the raw sensing signals to generate a standardized initial dataset.

[0016] The crack intelligent monitoring and prediction module is connected to the data acquisition module, receives the initial dataset, calculates the current propagation rate of the microcracks in the film and predicts their future evolution trend and remaining safety period through the built-in time series analysis model and crack propagation mechanics model, and outputs a crack status report.

[0017] The adaptive repair decision module is connected to the crack intelligent monitoring and prediction module. It receives the crack status report and, in conjunction with real-time environmental parameter data, runs a reinforcement learning algorithm to decide when, where, and with what intensity to initiate active repair, and generates a repair instruction that includes the required amount of repair fluid.

[0018] The precision repair execution module is connected to the adaptive repair decision module, receives the repair command, drives the repair execution unit to apply precisely controlled physical field stimulation to the target area of ​​the membrane, performs targeted and quantitative release of self-repairing microcapsules, and monitors the filling and solidification process of the repair fluid.

[0019] The repair performance evaluation module connects the precision repair execution module and the data acquisition module. By comparing the changes in sensor data before and after repair, it quantifies the efficiency and effectiveness of single and cumulative repair actions and generates a repair performance evaluation report.

[0020] The material and structure optimization analysis module is connected to the repair performance evaluation module, receives the repair performance evaluation report, uses a machine learning model to analyze the correlation between repair performance and material parameters and microcapsule distribution, and simulates and recommends optimized regional material formulations and structural parameters.

[0021] The system interaction and reporting module connects all the aforementioned modules, is responsible for summarizing the data output by each module, generating a visual report and displaying it through a human-machine interface, and receiving external configuration commands to adjust system operating parameters.

[0022] Furthermore, the data acquisition module specifically includes: a distributed fiber optic strain sensor array integrated into the membrane substrate for high spatial resolution sensing of the local strain distribution of the membrane under different gas pressure cycles; a piezoelectric thin-film acoustic emission sensor attached to the membrane surface for capturing weak stress wave signals generated during crack initiation and propagation; a temperature and humidity sensor and a gas flow meter installed inside the gas meter body for collecting the state parameters of the membrane's working environment; and a multi-channel data acquisition card responsible for synchronously acquiring the analog signals from all the above sensors, and executing digital signal processing programs including moving average filtering and wavelet denoising through an embedded processor to eliminate power frequency interference and random noise, and finally packaging the processed timestamp-synchronized strain data, acoustic emission event data, environmental temperature and humidity data, and instantaneous flow data into a structured data packet.

[0023] Furthermore, the intelligent crack monitoring and prediction module specifically includes: a data buffer for receiving and temporarily storing continuous time-series data streams from the data acquisition module; a feature extraction unit, whose algorithm calculates the local strain energy density change rate from strain data and extracts event count rate, amplitude, and rise time features from acoustic emission data; a crack condition diagnosis unit, which embeds a crack propagation dynamics model based on the modified formula of Paris's law. This model takes the extracted features as input and, combined with material fatigue characteristic parameters, calculates the propagation rate of the current dominant crack and the amplitude of the effective stress intensity factor in real time; and a prediction unit, which uses a Kalman filter or a long short-term memory neural network to extrapolate the calculated crack propagation rate over time, combines it with a preset critical crack length, estimates the remaining safe operating period of the membrane under the current working condition, and generates a structured status report containing crack location, current length, propagation rate, predicted risk level, and remaining safe period.

[0024] Furthermore, the adaptive repair decision module specifically includes: an environment fusion unit, used to receive real-time environmental data from the data acquisition module and align it spatiotemporally with the crack status report; a status assessment unit, used to calculate the current system health score based on crack propagation rate, remaining safety period, and environmental corrosion index; a decision core, which adopts a deep Q-network reinforcement learning algorithm, whose state space is defined as health score, crack location information, and historical repair records, and whose action space is defined as repair trigger commands of different intensity levels. The reward function design comprehensively considers repair cost, the improvement of health after repair, and the timeliness of repair actions. Through offline training and online fine-tuning, the decision core outputs the current optimal repair strategy; and an instruction generation unit, which transforms the optimal strategy into specific control commands, including the coordinates of the target repair area, the suggested stimulation method, and the theoretical demand for repair fluid calculated based on the crack size and repair fluid performance model.

[0025] Furthermore, the precise repair execution module specifically includes: an instruction parsing and path planning unit, used to receive repair instructions and decompose them into an executable sequence of steps; a repair execution driver, which connects to and controls an array of micro-actuators integrated on the non-working area of ​​the membrane or adjacent substrate, the micro-actuators including but not limited to micro-resistive heating elements, micro-focused ultrasonic transducers, or micro-current pulse generators, capable of generating local thermal, ultrasonic, or electric field stimulation; a release control unit, which precisely adjusts the stimulation intensity and duration applied to the microcapsules in the target area, so as to soften or rupture the shell of specific types of microcapsules in a controllable manner, thereby achieving the directional release of the repair fluid; and a process monitoring submodule, which monitors the outflow of the repair fluid, the crack filling process under capillary action, and the initiation of the curing reaction in real time by reading sensor feedback data near the target area, ensuring that the repair action is completed as expected.

[0026] Furthermore, the repair performance evaluation module specifically includes: a data comparison unit before and after repair, which triggers the data acquisition module to perform a new round of high-precision scanning on the repair area after the repair command is executed, obtains the strain field and acoustic emission background noise data after repair, and performs differential calculation with the baseline data before repair; a performance calculation unit, which calculates the crack closure rate, stiffness recovery rate of the repair area, and repair reaction completion rate by analyzing the reduction of strain concentration area and the silencing of acoustic emission activity in the differential data, and referring to the solidification reaction monitoring data; a comprehensive evaluation unit, which uses a weighted algorithm to integrate the above indicators into a comprehensive repair performance coefficient and evaluates the contribution of this repair to the extension of the overall remaining safety period; and a database for storing the command, process data, and performance evaluation results of each repair event, forming a historical performance dataset for use by the material and structure optimization analysis module.

[0027] Furthermore, the material and structure optimization analysis module specifically includes: a data warehouse for long-term archiving and storage of historical performance datasets from the repair performance evaluation module, long-term environmental spectrum data from the data acquisition module, and initial material parameters of the membrane; an association analysis engine that uses machine learning methods such as random forest or gradient boosting decision tree to mine the nonlinear relationship between the repair performance coefficient and numerous influencing factors, including key influencing factors such as the local volume fraction of microcapsules in the repair area, microcapsule shell thickness distribution, chemical composition of the repair fluid, and historical average and fluctuation amplitude of environmental temperature and humidity; a simulation optimization unit with a built-in generative adversarial network that uses current material parameters and typical environmental loads as input to generate multiple virtual microcapsule distribution schemes and repair fluid formulation variations, and predicts their repair performance; and a recommendation report generator that compares simulation results, selects optimization schemes that are expected to significantly improve performance under specific environmental or failure modes, and generates an optimization report containing recommendations for adjusting material parameters in specific areas.

[0028] Furthermore, the system interaction and reporting module specifically includes: a data aggregation and formatting unit, responsible for periodically pulling key output data from other modules, including real-time health status, early warning information, recent repair records, comprehensive performance trends, and material optimization suggestions; a visualization engine, which converts aggregated data into dashboards, trend graphs, membrane status heatmaps, and structured report documents; a human-computer interaction interface, implemented as a local LCD touchscreen or a remote web server, used to display visualized information and provide parameter setting windows, allowing users to adjust system sensitivity, early warning thresholds, repair strategy preferences, or confirm the execution of material optimization suggestions; and a log and communication unit, responsible for recording all key system operations and events, and supporting the uploading of reports and early warning information to the cloud platform via wired or wireless communication protocols.

[0029] Furthermore, the physical or chemical stimulus applied by the repair execution unit is at least one of a local thermal field, ultrasound of a specific frequency, a controllable current pulse, or light radiation of a specific wavelength, and the shell material or repair fluid component of the self-healing microcapsule contains a responsive substance sensitive to such stimulus.

[0030] (III) Beneficial Effects

[0031] This invention provides a diaphragm gas meter and system based on self-healing nanocomposite materials. It has the following beneficial effects:

[0032] (I) This diaphragm gas meter and system based on self-healing nanocomposite materials has a service life that can be extended by 3-5 times compared to the diaphragm of traditional diaphragm gas meters. This greatly reduces the frequency of diaphragm replacement, lowers the maintenance cost and manpower input of the gas meter, and at the same time, timely repair of diaphragm cracks can effectively avoid gas leakage and metering errors caused by diaphragm damage, ensuring that the diaphragm gas meter always maintains accurate metering performance, providing a guarantee for accurate gas billing. Moreover, the rapid self-healing function can prevent the diaphragm cracks from expanding further, reducing the risk of gas leakage, improving the safety of gas use, and reducing the occurrence of safety accidents. Furthermore, this diaphragm is particularly suitable for corrosive gas environments and can resist the erosion of chemical substances in gas. It can still maintain good performance and self-healing ability under harsh operating conditions.

[0033] (II) This diaphragm gas meter and system based on self-healing nanocomposite materials utilizes fiber optic strain sensors and acoustic emission sensors integrated into the diaphragm body to continuously collect high spatiotemporal resolution mechanical signals during gas meter operation. Through time-series analysis and modified Paris model in the crack intelligent monitoring and prediction module, the system can calculate and predict the propagation rate and remaining safety period of microcracks in real time. This allows for the identification of potential failure risks several days or even weeks in advance, before cracks cause substantial gas leakage or metering errors, even under harsh operating conditions such as extreme temperature fluctuations. This provides an accurate early warning window for planned maintenance, greatly improving the safety and long-term accuracy of gas meter use. Furthermore, it transforms the traditional emergency repair mode after a failure into a predictive maintenance mode, significantly reducing the risk of gas outages and emergency maintenance costs caused by sudden failures.

[0034] (III) This membrane gas meter and system based on self-healing nanocomposite materials, through an adaptive repair decision module, comprehensively analyzes crack prediction results, environmental parameters and historical repair records, and outputs the optimal repair strategy in real time by a deep Q-network model. Subsequently, the precision repair execution module drives the integrated micro thermal, acoustic or electrical stimulation unit to implement controlled stimulation at specific points, in quantitative and timed manner on high-risk areas of the membrane, thereby intelligently triggering the release of repair fluid from specific microcapsules. This on-demand and targeted repair mechanism overcomes the drawbacks of traditional self-healing materials that blindly rely on damage stress triggering and uneven distribution of repair resources, ensuring the high efficiency and timeliness of repair actions, effectively curbing crack propagation before it accelerates, greatly improving the reliability of self-healing function and the utilization rate of repair resources, reducing unnecessary repair agent consumption, and ensuring the continuous optimal maintenance of membrane structural integrity through intelligent intervention, thereby maintaining a more stable and longer-lasting high-precision metering performance of the gas meter under complex dynamic loads.

[0035] (IV) This membrane gas meter and system based on self-healing nanocomposite materials quantifies the effect of each repair action through a repair performance evaluation module. The material and structure optimization analysis module accumulates and analyzes performance data and environmental spectrum data over a long period of time. It uses machine learning models to explore the deep correlation between repair performance and material microstructure parameters. Then, through generative adversarial network simulation, the system can generate regional material optimization suggestions for specific corrosive environments, temperature cycling patterns, or load history. For example, it can adjust the local density of microcapsules, shell thickness distribution, or repair fluid formulation. This makes the membrane no longer a component with fixed performance after the initial state, but an intelligent material system that can learn and optimize itself according to actual service conditions. This enables the gas meter to be configured with personalized adaptability in diverse and harsh usage scenarios. Through dynamic optimization suggestions of material parameters, it continuously combats material aging and performance degradation, thereby achieving a step-by-step improvement in reliability and service life throughout the entire life cycle of the gas meter. It also provides a data-driven scientific basis for the remanufacturing and performance upgrade of core components. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the diaphragm structure of the diaphragm gas meter of the present invention;

[0037] Figure 2 This is a schematic diagram of the structure of the self-healing microcapsule of the present invention;

[0038] Figure 3 This is a schematic diagram of the self-repair process of cracks on the surface of the film in this invention;

[0039] Figure 4 This is a system flowchart of the present invention. Detailed Implementation

[0040] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] First embodiment: as follows Figures 1 to 4 As shown, the present invention provides a technical solution: a diaphragm gas meter based on self-healing nanocomposite materials, including a metering module, the metering module being the diaphragm gas meter body, the core metering component of the metering module being a diaphragm, the diaphragm being made of a self-healing nanocomposite material, the material including a polymer matrix and self-healing microcapsules dispersed therein, and further including:

[0042] The sensing unit is deployed on the diaphragm and is used to collect diaphragm deformation, strain or acoustic emission signals in real time.

[0043] A control and computing unit, connected to the sensing unit, is used to receive and process sensing data;

[0044] The repair execution unit is connected to the control and computing unit and applies physical or chemical stimulation to specific areas of the membrane in a controlled manner to trigger the response of the self-repairing microcapsules.

[0045] The matrix material is a polyurethane elastomer and silicone rubber polymer material with good flexibility and gas barrier properties. The self-healing microcapsule has a spherical structure, and its shell is made of a ruptureable polyurea formaldehyde and gelatin-gum arabic polymer material. The inside is encapsulated with a repair fluid that can repair membrane cracks.

[0046] The membrane can be prepared using the following two methods:

[0047] 1. Material Preparation: Polyurethane elastomer was selected as the matrix material, and polyurea formaldehyde was used as the shell material for the self-healing microcapsules. The microcapsules were encapsulated internally with a repair solution containing isocyanate prepolymer and polyol. Self-healing microcapsules with a particle size of 3 micrometers were prepared by emulsion polymerization and uniformly dispersed in a polyurethane elastomer solution at a mass fraction of 8%.

[0048] Film preparation: The uniformly mixed solution is poured onto a flat mold surface using a casting process, and then dried and cured under certain temperature and humidity conditions to form a film with a thickness of 0.5 mm.

[0049] Performance Testing: A simulated experiment was conducted on the prepared film, artificially creating micron-sized cracks on the film surface. Observations showed that the repair solution began to fill the cracks within 3 minutes and completely cured within 5 minutes, eliminating the cracks. After long-term aging and corrosive gas simulation tests, the service life of this film was extended by 4 times compared to traditional polyurethane films.

[0050] 2. Material preparation: Silicone rubber was selected as the matrix material, gelatin-gum arabic was used as the shell of the self-healing microcapsules, and the internal repair fluid was a mixture containing silane coupling agent and crosslinking agent to prepare self-healing microcapsules with a particle size of 5 micrometers, so that its mass fraction in silicone rubber reached 12%;

[0051] Film preparation: Using a compression molding process, the mixed materials are placed into a mold and cured under appropriate pressure and temperature to obtain a film;

[0052] Performance testing: The membrane was tested and found that when cracks appeared, the repair fluid released by the self-healing microcapsules could quickly repair the cracks. Moreover, the membrane maintained good performance in a highly corrosive gas environment, and its service life was extended by about 3.5 times.

[0053] A diaphragm gas meter system based on self-healing nanocomposite materials, comprising:

[0054] The data acquisition module is responsible for driving the sensing unit to acquire data in real time, and performing preliminary filtering, amplification and analog-to-digital conversion on the raw sensing signals to generate a standardized initial dataset.

[0055] The crack intelligent monitoring and prediction module connects to the data acquisition module, receives the initial dataset, calculates the current propagation rate of the microcracks in the film and predicts their future evolution trend and remaining safety period through the built-in time series analysis model and crack propagation mechanics model, and outputs a crack status report.

[0056] The adaptive repair decision module connects to the crack intelligent monitoring and prediction module, receives crack status reports, and combines them with real-time environmental parameter data to run a reinforcement learning algorithm to decide when, where, and with what intensity to initiate active repair, and generates repair instructions that include the required amount of repair fluid.

[0057] The precision repair execution module connects to the adaptive repair decision module, receives repair instructions, drives the repair execution unit to apply precisely controlled physical field stimulation to the target area of ​​the membrane, performs targeted and quantitative release of self-repairing microcapsules, and monitors the filling and solidification process of the repair fluid.

[0058] The repair performance evaluation module connects the precision repair execution module and the data acquisition module. By comparing the changes in sensor data before and after repair, it quantifies the efficiency and effectiveness of single and cumulative repair actions and generates a repair performance evaluation report.

[0059] The materials and structure optimization analysis module connects to the repair performance evaluation module, receives the repair performance evaluation report, uses machine learning models to analyze the correlation between repair performance and material parameters and microcapsule distribution, and simulates and recommends optimized regional material formulations and structural parameters.

[0060] The system interaction and reporting module connects all the aforementioned modules, is responsible for summarizing the data output by each module, generating a visual report and displaying it through a human-machine interface, and receiving external configuration commands to adjust system operating parameters.

[0061] The data acquisition module specifically includes: a distributed fiber optic strain sensor array integrated into the membrane substrate for high spatial resolution sensing of the local strain distribution of the membrane under different gas pressure cycles; a piezoelectric thin-film acoustic emission sensor attached to the membrane surface for capturing weak stress wave signals generated during crack initiation and propagation; a temperature and humidity sensor and a gas flow meter installed inside the gas meter body for collecting the state parameters of the membrane's working environment; and a multi-channel data acquisition card responsible for synchronously acquiring the analog signals from all the above sensors, and executing digital signal processing programs including moving average filtering and wavelet denoising through an embedded processor to eliminate power frequency interference and random noise, and finally packaging the processed timestamp-synchronized strain data, acoustic emission event data, environmental temperature and humidity data, and instantaneous flow data into a structured data package.

[0062] The intelligent crack monitoring and prediction module specifically includes: a data buffer for receiving and temporarily storing continuous time-series data streams from the data acquisition module; a feature extraction unit, whose algorithm calculates the local strain energy density change rate from strain data and extracts event count rate, amplitude, and rise time features from acoustic emission data; a crack condition diagnosis unit, which embeds a crack propagation dynamics model based on a modified formula of Paris's law. This model uses the extracted features as input and combines them with material fatigue characteristic parameters to calculate the propagation rate of the current dominant crack and the amplitude of the effective stress intensity factor in real time; and a prediction unit, which uses a Kalman filter or a long short-term memory neural network to extrapolate the calculated crack propagation rate over time, combines it with a preset critical crack length, estimates the remaining safe operating period of the membrane under the current operating conditions, and generates a structured status report containing crack location, current length, propagation rate, predicted risk level, and remaining safe operating period.

[0063] First, a distributed fiber optic strain sensor array is embedded in a grid pattern within a self-healing membrane composed of a polyurethane elastomer matrix and polyurea-formaldehyde microcapsules dispersed therein, and a piezoelectric acoustic emission sensor is attached to the surface. Then, when the gas meter operates in a cyclic thermal shock environment simulating a day-night temperature difference of 40°C, the data acquisition module drives a multi-channel acquisition card to synchronously acquire the strain timing signals and acoustic emission events of each sensor. Next, the crack intelligent monitoring and prediction module extracts features from the filtered signals and calculates the current main microcrack propagation rate of 0.2 micrometers per minute using a model based on the modified Paris law. It then uses an LSTM neural network to predict that the crack will reach the critical length in 15 days without intervention. Finally, a status report containing the crack location, propagation rate, and remaining safety period is output. Its beneficial effect is that it can predict the failure risk of the membrane under extreme temperature differences in advance, providing an accurate time window for proactive maintenance, thereby avoiding metering inaccuracies and leaks caused by sudden crack propagation.

[0064] Second embodiment: as follows Figures 1 to 4 As shown, the adaptive repair decision module specifically includes: an environment fusion unit, used to receive real-time environmental data from the data acquisition module and align it spatiotemporally with the crack status report; a status assessment unit, which calculates the current system health score based on crack propagation rate, remaining safety period, and environmental corrosion index; a decision core, which adopts a deep Q-network reinforcement learning algorithm, whose state space is defined as health score, crack location information, and historical repair records, and whose action space is defined as repair trigger commands of different intensity levels. The reward function design comprehensively considers repair cost, the improvement of health after repair, and the timeliness of repair actions. Through offline training and online fine-tuning, the decision core outputs the current optimal repair strategy; and an instruction generation unit, which transforms the optimal strategy into specific control commands, including the coordinates of the target repair area, the suggested stimulus method, and the theoretical demand for repair fluid calculated based on the crack size and repair fluid performance model.

[0065] The precision repair execution module specifically includes: an instruction parsing and path planning unit, used to receive repair instructions and decompose them into an executable sequence of steps; a repair execution driver, which connects to and controls an array of micro-actuators integrated on the non-working area of ​​the membrane or adjacent substrate. These micro-actuators include, but are not limited to, micro-resistive heating elements, micro-focused ultrasonic transducers, or micro-current pulse generators, capable of generating localized thermal, ultrasonic, or electric field stimulation; a release control unit, precisely adjusting the stimulation intensity and duration applied to the microcapsules in the target area to controllably soften or rupture the shells of specific types of microcapsules, achieving directional release of the repair fluid; and a process monitoring submodule, which reads sensor feedback data near the target area to monitor in real time the outflow of the repair fluid, the crack filling process under capillary action, and the initiation of the curing reaction, ensuring that the repair action is completed as expected. Specifically:

[0066] First, the adaptive repair decision module receives the crack status report output by Example 1 and the real-time collected gas humidity data. Then, its embedded deep Q-network reinforcement learning model evaluates the current health score and decides to initiate active repair for crack areas with high predicted risk. It generates control instructions including target coordinates, thermal stimulation using micro-resistive heating elements, and the required amount of repair fluid. Then, the precise repair execution module parses the instructions and drives the corresponding micro-heating element array integrated on the membrane backing plate to apply a local precise heat field of 60°C to the target area for 10 seconds, causing the temperature-sensitive microcapsule shell in the area to soften and rupture, releasing the repair fluid. Finally, by monitoring the drop in strain value in the area, it confirms that the repair fluid has been filled and solidified. Its beneficial effect is that it realizes the leap from passive repair to active intervention. By triggering repair on demand and at fixed points, it significantly improves the utilization efficiency of repair resources and the timeliness of repair actions, effectively curbing crack propagation before it accelerates.

[0067] Third embodiment: as follows Figures 1 to 4 As shown, the repair performance evaluation module specifically includes: a data comparison unit before and after repair, which triggers the data acquisition module to perform a new round of high-precision scanning on the repair area after the repair command is executed, obtains the strain field and acoustic emission background noise data after repair, and performs differential calculation with the baseline data before repair; a performance calculation unit, which calculates the crack closure rate, stiffness recovery rate of the repair area, and repair reaction completion rate by analyzing the reduction of strain concentration area and the silencing of acoustic emission activity in the differential data, and referring to the solidification reaction monitoring data; a comprehensive evaluation unit, which uses a weighted algorithm to integrate the above indicators into a comprehensive repair performance coefficient and evaluates the contribution of this repair to the extension of the overall remaining safety period; and a database, which stores the command, process data, and performance evaluation results of each repair event to form a historical performance dataset for use by the material and structure optimization analysis module.

[0068] The materials and structure optimization analysis module specifically includes: a data warehouse for long-term archiving and storage of historical performance datasets from the repair performance evaluation module, long-term environmental spectrum data from the data acquisition module, and initial material parameters of the membrane; an association analysis engine that uses machine learning methods such as random forests or gradient boosting decision trees to mine the nonlinear relationship between the repair performance coefficient and numerous influencing factors, including key influencing factors such as the local volume fraction of microcapsules in the repair area, microcapsule shell thickness distribution, chemical composition of the repair fluid, and historical average and fluctuation amplitude of environmental temperature and humidity; a simulation optimization unit with a built-in generative adversarial network that uses current material parameters and typical environmental loads as input to generate various virtual microcapsule distribution schemes and repair fluid formulation variations, and predicts their repair performance; and a recommendation report generator that compares simulation results, selects optimization schemes that are expected to significantly improve performance under specific environmental or failure modes, and generates an optimization report containing recommendations for adjusting material parameters in specific areas.

[0069] The system interaction and reporting module specifically includes: a data aggregation and formatting unit, responsible for periodically pulling key output data from other modules, including real-time health status, early warning information, recent repair records, overall performance trends, and material optimization suggestions; a visualization engine, which converts aggregated data into dashboards, trend graphs, membrane status heatmaps, and structured report documents; a human-computer interaction interface, implemented as a local LCD touchscreen or a remote web server, used to display visualized information and provide parameter setting windows, allowing users to adjust system sensitivity, early warning thresholds, repair strategy preferences, or confirm the execution of material optimization suggestions; and a log and communication unit, responsible for recording all key system operations and events, and supporting the uploading of reports and early warning information to the cloud platform via wired or wireless communication protocols.

[0070] The physical or chemical stimulus applied by the repair actuator is at least one of a local thermal field, ultrasound of a specific frequency, a controllable current pulse, or light radiation of a specific wavelength, and the shell material or repair fluid composition of the self-repairing microcapsule contains a responsive substance sensitive to such stimulus, specifically:

[0071] First, after the active repair in Example 2, the repair efficiency assessment module compared the strain energy density of the crack area before and after repair, which decreased by 85%, and calculated the repair efficiency coefficient to be 0.88. Next, the material and structure optimization analysis module called up a one-year historical repair efficiency dataset and corresponding environmental corrosion data. Through random forest model analysis, it was found that the repair efficiency of the current microcapsule distribution would decrease in a low-temperature and high-humidity environment. Then, its built-in generative adversarial network simulated a virtual scheme that would increase the local volume fraction of microcapsules in the target area from 8% to 12% and optimize the shell thickness distribution. It predicted that the repair efficiency coefficient could be stabilized above 0.90 under similar conditions. Finally, the system interaction and reporting module generated a visual report on the regional material parameter adjustment suggestions and pushed it to the maintenance personnel. Its beneficial effect is that it enables the membrane system to have continuous self-learning and optimization capabilities, and can evolve the optimal material configuration for specific usage environments, thereby improving the overall reliability and lifespan of the entire gas meter under complex and variable operating conditions in the long run.

[0072] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0073] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A diaphragm gas meter based on self-healing nanocomposite materials, comprising a metering module, characterized in that: The metering module is a diaphragm gas meter body. The core metering component of the metering module is a diaphragm, which is made of a self-healing nanocomposite material. This material includes a polymer matrix and self-healing microcapsules dispersed therein, and also includes: A sensing unit is disposed on the membrane and is used to collect membrane deformation, strain or acoustic emission signals in real time. A control and computing unit, connected to the sensing unit, is used to receive and process sensing data; A repair execution unit, connected to the control and computing unit, is controlled to apply physical or chemical stimulation to a specific area of ​​the membrane to trigger a response from the self-repairing microcapsule; The matrix material is a polyurethane elastomer or silicone rubber polymer material with good flexibility and gas barrier properties. The self-healing microcapsule has a spherical structure, and its outer shell is made of a ruptureable polyurea formaldehyde or gelatin-gum arabic polymer material. The inside is encapsulated with a repair fluid that can repair membrane cracks.

2. A diaphragm gas meter system based on self-healing nanocomposite materials, using the diaphragm gas meter of claim 1, characterized in that: include: The data acquisition module is responsible for driving the sensing unit to acquire data in real time, and performing preliminary filtering, amplification and analog-to-digital conversion on the raw sensing signals to generate a standardized initial dataset. The crack intelligent monitoring and prediction module is connected to the data acquisition module, receives the initial dataset, calculates the current propagation rate of the microcracks in the film and predicts their future evolution trend and remaining safety period through the built-in time series analysis model and crack propagation mechanics model, and outputs a crack status report. The adaptive repair decision module is connected to the crack intelligent monitoring and prediction module. It receives the crack status report and, in conjunction with real-time environmental parameter data, runs a reinforcement learning algorithm to decide when, where, and with what intensity to initiate active repair, and generates a repair instruction that includes the required amount of repair fluid. The precision repair execution module is connected to the adaptive repair decision module, receives the repair command, drives the repair execution unit to apply precisely controlled physical field stimulation to the target area of ​​the membrane, performs targeted and quantitative release of self-repairing microcapsules, and monitors the filling and solidification process of the repair fluid. The repair performance evaluation module connects the precision repair execution module and the data acquisition module. By comparing the changes in sensor data before and after repair, it quantifies the efficiency and effectiveness of single and cumulative repair actions and generates a repair performance evaluation report. The material and structure optimization analysis module is connected to the repair performance evaluation module, receives the repair performance evaluation report, uses a machine learning model to analyze the correlation between repair performance and material parameters and microcapsule distribution, and simulates and recommends optimized regional material formulations and structural parameters. The system interaction and reporting module connects all the aforementioned modules, is responsible for summarizing the data output by each module, generating a visual report and displaying it through a human-machine interface, and receiving external configuration commands to adjust system operating parameters.

3. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 2, characterized in that: The data acquisition module specifically includes: a distributed fiber optic strain sensor array integrated into the membrane substrate for high spatial resolution sensing of the local strain distribution of the membrane under different gas pressure cycles; a piezoelectric thin-film acoustic emission sensor attached to the membrane surface for capturing weak stress wave signals generated during crack initiation and propagation; a temperature and humidity sensor and a gas flow meter installed inside the gas meter body for collecting the state parameters of the membrane's working environment; and a multi-channel data acquisition card responsible for synchronously acquiring the analog signals from all the above sensors, and executing digital signal processing programs including moving average filtering and wavelet denoising through an embedded processor to eliminate power frequency interference and random noise, and finally packaging the processed timestamp-synchronized strain data, acoustic emission event data, environmental temperature and humidity data, and instantaneous flow data into a structured data packet.

4. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 3, characterized in that: The intelligent crack monitoring and prediction module specifically includes: a data buffer for receiving and temporarily storing continuous time-series data streams from the data acquisition module; a feature extraction unit, whose algorithm calculates the local strain energy density change rate from strain data and extracts event count rate, amplitude, and rise time features from acoustic emission data; a crack condition diagnosis unit, which embeds a crack propagation dynamics model based on a modified formula of Paris's law. This model takes the extracted features as input and, combined with material fatigue characteristic parameters, calculates the propagation rate of the current dominant crack and the amplitude of the effective stress intensity factor in real time; and a prediction unit, which uses a Kalman filter or a long short-term memory neural network to extrapolate the calculated crack propagation rate over time, and, combined with a preset critical crack length, estimates the remaining safe operating period of the membrane under the current operating conditions, and generates a structured status report containing crack location, current length, propagation rate, predicted risk level, and remaining safe operating period.

5. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 4, characterized in that: The adaptive repair decision-making module specifically includes: an environment fusion unit, used to receive real-time environmental data from the data acquisition module and align it spatiotemporally with the crack status report; a status assessment unit, used to calculate the current system health score based on crack propagation rate, remaining safety period, and environmental corrosion index; a decision core, which adopts a deep Q-network reinforcement learning algorithm, whose state space is defined as health score, crack location information, and historical repair records, and whose action space is defined as repair trigger commands of different intensity levels. The reward function design comprehensively considers repair cost, the improvement of health after repair, and the timeliness of repair actions. Through offline training and online fine-tuning, the decision core outputs the current optimal repair strategy; and an instruction generation unit, which transforms the optimal strategy into specific control commands, including the coordinates of the target repair area, the suggested stimulation method, and the theoretical demand for repair fluid calculated based on the crack size and repair fluid performance model.

6. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 5, characterized in that: The precise repair execution module specifically includes: an instruction parsing and path planning unit, used to receive repair instructions and decompose them into an executable sequence of steps; a repair execution driver, which connects to and controls an array of micro-actuators integrated on the non-working area of ​​the membrane or adjacent substrate, including but not limited to micro-resistive heating elements, micro-focused ultrasonic transducers, or micro-current pulse generators, capable of generating local thermal, ultrasonic, or electric field stimulation; a release control unit, which precisely adjusts the stimulation intensity and duration applied to the microcapsules in the target area to soften or rupture the shells of specific types of microcapsules in a controllable manner, achieving directional release of the repair fluid; and a process monitoring submodule, which reads sensor feedback data near the target area to monitor the outflow of the repair fluid, the crack filling process under capillary action, and the initiation of the curing reaction in real time, ensuring that the repair action is completed as expected.

7. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 6, characterized in that: The repair performance evaluation module specifically includes: a data comparison unit before and after repair, which triggers a data acquisition module to perform a new round of high-precision scanning on the repair area after the repair command is executed, obtains the strain field and acoustic emission background noise data after repair, and performs differential calculation with the baseline data before repair; a performance calculation unit, which calculates the crack closure rate, stiffness recovery rate of the repair area, and repair reaction completion rate by analyzing the reduction of strain concentration area and the silencing of acoustic emission activity in the differential data, and referring to the solidification reaction monitoring data; a comprehensive evaluation unit, which uses a weighted algorithm to integrate the above indicators into a comprehensive repair performance coefficient and evaluates the contribution of this repair to the extension of the overall remaining safety period; and a database for storing the command, process data, and performance evaluation results of each repair event, forming a historical performance dataset for use by the material and structure optimization analysis module.

8. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 7, characterized in that: The material and structure optimization analysis module specifically includes: a data warehouse for long-term archiving and storage of historical performance datasets from the repair performance evaluation module, long-term environmental spectrum data from the data acquisition module, and initial material parameters of the membrane; an association analysis engine that uses machine learning methods such as random forests or gradient boosting decision trees to mine the nonlinear relationship between the repair performance coefficient and numerous influencing factors, including key influencing factors such as the local volume fraction of microcapsules in the repair area, microcapsule shell thickness distribution, chemical composition of the repair fluid, and historical average and fluctuation amplitude of environmental temperature and humidity; a simulation optimization unit with a built-in generative adversarial network that uses current material parameters and typical environmental loads as input to generate multiple virtual microcapsule distribution schemes and repair fluid formulation variations, and predicts their repair performance; and a recommendation report generator that compares simulation results, selects optimization schemes that are expected to significantly improve performance under specific environmental or failure modes, and generates an optimization report containing recommendations for adjusting material parameters in specific areas.

9. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 8, characterized in that: The system interaction and reporting module specifically includes: a data aggregation and formatting unit, responsible for periodically pulling key output data from other modules, including real-time health status, early warning information, recent repair records, overall performance trends, and material optimization suggestions; a visualization engine, which converts aggregated data into dashboards, trend graphs, membrane status heatmaps, and structured report documents; a human-computer interaction interface, implemented as a local LCD touchscreen or a remote web server, used to display visualized information and provide parameter setting windows, allowing users to adjust system sensitivity, early warning thresholds, repair strategy preferences, or confirm the execution of material optimization suggestions; and a log and communication unit, responsible for recording all key system operations and events, and supporting the uploading of reports and early warning information to the cloud platform via wired or wireless communication protocols.

10. A diaphragm gas meter system based on self-healing nanocomposite materials according to claim 9, characterized in that: The physical or chemical stimulus applied by the repair execution unit is at least one of a local thermal field, ultrasound of a specific frequency, a controllable current pulse, or light radiation of a specific wavelength, and the shell material or repair fluid component of the self-healing microcapsule contains a responsive substance sensitive to such stimulus.

Images