Intelligent remote monitoring system and method based on stop valve

Abstract

The application relates to the technical field of stop valve monitoring, and discloses an intelligent remote monitoring system and method based on a stop valve. The system comprises a parameter acquisition module, a valve core matching module, a condition initialization module, a flow rate matching optimization module, a control optimization module and a remote monitoring module. The parameter acquisition module acquires fluid characteristic parameters and operation requirement parameters of the stop valve; the valve core matching module matches valve core types in combination with fluid component information and operation precision constraints and configures a target valve core; the condition initialization module initializes operation conditions according to structure characteristics of the target valve core and selection results, and outputs initial control constraints containing a pressure difference threshold value and a maximum allowable wear rate; the flow rate matching optimization module optimizes flow control speed in combination with flow volume and time constraints under the condition of the initial constraints; the control optimization module analyzes valve core wear trends and optimizes control parameters; and the remote monitoring module intermittently and remotely monitors the target valve core in operation by using the control parameters.

Description

Technical Field

[0001] This invention relates to the field of gate valve monitoring technology, specifically to an intelligent remote monitoring system and method based on gate valves. Background Technology

[0002] In industrial production processes, gate valves, as key control components in fluid transport systems, are widely used in petroleum, chemical, and water treatment industries. Their operating status directly affects the safety, stability, and economy of the entire production system. Currently, traditional gate valve control and monitoring methods generally have significant limitations and cannot meet the demands of modern industry for high-precision and intelligent operation.

[0003] Traditional gate valve operation relies heavily on manual experience for valve core selection and parameter setting, failing to adequately consider the impact of fluid composition differences on valve core compatibility. For example, when conveying corrosive fluids, using a standard metal valve core can lead to rapid corrosion due to material incompatibility with the fluid composition, shortening the valve core's lifespan and potentially causing safety hazards such as fluid leakage. Conversely, when conveying high-viscosity fluids, an improperly designed valve core structure can increase fluid flow resistance and reduce flow control accuracy. Furthermore, during manual selection, operational time constraints and operational accuracy constraints are often considered separately, failing to achieve synergistic optimization. This results in gate valves either sacrificing control accuracy to meet time requirements or extending operating cycles to maintain accuracy, impacting production efficiency.

[0004] In the monitoring of gate valve operation, traditional methods often employ periodic on-site inspections or local monitoring of single parameters. This approach suffers from untimely monitoring and incomplete data. On-site inspections are spaced far apart, making it difficult to capture sudden conditions such as valve core wear and pressure fluctuations in real time. By the time a fault is discovered, it often indicates a production interruption or equipment damage. Local monitoring typically focuses only on a few key parameters such as flow rate and pressure, lacking dynamic analysis and prediction of valve core wear trends. This prevents proactive maintenance measures, forcing repairs and replacements only after a fault occurs, increasing equipment maintenance costs and production risks. Furthermore, with the expansion of industrial production scale and the improvement of automation levels, traditional local monitoring methods struggle to achieve centralized management and remote control of multiple gate valves, hindering the intelligent scheduling and collaborative operation of the overall production system.

[0005] The limitations of traditional gate valve control and monitoring methods have become a major factor restricting the improvement of production system efficiency and safety. There is an urgent need for an integrated system that can achieve precise valve core selection, operating parameter optimization, and remote intelligent monitoring to solve the problems existing in the current technology. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent remote monitoring system and method based on a shut-off valve to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides an intelligent remote monitoring system based on a shut-off valve, the system comprising:

[0008] The parameter acquisition module is used to acquire the fluid characteristic parameters and operation requirement parameters of the shut-off valve. The fluid characteristic parameters include fluid composition information and flow rate / volume information, and the operation requirement parameters include operation time constraints and operation accuracy constraints.

[0009] The valve core matching module is used to match valve core types based on the fluid composition information and operating accuracy constraints. After obtaining the valve core selection, the valve core of the shut-off valve is configured based on the valve core selection to obtain the target valve core.

[0010] The condition initialization module is used to obtain the structural features of the target valve core, and initialize the operating conditions according to the structural features and valve core selection, and output the initial operating control constraints, wherein the initial operating control constraints include the pressure difference threshold and the maximum allowable wear rate;

[0011] The flow rate optimization module is used to take the initial operation control constraints as optimization conditions, perform flow rate optimization based on the flow volume information and operation time constraints, and determine the flow control speed.

[0012] The control optimization module is used to perform wear trend analysis of the target valve core based on the flow control speed, and to perform control optimization based on the wear trend analysis results to obtain control parameters;

[0013] The remote monitoring module is used to intermittently remotely monitor the target valve core using the control parameters during the operation of the shut-off valve at the flow control speed.

[0014] Preferably, the valve core matching module includes:

[0015] Based on the fluid type, the fluid component information is decomposed to obtain multiple fluid particle size distribution characteristics for various fluid types;

[0016] The multiple fluid particle size distribution features are updated according to the operational accuracy constraints to obtain multiple operational particle size distribution features;

[0017] By fusing the multiple operational particle size distribution features, a global particle size distribution feature is obtained;

[0018] Based on the global particle size distribution characteristics, multi-level valve core matching is performed to obtain a multi-layer valve core element, wherein the multi-layer valve core element has multiple operating dimensions, valve layer thickness and valve layer material identification.

[0019] Based on the multiple operating dimensions, the multi-layer valve core components are sequentially assembled to obtain the valve core selection.

[0020] Preferably, the condition initialization module includes:

[0021] Obtain material information for multiple components of the multilayer valve core element;

[0022] The effective operating area is obtained by calculating the operating area based on the structural features described above.

[0023] Based on the material information of the multiple components and the effective operating area, network data is called to obtain multiple operation simulation models. Then, the valve core model is obtained by fusing the multiple operation simulation models.

[0024] After obtaining the standard flow rate by calling local data, the fluid dynamics simulation of the valve core model is performed using the standard flow rate, and the structural pressure difference threshold is output. The fluid dynamics simulation of the multiple operation simulation models is also performed using the standard flow rate, and multiple individual pressure difference thresholds are output.

[0025] The pressure difference threshold is obtained by solving the intersection of the structural pressure difference threshold and multiple individual pressure difference thresholds.

[0026] The maximum value of the pressure difference threshold is used as a constraint to perform a fluid dynamics simulation of the valve core model, and the maximum allowable wear rate is obtained by calling the valve core model based on the simulation results.

[0027] Preferably, the flow rate optimization module includes:

[0028] Pre-built traffic correlation prediction model;

[0029] The reciprocal of the standard flow rate is used as the speed change scale to update the standard flow rate data, thereby obtaining the first backup flow rate.

[0030] Input the first backup flow rate into the flow correlation prediction model to obtain the first wear rate time limit, the first pressure difference time limit, and the first prediction operation efficiency.

[0031] The first predicted operation time is calculated using the flow volume information and the first predicted operation efficiency. If the first predicted operation time meets the operation time constraint, the first backup flow rate is retained, and the first difference between the first wear rate time limit and the first pressure difference time limit is used as the first control reliability coefficient of the first backup flow rate.

[0032] Similarly, by continuously updating the standard flow rate and evaluating the update results using the flow correlation prediction model and operation time constraints, multiple backup flow rates and multiple control reliability coefficients that meet the preset update frequency are obtained.

[0033] The plurality of control reliability coefficients are serialized, and the flow control speed is located from the plurality of backup flow speeds based on the sorting results.

[0034] Preferably, the traffic grading optimization module pre-builds a traffic correlation prediction model, including:

[0035] Using the valve core selection and fluid composition information as constraints for historical data retrieval, historical flow correlation data is obtained, wherein the historical flow correlation data includes multiple sample flow rates, multiple sample operating efficiencies, multiple wear rate change rates, and multiple pressure difference change rates;

[0036] Multiple pressure difference change time limits are calculated based on the pressure difference threshold and multiple pressure difference change rates;

[0037] Multiple wear rate change time limits are calculated based on the maximum permissible wear rate and multiple wear rate change rates;

[0038] The multiple sample flow rates, multiple sample operating efficiencies, multiple pressure difference time limits, and multiple wear rate change time limits are used as training parameters to train the flow correlation prediction model based on a neural network.

[0039] Preferably, the control optimization module includes:

[0040] The directional data of the flow control speed is retrieved to obtain the control wear rate time limit and the control pressure difference time limit;

[0041] The monitoring cycle is determined by comparing the time limit for controlling the wear rate and the time limit for controlling the pressure difference, and selecting the smaller value.

[0042] A fitness weight configuration is predefined, and a fitness evaluation function is constructed based on the fitness weight configuration, wherein the fitness weight configuration includes energy consumption weight, lifetime loss weight, and fluid consumption weight;

[0043] The initial operation control constraint is used as a data call constraint to filter and call historical control records to obtain multiple sample control records. Each sample control record includes sample control parameters and sample monitoring consumption.

[0044] The fitness evaluation function is used to evaluate the fitness of the multiple sample control records, and the initial control parameters are located based on the evaluation results. The initial control parameters consist of the monitored water flow rate, the monitored water duration, the monitored air flow rate, the monitored air duration, and the comprehensive monitoring duration.

[0045] The monitoring period is added to the initial control parameters to obtain the control parameters.

[0046] Preferably, the sample control parameters consist of sample water flow rate, sample water duration, sample gas flow rate, sample gas duration, and overall sample duration.

[0047] Preferably, the sample monitoring consumption consists of sample monitoring energy consumption, sample component lifespan loss, and sample water consumption.

[0048] Preferably, it also includes a safety early warning module, used to execute a safety early warning based on the wear trend analysis results, obtain early warning parameters, and transmit the early warning parameters to a remote monitoring center.

[0049] Preferably, the present invention also includes an intelligent remote monitoring method based on a gate valve, the method comprising all the modules and method flow of the above-mentioned intelligent remote monitoring system based on a gate valve.

[0050] Compared with the prior art, the beneficial effects of the present invention are:

[0051] In the valve core selection process, the system's valve core matching module combines fluid composition information with operational precision constraints to match valve core types, achieving precise valve core selection. This module no longer relies on manual experience but selects valve core types with suitable materials and structures based on the specific characteristics of the fluid components, such as corrosivity, viscosity, and purity, ensuring a high degree of matching between the valve core and fluid characteristics. For example, for corrosive fluids, it can automatically match corrosion-resistant ceramic or alloy valve cores, reducing corrosion damage caused by unsuitable materials; for high-viscosity fluids, it can select valve cores with more reasonable flow channel designs, reducing fluid flow resistance and ensuring the achievement of operational precision constraints. Furthermore, after valve core selection, the system configures the valve core based on the selection results, forming a more adaptable target valve core. This avoids various problems caused by mismatches between the valve core and actual operating requirements, extends valve core lifespan, and reduces safety hazards caused by valve core failures.

[0052] The condition initialization module initializes operating conditions by acquiring the structural characteristics of the target valve core and combining this with the valve core selection results. It outputs initial operating control constraints including pressure difference thresholds and maximum permissible wear rates, providing a scientific basis for subsequent flow rate optimization of the gate valve. This module fully considers the structural characteristics of the target valve core, such as the valve core sealing surface type and opening / closing stroke, combining these structural characteristics with the fluid characteristics and accuracy requirements corresponding to the valve core selection. It formulates initial constraints that conform to the actual operating capabilities of the valve core, preventing the valve core from operating beyond its safe range due to unreasonable initial parameter settings, and ensuring that the gate valve starts and operates within a safe and stable range.

[0053] The flow rate optimization module uses initial operational control constraints as optimization conditions, combining flow volume information and operational time constraints to perform flow rate optimization, achieving precise positioning of the flow control speed. This module, while meeting the pressure difference threshold and maximum allowable wear rate, balances flow volume requirements and operational time requirements, finding the optimal flow control speed through algorithmic optimization. For example, when rapidly conveying a certain volume of fluid, the flow control speed can be appropriately increased within the range of the pressure difference threshold and maximum allowable wear rate to meet operational time constraints. In scenarios requiring high flow control accuracy, the speed can be finely adjusted to ensure that the flow control accuracy meets the requirements. This optimization method achieves synergy between operational time and operational accuracy, avoiding the efficiency and accuracy imbalance caused by the separation of these two aspects in traditional methods, thus improving the operating efficiency and control quality of the shut-off valve.

[0054] The control optimization module analyzes the wear trend of the target valve core based on the flow control speed and performs control optimization based on the analysis results to obtain suitable control parameters. This module can collect relevant data during valve core operation in real time and dynamically analyze the wear state and development trend of the valve core in conjunction with the flow control speed. By predicting the wear trend, control parameters are adjusted, such as appropriately reducing the valve core opening and closing speed to slow down the wear rate, or optimizing the fluid flow path to reduce localized wear of the valve core. This slows down the valve core wear process while ensuring the normal operation of the shut-off valve, reducing the risk of failure due to valve core wear and lowering equipment maintenance frequency and costs.

[0055] The remote monitoring module uses control parameters to intermittently monitor the target valve core during the flow-controlled speed operation of the gate valve, overcoming the limitations of traditional on-site inspections and local monitoring. This module enables real-time remote monitoring of the gate valve's operating status. Personnel can obtain key data such as valve core wear, pressure, and flow rate through a remote terminal without needing to go to the site, allowing for timely understanding of the gate valve's operation. The intermittent monitoring mode reduces data transmission volume and energy consumption while ensuring monitoring effectiveness, achieving highly efficient and energy-saving monitoring operations. Furthermore, the remote monitoring function supports centralized management of multiple gate valves. Personnel can uniformly monitor and control multiple devices through a remote platform, facilitating coordinated scheduling of the overall production system. When an anomaly is detected, control commands can be remotely issued to adjust the gate valve's operating parameters in a timely manner, preventing the escalation of the fault and ensuring the continuous and stable operation of the production system. Attached Figure Description

[0056] Figure 1 This is a schematic diagram illustrating the working principle of the intelligent remote monitoring system based on a shut-off valve as described in this invention.

[0057] Figure 2 This is a flowchart illustrating the working principle of the valve core matching module.

[0058] Figure 3 Flowchart illustrating the working principle of the condition initialization module;

[0059] Figure 4 A flowchart illustrating the working principle of the traffic grading optimization module. Detailed Implementation

[0060] 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.

[0061] Please see Figure 1 This invention provides an intelligent remote monitoring system and method based on a shut-off valve. The system includes a parameter acquisition module, a valve core matching module, a condition initialization module, a flow rate optimization module, a control optimization module, and a remote monitoring module operating in coordination. Specific implementation details are as follows:

[0062] The parameter acquisition module is responsible for collecting the fluid characteristic parameters and operational requirement parameters of the shut-off valve. The fluid characteristic parameters include fluid composition information and flow rate / volume information, while the operational requirement parameters include operational time constraints and operational accuracy constraints. The valve core matching module completes valve core type matching based on the fluid composition information and operational accuracy constraints, determining the target valve core through valve core selection. The condition initialization module acquires the structural characteristics of the target valve core, combines these characteristics with the valve core selection to initialize operational conditions, generating initial operational control constraints including pressure difference thresholds and maximum allowable wear rates. The flow rate optimization module uses the initial operational control constraints as optimization conditions, performing flow rate optimization based on flow rate / volume information and operational time constraints to determine the flow control speed. The control optimization module analyzes the wear trend of the target valve core based on the flow control speed and performs control optimization to obtain control parameters. The remote monitoring module uses the control parameters to intermittently remotely monitor the target valve core during the shut-off valve's operation at the flow control speed.

[0063] Example 1: See Figure 2 The implementation of the valve core matching module begins with in-depth analysis of fluid composition information, typically derived from real-time data streams from upstream process control systems or online analyzers. The module's built-in analytical algorithm identifies different chemical substances or mixture types present in the fluid and automatically classifies them based on their physicochemical properties, such as distinguishing between liquid water, oil, gas, or slurries containing solid particles. Each identified fluid type triggers an independent analysis thread to extract its corresponding particle size distribution characteristics. The acquisition of these fluid particle size distribution characteristics relies on a particle size analysis database or real-time sensor network connected to the module. The database contains a large number of pre-stored typical particle size distribution patterns for various fluids, while real-time sensors provide actual particulate data for the current flowing medium. The module combines historical typical data with real-time monitoring data through data fusion technology to generate a set of distribution curves or datasets reflecting the particle characteristics of various fluid types. Each distribution feature includes key parameters such as the size range, distribution density, and shape factor of the particles in that fluid component.

[0064] The next crucial step is to dynamically update the initially obtained fluid particle size distribution features based on operational accuracy constraints. These constraints, typically set by the user or issued from a higher-level control strategy, define the allowable error range or resolution requirements for valve operation. The module's processing unit translates these constraints into adjustment instructions for the particle size distribution feature data resolution. For example, it might use resampling algorithms to increase or decrease the granularity of the distribution data, or apply filtering techniques to eliminate noise data exceeding the accuracy range. This ensures that the updated operational particle size distribution features retain the original fluid's particle characteristics while meeting the actual control accuracy requirements. The process of fusing multiple operational particle size distribution features to generate a global particle size distribution feature employs a multi-source data integration method. The module may use a weighted average algorithm to assign weights to each distribution feature based on the flow rate ratio or importance index of each fluid component, or it may use statistical techniques such as principal component analysis to extract the most representative comprehensive features from multiple distributions. The final output global feature is a unified model representing the overall particle behavior of the entire fluid mixture. This model typically exists in the form of a probability distribution function or eigenvector.

[0065] When performing multi-stage valve core matching based on global particle size distribution characteristics, the module accesses an internal or cloud-based database of valve core materials and structures. This database stores detailed specifications of various valve core components (such as sealing rings, valve discs, and valve seats), including their material hardness, wear resistance, corrosion resistance, and geometric dimensions such as pore size gradient and layer thickness sequence. The matching algorithm compares the global features with the records in the database to find valve core combinations that can effectively handle the particle distribution characteristics. For example, for fluids with a large number of fine particles, a valve layer with a smaller pore size and a smoother surface may be matched, while for fluids containing abrasive particles, a material with higher hardness will be selected. The matching result is a list of selected multi-layer valve core components. Each component is labeled with its specific operating dimensions (such as effective diameter range), valve layer thickness (such as millimeter-level dimensions), and valve layer material identification (such as alloy grade or polymer code).

[0066] The sequential assembly of multi-layer valve core components based on multiple operating scales is a digital mapping of the physical valve core construction. The virtual assembly engine in the module determines the installation sequence and relative position of each component according to its operating scale. For example, valve plates with gradually changing orifice diameters are arranged in ascending order to achieve graded throttling, or valve layers of different materials are placed alternately to balance wear resistance and sealing requirements. The assembly logic may be based on fluid dynamics simulation results or a historical optimization case library to ensure that the assembled valve core structure exhibits the expected flow control characteristics and mechanical strength in the digital model. The final output valve core selection is a complete specification document, which details the model, material, size, and overall assembly diagram of each component. This specification document can be directly sent to the intelligent manufacturing system or inventory management system to guide the preparation or allocation of physical valve cores.

[0067] The operation of the entire valve core matching module relies on continuous data inflow and real-time processing capabilities. It maintains a close interface with the parameter acquisition module to receive the latest fluid characteristics, and at the same time, it is interconnected with the enterprise's resource management system to obtain available valve core inventory information. Its internal algorithm also has a self-learning function, which can extract knowledge from each matching result and subsequent operation performance to continuously optimize the matching rules, so that the system can always recommend the optimal valve core configuration scheme when facing complex and ever-changing industrial fluids.

[0068] Example 2: See Figure 3 The execution flow of the condition initialization module begins with acquiring material information for multiple components of the multi-layer valve core element. This information typically originates from the valve core selection data package output by the valve core matching module or from a connected material database. The material information includes not only basic chemical compositions such as the identification of carbon steel, 304 stainless steel, or PTFE, but also the material's heat treatment state, surface treatment process, and detailed performance parameter tables provided by the supplier. The module's data parsing unit accurately extracts key attributes related to mechanical properties and wear resistance, such as elastic modulus, yield strength, and coefficient of friction, from this massive amount of information. Calculating the operating area based on structural features is a multi-dimensional geometric analysis process. The structural features include the three-dimensional geometric data of the valve core derived from the computer-aided design model, including the precise dimensions of each valve plate layer, surface contours, and their assembly relationships. The calculation engine applies geometric algorithms to identify the projected areas of the sealing surface and throttling surface that actually participate in the operation under fluid action, while also considering the dynamic area changes caused by valve stem movement. The final output effective operating area is a function that varies with the opening degree or a matrix of area values ​​under a series of discrete operating conditions.

[0069] When calling network data based on multiple component material information and effective operating area, the module sends data requests to a simulation service cluster deployed in the cloud or a high-performance computing platform within the enterprise. The requests encapsulate material and geometric parameters as query conditions. Upon receiving the request, the cloud system searches its vast model library for matching operational simulation models. These models may be pre-calibrated mathematical models based on physical experiments or numerical simulations for specific material combinations and flow conditions. Each model can predict the dynamic response behavior of the valve core component in the fluid under those material properties and area conditions. After obtaining multiple operational simulation models, the module's data fusion unit initiates a multi-model integration algorithm. This algorithm may employ confidence-based weighted fusion or physical law-based model stitching techniques to integrate multiple simulation models targeting a single material or local structure into a comprehensive valve core model that reflects the behavior of the entire valve core assembly. This comprehensive model fully considers the combined effects of interlayer interactions and material differences.

[0070] When obtaining the standard flow rate through local data retrieval, the module queries the historical operation database or default parameter table. The standard flow rate is typically taken from the typical flow velocity value of this type of shut-off valve under normal operating conditions or a user-preset benchmark value. It is an important reference quantity for initializing the simulation. When performing fluid dynamics simulation of the valve core model using the standard flow rate, the module activates the built-in CFD solver or calls the external simulation software interface to apply the flow rate boundary condition to the constructed valve core model. The simulation calculation simulates the pressure field, velocity field, and stress distribution of the fluid flowing through the valve core in a virtual environment. After iterative solution, it outputs a structural pressure difference threshold representing the pressure-bearing capacity of the entire valve core assembly. Fluid dynamics simulation of multiple operational simulation models using the standard flow rate is a parallel computation process. Each individual model runs the simulation independently and outputs its corresponding individual pressure difference threshold. These thresholds reflect the pressure-bearing limit of each individual material element in an isolated state.

[0071] When solving for the intersection of the structural pressure difference threshold and multiple individual unit pressure difference thresholds, the module's optimization algorithm analyzes the distribution range of these threshold data and seeks their common safe interval. For example, it finds overlapping numerical ranges by comparing the maximum and minimum value sequences, or applies statistical methods to determine an upper limit of pressure difference that allows all models to operate safely. The resulting pressure difference threshold is a conservative and reliable design limit. Using the maximum value of the pressure difference threshold as a constraint, when performing fluid dynamics simulation of the valve core model again, the module uses this ultimate pressure condition as input parameters to drive the simulation model. This simulation focuses on the valve core's performance under extreme conditions. By analyzing the surface stress distribution and relative velocity data in the simulation results, the module can interpolate and query the wear rate data under this stress state and material pair from a pre-set material wear database, and then output the key parameter of the maximum allowable wear rate, providing boundary conditions for subsequent optimization control.

[0072] Taking a shut-off valve control system for conveying corrosive slurry in a chemical plant as an example, when the condition initialization module starts execution, it first receives a valve core selection data packet from the upstream valve core matching module. This data packet shows that the matched multi-layer valve core element includes: a base layer made of 316 stainless steel, a middle layer of tungsten carbide wear-resistant coating, and a sealing layer made of reinforced polytetrafluoroethylene (PTFE). The module's material analysis unit immediately retrieves detailed performance parameters of these materials from the enterprise resource management system, including the corrosion resistance index of 316 stainless steel, the Rockwell hardness value of tungsten carbide, and the friction coefficient characteristics of PTFE. These data are all timestamped and have supplier batch numbers to ensure accuracy. When calculating the operating area based on structural characteristics, the module loads the 3D model file of the valve core. This model shows that the valve seat inner diameter is 150mm and the valve disc stroke is 80mm. The calculation engine obtains the effective sealing area variation curve under different opening degrees through integral calculation, and finally determines that the effective operating area in the fully open state is 176.7 square centimeters.

[0073] When calling network data based on multiple component material information and effective operating area, the module sends encrypted data packets to the cooperating simulation service platform. This platform matches three operational simulation models in its material database: a 316 stainless steel corrosion model, a tungsten carbide wear model, and a polytetrafluoroethylene deformation model. After obtaining these models, the module's data fusion algorithm uses multiphysics coupling technology to integrate the three independent models into a unified valve core assembly simulation model. This model can simulate the collaborative working behavior of different material layers under slurry flow conditions.

[0074] When retrieving the standard flow rate from local data, the module queries the pipeline's historical operation database and finds that the average flow rate over the past month is 2.8 m / s. This value is then set as the standard flow rate. During the fluid dynamics simulation of the valve core model using the standard flow rate, the module initiates transient simulation calculations to simulate the flow state of the slurry medium passing through the valve core. After 156 iterations, the output structural pressure difference threshold is 0.85 MPa, representing the maximum pressure difference limit that the entire valve core assembly can withstand. Simultaneously, fluid dynamics simulations of multiple operational models using the standard flow rate are performed. The 316 stainless steel corrosion model outputs 0.92 MPa, the tungsten carbide wear model outputs 1.15 MPa, and the polytetrafluoroethylene deformation model outputs 0.78 MPa—three individual pressure difference thresholds. These values ​​reflect the pressure-bearing capacity of each material when operating independently.

[0075] When solving for the intersection of the structural pressure difference threshold and multiple individual pressure difference thresholds, the module's comparison algorithm identifies a common safe range of 0.78-0.85 MPa. Ultimately, 0.78 MPa is chosen as the system's pressure difference threshold to ensure the safety of the weakest link. Using the maximum pressure difference threshold of 0.85 MPa as a constraint, the module again performs a fluid dynamics simulation of the valve core model. This simulation focuses on the stress distribution under extreme conditions. The simulation results show a maximum stress concentration point in the valve seat sealing area. By querying the stress-wear relationship curve in the material wear database, the module calculates the maximum allowable wear rate under this condition to be 0.12 mm / kWh. The entire initialization process provides precise safety boundary parameters for subsequent control, ensuring that the valve can meet process requirements and operate within a safe range when conveying corrosive slurries.

[0076] Example 3: See Figure 4 The implementation of the flow rate optimization module first requires the pre-construction of a predictive model that reflects the complex relationship between flow rate and key parameters. This model relies on retrieving a large number of operational records from a historical database system that match the current valve core selection and fluid composition. These historical flow-related data include actual sample flow rate values ​​recorded under different flow rate conditions, corresponding valve regulation efficiency indicators, and time-series data on wear rate changes and pressure difference changes obtained through sensor monitoring. During the retrieval process, the system uses the valve core material code and fluid composition hash value as a joint query key to ensure that the retrieved data is highly relevant to the current task in terms of physical characteristics and operating conditions. The data cleaning unit removes abnormal records and imputes missing values, forming a complete multi-dimensional training dataset.

[0077] When calculating the pressure difference change time limit based on the pressure difference threshold and multiple pressure difference change rates, the module uses a combination of linear extrapolation and threshold comparison. For each historical data point's pressure difference change rate value, the algorithm calculates the time required for the initial pressure difference to rise to a preset threshold. This calculation process considers the nonlinearity of fluid acceleration and introduces a correction coefficient to enhance the accuracy of the prediction. Similarly, based on the maximum allowable wear rate and multiple wear rate change rates, the system calculates the wear rate change time limit. It analyzes the cumulative trend of wear rate in each historical data point and estimates the time point when the maximum allowable wear rate is reached. These calculations provide crucial time-stamped data for subsequent model training.

[0078] When multiple sample flow rates, multiple sample operational efficiencies, multiple pressure difference time limits, and multiple wear rate change time limits are used as training parameters, the feature engineering unit of the module organizes these parameters into a correspondence between feature vectors and target variables. The neural network model adopts a deep learning architecture with multiple hidden layers, and iteratively adjusts the connection weights through forward and backward propagation. During training, an early stopping strategy and regularization techniques are used to prevent overfitting. The resulting flow correlation prediction model can accurately predict the wear rate time limit, pressure difference time limit, and operational efficiency under any given flow rate.

[0079] The module updates data using the reciprocal of the standard flow rate as a measure of velocity change. This reciprocal relationship reflects the granularity of flow rate adjustment. The generated first backup flow rate is typically slightly higher or lower than the standard value. This velocity value is fed into a trained prediction model for inference calculations. The model outputs three key predicted values: the first wear rate time limit, the first pressure difference time limit, and the first predicted operating efficiency. When calculating the first predicted operating time using flow volume information and the first predicted operating efficiency, the system divides the total flow volume by the predicted efficiency value to obtain the theoretical operating time. This time is then compared with the user-defined operating time constraint. If the constraint is met, the backup velocity value is retained. Simultaneously, the absolute value of the difference between the wear rate time limit and the pressure difference time limit is used as the first control reliability coefficient. This coefficient reflects the size of the system's safety margin at that flow rate.

[0080] This test and evaluation cycle is repeated until the preset update frequency is reached. Each update fine-tunes the previous speed value, generating a series of gradually changing backup flow rates and corresponding control reliability coefficients. These coefficient sequences are sorted in descending order using a sorting algorithm, and the backup speed with the highest reliability coefficient is selected as the final determined flow control speed. This optimization process ensures that the selected speed meets both operational time requirements and has the maximum safety margin. The entire flow rate optimization module, through a systematic data-driven approach, finds the optimal balance between safety and efficiency in valve control.

[0081] Taking a heavy oil pipeline control system in an oil refinery as an example, the flow rate optimization module first needs to build a predictive model upon startup. The pipeline transports high-viscosity heavy oil mixed with a small amount of catalyst particles, and the valve core is made of a special ceramic composite material. The module uses "ceramic composite material - heavy oil catalyst" as its characteristic combination and retrieves 415 operation records from the plant's data center based on similar operating conditions over the past two years. This historical data includes sample flow rate velocity data ranging from 1.2 m / s to 3.8 m / s. Each record is accompanied by records of valve regulation efficiency, wear rate monitoring data, and pressure differential sensor readings at that time. The data preprocessing unit first excludes 38 abnormal data points from equipment maintenance periods and smooths the remaining 377 records using a moving average method to form a standardized training dataset.

[0082] Based on the known system pressure difference threshold of 1.2 MPa and the pressure difference change rate in each historical record, the module calculates the estimated time to reach this threshold under different flow rates. For example, when a record shows that the pressure difference increases at a rate of 0.08 MPa / hour, the system calculates the corresponding pressure difference change time limit to be 15 hours. Similarly, based on the maximum allowable wear rate of 0.08 mm / thousand hours and the wear rate change data, the module calculates the corresponding wear rate change time limit. For example, when the wear rate in a record is 0.10 mm / thousand hours, the system calculates that it will take 800 hours of operation to reach the allowable wear rate.

[0083] After integrating these time parameters with the corresponding sample flow rate and sample operation efficiency data into training samples, the module starts the neural network training process. The network structure adopts a four-layer hidden layer design, with 5 neurons in the input layer corresponding to the feature parameters and 3 neurons in the output layer corresponding to the prediction target. The training cycle is set to 3000 iterations. The final flow correlation prediction model can accurately output three predicted values: wear rate time limit, pressure difference time limit, and operation efficiency, based on the input flow rate.

[0084] The module uses 0.4, the reciprocal of the standard flow rate of 2.5 m / s, as the adjustment benchmark. It first generates a backup flow rate of 2.9 m / s and inputs this value into the prediction model. The model outputs the following predictions: first wear rate time limit is 1200 hours, first pressure difference time limit is 1000 hours, and first predicted operating efficiency is 85%. Assuming the total flow volume of the current transport task is 600 cubic meters, the calculated first predicted operating time is 600 / (2.9)... 85%) = 243 minutes. This value meets the system's required 240-minute operating time constraint (within the allowable 5% error range). Therefore, this backup speed is retained, and the first control reliability coefficient is calculated to be 1200-1000=200.

[0085] Next, a second backup flow rate of 2.1 m / s was generated with a step size of 0.4. The model predicted a second wear rate time limit of 1500 hours, a second pressure difference time limit of 1300 hours, and a second predicted operating efficiency of 78%. The second predicted operating time was calculated to be 600 / (2.1). 78%) = 366 minutes, which significantly exceeds the operation time constraint and is therefore excluded. The module continues to generate backup velocities in steps of 0.4 within the range of 2.1 m / s to 2.9 m / s, ultimately obtaining three sets of backup flow rates that meet the time constraints and their corresponding control reliability coefficients.

[0086] After sorting these reliability coefficients in descending order, the velocity value corresponding to the maximum value is selected as the final determined flow control velocity. The entire optimization process, through systematic data analysis and model prediction, finds the optimal flow rate setting with the largest safety margin for heavy oil transportation while meeting process time requirements.

[0087] Example 4: The implementation process of the control optimization module begins with a deep analysis of the flow control speed to obtain its directional characteristics. This speed value typically comes from the output of the flow rate optimization module and carries vector information. The module obtains the flow direction parameters corresponding to this speed, such as forward flow or reverse flushing mode, by querying the internally stored speed-direction mapping table or real-time sensor data stream. Based on the flow direction parameters, the module calls the historical operating condition database to obtain the typical wear development patterns and pressure change patterns in that direction. Through interpolation, it calculates two key time indicators: the estimated control wear rate time limit and the control pressure difference time limit under the current flow control speed. The system compares these two time limit values ​​and selects the smaller value as the basic monitoring cycle to ensure that monitoring is initiated before more pressing limits arrive.

[0088] The process of predefining fitness weight configuration involves the application of multi-objective optimization theory. System administrators can input specific values ​​for energy consumption weight, lifetime loss weight, and fluid consumption weight through a human-machine interface. These weights are usually expressed as percentages and sum to 100%. The settings of these values ​​vary depending on the priority considerations of energy costs, equipment maintenance costs, and media costs in actual production. When constructing the fitness evaluation function based on the fitness weight configuration, the module uses a linear weighted sum model. It multiplies the energy consumption data, lifetime loss data, and fluid consumption data in the sample records by their corresponding weights and then sums them to obtain the comprehensive fitness score. The lower the score, the better the overall performance of the control record.

[0089] When using initial operational control constraints as data retrieval constraints to filter and retrieve historical control records, the module uses pressure difference threshold and maximum permissible wear rate as key filtering conditions to retrieve all control records that have operated under similar operational boundaries from the historical database, forming an initial candidate set. Then, a secondary filtering is performed based on the current flow control speed range to obtain multiple sample control records. Each sample control record contains two types of data: sample control parameters and sample monitoring consumption. The sample control parameters detail the specific control strategy used in this historical operation, including sample water flow rate, sample water duration, sample gas flow rate, sample gas duration, and overall sample duration. Sample monitoring consumption quantifies the actual costs incurred in this operation, including sample monitoring energy consumption (in kilowatt-hours), sample component lifespan loss (expressed as a percentage), and sample water consumption (in cubic meters).

[0090] When evaluating the fitness of multiple sample control records using the fitness evaluation function, the module's calculation unit calculates and sorts the comprehensive fitness score for each record. From the sorted results, the control parameters with the lowest fitness scores are selected as the initial control parameters. This set of parameters includes historically validated optimal values ​​for monitored water flow rate, monitored water duration, monitored air flow rate, monitored air duration, and comprehensive monitoring duration. When the monitoring period is added to the initial control parameters, the system integrates this period value as a time-dimensional constraint into the control parameters, forming the final set of executable control parameters. This set guides the specific operational rhythm and intensity of the remote monitoring module. Refer to Table 1, which shows a set of exemplary historical control record data used for fitness evaluation calculations.

[0091] Table 1 is the historical control record table:

[0092]

[0093] Assuming the current fitness weight configuration is: energy consumption weight 40%, lifetime loss weight 35%, and fluid consumption weight 25%, the module will calculate a fitness score for each record. For example, for record RC-2023-001, the calculation process is: (5.2×0.4)+(0.15×0.35)+(2.63×0.25)=2.08+0.0525+0.6575=2.79. Similarly, after calculating the scores for other records, the system will select the record with the lowest score as the initial control parameters. Finally, the previously determined monitoring period (e.g., 15 minutes) will be integrated into these parameters to form the final control strategy. The entire control optimization module provides the system with a validated and reliable control scheme by mining the best practices from historical data.

[0094] Example 5: The implementation of the safety early warning module is based on the wear trend analysis results received from the control optimization module. These results typically include dynamic data such as the current wear rate, cumulative wear amount, and predicted remaining service life. The module's internal data processing unit continuously monitors these parameters and compares them with pre-stored safety thresholds. When a sudden increase in the wear rate or the cumulative wear amount approaches the design limit is detected, the system automatically triggers the early warning generation process. The generation of early warning parameters employs a multi-level evaluation mechanism. First, the early warning level is determined based on the degree of wear deviation (e.g., Level 1 observation, Level 2 attention, Level 3 alarm). Then, the expected early warning time is calculated by combining the equipment's operating timeline. Finally, targeted recommended operational measures are generated through knowledge base matching.

[0095] Upon receiving the warning parameters, the module initiates a data transmission protocol. The warning parameters are encapsulated into a standard data packet format with an appended timestamp and device identifier. The data packet is transmitted to the remote monitoring center via industrial Ethernet or a wireless communication module. After parsing the data packet content, the data receiving end at the remote monitoring center activates different handling procedures based on the warning level: Level 1 observation warnings may only generate a log record and send it to the duty personnel's workstation; Level 2 attention warnings will trigger audible and visual alerts and send an SMS notification to the maintenance personnel's mobile terminal; Level 3 alarms will directly activate the emergency response protocol, which may include automatically reducing system load or activating backup equipment.

[0096] The specific classification of warning levels depends on the severity of the wear trend. For example, a wear rate 10%-25% higher than normal is defined as Level 1 Observation, 25%-50% higher as Level 2 Attention, and over 50% as Level 3 Alarm. The warning time calculation considers not only the current wear rate but also a comprehensive analysis of the equipment's continuous operating time and load variation history, providing the time interval from the current moment to the expected failure. The recommended operating measures knowledge base contains extensive experience from equipment maintenance experts. The system will recommend the most suitable handling plan based on the specific wear type (e.g., uniform wear, localized corrosion, fracture wear), including specific instructions such as adjusting operating parameters, scheduling maintenance, or immediate shutdown for inspection.

[0097] The entire early warning transmission process employs an encrypted communication protocol to ensure data security, while a retransmission mechanism guarantees transmission reliability. Upon receiving the early warning information, the remote monitoring center automatically generates a tracking work order and assigns processing permissions. Maintenance personnel can view detailed wear trend curves and early warning analysis reports through the monitoring center's interface. The system records the handling process and results of all early warning events, and this historical data is fed back into the knowledge base to optimize future early warning judgments. The safety early warning module provides crucial safeguards for the safe operation of equipment through real-time monitoring and intelligent analysis.

[0098] 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 process, method, article, or apparatus.

[0099] 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. An intelligent remote monitoring system based on a gate valve, characterized in that, include: The parameter acquisition module is used to acquire the fluid characteristic parameters and operation requirement parameters of the shut-off valve. The fluid characteristic parameters include fluid composition information and flow rate / volume information, and the operation requirement parameters include operation time constraints and operation accuracy constraints. The valve core matching module is used to match valve core types based on the fluid composition information and operating accuracy constraints. After obtaining the valve core selection, the valve core of the shut-off valve is configured based on the valve core selection to obtain the target valve core. The condition initialization module is used to obtain the structural features of the target valve core, and initialize the operating conditions according to the structural features and valve core selection, and output the initial operating control constraints, wherein the initial operating control constraints include the pressure difference threshold and the maximum allowable wear rate; The flow rate optimization module is used to take the initial operation control constraints as optimization conditions, perform flow rate optimization based on the flow volume information and operation time constraints, and determine the flow control speed. The control optimization module is used to perform wear trend analysis of the target valve core based on the flow control speed, and to perform control optimization based on the wear trend analysis results to obtain control parameters; A remote monitoring module is used to intermittently remotely monitor the target valve core using the control parameters during the operation of the shut-off valve at a flow control speed. The flow rate optimization module includes: Pre-built traffic correlation prediction model; The standard flow rate is updated by using the reciprocal of the standard flow rate as a measure of the rate change, thus obtaining the first backup flow rate. Input the first backup flow rate into the flow correlation prediction model to obtain the first wear rate time limit, the first pressure difference time limit, and the first prediction operation efficiency. The first predicted operation time is calculated using the flow volume information and the first predicted operation efficiency. If the first predicted operation time meets the operation time constraint, the first backup flow rate is retained, and the first difference between the first wear rate time limit and the first pressure difference time limit is used as the first control reliability coefficient of the first backup flow rate. Similarly, by continuously updating the standard flow rate and evaluating the update results using the flow correlation prediction model and operation time constraints, multiple backup flow rates and multiple control reliability coefficients that meet the preset update frequency are obtained. The plurality of control reliability coefficients are serialized, and the flow control speed is located from the plurality of backup flow speeds based on the sorting results.

2. The intelligent remote monitoring system based on a shut-off valve as described in claim 1, characterized in that, The valve core matching module includes: Based on the fluid type, the fluid component information is decomposed to obtain multiple fluid particle size distribution characteristics for various fluid types; The multiple fluid particle size distribution features are updated according to the operational accuracy constraints to obtain multiple operational particle size distribution features; By fusing the multiple operational particle size distribution features, a global particle size distribution feature is obtained; Based on the global particle size distribution characteristics, multi-level valve core matching is performed to obtain a multi-layer valve core element, wherein the multi-layer valve core element has multiple operating dimensions, valve layer thickness and valve layer material identification. Based on the multiple operating dimensions, the multi-layer valve core components are sequentially assembled to obtain the valve core selection.

3. The intelligent remote monitoring system based on a shut-off valve as described in claim 2, characterized in that, The condition initialization module includes: Obtain material information for multiple components of the multilayer valve core element; The effective operating area is obtained by calculating the operating area based on the structural features described above. Based on the material information of the multiple components and the effective operating area, network data is called to obtain multiple operation simulation models. Then, the valve core model is obtained by fusing the multiple operation simulation models. After obtaining the standard flow rate by calling local data, the fluid dynamics simulation of the valve core model is performed using the standard flow rate, and the structural pressure difference threshold is output. The fluid dynamics simulation of the multiple operation simulation models is also performed using the standard flow rate, and multiple individual pressure difference thresholds are output. The pressure difference threshold is obtained by solving the intersection of the structural pressure difference threshold and multiple individual pressure difference thresholds. The maximum value of the pressure difference threshold is used as a constraint to perform a fluid dynamics simulation of the valve core model, and the maximum allowable wear rate is obtained by calling the valve core model based on the simulation results.

4. The intelligent remote monitoring system based on a shut-off valve as described in claim 1, characterized in that, The traffic grading optimization module pre-builds a traffic correlation prediction model, including: Using the valve core selection and fluid composition information as constraints for historical data retrieval, historical flow correlation data is obtained, wherein the historical flow correlation data includes multiple sample flow rates, multiple sample operating efficiencies, multiple wear rate change rates, and multiple pressure difference change rates; Multiple pressure difference change time limits are calculated based on the pressure difference threshold and multiple pressure difference change rates; Multiple wear rate change time limits are calculated based on the maximum permissible wear rate and multiple wear rate change rates; The multiple sample flow rates, multiple sample operating efficiencies, multiple pressure difference time limits, and multiple wear rate change time limits are used as training parameters to train the flow correlation prediction model based on a neural network.

5. The intelligent remote monitoring system based on a shut-off valve as described in claim 1, characterized in that, The control optimization module includes: The directional data of the flow control speed is retrieved to obtain the control wear rate time limit and the control pressure difference time limit; The monitoring cycle is determined by comparing the time limit for controlling the wear rate and the time limit for controlling the pressure difference, and selecting the smaller value. A fitness weight configuration is predefined, and a fitness evaluation function is constructed based on the fitness weight configuration, wherein the fitness weight configuration includes energy consumption weight, lifetime loss weight, and fluid consumption weight; The initial operation control constraint is used as a data call constraint to filter and call historical control records to obtain multiple sample control records. Each sample control record includes sample control parameters and sample monitoring consumption. The fitness evaluation function is used to evaluate the fitness of the multiple sample control records, and the initial control parameters are located based on the evaluation results. The initial control parameters consist of the monitored water flow rate, the monitored water duration, the monitored air flow rate, the monitored air duration, and the comprehensive monitoring duration. The monitoring period is added to the initial control parameters to obtain the control parameters.

6. The intelligent remote monitoring system based on a shut-off valve as described in claim 5, characterized in that, The sample control parameters consist of sample water flow rate, sample water duration, sample gas flow rate, sample gas duration, and overall sample duration.

7. The intelligent remote monitoring system based on a shut-off valve as described in claim 6, characterized in that, The sample monitoring consumption consists of sample monitoring energy consumption, sample component lifespan loss, and sample water consumption.

8. The intelligent remote monitoring system based on a shut-off valve as described in claim 1, characterized in that, It also includes a safety early warning module, which is used to perform safety early warning based on the wear trend analysis results, obtain early warning parameters, and transmit the early warning parameters to the remote monitoring center.

9. A method for intelligent remote monitoring of a shut-off valve, characterized in that, It includes all modules and method flows of the intelligent remote monitoring system based on the shut-off valve as described in any one of claims 1 to 8.

Images