A method and system for constructing a digital twin intelligent pump station knowledge engine

Abstract

The application provides a construction method and system of a digital twin intelligent pump station knowledge engine, relates to the field of intelligent water affairs, and solves the technical problem that the prior art is limited by the perspective of a single field, cannot form a complete "perception-analysis-decision-verification-feedback" closed loop system, is difficult to comprehensively evaluate complex working conditions, and cannot guarantee stable operation of the pump station. The method comprises the following steps: deploying a pump station monitoring network, collecting operation data of the pump station, and analyzing characteristic parameters of the operation data; constructing a pump station digital twin, a pump station field knowledge graph, and a health state evaluation model; the pump station field knowledge graph outputs a solution to a pump station fault; the solution pushed by the pump station field knowledge graph is verified for feasibility by a verification module of the pump station digital twin; the pump station digital twin feeds back a solution result to the pump station field knowledge graph, and updates the pump station field knowledge graph. The application is used in the process of pump station operation and maintenance.

Description

Technical Field

[0001] This application relates to the field of smart water management, and in particular to a method and system for constructing a digital twin intelligent pumping station knowledge engine. Background Technology

[0002] Currently, large and medium-sized pumping stations across the country generally suffer from serious equipment aging, high energy consumption, and low operation and maintenance efficiency. While existing technologies can address specific types of problems, they are limited by a single-domain perspective and cannot form a complete "perception-analysis-decision-verification-feedback" closed-loop system. This makes it difficult to comprehensively assess complex operating environments and effectively improve the operating efficiency of pumping stations and reduce energy consumption. Summary of the Invention

[0003] This application provides a method and apparatus for constructing a digital twin intelligent pumping station knowledge engine, which solves the technical problem that existing technologies are limited by a single field perspective, making it impossible to form a complete "perception-analysis-decision-verification-feedback" closed-loop system, and making it difficult to comprehensively assess complex working conditions and ensure the stable operation of the pumping station.

[0004] To achieve the above objectives, this application adopts the following technical solution:

[0005] Firstly, a method for constructing a digital twin intelligent pumping station knowledge engine is provided, including:

[0006] Deploy a pump station monitoring network to collect pump station operation data and analyze the characteristic parameters of the operation data; among which, the characteristic parameters include peak value, root mean square value and kurtosis coefficient;

[0007] A digital twin of the pumping station is constructed based on the collected operational data and corresponding feature parameters; the digital twin of the pumping station integrates a prediction module, an optimization module, and a verification module;

[0008] Constructing a knowledge graph for the pumping station domain based on ontology;

[0009] In addition, a health status assessment model is constructed based on historical operating data and characteristic parameters of the operating data to assess the health status of the equipment; the health status is determined based on the health index and remaining service life.

[0010] Based on the output of the health status prediction model, the knowledge graph of the pumping station domain outputs a solution; the solution pushed by the knowledge graph of the pumping station domain verifies its feasibility through the verification module of the pumping station digital twin; the pumping station digital twin feeds back the solution results to the knowledge graph of the pumping station domain, updating the knowledge graph of the pumping station domain.

[0011] Based on the above technical solutions, the method for constructing a digital twin intelligent pumping station knowledge engine provided in this application acquires operational data by deploying a pumping station monitoring network, extracts key feature parameters, and constructs a digital twin and a health status assessment model, realizing full-process digital management from data collection and status assessment to fault prediction. Simultaneously, combined with a pumping station domain knowledge graph constructed based on ontology, solutions can be intelligently generated based on health status prediction results, and feasibility verification is performed through the digital twin's verification module to ensure the scientific validity and effectiveness of the solutions. Finally, the solution results are fed back to the knowledge graph, enabling continuous knowledge updates and optimization, forming a closed-loop system of "perception-analysis-decision-verification-feedback," ensuring the safety, stability, and intelligence level of pumping station operation.

[0012] In conjunction with the first aspect above, in one possible implementation, the construction of the pump station digital twin includes:

[0013] Acquire multi-source heterogeneous operating data and key characteristic parameters during the operation of the pumping station;

[0014] The structural model, behavioral model, and parameter model of the pumping station system are established using the MBSE method to obtain a digital twin of the pumping station.

[0015] The structural model is used to reflect the connection relationships and topology of the pump station equipment;

[0016] The behavioral model is used to describe the dynamic response characteristics of each component of the pumping station;

[0017] The parameter model is used to construct the pump performance curve;

[0018] The prediction module is used to predict the future operating status of the pumping station based on historical and real-time data.

[0019] The optimization module is used to dynamically optimize the pump station operation strategy under different operating conditions;

[0020] The verification module is used to compare and verify the output results of the pump station digital twin with the actual physical pump station operation results, obtain error information, and feed the error information back to at least one of the structural model, behavioral model, parameter model, prediction model or optimization model for feedback correction and iterative update.

[0021] In conjunction with the first aspect above, in one possible implementation, the construction of the pump performance curve includes:

[0022] Collect historical operating data, which includes at least the pipeline head H and flow rate Q;

[0023] The collected pipeline parameters are fitted using the least squares method to obtain the optimal fitting parameters D and S; where D and S are constants.

[0024] Based on the optimal fitting parameters D and S, and the flow rate Q, the formula H = D + SQ is used. 2 Calculate the head H of the pipeline.

[0025] In conjunction with the first aspect above, in one possible implementation, the optimal fitting parameter D and the fitting parameter S include:

[0026] The historical operational data is normalized using the following formula:

[0027] ;

[0028] Where k is the sequence number of the running data. For the value of the kth running data, and These are the maximum and minimum values ​​of the k-th running data, respectively;

[0029] Substitute the normalized historical operating data into the following objective function. To obtain the optimal fitting parameters D and S:

[0030] .

[0031] In conjunction with the first aspect above, in one possible implementation, obtaining the error information includes:

[0032] Acquire real-time operational data of physical pump stations during operation;

[0033] Obtain simulation output data of the pump station's digital twin under the same operating conditions;

[0034] The real-time running data and simulation output data are timestamped and preprocessed.

[0035] Error indicators are calculated based on the aligned data to obtain error information;

[0036] The error information includes at least one of numerical error, time delay error, trend deviation, or state error.

[0037] In conjunction with the first aspect above, in one possible implementation, the construction of a knowledge graph for the pumping station domain based on ontology includes:

[0038] Collect multi-source heterogeneous data in the field of pumping stations;

[0039] An ontology model for the pumping station domain is constructed based on an ontology modeling language. The ontology model includes concept classes, object attributes, data attributes, instances, and axiom constraints.

[0040] The concept class is used to classify the core entities in the pumping station field;

[0041] The object attributes are used to describe the semantic relationships between different concept classes;

[0042] The data attributes are used to represent specific quantifiable features or static information of a concept class;

[0043] The instance is used to map abstract concept classes to concrete objects or events in the real world;

[0044] The axiomatic constraints are used to formally define the structural and semantic relationships in the ontology model;

[0045] Map multi-source heterogeneous data from the pumping station field to concept classes, object attributes, and data attributes in the ontology model;

[0046] Structured triple information is generated based on the mapping results, and the triple information includes the semantic relationship expression of subject, predicate, and object;

[0047] By importing the triplet information into a graph database, a knowledge graph for the pump station domain is constructed.

[0048] In conjunction with the first aspect above, in one possible implementation, the health status prediction model includes:

[0049] Extract the time-domain, frequency-domain, and nonlinear features of historical operating data to obtain a comprehensive feature vector;

[0050] A health status prediction model is constructed based on a hybrid neural network model with LSTM-attention mechanism, and a multi-task output layer is set up. The input is a comprehensive feature vector, and the outputs are the health index and remaining service life of the device.

[0051] In conjunction with the first aspect above, in one possible implementation, the health index ,include:

[0052] Health Index ,include:

[0053] ;

[0054] Where k is the sequence number of the running data, k [1,n], where t is the time step. Let k be the feature weight coefficients of the running data, 0 ≤ ≤1, Let k be the value collected at time step t. The fault threshold for running data k is determined by the 3σ principle. This serves as the health baseline value for the running data k;

[0055] The remaining service life ,include:

[0056] LSTM layer computation:

[0057] ;

[0058] in, The input feature vector at time step t, The hidden state at time step t-1, The hidden state at time step t; This refers to the operational functions of a long short-term memory neural network.

[0059] Fully connected layer computation:

[0060] ;

[0061] in, It is a fully connected layer. This represents the predicted remaining useful life.

[0062] In conjunction with the first aspect above, in one possible implementation, the feature weight coefficients ,include:

[0063] Extract the associated operational data k during equipment failure, count the number of times each associated operational data k occurs during equipment failure, and mark it as... And obtain the total number of equipment failures, marked as ;

[0064] ;

[0065] in, λ is the time decay factor, t is the decay rate, and t is the time interval between the time of the most recent fault related to the associated running data k and the present.

[0066] Secondly, a device for constructing a digital twin intelligent pumping station knowledge engine is provided, comprising: a communication unit and a processing unit; the communication unit is used to deploy a pumping station monitoring network, collect pumping station operation data, and analyze the characteristic parameters of the operation data; the processing unit is used to output solutions based on the construction of a pumping station digital twin, a pumping station domain knowledge graph, and a health status assessment model; and based on the output results of the health status prediction model, the pumping station domain knowledge graph outputs solutions; the solutions pushed by the pumping station domain knowledge graph are verified for feasibility through the verification module of the pumping station digital twin; the pumping station digital twin feeds back the solution results to the pumping station domain knowledge graph, updating the pumping station domain knowledge graph.

[0067] Thirdly, this application provides an apparatus for constructing a digital twin intelligent pumping station knowledge engine, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is used to execute the instructions to implement the methods described in the first aspect and any possible implementation thereof. This apparatus for constructing a digital twin intelligent pumping station knowledge engine can be an electronic device or a chip within an electronic device.

[0068] Fourthly, this application provides a system for constructing a digital twin intelligent pumping station knowledge engine, comprising: a data acquisition module, a construction module, and a decision-making module; wherein, the data acquisition module is used to deploy a pumping station monitoring network, collect pumping station operation data, and analyze the characteristic parameters of the operation data; the construction module constructs a digital twin of the pumping station based on the collected operation data and corresponding characteristic parameters; constructs a pumping station domain knowledge graph based on ontology; and constructs a health status assessment model based on historical operation data and the characteristic parameters of the operation data to assess the health status of the equipment; the decision-making module outputs a solution based on the output results of the health status prediction model and the pumping station domain knowledge graph; the solution pushed by the pumping station domain knowledge graph is verified for feasibility through the verification module of the pumping station digital twin; the pumping station digital twin feeds back the solution results to the pumping station domain knowledge graph, updating the pumping station domain knowledge graph.

[0069] Fifthly, this application provides a computer-readable storage medium storing instructions that, when executed on a construction apparatus for a digital twin intelligent pumping station knowledge engine, cause the construction apparatus to perform the methods described in the first aspect and any possible implementation thereof.

[0070] Sixthly, this application provides a computer program product containing instructions that, when run on a construction apparatus for a digital twin intelligent pumping station knowledge engine, causes the construction apparatus for the digital twin intelligent pumping station knowledge engine to perform the methods described in the first aspect and any possible implementation thereof.

[0071] This application provides a method and apparatus for constructing a digital twin intelligent pumping station knowledge engine, establishing a closed-loop system for pumping station management that is interconnected, highly efficient, collaborative, and continuously evolving. First, a deployed monitoring network extensively collects pumping station operation data and deeply analyzes its characteristic parameters, laying a solid data foundation for building a realistic digital twin of the pumping station and achieving a high degree of consistency between the physical entity and the virtual model. Second, a health status assessment model built based on historical operation data and its characteristic parameters acts like a "health guardian," accurately assessing the health status of equipment in real time. Third, the output of the health status prediction model can quickly trigger the generation of targeted solutions from a knowledge graph in the pumping station field. The verification module of the digital twin acts as a "quality inspector," rigorously verifying the feasibility of the solutions and ensuring scientific and reasonable decision-making. Finally, the solution results are promptly fed back to the knowledge graph, enabling dynamic updates and optimization of knowledge. The entire process, starting from data collection, goes through model analysis, solution generation, verification feedback, and finally returns to knowledge updates, forming a complete, coherent, and self-improving closed loop, powerfully promoting pumping station management towards intelligence and refinement.

[0072] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0073] Figure 1 A system architecture diagram of a digital twin intelligent pumping station knowledge engine construction system provided in this application embodiment;

[0074] Figure 2 A flowchart illustrating a method for constructing a digital twin intelligent pumping station knowledge engine, as provided in an embodiment of this application;

[0075] Figure 3 A schematic diagram of the structure of a device for constructing a digital twin intelligent pumping station knowledge engine, provided in an embodiment of this application;

[0076] Figure 4This is a schematic diagram of the hardware structure of a device for constructing a digital twin intelligent pumping station knowledge engine, provided in an embodiment of this application. Detailed Implementation

[0077] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0078] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0079] The method for constructing a digital twin intelligent pumping station knowledge engine provided in this application embodiment can be applied to, for example... Figure 1 The system shown is a construction system for a digital twin intelligent pumping station knowledge engine, such as... Figure 1 As shown, the system includes: a data acquisition module, a data construction module, and a decision-making module. The data acquisition module is used to deploy the pump station monitoring network, collect pump station operating data, and analyze the characteristic parameters of the operating data.

[0080] The construction module builds a digital twin of the pump station based on the collected operational data and corresponding feature parameters; it constructs a knowledge graph of the pump station domain based on ontology; and it constructs a health status assessment model based on historical operational data and feature parameters of the operational data to assess the health status of the equipment.

[0081] The decision-making module outputs a solution based on the health status prediction model and the knowledge graph of the pumping station domain. The solution pushed by the knowledge graph of the pumping station domain verifies its feasibility through the verification module of the pumping station digital twin. The pumping station digital twin feeds back the solution to the knowledge graph of the pumping station domain, updating the knowledge graph of the pumping station domain.

[0082] To address the limitations of existing technologies, which are confined to a single-domain perspective and unable to form a complete "perception-analysis-decision-verification-feedback" closed-loop system, thus hindering comprehensive assessment of complex operating environments and ensuring stable pump station operation, this application provides a method for constructing a digital twin intelligent pump station knowledge engine. This method includes: deploying a pump station monitoring network, collecting pump station operating data, and analyzing the characteristic parameters of the operating data; constructing a digital twin of the pump station based on the collected operating data and corresponding characteristic parameters; constructing a health status assessment model based on historical operating data and the characteristic parameters of the operating data to assess the health status of the equipment; outputting a pump station domain knowledge graph solution based on the output results of the health status prediction model; verifying the feasibility of the pump station domain knowledge graph push solution through the verification module of the pump station digital twin; and finally, the pump station digital twin... The solution feeds back the solution results to the pump station domain knowledge graph, updating the knowledge graph. Based on this, the solution constructs a tightly closed-loop and coherent pump station management ecosystem: by deploying a monitoring network to collect operational data and analyze characteristic parameters, it provides a precise data foundation for building a digital twin of the pump station, realizing a virtual mapping of the physical pump station; a health status assessment model built based on historical data and characteristic parameters can provide real-time insight into the health status of equipment; combined with the output of the health status prediction model, the knowledge graph quickly outputs targeted solutions; the verification module of the digital twin strictly controls the feasibility of the solution to ensure its practicality; finally, the solution results are fed back to the knowledge graph for updates and optimizations, forming a complete closed loop of "data collection - model construction - solution output - verification feedback - knowledge update", continuously improving the intelligence and precision of pump station management.

[0083] like Figure 2 As shown in the embodiment of this application, a method for constructing a digital twin intelligent pumping station knowledge engine includes:

[0084] S201. Deploy a pump station monitoring network to collect pump station operation data; analyze the characteristic parameters of the operation data.

[0085] The characteristic parameters include peak value, root mean square value, and kurtosis coefficient.

[0086] It should be noted that the operational data can be time-series signals, such as vibration signals, current signals, pressure fluctuations, etc.

[0087] For example, the process of deploying a pump station monitoring network is as follows:

[0088] Multiple types of sensors are deployed in the pumping station to collect operating data of key equipment, such as water pumps, motors, and valves;

[0089] Sensor types include, but are not limited to: vibration sensors (accelerometers): used to collect mechanical vibration signals; current transformers: used to collect motor current waveforms; pressure transmitters: used to collect inlet and outlet pressures; temperature sensors: used to collect the temperature of key components; all sensors are connected to an edge computing gateway or central monitoring system via industrial communication protocols such as Modbus, CAN, and OPC UA; the data acquisition frequency is set according to the signal characteristics, for example, the sampling rate of vibration signals is usually set to 1kHz to 10kHz; data can be stored locally or uploaded to a cloud platform for further processing and modeling.

[0090] Data is continuously collected at fixed intervals to form a continuous time series data stream; the collected data enters the feature extraction stage after preprocessing steps such as filtering, denoising, and normalization.

[0091] The following three typical time-domain feature parameters are extracted from the preprocessed data to characterize the device's operating status:

[0092] Peak value, which is the maximum absolute value of a signal over a period of time, is often used to identify sudden anomalies, such as bearing pitting or gear breakage.

[0093] The root mean square (RMS) value reflects the overall energy level of the signal and is suitable for evaluating the stability of equipment operation.

[0094] The kurtosis coefficient is a statistical measure of the "sharpness" of the tail of a signal distribution, indicating that there are many pulse components in the signal, making it suitable for detecting minor faults.

[0095] S202. Construct a digital twin of the pump station based on the collected operational data and corresponding feature parameters.

[0096] The pump station digital twin integrates a prediction module, an optimization module, and a verification module.

[0097] The prediction module is used to predict the future operating status of the pumping station based on historical and real-time data;

[0098] The optimization module is used to dynamically optimize the pump station operation strategy under different operating conditions;

[0099] The verification module is used to compare and verify the output results of the pump station digital twin with the actual physical pump station operation results, obtain error information, and feed the error information back to at least one of the structural model, behavioral model, parameter model, prediction model or optimization model for feedback correction and iterative update.

[0100] The error information is obtained in the following ways:

[0101] Acquire real-time operational data of physical pump stations during operation;

[0102] Obtain simulation output data of the pump station's digital twin under the same operating conditions;

[0103] Timestamp alignment and data preprocessing are performed on real-time running data and simulation output data;

[0104] Error indicators are calculated based on the aligned data to obtain error information;

[0105] Error information includes at least one of numerical error, time delay error, trend deviation, or state error.

[0106] It should be noted that numerical error is the difference between the predicted value and the actual value, relative error, etc.; time delay error is the gap between the simulated response time and the actual response time; trend error is the accuracy of predicting the development trend of the system in long-term observation; and state error is the difference between the predicted state and the actual state under different states, such as the working state of the pump and the opening degree of the valve.

[0107] S203. Construct a knowledge graph for the pumping station domain based on ontology.

[0108] S204. Based on historical operating data and characteristic parameters of the operating data, construct a health status assessment model to assess the health status of the equipment.

[0109] Among them, health status is determined based on health index and remaining life expectancy.

[0110] S205. Based on the output results of the health status prediction model, the pump station domain knowledge graph outputs a solution; the solution pushed by the pump station domain knowledge graph is verified for feasibility through the verification module of the pump station digital twin; the pump station digital twin feeds back the solution results to the pump station domain knowledge graph, updating the pump station domain knowledge graph.

[0111] Based on the above technical solutions, this application provides a method for constructing a digital twin intelligent pumping station knowledge engine. This method involves deploying a monitoring network to accurately collect pumping station operation data and extracting key characteristic parameters such as peak value, root mean square value, and kurtosis coefficient, laying the foundation for subsequent analysis. It constructs a digital twin integrating prediction, optimization, and verification modules, enabling early prediction of operating status, dynamic optimization of strategies, and verification of decision-making feasibility, enhancing the foresight and scientific nature of management. A knowledge graph based on ontology enables structured storage and intelligent reasoning of knowledge, quickly outputting targeted solutions based on health assessment results. A health status assessment model combining historical and real-time data comprehensively evaluates equipment status using health indices and remaining service life, providing a reliable basis for maintenance decisions. Furthermore, through the interactive feedback mechanism between the knowledge graph and the digital twin, a closed-loop optimization system is formed, continuously improving the intelligence and precision of pumping station management and ensuring the safe and efficient operation of the pumping station.

[0112] In one possible implementation of this application embodiment, the above-mentioned S202 can be specifically implemented by the following S301 and S302, which are described in detail below:

[0113] S301. Acquire multi-source heterogeneous operation data and key characteristic parameters during the operation of the pumping station;

[0114] S302. Use the MBSE method to establish the structural model, behavioral model and parameter model of the pumping station system to obtain a digital twin of the pumping station;

[0115] Among them, the structural model is used to reflect the connection relationship and topology of the pump station equipment; the behavioral model is used to describe the dynamic response characteristics of each component of the pump station; and the parametric model is used to construct the pump performance curve.

[0116] For example, using MBSE methods, such as the SysML modeling language, a systematic model of the pumping station system can be constructed, resulting in the following three types of models:

[0117] I. Structural Model:

[0118] Use the Block Definition Diagram (BDD) in SysML to define components such as pumps, motors, valves, pipes, control systems, and sensors.

[0119] The Internal Block Diagram (IBD) is used to represent the physical connections and data flow between components: for example, a water pump is connected to a motor to provide power, and a sensor collects the vibration signal of the water pump; the components interact with each other through water flow, electrical signals, and control signals.

[0120] II. Behavioral Model

[0121] Use SysML’s Activity Diagram or State Machine Diagram to describe the pump station workflow.

[0122] Use Parametric Diagram in conjunction with simulation tools such as MATLAB / Simulink to simulate system behavior;

[0123] This includes, but is not limited to: current surges during startup; pressure response during flow regulation; and behavior under fault conditions, such as blockage or idling.

[0124] III. Parametric Model

[0125] The pump performance curve (HQ curve) is fitted using measured data; the model is embedded into the Parametric Diagram to achieve parameter constraints and dynamic calculation.

[0126] Based on the above technical solution, multi-source heterogeneous operational data encompasses rich information from different equipment and sensors during pump station operation. Characteristic parameters such as peak value, root mean square value, and kurtosis coefficient are extracted from the operational data. By extracting these characteristic parameters, the core operational information of the equipment can be grasped quickly and effectively, reducing data processing complexity and improving analysis efficiency. The MBSE method organically integrates structural, behavioral, and parametric models to form a complete pump station system model. In this integrated model, the various models are interconnected and mutually influential, comprehensively and accurately reflecting the operating status and performance characteristics of the pump station system. Through collaborative analysis of the integrated model, factors such as equipment connection relationships, dynamic response characteristics, and parameter settings can be comprehensively considered to optimize the design and operation control of the pump station system.

[0127] In one possible implementation, the parameter model in S302 above can be specifically implemented through the following S401, S402 and S403, which are explained in detail below:

[0128] S401. Collect historical operating data, which should include at least the pipeline head H and flow rate Q.

[0129] S402. Normalize the historical operating data. Normalization formula:

[0130] ;

[0131] Where k is the sequence number of the running data. For the value of the kth running data, and These are the maximum and minimum values ​​of the k-th running data, respectively;

[0132] Substitute the normalized historical operating data into the following objective function. To obtain the optimal fitting parameters D and S:

[0133] ;

[0134] S403. Based on the optimal fitting parameter D and the fitting parameter S, and the flow rate Q, the formula H = D + SQ is used to calculate the flow rate Q. 2 Calculate the head H of the pipeline.

[0135] The goal is to find The smallest parameters D and S.

[0136] Based on the above technical solution, historical operating data of the pump station pipeline system (such as key parameters head H and flow rate Q) are comprehensively collected and normalized to eliminate the influence of dimensions and improve calculation efficiency. Then, the optimal fitting parameters D and S are solved using a reasonable parameter objective function. Finally, based on these optimal parameters, the pipeline head H is accurately calculated, providing reliable data support for pump station operation and scheduling. This helps pump station managers to reasonably adjust pump parameters and plan maintenance schedules in advance. It also provides an analytical basis for pipeline system performance evaluation and optimization, ensuring the safe, stable and efficient operation of the pump station.

[0137] In one possible implementation, the parameter model in S203 above can be specifically implemented through the following S401, S402 and S403, which are explained in detail below:

[0138] S401. Collect multi-source heterogeneous data in the field of pumping stations;

[0139] S402. Construct an ontology model for the pumping station domain based on an ontology modeling language. The ontology model includes concept classes, object attributes, data attributes, instances, and axiom constraints.

[0140] Among them, the concept class is used to classify the core entities in the pumping station field; including at least one of the following: pumping station entity class, function class, status class, fault class, and operation class;

[0141] Object attributes are used to describe the semantic relationships between different concept classes; including at least one of the following: composition relationship, causal relationship, influence relationship, and control relationship; such as "equipment-fault", "fault-solution", etc.

[0142] Data attributes are used to represent specific quantifiable features or static information of a concept class;

[0143] Instances are used to map abstract conceptual classes to concrete objects or events in the real world; for example, representing actual pumping station equipment, events, or operating states.

[0144] Axiom constraints are used to formally define the structural and semantic relationships in an ontology model; they include at least one of type constraints, cardinality constraints, range constraints, relation constraints, and integrity constraints.

[0145] S403. Map multi-source heterogeneous data in the pumping station field to concept classes, object attributes, and data attributes in the ontology model;

[0146] S404. Generate structured triple information based on the mapping results. The triple information includes the semantic relationship expression of the subject, predicate, and object.

[0147] The subject represents a conceptual instance in the field of pumping stations, including equipment, status, operation, or event;

[0148] Predicates represent object attributes or data attributes, describing the semantic relationship or characteristics between the subject and the object;

[0149] An object represents another instance of a conceptual class or a specific numerical / textual data that is associated with a subject.

[0150] S405. Import the triplet information into the graph database to construct a knowledge graph for the pump station domain.

[0151] Based on the above technical solution, by collecting multi-source heterogeneous data from pumping stations, an ontology model containing various elements such as concept classes and object attributes is constructed, which can accurately define the core entities and their relationships in the pumping station domain. By mapping the data to the ontology model and generating structured triple information, the data is standardized and semantically expressed. Finally, the data is imported into a graph database to construct a knowledge graph, which can efficiently store and query complex knowledge, providing intelligent decision support for pumping station fault diagnosis and operation optimization, and improving the scientific and accurate management of pumping stations.

[0152] In one possible implementation, the parameter model in S204 above can be specifically implemented through the following S501, S502 and S503, which are explained in detail below:

[0153] S501. Extract the time domain, frequency domain, and nonlinear features of historical operating data to obtain a comprehensive feature vector;

[0154] S502. A health status prediction model is constructed based on the LSTM-attention mechanism hybrid neural network model, and a multi-task output layer is set up. The comprehensive feature vector is input, and the health index and remaining service life of the device are output respectively.

[0155] Among them, the health index ,include:

[0156] ;

[0157] Where k is the sequence number of the running data, k [1,n], where t is the time step. Let k be the feature weight coefficients of the running data, 0 ≤ ≤1, Let k be the value collected at time step t. The fault threshold for running data k is determined by the 3σ principle. This is the health baseline value for the running data k.

[0158] The fault threshold is μ±3σ, which means that the monitoring data of the equipment under normal conditions, such as vibration value, temperature, current, etc., follow a normal distribution, that is, the data are concentrated near the mean, and the further away from the mean, the lower the probability.

[0159] Remaining service life ,include:

[0160] LSTM layer computation:

[0161] ;

[0162] in, The input feature vector at time step t, The hidden state at time step t-1, The hidden state at time step t; This refers to the operational functions of a long short-term memory neural network.

[0163] Fully connected layer computation:

[0164] ;

[0165] in, It is a fully connected layer. This represents the predicted remaining useful life.

[0166] Based on the above technical solution, multi-dimensional features of historical operating data are extracted to form a comprehensive feature vector, which can fully explore equipment operating information. Then, a health status prediction model is built based on a hybrid neural network of LSTM and attention mechanism. LSTM is good at processing time series data, and the attention mechanism can focus on key features. The multi-task output layer can output the health index and remaining service life at the same time. Moreover, the health index calculation takes into account the weight of multiple factors. The remaining service life prediction combines LSTM and fully connected layer to accurately assess the equipment status, provide strong support for equipment maintenance decisions, and reduce operation and maintenance costs and failure risks.

[0167] In one possible implementation, the feature weight coefficients in S502 above can be specifically explained as follows:

[0168] Extract the associated operational data k during equipment failure, count the number of times each associated operational data k occurs during equipment failure, and mark it as... And obtain the total number of equipment failures, marked as ;

[0169] ;

[0170] in, λ is the time decay factor, t is the decay rate, and t is the time interval between the time of the most recent fault related to the associated running data k and the present.

[0171] It should be noted that the attenuation rate was set by the technicians based on historical data for each feature.

[0172] For example, historical operating data and fault records of the pumping station are collected. Each fault type is categorized, and the associated operating data leading to the fault, such as vibration, temperature, and pressure, are identified.

[0173] Suppose that the fault records of a certain device are shown in Table 1 below;

[0174] Table 1 Fault Records

[0175]

[0176] In this embodiment, Table 1 only lists one associated running data, but there are actually several associated running data.

[0177] Analyzing the weighting coefficients for the feature "vibration," as shown in Table 1, it was determined that vibration caused equipment failure twice. The total number of historical failure records for this equipment is four. Assuming λ = 0.1, and the most recent vibration-related failure occurred 10 days ago; then... =0.5 0.368≈0.184, meaning the characteristic weighting coefficient of the vibration characteristics of the equipment is 0.184.

[0178] Based on the above technical solution, by statistically analyzing the ratio of the number of occurrences of associated operational data during equipment failures to the total number of failures, and introducing a time decay factor to dynamically adjust feature weights, this approach considers both the correlation strength between each operational data point and the failure in historical failure data, and can reasonably reduce the impact of old failure data on the current weights over time. This makes the weight calculation more closely reflect the actual changes in the equipment's operating status, helps to accurately locate key failure characteristics, improves the accuracy and timeliness of equipment failure diagnosis, and provides more valuable reference for equipment maintenance.

[0179] The above primarily describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as a construction apparatus for a digital twin intelligent pumping station knowledge engine, includes at least one of the hardware structures and software modules corresponding to each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0180] This application embodiment can divide a digital twin intelligent pump station knowledge engine construction device into functional units based on the above method example. For example, each function can be divided into different functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or software. It should be noted that the unit division in this application embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0181] When using integrated units, Figure 3 A possible structural schematic diagram of a digital twin intelligent pump station knowledge engine construction device (referred to as construction device 50) involved in the above embodiments is shown. The construction device 50 includes a processing unit 501 and a communication unit 502, and may also include a storage unit 503. Figure 3 The structural diagram shown can be used to illustrate the structure of a construction device for a digital twin intelligent pump station knowledge engine involved in the above embodiments.

[0182] when Figure 3 The schematic diagram shown illustrates the structure of a digital twin intelligent pumping station knowledge engine construction device involved in the above embodiments. The processing unit 501 is used to control and manage the operation of the digital twin intelligent pumping station knowledge engine construction device, the communication unit 502 is used for the digital twin intelligent pumping station knowledge engine construction device to communicate with other devices, and the storage unit 503 is used to store the program code and data of the digital twin intelligent pumping station knowledge engine construction device.

[0183] For example, communication unit 502 is used to deploy a pump station monitoring network, collect pump station operating data, and analyze the characteristic parameters of the operating data;

[0184] Processing unit 501 is used to construct a digital twin of the pumping station, a knowledge graph of the pumping station domain, and a health status assessment model;

[0185] In one possible implementation, the processing unit 502 is further configured to output a solution based on the output of the health status prediction model and the knowledge graph of the pumping station domain; the solution pushed by the knowledge graph of the pumping station domain is verified for feasibility by the verification module of the pumping station digital twin; the pumping station digital twin feeds back the solution to the knowledge graph of the pumping station domain, and updates the knowledge graph of the pumping station domain.

[0186] The processing unit 501 can be a processor or a controller, and the communication unit 502 can be a communication interface, transceiver, transceiver circuit, transceiver device, etc. The term "communication interface" is a general term and may include one or more interfaces. The storage unit 503 can be a memory. When the device 50 is a chip, the processing unit 501 can be a processor or a controller, and the communication unit 502 can be an input interface and / or an output interface, pins, or circuits, etc. The storage unit 503 can be a storage unit within the chip (e.g., a register, cache, etc.) or a storage unit located outside the chip (e.g., read-only memory (ROM), random access memory (RAM, etc.).

[0187] The communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the construction device 50 can be considered as the communication unit 502 of the construction device 50, and the processor with processing functions can be considered as the processing unit 501 of the construction device 50. Optionally, the device in the communication unit 502 that implements the receiving function can be considered as a communication unit. The communication unit is used to execute the receiving steps in the embodiments of this application, and the communication unit can be a receiver, a receiver circuit, etc. The device in the communication unit 502 that implements the transmitting function can be considered as a transmitting unit. The transmitting unit is used to execute the transmitting steps in the embodiments of this application, and the transmitting unit can be a transmitter, a transmitter, a transmitting circuit, etc.

[0188] Figure 3 If the integrated units in the process are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. Storage media for storing computer software products include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0189] Figure 3 The units in the process can also be called modules; for example, a processing unit can be called a processing module.

[0190] This application embodiment also provides a hardware structure diagram of a construction device (denoted as construction device 60) for a digital twin intelligent pumping station knowledge engine, see [link to diagram]. Figure 4The building device 60 includes a processor 601, and optionally, a memory 602 connected to the processor 601.

[0191] In the first possible implementation, see Figure 4 The construction device 60 also includes a transceiver 603. The processor 601, memory 602, and transceiver 603 are connected via a bus. The transceiver 603 is used to communicate with other devices or communication networks. Optionally, the transceiver 603 may include a transmitter and a receiver. The device in the transceiver 603 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of this application. The device in the transceiver 603 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of this application.

[0192] Based on the first possible implementation method Figure 4 The structural diagram shown can be used to illustrate the structure of a construction device for a digital twin intelligent pump station knowledge engine involved in the above embodiments.

[0193] in, Figure 4 This can also be illustrated by a system chip in a device for building a digital twin intelligent pumping station knowledge engine. In this case, the actions performed by the aforementioned device for building a digital twin intelligent pumping station knowledge engine can be implemented by this system chip. The specific actions performed are described above and will not be repeated here.

[0194] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0195] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a separate semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a SoC (System-on-a-Chip), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0196] The memory in the embodiments of this application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto.

[0197] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0198] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0199] This application also provides a chip including a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0200] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0201] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0202] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A method for constructing a knowledge engine for a digital twin intelligent pumping station, characterized in that, include: Deploy a pump station monitoring network to collect pump station operation data and analyze the characteristic parameters of the operation data; among which, the characteristic parameters include peak value, root mean square value and kurtosis coefficient; A digital twin of the pumping station is constructed based on the collected operational data and corresponding feature parameters, including: Acquire multi-source heterogeneous operating data and key characteristic parameters during the operation of the pumping station; The structural model, behavioral model, and parameter model of the pumping station system are established using the MBSE method to obtain a digital twin of the pumping station. The structural model is used to reflect the connection relationships and topology of the pump station equipment; The behavioral model is used to describe the dynamic response characteristics of each component of the pumping station; The parameter model is used to construct the pump performance curve, including: Collect historical operating data, which includes at least the pipeline head H and flow rate Q; The collected pipeline parameters are fitted using the least squares method to obtain the optimal fitting parameters D and S; where D and S are constants. Based on the optimal fitting parameters D and S, and the flow rate Q, the formula H = D + SQ is used. 2 Calculate the head H of the pipeline; The digital twin of the pump station also integrates a prediction module, an optimization module, and a verification module; The prediction module is used to predict the future operating status of the pumping station based on historical and real-time data; The optimization module is used to dynamically optimize the pump station operation strategy under different operating conditions; The verification module is used to compare and verify the output results of the pump station digital twin with the actual physical pump station operation results, obtain error information, and feed the error information back to at least one of the structural model, behavioral model, parameter model, prediction model or optimization model for feedback correction and iterative update. Constructing a knowledge graph for the pumping station domain based on ontology; In addition, a health status assessment model is constructed based on historical operating data and characteristic parameters of the operating data to assess the health status of the equipment; the health status is determined based on the health index and remaining service life. Based on the output of the health status prediction model, the knowledge graph of the pumping station domain outputs a solution; the solution pushed by the knowledge graph of the pumping station domain verifies its feasibility through the verification module of the pumping station digital twin; the pumping station digital twin feeds back the solution results to the knowledge graph of the pumping station domain, updating the knowledge graph of the pumping station domain.

2. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 1, characterized in that, The optimal fitting parameters D and fitting parameters S include: The historical operational data is normalized using the following formula: ； Where k is the sequence number of the running data. For the value of the kth running data, and These are the maximum and minimum values ​​of the k-th running data, respectively; Substitute the normalized historical operating data into the following objective function. To obtain the optimal fitting parameters D and S: 。 3. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 1, characterized in that, The acquisition of error information includes: Acquire real-time operational data of physical pump stations during operation; Obtain simulation output data of the pump station's digital twin under the same operating conditions; The real-time running data and simulation output data are timestamped and preprocessed. Error indicators are calculated based on the aligned data to obtain error information; The error information includes at least one of numerical error, time delay error, trend deviation, or state error.

4. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 1, characterized in that, The knowledge graph for the pumping station domain constructed based on ontology includes: Collect multi-source heterogeneous data in the field of pumping stations; An ontology model for the pumping station domain is constructed based on an ontology modeling language. The ontology model includes concept classes, object attributes, data attributes, instances, and axiom constraints. The concept class is used to classify the core entities in the pumping station field; The object attributes are used to describe the semantic relationships between different concept classes; The data attributes are used to represent specific quantifiable features or static information of a concept class; The instance is used to map abstract concept classes to concrete objects or events in the real world; The axiomatic constraints are used to formally define the structural and semantic relationships in the ontology model; Map multi-source heterogeneous data from the pumping station field to concept classes, object attributes, and data attributes in the ontology model; Structured triple information is generated based on the mapping results, and the triple information includes the semantic relationship expression of subject, predicate, and object; By importing the triplet information into a graph database, a knowledge graph for the pump station domain is constructed.

5. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 1, characterized in that, The health status prediction model includes: Extract the time-domain, frequency-domain, and nonlinear features of historical operating data to obtain a comprehensive feature vector; A health status prediction model is constructed based on a hybrid neural network model with LSTM-attention mechanism, and a multi-task output layer is set up. The input is a comprehensive feature vector, and the outputs are the health index and remaining service life of the device.

6. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 5, characterized in that, The health index : ； Where k is the sequence number of the running data, k [1,n], where t is the time step. Let k be the feature weight coefficients of the running data, 0 ≤ ≤1, Let k be the value collected at time step t. The fault threshold for running data k is determined by the 3σ principle. This serves as the health baseline value for the running data k; The remaining service life : ； ； in, It is a fully connected layer. For the predicted remaining useful life, Let be the input feature vector at time step t. The hidden state at time step t-1, The hidden state at time step t, This is a function for operating long short-term memory neural networks.

7. The method for constructing a digital twin intelligent pumping station knowledge engine according to claim 6, characterized in that, The feature weight coefficient ,include: Extract the associated operational data k during equipment failure, count the number of times each associated operational data k occurs during equipment failure, and mark it as... And obtain the total number of equipment failures, marked as ; ； in, λ is the time decay factor, t is the decay rate, and t is the time interval between the time of the most recent fault related to the associated running data k and the present.

8. A system for constructing a digital twin intelligent pumping station knowledge engine, operating based on the method for constructing a digital twin intelligent pumping station knowledge engine according to any one of claims 1-7, characterized in that, include: The module consists of a data acquisition module, a data construction module, and a decision-making module. The acquisition module is used to deploy a pump station monitoring network, collect the pump station's operating data, and analyze the characteristic parameters of the operating data. The construction module is used to construct a digital twin of the pumping station based on the collected operational data and corresponding feature parameters, including: Acquire multi-source heterogeneous operating data and key characteristic parameters during the operation of the pumping station; The structural model, behavioral model, and parameter model of the pumping station system are established using the MBSE method to obtain a digital twin of the pumping station. The structural model is used to reflect the connection relationships and topology of the pump station equipment; The behavioral model is used to describe the dynamic response characteristics of each component of the pumping station; The parameter model is used to construct the pump performance curve, including: Collect historical operating data, which includes at least the pipeline head H and flow rate Q; The collected pipeline parameters are fitted using the least squares method to obtain the optimal fitting parameters D and S; where D and S are constants. Based on the optimal fitting parameters D and S, and the flow rate Q, the formula H = D + SQ is used. 2 Calculate the head H of the pipeline; The digital twin of the pump station also integrates a prediction module, an optimization module, and a verification module; The prediction module is used to predict the future operating status of the pumping station based on historical and real-time data; The optimization module is used to dynamically optimize the pump station operation strategy under different operating conditions; The verification module is used to compare and verify the output results of the pump station digital twin with the actual physical pump station operation results, obtain error information, and feed the error information back to at least one of the structural model, behavioral model, parameter model, prediction model or optimization model for feedback correction and iterative update. Furthermore, a knowledge graph for the pumping station domain is constructed based on ontology; Based on historical operating data and characteristic parameters of the operating data, a health status assessment model is constructed to assess the health status of the equipment. The decision-making module outputs a solution based on the output of the health status prediction model and the knowledge graph of the pumping station domain. The solution pushed by the knowledge graph of the pumping station domain is verified for feasibility by the verification module of the pumping station digital twin. The pumping station digital twin feeds back the solution to the knowledge graph of the pumping station domain, updating the knowledge graph of the pumping station domain.

Images