Pump system performance monitoring and fault diagnosis method and system based on digital twinning

Abstract

The invention relates to the technical field of pump system monitoring and diagnosis, and discloses a pump system performance monitoring and fault diagnosis method and system based on digital twinning, and the method comprises the steps: carrying out the topological correlation analysis of target pump equipment, and obtaining a working logic rule and component relation; performing digital three-dimensional mapping on the target pump equipment to construct a digital twin frame; performing operation simulation deduction on the digital twin frame to obtain simulation operation data; performing synchronous deviation analysis on the simulation operation data and the actual operation data, and performing evolution trend fitting on the analyzed deviation data in combination with historical operation data to obtain a degradation track; performing fault risk study and judgment on the degradation track to obtain a fault symptom category and a fault severity level; performing health state evaluation on the target pump equipment to generate a diagnosis report; according to the invention, the efficiency of performance monitoring and fault diagnosis of the pump system based on digital twinning can be improved.

Description

Technical Field

[0001] This invention relates to the field of pump system monitoring and diagnosis technology, and in particular to a method and system for pump system performance monitoring and fault diagnosis based on digital twins. Background Technology

[0002] Existing methods for monitoring and diagnosing pump system performance lack depth in analyzing the topological relationships and operational logic of equipment, failing to fully deconstruct the coupling relationships and control dependencies between components, resulting in limitations in monitoring dimensions. Their data processing focuses solely on the analysis of actual operational data, without establishing a collaborative comparison mechanism between virtual simulation and actual data, leading to insufficient completeness and timeliness in deviation identification.

[0003] Traditional methods struggle to effectively fit performance degradation trends using historical operating data, lack precise feature support for judging fault symptom categories and severity levels, and suffer from insufficient scientific rigor in health status assessments and remaining life predictions, thus failing to provide comprehensive and reliable decision-making basis for equipment operation and maintenance. Therefore, improving the comprehensiveness of pump system performance monitoring, the accuracy of fault diagnosis, and the scientific rigor of health assessments has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for monitoring the performance and diagnosing faults of pump systems based on digital twins, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a pump system performance monitoring and fault diagnosis method based on digital twins, comprising: S1. Perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment; S2. Based on the working logic rules and component relationships, perform digital three-dimensional mapping on the target pump equipment to construct a digital twin framework for the target pump equipment; S3. Input the actual operating data of the target pump equipment into the digital twin framework, and perform an operational simulation of the digital twin framework to obtain the simulation operating data of the target pump equipment. S4. Perform synchronous deviation analysis on the simulation operation data and the actual operation data, and combine the historical operation data of the target pump equipment to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. S5. Conduct fault risk assessment on the degradation trajectory to obtain the fault symptom categories and fault severity levels of the target pump equipment; S6. Based on the fault symptom category and fault severity level, perform a health status assessment on the target pump equipment to generate a diagnostic report for the target pump equipment.

[0006] In a preferred embodiment, topological association parsing is performed on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment, including: Obtain the mechanical assembly drawings and electrical control schematic diagrams of the target pump equipment; Based on the mechanical assembly drawings, the coupling relationships of the rotating parts, sealing parts and pipelines of the target pump equipment are deconstructed to obtain the component relationships of the target pump equipment. Based on the electrical control schematic diagram, the control signals, sensor signals and power execution components of the target pump equipment are analyzed for control dependency, and the working logic rules of the target pump equipment are obtained.

[0007] In a preferred embodiment, based on working logic rules and component relationships, a digital 3D mapping is performed on the target pump equipment to construct a digital twin framework of the target pump equipment, including: Based on the component relationships, the target pump equipment is reconstructed in three-dimensional space to obtain the static geometric structure of the target pump equipment. Based on logical rules, timing rules are injected into the target pump equipment to obtain the dynamic control logic of the target pump equipment. By coupling and associating the static geometric structure with the dynamic control logic, a digital twin framework of the target pump equipment is obtained.

[0008] In a preferred embodiment, the actual operating data of the target pump equipment is input into the digital twin framework, and the digital twin framework is used for operational simulation to obtain the simulation operating data of the target pump equipment, including: By decoupling the parameters of the actual operating data of the target pump equipment, multi-dimensional operating parameters of the target pump equipment are obtained. By mapping multidimensional operating condition parameters to behavioral commands, virtual control commands for the digital twin framework are obtained. The digital twin framework is driven by virtual control commands to achieve collaborative state evolution; During the cooperative state evolution process, the digital twin framework is synchronously sampled to obtain the virtual physical quantities of the digital twin framework; According to the operating sequence of the target pump equipment, the virtual physical quantities are reorganized into the simulation operating data of the target pump equipment.

[0009] In a preferred embodiment, a synchronization deviation analysis is performed between the simulation data and the actual data, including: The simulation data and the actual data are timestamped to obtain the aligned simulation data and the aligned actual data of the target pump equipment. The aligned simulation data and the aligned actual data are correlated and matched to obtain the virtual and real data pairs of the target pump equipment. By performing a dimension-by-dimensional difference analysis on the virtual and real data pairs, the deviation data of the target pump equipment can be obtained.

[0010] In a preferred embodiment, by combining historical operating data of the target pump equipment, the analyzed deviation data is fitted with an evolution trend to obtain the degradation trajectory of the target pump equipment, including: Health status analysis is performed on the historical operating data of the target pump equipment to obtain the historical performance benchmark of the target pump equipment; Based on historical performance benchmarks, a relative degradation assessment is performed on the analyzed deviation data to obtain the performance degradation degree of the target pump equipment. By fitting the time-series trajectory of the performance degradation, the degradation trajectory of the target pump equipment is obtained.

[0011] In a preferred embodiment, the formula for calculating performance degradation is as follows: ; In the formula, Indicates at time Performance degradation, This indicates the total number of dimensions of performance parameters monitored for the target pump equipment. Indicates at time No. Deviation data values ​​for each performance parameter, Indicates the first The health status baseline value of each performance parameter Indicates the first The health status benchmark variance of each performance parameter This represents the summation operation. This represents the square root operation.

[0012] In a preferred embodiment, the degradation trajectory is analyzed for fault risk to determine the fault symptom category and fault severity level of the target pump equipment, including: Morphological analysis of the degenerate trajectory yields information on its slope, curvature characteristics, and key inflection points. Heterogeneous feature decomposition is performed on the slope, curvature features and key inflection point information to obtain the multidimensional feature vector of the degenerate trajectory. Based on a pre-defined fault mode knowledge base, similarity matching is performed on multi-dimensional feature vectors to identify the fault symptom categories of degradation trajectories. Risk level determination is performed on the multidimensional feature vector to obtain the fault severity level of the target pump equipment.

[0013] In a preferred embodiment, a health status assessment of the target pump equipment is performed based on the fault symptom category and fault severity level to generate a diagnostic report for the target pump equipment, including: By comprehensively correlating the fault symptom categories, fault severity levels, and historical maintenance records of the target pump equipment, a comprehensive diagnostic evidence body for the target pump equipment is obtained. Based on the comprehensive diagnostic evidence, the life curve of the target pump equipment is extrapolated to obtain the remaining service life data of the target pump equipment. Health status rating is performed on the remaining service life data to obtain the health status score of the target pump equipment; The diagnostic report for the target pump equipment integrates fault symptom categories, fault severity levels, remaining service life data, and health status scores.

[0014] To address the aforementioned problems, this invention also provides a pump system performance monitoring and fault diagnosis system based on digital twins, the system comprising: The equipment mechanism and topology parsing module is used to perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment. The digital twin construction module is used to perform digital three-dimensional mapping of the target pump equipment based on working logic rules and component relationships, so as to construct the digital twin framework of the target pump equipment; The virtual-real synchronous simulation and deduction module is used to input the actual operating data of the target pump equipment into the digital twin framework, and to perform operation simulation and deduction on the digital twin framework to obtain the simulation operating data of the target pump equipment. The performance degradation trajectory evaluation module is used to perform synchronous deviation analysis between simulation running data and actual running data, and combined with the historical running data of the target pump equipment, to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. The intelligent fault diagnosis module is used to assess the fault risk of the degradation trajectory and obtain the fault symptom category and fault severity level of the target pump equipment; The health assessment and report generation module is used to assess the health status of the target pump equipment based on the type of fault symptoms and the severity level of the fault, so as to generate a diagnostic report for the target pump equipment.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention accurately obtains the working logic rules and component relationships by performing topological association analysis on the target pump equipment. Based on this, a digital twin framework integrating static geometric structure and dynamic control logic is constructed to realize synchronous simulation and deduction of virtual and real data and dimension-by-dimensional deviation analysis, which significantly improves the comprehensiveness of performance monitoring and the accuracy of data analysis, making the acquisition of deviation data more timely and complete.

[0016] 2. This invention combines historical operating data to fit the performance degradation trajectory, accurately judges the type and severity of fault symptoms through morphological analysis and feature decomposition, integrates fault information and historical maintenance records to complete health status rating and remaining life prediction, and generates a comprehensive diagnostic report, effectively improving the accuracy of fault diagnosis and the scientific nature of health assessment, and providing reliable decision support for equipment operation and maintenance. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for monitoring and diagnosing pump system performance based on digital twins, provided in an embodiment of the present invention. Figure 2 A functional block diagram of a pump system performance monitoring and fault diagnosis system based on digital twin provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for monitoring and diagnosing pump system performance based on digital twins. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for monitoring and diagnosing pump system performance based on digital twins can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be a standalone server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a pump system performance monitoring and fault diagnosis method based on digital twins according to an embodiment of the present invention. In this embodiment, the pump system performance monitoring and fault diagnosis method based on digital twins includes: S1. Perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment; In this embodiment of the invention, topological association parsing is performed on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment, including: Obtain the mechanical assembly drawings and electrical control schematic diagrams of the target pump equipment; Based on the mechanical assembly drawings, the coupling relationships of the rotating parts, sealing parts and pipelines of the target pump equipment are deconstructed to obtain the component relationships of the target pump equipment. Based on the electrical control schematic diagram, the control signals, sensor signals and power execution components of the target pump equipment are analyzed for control dependency, and the working logic rules of the target pump equipment are obtained.

[0021] The unique serial number is read from the nameplate of the target pump equipment. The corresponding production record is retrieved from the equipment production management system to obtain the assembly batch information. These two key pieces of information are accurately entered into the design file management system for the target pump equipment. In the design file management system, a special search function is selected. Following the system's file classification options, the system locates the technical drawings category and triggers the search command. The system then performs precise matching and filtering based on the entered serial number and assembly batch. After the search results are generated, each candidate drawing is checked against its title bar, equipment identification, technical parameter summary, and other key information to ensure that the selected mechanical assembly drawings and electrical control schematics completely correspond to the serial number and assembly batch of the target pump equipment and are not confused with drawings of other equipment. Finally, the complete mechanical assembly drawings and electrical control schematics specific to the target pump equipment are retrieved.

[0022] Unfold the retrieved mechanical assembly drawings and examine them step by step, starting with the overall equipment and moving to individual components. First, clarify the overall structural layout of the equipment, then focus on core components such as rotating parts, sealing components, and piping. For rotating parts, focus on analyzing pump shafts, impellers, and bearings, clarifying the length, diameter, and surface finish of the pump shaft; the number, shape, and installation orientation of the impeller blades; and the installation position of the bearings and their fit with the pump shaft. For sealing components, carefully observe the dimensions, materials, and fit of the mounting slots for sealing rings, gaskets, and mechanical seals, as well as their contact methods with adjacent components. For piping, follow the direction of media flow to trace the piping's path, arranging pipe diameter variations and branch nodes, and clarifying the connection points of each pipe section. Each component's specific form is recorded one by one, and different connection methods such as threaded connections, flange connections, and welded connections are distinguished in detail. The coupling relationship between each component is analyzed in depth. For example, the pump shaft is fixed to the impeller by a key connection. When the pump shaft is running, it drives the impeller to rotate synchronously. The sealing ring tightly fits the pump shaft and the end face of the casing to form a sealing barrier to prevent media leakage. The pipeline is fixed to the pump body outlet by a flange and forms a complete flow channel with the subsequent pipeline. The names, morphological characteristics, connection positions, connection methods, and coupling effects of all components are classified and organized according to component categories, and finally a comprehensive system of component relationships is formed.

[0023] A careful study of the electrical control schematic diagrams revealed that the signals were categorized into two main types: control signals and sensor signals. A comprehensive analysis was conducted, proceeding from the signal source to the execution terminal. For control signals, the focus was on identifying different signal types for functions such as start, stop, speed regulation, and reversal. The origin of each control signal was clearly identified as a specific output port of the dedicated controller, the intermediate components such as terminals and relays along the transmission path, and the final destination of the power actuator. For sensor signals, the precise location of the acquisition source for different types of sensor signals, such as pressure, temperature, and flow rate, was identified as the signal output terminal of the corresponding sensor, the wiring nodes of the signal conversion module in the transmission path, and the input port of the controller as the signal receiving terminal. The correspondence between the various actions of each power actuator and the control and sensor signals was analyzed one by one. For example, when the flow rate sensor detects that the medium flow rate is lower than the set standard, the signal is transmitted to the controller along a preset path. After signal recognition, the controller issues a speed control signal, which is transmitted to the motor speed control module, driving the motor to increase its speed to increase the medium flow rate. When the temperature sensor detects that the equipment operating temperature exceeds the set limit, the controller immediately issues a stop control signal, cutting off the motor power circuit and stopping the motor. According to the action function category, the system records the signal type, signal transmission path, triggering condition, execution result and other information corresponding to each action, and organizes these control dependency logic systems into clear and complete working logic rules.

[0024] The beneficial effects include: a detailed and standardized process for obtaining factory serial numbers and assembly batches, combined with specialized searches and multi-dimensional information verification through the design document management system, ensuring that the retrieved mechanical assembly drawings and electrical control schematics are valid documents specific to the target pump equipment. This provides accurate, comprehensive, and unbiased basic technical data for subsequent topology correlation analysis. Detailed component-by-component analysis based on the mechanical assembly drawings covers the morphological characteristics, connection methods, and coupling relationships of rotating components, sealing components, and pipelines. This ensures that the component relationships fully present the actual structural attributes and interaction mechanisms of each component, without any omissions or ambiguities. Through classification and analysis by signal type and action-by-action correlation, the transmission paths and trigger conditions of control and sensor signals, along with the corresponding logic of the power execution component actions, are clarified, allowing the working logic rules to accurately match the actual control and operation of the target pump equipment. These detailed and systematically organized component relationships and working logic rules provide comprehensive and reliable technical support for the subsequent static geometric reconstruction and dynamic control logic injection of the digital twin framework. From the source, it ensures that the digital twin framework is highly consistent with the actual state of the target pump equipment in terms of structural form, connection relationship, control mechanism, and other aspects, laying a solid foundation for subsequent core links such as simulation and deviation analysis.

[0025] S2. Based on the working logic rules and component relationships, perform digital three-dimensional mapping on the target pump equipment to construct a digital twin framework for the target pump equipment; In this embodiment of the invention, based on logical rules and component relationships, a digital three-dimensional mapping is performed on the target pump equipment to construct a digital twin framework of the target pump equipment, including: Based on the component relationships, the target pump equipment is reconstructed in three-dimensional space to obtain the static geometric structure of the target pump equipment. Based on logical rules, timing rules are injected into the target pump equipment to obtain the dynamic control logic of the target pump equipment. By coupling and associating the static geometric structure with the dynamic control logic, a digital twin framework of the target pump equipment is obtained.

[0026] Based on the structural dimensions, installation locations, connection methods, and coupling information of each component as defined in the component relationships, the static geometric structure is constructed. Regarding structural dimensions, the specific data recorded in the component relationships, such as pump shaft length, diameter, surface precision, impeller blade quantity, shape, thickness, sealing component dimensions, and pipe diameter and wall thickness, are strictly referenced. Professional 3D modeling technology is used to create individual digital models for each component. During modeling, not only is the external outline of the component restored, but the internal structure is also meticulously replicated, such as the keyway dimensions and locations of the pump shaft, the impeller hub structure, and the mounting groove structure of the sealing components—key feature details—ensuring that the dimensional parameters of the virtual components are completely consistent with the actual components. When determining the installation location, a spatial coordinate system is established using the equipment reference plane defined in the component relationships as a reference. The installation coordinates of each component are accurately input into the modeling system. For example, the 3D coordinate values ​​of components such as the pump body, bearing seat, and impeller are determined with the center of the equipment base as the coordinate origin, ensuring that the spatial position of the virtual components matches the actual equipment. When implementing the connection method, according to the types of connections (threaded, flange, welded, etc.) specified in the component relationships, the corresponding connection tools are selected in the modeling system to simulate the actual assembly process. For example, for threaded connections, the specifications, quantity, and tightening torque parameters of the bolts are set; for flange connections, the flange hole diameter, bolt hole distribution, and gasket placement are restored. When recreating the coupling effect, based on the interaction mechanism of each component in the component relationships, corresponding constraint relationships are set in the virtual model. For example, the fixed constraint between the pump shaft and the impeller ensures synchronous rotation, and the contact constraint between the sealing ring and the pump shaft housing achieves the sealing effect. After completing the modeling of a single component, virtual assembly is performed one by one, starting from the basic components of the equipment, according to the assembly sequence specified in the component relationships. First, the pump body base is fixed, then the bearing housing is installed, and the pump shaft is accurately installed into the bearing housing. Next, the impeller is assembled and fixed by key connection. Then, the sealing components are installed, and finally, the pipeline sections are connected and fixed according to the pipeline route. After assembly, the spatial verification process is initiated. Collision detection technology is used to check the spatial distance between all virtual components to ensure that the gaps between each component meet the component relationship requirements and there is no mutual interference. At the same time, the connection method and connection parameters of each connection point are checked one by one to ensure that they are consistent with the component relationship. After confirming that there are no errors, a static geometric structure that is completely matched with the physical form and structure of the target pump equipment is formed.

[0027] This comprehensive analysis of the operational logic rules, including the transmission paths of control and sensor signals, trigger conditions, and the response sequence of each component, lays the foundation for the construction of dynamic control logic. Regarding control signals, different functional types such as start, stop, speed regulation, and reverse are differentiated. The origin, intermediate components, and final power execution components of each control signal are clearly defined. For example, the start control signal originates from a specific output port of the controller and is transmitted to the motor control module via intermediate components such as terminals and relays. Regarding sensor signals, the acquisition sources, signal conversion methods, transmission paths, and receiving terminals of different types of sensor signals, such as pressure, temperature, and flow, are categorized and analyzed. For example, pressure sensor signals are acquired by pressure sensors, converted into standard electrical signals, and transmitted to the controller input port via signal cable terminals. When analyzing trigger conditions, the sensor signal threshold or control signal command corresponding to each component's action is clearly defined. For example, the trigger condition for motor start-up is that the pressure sensor signal value reaches a set standard and the temperature sensor signal value is within the allowable range. When analyzing the response sequence of each component, the order of actions of each component after a signal is issued is clearly defined according to the equipment operation process. For example, after receiving a start command, the controller first collects sensor signals. After the signals are verified, a motor start signal is issued. After the motor starts, it drives the pump shaft impeller to rotate and simultaneously feeds back the operating status signal to the controller. After the analysis is completed, these logical relationships are transformed into a sequence of timing instructions that can be recognized by the digital twin framework. The instruction sequence is divided into multiple modules according to the equipment operation stage, such as start sequence, running sequence, speed regulation sequence, and stop sequence. In each timing module, the execution order of the instructions is clearly defined. For example, the instruction sequence of the start sequence is: sensor signal acquisition instruction, signal verification instruction, motor start instruction, and status feedback instruction; the instruction sequence of the speed regulation sequence is: sensor signal monitoring instruction, signal analysis instruction, speed control instruction, action execution instruction, and status feedback instruction. Each instruction is assigned a unique identifier, clearly defining the trigger conditions and execution time of the executing component's action parameters. Trigger relationships between instructions are defined: after the previous instruction completes and sends a success signal, the next instruction is triggered. Simultaneously, the execution time interval is set; for example, a signal verification instruction is triggered within 0.1 seconds after the completion of the sensor signal acquisition instruction, ensuring that the timing of instruction execution is consistent with the actual equipment control logic. These timing instructions are then injected into the digital twin framework one by one according to the chronological order of the equipment's operation, completing the construction of dynamic control logic and enabling the framework to accurately match the actual control mechanism of the target pump equipment.

[0028] A dedicated data interaction interface based on a standard communication protocol is established to ensure stable and efficient data transmission between the static geometry and dynamic control logic. First, a unique digital identifier is assigned to each virtual component in the static geometry. Simultaneously, the digital identifier of the execution object is explicitly defined in each instruction of the dynamic control logic. A dedicated association mapping table is established to map the component identifiers of the static geometry to the execution object identifiers of the dynamic control logic instructions, clearly defining the execution component corresponding to each instruction and the instruction types that each component can respond to. The association mapping table also includes the correspondence between component actions and control instruction parameters. For example, the speed value in a motor speed instruction is directly associated with the speed parameter of the virtual motor, and the pressure value in a pressure regulation instruction corresponds to the pressure parameter of the virtual pipeline. When an instruction in the dynamic control logic is triggered, the instruction is packaged according to a preset data format, including information such as the instruction identifier, execution component identifier, action parameters, and trigger conditions, and transmitted to the control module of the static geometry through the dedicated data interaction interface. After receiving an instruction, the control module parses the corresponding virtual component and the action to be performed based on the association mapping table. It then drives the component to perform the corresponding action as required by the instruction. For example, upon receiving a motor start instruction, the control module parses the corresponding virtual motor and controls it to accelerate from a standstill according to the speed parameters in the instruction, gradually increasing the speed to the set value. Simultaneously, it drives the associated virtual pump shaft and virtual impeller to rotate synchronously, maintaining a transmission ratio consistent with the component's relationship during rotation. During the motion of the static geometric structure, a real-time data acquisition module is set up to collect motion state data of each virtual component at a fixed frequency, including key parameters such as speed, displacement, pressure, temperature, and flow rate. The collected data is packaged in a unified format and fed back to the dynamic control logic in real time through a dedicated data interaction interface. After receiving the feedback data, the dynamic control logic compares it with the preset parameters in the instruction. If a difference is found between the actual motion state data and the preset parameters, an adjustment instruction is immediately issued and transmitted to the static geometric structure through the data interaction interface to adjust the action of the corresponding component, forming a complete two-way data interaction closed loop. Through this continuous transmission of instructions and feedback of status, the motion of the static geometric structure and the instructions of the dynamic control logic are highly coordinated and operate synchronously without deviation. Finally, the coupling and association binding of the static geometric structure and the dynamic control logic are completed, resulting in the digital twin framework of the target pump equipment.

[0029] The beneficial effects include detailed 3D modeling based on component relationships, which comprehensively restores the structural dimensions, internal structure, key features, and spatial positions of each component. Combined with virtual assembly implemented according to actual assembly processes and rigorous spatial verification, the static geometric structure can accurately reproduce the physical form and component relationships of the target pump equipment, without dimensional deviations or connection errors, providing a solid and accurate physical foundation for the digital twin framework. Through meticulous analysis of the working logic rules, the signal transmission path trigger condition response sequence is transformed into an executable digital instruction sequence containing multiple timing modules, clarifying the relationships and time intervals between instructions. This allows the dynamic control logic to completely replicate the actual control mechanism of the equipment, ensuring that the response to various operating condition changes is consistent with the actual equipment. The establishment of a dedicated data interaction interface and the creation of an association mapping table achieves precise docking between the static geometric structure and the dynamic control logic. The two-way data interaction closed loop ensures the coordinated operation of both, allowing the digital twin framework to possess a physical structure consistent with the actual equipment while accurately reproducing the equipment's operation and control process, without any disconnect between structure and function. This digital twin framework, which closely resembles the actual equipment, provides a reliable virtual platform for inputting actual operating data and conducting simultaneous virtual-real simulations. It ensures that the simulation process can realistically simulate the operating state of the equipment and provides accurate and effective data support for subsequent deviation analysis, degradation trajectory fitting, and other processes.

[0030] S3. Input the actual operating data of the target pump equipment into the digital twin framework, and perform an operational simulation of the digital twin framework to obtain the simulation operating data of the target pump equipment. In this embodiment of the invention, the actual operating data of the target pump equipment is input into a digital twin framework, and the digital twin framework is used for operational simulation to obtain the simulated operating data of the target pump equipment, including: By decoupling the parameters of the actual operating data of the target pump equipment, multi-dimensional operating parameters of the target pump equipment are obtained. By mapping multidimensional operating condition parameters to behavioral commands, virtual control commands for the digital twin framework are obtained. The digital twin framework is driven by virtual control commands to achieve collaborative state evolution; During the cooperative state evolution process, the digital twin framework is synchronously sampled to obtain the virtual physical quantities of the digital twin framework; According to the operating sequence of the target pump equipment, the virtual physical quantities are reorganized into the simulation operating data of the target pump equipment.

[0031] Complete operational data from various sensors and data acquisition modules are collected during the actual operation of the target pump equipment. This data encompasses a variety of comprehensive data types, including pressure, temperature, speed, flow rate, power, and vibration frequency. The collected operational data is categorized and organized according to the equipment operating status attributes reflected by the data, into different dimensions. For example, data reflecting the characteristics of medium transport is categorized into pressure and flow rate dimensions, data reflecting the equipment's heating state is categorized into temperature dimensions, data reflecting power output state is categorized into speed and power dimensions, and data reflecting equipment stability is categorized into vibration frequency dimensions. Data for each dimension is extracted independently, removing interfering information. The data source, numerical range, unit, and acquisition time interval for each dimension parameter are clearly defined to ensure that each parameter accurately reflects the equipment status under the corresponding operating conditions. The final result is a multi-dimensional operating condition parameter set containing multiple independent dimensions, complete data, and clearly defined attributes.

[0032] A pre-defined mapping rule is established between multi-dimensional operating condition parameters and actions within the digital twin framework. This rule clearly defines the framework's execution component, action type, action parameter range, and response priority for each operating condition parameter. For example, pressure parameters correspond to pipeline pressure regulation actions, temperature parameters correspond to the start / stop or speed regulation actions of cooling components, and speed parameters correspond to motor speed control actions. Each obtained multi-dimensional operating condition parameter is compared with the pre-defined mapping rule, and the corresponding action requirement is determined based on the parameter's specific value. For instance, when the pressure parameter reaches a specific value, it corresponds to the adjustment range of the pipeline valve opening; when the temperature parameter reaches a certain value, it corresponds to the starting speed of the cooling fan. The action requirements, execution components, and parameter thresholds for each operating condition parameter are integrated into a standardized instruction format. Each instruction includes key information such as the operating condition parameter identifier, execution component identifier, action type, action parameters, and execution time. The behavioral instruction conversion for all multi-dimensional operating condition parameters is completed one by one, resulting in virtual control instructions that the digital twin framework can recognize and execute.

[0033] The generated virtual control commands are sequentially input into the command receiving module of the digital twin framework according to execution priority and equipment operation logic. The command receiving module parses each virtual control command, extracting the execution component identifier and action requirements, and transmits the parsing results to the dynamic control logic module of the digital twin framework. The dynamic control logic module triggers the corresponding control flow according to the command requirements, driving the relevant virtual components in the static geometry to perform actions. For example, upon receiving a motor speed adjustment command, the dynamic control logic module activates the speed adjustment program of the motor virtual component, controlling the virtual motor to operate according to the speed parameters in the command, while simultaneously driving the associated virtual pump shaft and impeller to adjust their speeds synchronously; upon receiving a pipeline pressure adjustment command, it drives the virtual valve to operate at a specified opening degree, changing the flow cross-sectional area of ​​the pipeline virtual model, thereby adjusting the virtual pressure within the pipeline. Throughout the entire driving process, the dynamic control logic module coordinates the action sequence of each virtual component in real time, ensuring that the actions of different components cooperate without conflict, realizing the coordinated state evolution of each module of the digital twin framework, and accurately simulating the actual operation process of the target pump equipment.

[0034] During the collaborative state evolution process within the digital twin framework, a sampling interval is set that is completely consistent with the actual data acquisition frequency of the target pump equipment. For example, if the actual data is acquired every 0.01 seconds, the sampling interval is also set to 0.01 seconds. A synchronous state sampling program is initiated, collecting key physical quantity data at each sampling moment for each virtual component of the static geometric structure within the digital twin framework. This includes pressure, temperature, rotational speed, displacement, vibration frequency, and power consumption of each component, while simultaneously recording the acquisition time point for each physical quantity. During the sampling process, it is ensured that all relevant physical quantities of each virtual component are comprehensively acquired without missing any key parameters. The type and unit of the acquired virtual physical quantity data are consistent with those of the actual equipment physical quantities; for example, pressure is uniformly measured in megapascals, temperature in degrees Celsius, and rotational speed in revolutions per minute. All virtual physical quantity data from each sampling moment are temporarily stored in chronological order of acquisition time, forming a complete virtual physical quantity dataset.

[0035] Extract the acquisition time point of each data point in the virtual physical quantity dataset and use it as a timestamp to match the timestamp format of the actual operating data of the target pump equipment. Based on the timestamp, sort all virtual physical quantity data in chronological order to ensure that each time point corresponds to a complete set of virtual physical quantity data. For example, virtual physical quantity data such as pressure, temperature, speed, and flow rate at a certain moment form a set of correlated data. Referring to the storage format and data arrangement of the actual operating data of the target pump equipment, reorganize the sorted virtual physical quantity data. Different types of virtual physical quantities at the same time point are arranged in a preset order, and datasets from different time points are sequentially linked by timestamps, forming a dataset whose structure, format, and timing completely match the actual operating data. This ultimately yields the simulation operating data of the target pump equipment.

[0036] The beneficial effects are as follows: By decoupling refined parameters from actual operating data, multi-dimensional operating condition parameters can comprehensively and accurately reflect the operating status of various aspects of the equipment, providing a high-quality data foundation for subsequent command mapping; by pre-setting clear mapping rules to complete the conversion of multi-dimensional operating condition parameters into virtual control commands, the virtual control commands can accurately correspond to the actual operating needs of the equipment, ensuring the effectiveness and relevance of the commands; by using virtual control commands to drive the digital twin framework for collaborative state evolution, accurate simulation of the actual operation process of the target pump equipment is achieved, ensuring that the evolution state is highly consistent with the actual operating state of the equipment; synchronous state sampling at a unified frequency ensures the real-time performance and integrity of virtual physical quantity data; and by recombining virtual physical quantities according to the running sequence to obtain simulation operation data, the simulation data is consistent with the actual operation data in terms of structure and timing, providing reliable and adaptable data support for subsequent synchronous deviation analysis between simulation operation data and actual operation data, improving the accuracy and efficiency of deviation analysis.

[0037] S4. Perform synchronous deviation analysis on the simulation operation data and the actual operation data, and combine the historical operation data of the target pump equipment to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. In this embodiment of the invention, a synchronization deviation analysis is performed between the simulation running data and the actual running data, including: The simulation data and the actual data are timestamped to obtain the aligned simulation data and the aligned actual data of the target pump equipment. The aligned simulation data and the aligned actual data are correlated and matched to obtain the virtual and real data pairs of the target pump equipment. By performing a dimension-by-dimensional difference analysis on the virtual and real data pairs, the deviation data of the target pump equipment can be obtained.

[0038] By combining historical operating data of the target pump equipment, the evolution trend of the analyzed deviation data is fitted to obtain the degradation trajectory of the target pump equipment, including: Health status analysis is performed on the historical operating data of the target pump equipment to obtain the historical performance benchmark of the target pump equipment; Based on historical performance benchmarks, a relative degradation assessment is performed on the analyzed deviation data to obtain the performance degradation degree of the target pump equipment. By fitting the time-series trajectory of the performance degradation, the degradation trajectory of the target pump equipment is obtained.

[0039] The formula for calculating performance degradation is as follows: ; In the formula, Indicates at time Performance degradation, This indicates the total number of dimensions of performance parameters monitored for the target pump equipment. Indicates at time No. Deviation data values ​​for each performance parameter, Indicates the first The health status baseline value of each performance parameter Indicates the first The health status benchmark variance of each performance parameter This represents the summation operation. This represents the square root operation.

[0040] The time stamps recorded in the simulation data and the actual data are extracted. The time stamps adopt a unified format of year-month-day-hour-minute-second-millisecond. The time axis of the simulation data is adjusted based on the time stamps of the actual data. The time stamps of the simulation data and the actual data are compared one by one. Entries in the simulation data with inconsistent time stamps with the actual data are shifted and adjusted to ensure that the simulation data and the actual data have corresponding datasets at every identical time point. Finally, the aligned simulation data and aligned actual data are obtained.

[0041] The aligned simulation data and aligned actual data are categorized according to the type of performance parameters, into different parameter categories such as pressure, temperature, speed, flow rate, and vibration frequency. Within the same parameter category, the aligned simulation data and aligned actual data at the same time point are paired one-to-one. For example, simulated pressure data at a certain moment is paired with actual pressure data at the same moment, and simulated temperature data at a certain moment is paired with actual temperature data at the same moment. This ensures that the parameter type and time point of each paired data set are completely consistent, forming a virtual-real data pair for the target pump equipment.

[0042] For each pair of virtual and real data, the difference is calculated by subtracting numerical values. The aligned simulated data value of the same parameter category at the same time point is subtracted from the aligned actual data value, and the result of each calculation is recorded. The difference calculations for all virtual and real data pairs are completed one by one according to parameter category and time order, and the difference results of each parameter category at each time point are recorded in detail, finally obtaining the deviation data of the target pump equipment.

[0043] Historical operating data of the target pump equipment was retrieved, and datasets showing the equipment in a healthy operating state were selected. The criteria for healthy operating state were: no fault records within 1000 hours of continuous operation, all operating parameters within the design rated range, and fluctuations not exceeding 3% of the design value. The selected healthy operating data were categorized and statistically analyzed by parameter type. The average value of all data for each parameter type was calculated, and this average value became the health status benchmark value for that parameter type. The differences between all data for each parameter type and the health status benchmark value were calculated, and the squared average of these differences was calculated. Finally, the square root of the squared average was taken to obtain the health status benchmark variance for the corresponding parameter type. The health status benchmark values ​​and health status benchmark variances for all parameter types were integrated to form the historical performance benchmark of the target pump equipment.

[0044] For each time point, the deviation data value for each parameter category is collected. This deviation data value is then subtracted from the corresponding parameter category's health status benchmark value to obtain the difference between the deviation and the benchmark. This difference is divided by the variance of the corresponding parameter category's health status benchmark to obtain the relative deviation value. Each relative deviation value is squared to prevent positive and negative deviations from canceling each other out. The squared relative deviation values ​​for all parameter categories are summed, and the sum is divided by the total number of performance parameter dimensions monitored by the target pump equipment to obtain the average relative deviation squared value. The square root of the average relative deviation squared value is then performed to obtain the performance degradation degree at that time point. Calculations are completed sequentially for all time points to obtain a series of performance degradation data.

[0045] The system uses time as the horizontal axis, with units in hours, and each scale mark representing a fixed time interval consistent with the data collection interval. Performance degradation is plotted on the vertical axis, with units dimensionless, and scale marks evenly distributed across the possible range of performance degradation values. The performance degradation data for each time point is used as coordinate points, marked sequentially on the coordinate system. A continuous curve connects all coordinate points, ensuring a smooth curve that accurately reflects the transition in performance degradation between adjacent time points, ultimately forming a visually compelling degradation trajectory that illustrates the performance changes of the target pump over time.

[0046] The beneficial effects include: ensuring the time synchronization of aligned simulation and actual data through a unified time stamp and benchmark alignment method, providing a precise time basis for subsequent difference analysis; ensuring the accuracy of the correspondence between virtual and real data pairs through parameter category-based correlation matching, avoiding analysis errors caused by parameter confusion; obtaining comprehensive and accurate deviation data through a clear numerical calculation method for dimension-by-dimensional difference analysis, fully reflecting the differences between simulation and actual operation; ensuring the historical performance benchmark has reliable reference value based on health data selected according to clear standards, and ensuring the objectivity of the benchmark data through the calculation method of health status benchmark value and benchmark variance; integrating deviation information of all parameters in the performance degradation calculation process, realizing a comprehensive quantitative assessment of equipment performance degradation, and avoiding the one-sidedness of single parameter assessment; and clearly showing the evolution trend of equipment performance through time-series trajectory fitting, providing intuitive and accurate data support for subsequent fault risk assessment. The entire process is clear in steps and transparent in calculation, ensuring the reproducibility of the technical solution and the reliability of the results.

[0047] S5. Conduct fault risk assessment on the degradation trajectory to obtain the fault symptom categories and fault severity levels of the target pump equipment; In this embodiment of the invention, the degradation trajectory is analyzed for fault risk to obtain the fault symptom category and fault severity level of the target pump equipment, including: Morphological analysis of the degenerate trajectory yields information on its slope, curvature characteristics, and key inflection points. Heterogeneous feature decomposition is performed on the slope, curvature features and key inflection point information to obtain the multidimensional feature vector of the degenerate trajectory. Based on a pre-defined fault mode knowledge base, similarity matching is performed on multi-dimensional feature vectors to identify the fault symptom categories of degradation trajectories. Risk level determination is performed on the multidimensional feature vector to obtain the fault severity level of the target pump equipment.

[0048] The degradation trajectory is segmented by time intervals, which are consistent with the data collection interval for performance degradation, ensuring that each segment corresponds to a fixed duration of equipment operation. The difference in performance degradation values ​​between the two ends of each segment is calculated, and this difference is divided by the corresponding time length to obtain the slope of each segment. The slope value directly reflects the rate of equipment performance degradation within that time period. The curvature of each segment is analyzed by comparing the slopes of three adjacent segments and calculating the difference between adjacent slopes. The magnitude of the difference corresponds to the curvature characteristic; a larger difference indicates more pronounced trajectory curvature and a more drastic change in the rate of performance degradation. A threshold for the change in performance degradation is set, based on the normal fluctuation range of performance parameters in historical performance benchmarks. When the difference between the performance degradation value at a certain moment and the value at the previous moment exceeds this threshold, that moment is marked as a critical inflection point. The performance degradation value and time point corresponding to the critical inflection point are recorded, ultimately yielding the slope, curvature characteristics, and critical inflection point information of the degradation trajectory.

[0049] The slope, curvature, and key inflection point information are divided into three independent feature categories, each serving as a feature dimension. The slope feature is quantized, divided into different intervals based on numerical value. Each interval corresponds to a fixed quantized value; for example, a slope within a certain range corresponds to a quantized value of 1, another within a different range to quantized value 2, and so on. Curvature is quantized, with different levels assigned based on the difference between adjacent slopes. Each level is assigned a corresponding quantized value; a smaller difference results in a lower quantized value, and a larger difference results in a higher quantized value. Key inflection point information is quantized, converting the number of key inflection points, their corresponding performance degradation values, and their occurrence time into quantified indicators. For example, no inflection point corresponds to a quantized value of 0, one inflection point corresponds to a quantized value of 3, and earlier inflection points result in higher quantized values. The quantized values ​​of the three feature dimensions are arranged in a preset order to form a sequence containing multiple quantized feature values; this sequence constitutes the multidimensional feature vector of the degradation trajectory.

[0050] The pre-defined fault mode knowledge base stores multi-dimensional feature vector templates corresponding to various known faults. Each template explicitly corresponds to a fault symptom category, and each feature dimension in the template has a fixed quantization value range. The obtained multi-dimensional feature vectors are compared one by one with all templates in the knowledge base. During the comparison, the quantization value differences between the vector and the template are compared sequentially according to the feature dimensions. The absolute difference of the quantization value for each dimension is calculated, and the absolute differences of all dimensions are added together to obtain the total difference value. The template with the smallest total difference value is identified; the fault symptom category corresponding to this template is the fault symptom category corresponding to the current degradation trajectory. If the total difference value is zero, the fault symptom category corresponding to this template is directly matched, ensuring accurate identification of the fault symptom category.

[0051] Risk level assessment criteria are established, classifying risks into three levels: minor, moderate, and severe. Each level corresponds to a specific range of multidimensional feature vector quantization values. The criteria for a minor level are: the quantization value of the slope change in the multidimensional feature vector is in the lowest range; the quantization value of the curvature feature is in the lowest level; there are no critical inflection points or only one critical inflection point that appears relatively late. The criteria for a moderate level are: the quantization value of the slope change is in the middle range; the quantization value of the curvature feature is in the middle level; there are one to two critical inflection points; and the occurrence time is in the middle stage of the equipment's operating cycle. The criteria for a severe level are: the quantization value of the slope change is in the highest range; the quantization value of the curvature feature is in the highest level; there are at least three critical inflection points or the occurrence time is relatively early. The current multidimensional feature vector is comprehensively compared with the established assessment criteria. Based on the range within which the vector quantization value falls, the severity level of the target pump equipment's failure is determined.

[0052] The beneficial effects are as follows: Clear time segmentation, numerical calculation, and threshold setting provide precise quantitative basis for the slope, curvature features, and key inflection point information obtained from morphological analysis, ensuring the reliability of feature extraction; heterogeneous feature decomposition, through multi-dimensional quantification, transforms different types of features into multi-dimensional feature vectors in a unified format, providing a standardized data foundation for subsequent matching and judgment; similarity matching based on a preset fault mode knowledge base achieves accurate identification of fault symptom categories through dimension-by-dimensional difference calculation, avoiding misjudgment of faults; clear risk level judgment criteria make the classification of fault severity levels operable and consistent, ensuring objective and accurate judgment results. The entire process is clear in steps and quantitative standards, providing accurate and comprehensive fault information support for subsequent equipment health status assessment.

[0053] S6. Based on the fault symptom category and fault severity level, perform a health status assessment on the target pump equipment to generate a diagnostic report for the target pump equipment.

[0054] In this embodiment of the invention, a health status assessment of the target pump equipment is performed based on the fault symptom category and fault severity level to generate a diagnostic report for the target pump equipment, including: By comprehensively correlating the fault symptom categories, fault severity levels, and historical maintenance records of the target pump equipment, a comprehensive diagnostic evidence body for the target pump equipment is obtained. Based on the comprehensive diagnostic evidence, the life curve of the target pump equipment is extrapolated to obtain the remaining service life data of the target pump equipment. Health status rating is performed on the remaining service life data to obtain the health status score of the target pump equipment; The diagnostic report for the target pump equipment integrates fault symptom categories, fault severity levels, remaining service life data, and health status scores.

[0055] Retrieve historical maintenance records for the target pump equipment. These records include complete information such as the time of each maintenance, the corresponding fault type, the repair measures taken, the name and model of the replaced parts, post-repair performance test data, and the fault recurrence interval. Match the fault symptom category with the fault types in the historical maintenance records, filtering out all historical maintenance entries that match the current fault symptom category. Then, match records of the same or similar severity under the same fault type based on the fault severity level, clarifying the repair effectiveness and fault development cycle in these records. Comprehensively integrate the current equipment status data corresponding to the fault symptom category and fault severity level with the repair measures, part replacement information, performance recovery data, and recurrence interval in the filtered historical maintenance records, ensuring a clear correspondence for each dimension. This ultimately forms a comprehensive diagnostic evidence body containing fault information, historical maintenance experience, and performance change patterns.

[0056] Based on comprehensive diagnostic evidence, historical lifespan data of equipment under similar fault symptom categories and severity levels is extracted. This includes the time period from the appearance of fault symptoms to the inability to operate normally, the extended service life after maintenance, and the performance degradation rate at different stages of use. A lifespan curve of the historical equipment is plotted with time on the horizontal axis and equipment performance status on the vertical axis, clearly marking key locations such as fault occurrence points, maintenance nodes, and performance degradation inflection points. Combining the current target pump equipment's performance degradation rate, length of service, and current fault severity level with the trend of the historical lifespan curve, the corresponding position of the current equipment on the lifespan curve is determined. Using a linear extrapolation method, the curve is extended along the degradation trend of the historical lifespan curve to calculate the time period from the current time point until the equipment performance deteriorates to the point where it can no longer meet operational requirements. This time period is the remaining service life data of the target pump equipment.

[0057] The health status rating is set at a maximum score of 100 points, divided into three rating ranges: 80 points and above is excellent, 60 to 79 points is good, and below 60 points is unsatisfactory. Base scores are set based on remaining service life data: 80 points for remaining service life greater than 1000 hours, 60 points for remaining service life between 500 and 1000 hours, and 40 points for remaining service life less than 500 hours. The scores are adjusted based on the severity of the fault: no adjustment for minor faults, 10 points deducted from the base score for moderate faults, and 20 points deducted for severe faults. The base score and the adjusted score are added together to obtain the health status score of the target pump equipment. The score must accurately correspond to the relevant health rating range.

[0058] A standardized format for the diagnostic report is established, comprising four core sections. The first section clearly identifies the fault symptom category and describes the specific characteristics of the fault. The second section indicates the fault severity level, explaining its impact on equipment operation. The third section presents remaining service life data, specifying the timeframe during which the equipment can continue to operate normally. The fourth section lists the health status score and corresponding health rating, visually reflecting the equipment's current health level. Following this format, the specific details of the fault symptom category, fault severity level, remaining service life data, and health status score are entered into the corresponding sections, ensuring completeness, clarity, and no omission of key data in each section. This process ultimately integrates the information to form a well-structured and comprehensive diagnostic report for the target pump equipment.

[0059] The beneficial effects include: by comprehensively linking fault-related information with historical maintenance records, the integrated diagnostic evidence is supported by rich historical data and practical evidence, ensuring the reliability of subsequent lifespan estimation and health assessment; the extrapolation method based on historical lifespan curves, combined with the current actual state of the equipment, provides clear reference standards and scientific logic for the estimation of remaining service life data, resulting in accurate and reliable results; the health status rating, through clear score settings and adjustment rules, achieves quantitative assessment of health status, resulting in objective and consistent results; and the diagnostic report, integrated in a fixed format, comprehensively presents key information such as faults, lifespan, and health, providing clear, comprehensive, and directly referable basis for equipment operation and maintenance decisions, improving the pertinence and efficiency of operation and maintenance work.

[0060] like Figure 2 The diagram shown is a functional block diagram of a pump system performance monitoring and fault diagnosis system based on digital twins provided in an embodiment of the present invention.

[0061] The pump system performance monitoring and fault diagnosis system 100 based on digital twins of this invention can be installed in electronic devices. Depending on the functions implemented, the pump system performance monitoring and fault diagnosis system 100 based on digital twins may include a device mechanism and topology analysis module 101, a digital twin construction module 102, a virtual-real synchronous simulation and deduction module 103, a performance degradation trajectory evaluation module 104, a fault intelligent diagnosis module 105, and a health assessment and report generation module 106. The modules of this invention can also be referred to as units, which refer to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0062] In this embodiment, the functions of each module / unit are as follows: The equipment mechanism and topology parsing module 101 is used to perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment. The digital twin construction module 102 is used to perform digital three-dimensional mapping of the target pump equipment based on working logic rules and component relationships, so as to construct a digital twin framework of the target pump equipment. The virtual-real synchronous simulation and deduction module 103 is used to input the actual operating data of the target pump equipment into the digital twin framework, and to perform operation simulation and deduction on the digital twin framework to obtain the simulation operating data of the target pump equipment. The performance degradation trajectory evaluation module 104 is used to perform synchronous deviation analysis between simulation operation data and actual operation data, and combine the historical operation data of the target pump equipment to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. The intelligent fault diagnosis module 105 is used to assess the fault risk of the degradation trajectory and obtain the fault symptom category and fault severity level of the target pump equipment. The health assessment and report generation module 106 is used to assess the health status of the target pump equipment based on the fault symptom category and fault severity level, so as to generate a diagnostic report for the target pump equipment.

[0063] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0064] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0065] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0066] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0067] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for performance monitoring and fault diagnosis of pump systems based on digital twins, characterized in that, The methods include: S1. Perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment; S2. Based on the working logic rules and component relationships, perform digital three-dimensional mapping on the target pump equipment to construct a digital twin framework for the target pump equipment; S3. Input the actual operating data of the target pump equipment into the digital twin framework, and perform an operational simulation of the digital twin framework to obtain the simulation operating data of the target pump equipment. S4. Perform synchronous deviation analysis on the simulation operation data and the actual operation data, and combine the historical operation data of the target pump equipment to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. S5. Conduct fault risk assessment on the degradation trajectory to obtain the fault symptom categories and fault severity levels of the target pump equipment; S6. Based on the fault symptom category and fault severity level, perform a health status assessment on the target pump equipment to generate a diagnostic report for the target pump equipment.

2. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 1, characterized in that, Perform topology relation parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment, including: Obtain the mechanical assembly drawings and electrical control schematic diagrams of the target pump equipment; Based on the mechanical assembly drawings, the coupling relationships of the rotating parts, sealing parts and pipelines of the target pump equipment are deconstructed to obtain the component relationships of the target pump equipment. Based on the electrical control schematic diagram, the control signals, sensor signals and power execution components of the target pump equipment are analyzed for control dependency, and the working logic rules of the target pump equipment are obtained.

3. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 1, characterized in that, Based on logical rules and component relationships, a digital 3D mapping is performed on the target pump equipment to construct a digital twin framework for the target pump equipment, including: Based on the component relationships, the target pump equipment is reconstructed in three-dimensional space to obtain the static geometric structure of the target pump equipment. Based on logical rules, timing rules are injected into the target pump equipment to obtain the dynamic control logic of the target pump equipment. By coupling and associating the static geometric structure with the dynamic control logic, a digital twin framework of the target pump equipment is obtained.

4. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 1, characterized in that, The actual operating data of the target pump equipment is input into the digital twin framework, and the operation of the digital twin framework is simulated and extrapolated to obtain the simulation operating data of the target pump equipment, including: By decoupling the parameters of the actual operating data of the target pump equipment, multi-dimensional operating parameters of the target pump equipment are obtained. By mapping multidimensional operating condition parameters to behavioral commands, virtual control commands for the digital twin framework are obtained. The digital twin framework is driven by virtual control commands to achieve collaborative state evolution; During the cooperative state evolution process, the digital twin framework is synchronously sampled to obtain the virtual physical quantities of the digital twin framework; According to the operating sequence of the target pump equipment, the virtual physical quantities are reorganized into the simulation operating data of the target pump equipment.

5. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 1, characterized in that, Synchronization deviation analysis was performed between simulation data and actual operation data, including: The simulation data and the actual data are timestamped to obtain the aligned simulation data and the aligned actual data of the target pump equipment. The aligned simulation data and the aligned actual data are correlated and matched to obtain the virtual and real data pairs of the target pump equipment. By performing a dimension-by-dimensional difference analysis on the virtual and real data pairs, the deviation data of the target pump equipment can be obtained.

6. The method for performance monitoring and fault diagnosis of a pump system based on digital twin as described in claim 1, characterized in that, By combining historical operating data of the target pump equipment, the evolution trend of the analyzed deviation data is fitted to obtain the degradation trajectory of the target pump equipment, including: Health status analysis is performed on the historical operating data of the target pump equipment to obtain the historical performance benchmark of the target pump equipment; Based on historical performance benchmarks, a relative degradation assessment is performed on the analyzed deviation data to obtain the performance degradation degree of the target pump equipment. By fitting the time-series trajectory of the performance degradation, the degradation trajectory of the target pump equipment is obtained.

7. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 6, characterized in that, The formula for calculating performance degradation is as follows: ; In the formula, Indicates at time Performance degradation, This indicates the total number of dimensions of performance parameters monitored for the target pump equipment. Indicates at time No. Deviation data values ​​for each performance parameter, Indicates the first The health status baseline value of each performance parameter Indicates the first The health status benchmark variance of each performance parameter This represents the summation operation. This represents the square root operation.

8. The method for monitoring and diagnosing pump system performance based on digital twins as described in claim 1, characterized in that, Fault risk assessment is performed on the degradation trajectory to obtain the fault symptom categories and fault severity levels of the target pump equipment, including: Morphological analysis of the degenerate trajectory yields information on its slope, curvature characteristics, and key inflection points. Heterogeneous feature decomposition is performed on the slope, curvature features and key inflection point information to obtain the multidimensional feature vector of the degenerate trajectory. Based on a pre-defined fault mode knowledge base, similarity matching is performed on multi-dimensional feature vectors to identify the fault symptom categories of degradation trajectories. Risk level determination is performed on the multidimensional feature vector to obtain the fault severity level of the target pump equipment.

9. The method for performance monitoring and fault diagnosis of a pump system based on digital twin as described in claim 1, characterized in that, Based on the types of fault symptoms and the severity level of the fault, a health status assessment is performed on the target pump equipment to generate a diagnostic report for the target pump equipment, including: By comprehensively correlating the fault symptom categories, fault severity levels, and historical maintenance records of the target pump equipment, a comprehensive diagnostic evidence body for the target pump equipment is obtained. Based on the comprehensive diagnostic evidence, the life curve of the target pump equipment is extrapolated to obtain the remaining service life data of the target pump equipment. Health status rating is performed on the remaining service life data to obtain the health status score of the target pump equipment; The diagnostic report for the target pump equipment integrates fault symptom categories, fault severity levels, remaining service life data, and health status scores.

10. A pump system performance monitoring and fault diagnosis system based on digital twins, characterized in that, The system for implementing the digital twin-based pump system performance monitoring and fault diagnosis method of claim 1 includes: The equipment mechanism and topology parsing module is used to perform topology association parsing on the target pump equipment to obtain the working logic rules and component relationships of the target pump equipment. The digital twin construction module is used to perform digital three-dimensional mapping of the target pump equipment based on working logic rules and component relationships, so as to construct the digital twin framework of the target pump equipment; The virtual-real synchronous simulation and deduction module is used to input the actual operating data of the target pump equipment into the digital twin framework, and to perform operation simulation and deduction on the digital twin framework to obtain the simulation operating data of the target pump equipment. The performance degradation trajectory evaluation module is used to perform synchronous deviation analysis between simulation running data and actual running data, and combined with the historical running data of the target pump equipment, to fit the evolution trend of the analyzed deviation data to obtain the degradation trajectory of the target pump equipment. The intelligent fault diagnosis module is used to assess the fault risk of the degradation trajectory and obtain the fault symptom category and fault severity level of the target pump equipment; The health assessment and report generation module is used to assess the health status of the target pump equipment based on the type of fault symptoms and the severity level of the fault, so as to generate a diagnostic report for the target pump equipment.

Images