Remote fault diagnosis method and system for intelligent temperature transmitter

Abstract

The invention relates to the technical field of remote control, and discloses a remote fault diagnosis method and system for an intelligent temperature transmitter, and the method comprises the steps: defining a confidence distribution function of multi-source sensing data, carrying out the conflict elimination of the multi-source sensing data, and obtaining the conflict elimination data; extracting a cross-modal correlation feature matrix of the conflict elimination data, determining a thermodynamic transfer sub-equation of the temperature transmitter and an LSTM-PHY joint modeling framework, and constructing a hybrid digital twinborn model of the temperature transmitter; the normal working condition of the temperature transmitter is simulated through a mixed digital twin model, and the behavior deviation of the temperature transmitter is analyzed; constructing a hierarchical collaborative network of the temperature transmitters, and analyzing a multi-layer fault analysis result of the temperature transmitters by using the hierarchical collaborative network; fault optimization parameters of the temperature transmitter are determined, and fault diagnosis of the intelligent temperature transmitter is executed based on the fault optimization parameters. According to the invention, the fault analysis efficiency of the temperature transmitter can be improved.

Description

Technical Field

[0001] The present invention relates to a remote fault diagnosis method and system for an intelligent temperature transmitter, belonging to the technical field of remote control. Background Art

[0002] Remote fault diagnosis of a temperature transmitter refers to a technical system that, through technical means such as the Internet of Things and artificial intelligence, monitors the operating state of the temperature transmitter in real time, analyzes data, and judges faults, and realizes early warning, accurate positioning, and remote disposal of faults without on-site manual intervention.

[0003] Traditional remote fault diagnosis of temperature transmitters usually relies on simple monitoring and regular inspections. The device status is judged by comparing real-time data with preset thresholds. Once an anomaly is found, technicians are notified to conduct on-site inspections and repairs. This diagnostic method relies on limited single signals, lacks comprehensive analysis of multi-source data, and is difficult to achieve accurate fault prediction and positioning, resulting in low diagnostic efficiency. Summary of the Invention

[0004] The present invention provides a remote fault diagnosis method and system for an intelligent temperature transmitter, and its main purpose is to improve the fault analysis efficiency of the temperature transmitter.

[0005] To achieve the above purpose, a remote fault diagnosis method for an intelligent temperature transmitter provided by the present invention includes: Collect multi-source sensing data of the temperature transmitter, where the multi-source sensing data includes temperature signal time-series data, power supply current harmonic characteristic data, device vibration spectrum data, and environmental parameters. Define the confidence assignment function of the multi-source sensing data, and eliminate conflicts of the multi-source sensing data through the confidence assignment function to obtain conflict-eliminated data; Extract the cross-modal correlation feature matrix of the conflict-eliminated data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct a hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework; Simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix; Construct a hierarchical collaborative network of the temperature transmitter, where the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer, and analyze the multi-layer fault analysis results of the temperature transmitter by using the hierarchical collaborative network according to the behavior deviation; Determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

[0006] Optionally, the confidence assignment function for defining the multi-source sensing data includes: Preprocess the multi-source sensing data to obtain processed multi-source sensing data; Analyze the confidence influence factors of the processed multi-source sensing data, where the confidence influence factors include data source accuracy, data source stability, data source reliability, and environmental impact coefficient; Define the factor weights of the confidence influence factors; Based on the confidence influence factors and the factor weights, use the following formula to construct the confidence assignment function for the multi-source sensing data, where the confidence assignment function includes: ; Wherein, represents the confidence of the c-th data source of the multi-source sensing data, represents the value of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor, represents the maximum value of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor, represents the factor weight of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor.

[0007] Optionally, the conflict elimination of the multi-source sensing data through the confidence assignment function to obtain conflict elimination data includes: Analyze the confidence of the multi-source sensing data through the confidence assignment function; Based on the multi-source sensing data, determine the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term of the temperature transmitter corresponding to the multi-source sensing data; Analyze the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term; Analyze the data conflict of the multi-source sensing data according to the energy deviation value; Perform conflict elimination on the multi-source sensing data through the data conflict and the confidence to obtain conflict elimination data.

[0008] Optionally, the analysis of the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term includes: According to the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term, use the following formula to calculate the heat conversion rate of the temperature transmitter: ; Among them, represents the heat conversion rate of the temperature transmitter, represents the heat transfer coefficient, represents the effective heat transfer area, represents the temperature difference, represents the radiative heat transfer term, represents the Stefan-Boltzmann constant, represents the surface emissivity, represents the radiative surface area, represents the absolute surface temperature; Based on the actually measured heat change rate and the heat conversion rate, calculate the energy deviation value of the temperature transmitter.

[0009] Optionally, the cross-modal correlation feature matrix for extracting the conflict resolution data includes: Establish a coordinate system transformation matrix for the conflict resolution data; According to the coordinate system transformation matrix, perform spatial registration on the conflict resolution data to obtain registered data; Extract the single-modal features of the registered data; Construct a graph structure for the single-modal features; Through the graph structure, construct the cross-modal correlation feature matrix for the conflict resolution data.

[0010] Optionally, the determination of the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter includes: Divide the temperature transmitter into multiple control volumes; Determine the heat balance equations of the multiple control volumes; According to the heat balance equations, construct the thermodynamic transfer sub-equation of the temperature transmitter; Determine the LSTM-PHY network structure of the temperature transmitter; Integrate the thermodynamic transfer sub-equation into the LSTM-PHY network structure to obtain the LSTM-PHY joint modeling framework of the temperature transmitter.

[0011] Optionally, the construction of the thermodynamic transfer sub-equation of the temperature transmitter according to the heat balance equations includes: According to the heat balance equations, analyze the time constant, pure delay time, thermo-elastic coupling coefficient, mechanical vibration natural frequency, and damping ratio of the temperature transmitter; Based on the time constant, pure delay time, thermo-elastic coupling coefficient, mechanical vibration natural frequency, and damping ratio, use the following formula to construct the thermodynamic transfer sub-equation of the temperature transmitter, where the thermodynamic transfer sub-equation: ; wherein, represents the thermodynamic transfer sub - equation, represents the time constant, represents the pure delay time, represents the Laplace variable, represents the exponential function, represents the thermo - elastic coupling coefficient, represents the damping ratio, represents the natural frequency of mechanical vibration.

[0012] Optionally, analyzing the behavior deviation of the temperature transmitter based on the normal condition and the cross - modal correlation feature matrix includes: Analyzing the feature distribution of the normal condition; Normalizing the cross - modal correlation feature matrix to obtain a normalized correlation feature matrix; Defining the deviation index of the temperature transmitter; Analyzing the behavior deviation of the temperature transmitter based on the deviation index, the normalized correlation feature matrix, and the feature distribution.

[0013] Optionally, constructing the hierarchical collaborative network of the temperature transmitter includes: Constructing the hierarchical collaborative structure of the temperature transmitter, wherein the hierarchical collaborative structure includes an edge - layer structure, a fog - node - layer structure, and a cloud - layer structure; Performing network integration on the edge - layer structure, the fog - node - layer structure, and the cloud - layer structure to obtain a hierarchical collaborative network structure; Defining the collaborative analysis rules of the hierarchical collaborative network structure; Analyzing the inter - layer interoperability of the hierarchical collaborative network structure; When the inter - layer interoperability meets the preset inter - layer interoperability threshold, constructing the hierarchical collaborative network of the temperature transmitter based on the hierarchical collaborative network structure and the collaborative analysis rules.

[0014] To solve the above problems, the present invention further provides an intelligent temperature transmitter remote fault diagnosis system, and the system includes: A data conflict elimination module, configured to collect multi - source sensing data of the temperature transmitter, wherein the multi - source sensing data includes temperature signal time - series data, power supply current harmonic feature data, equipment vibration spectrum data, and environmental parameters, define a confidence assignment function for the multi - source sensing data, and eliminate conflicts of the multi - source sensing data through the confidence assignment function to obtain conflict - eliminated data; The twin model construction module is used to extract the cross-modal correlation feature matrix of the conflict elimination data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct the hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework; The behavior deviation analysis module is used to simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix; The multi-layer fault analysis module is used to construct a hierarchical collaborative network of the temperature transmitter, where the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer, and analyze the multi-layer fault analysis results of the temperature transmitter according to the behavior deviation by using the hierarchical collaborative network; The fault parameter optimization module is used to determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform the fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

[0015] Compared with the problems in the background technology, first, the confidence assignment function of multi-source sensing data effectively eliminates data conflicts, improves data quality and reliability, and obtains more comprehensive and accurate device status information by integrating temperature signal time-series data, power supply current harmonic feature data, equipment vibration spectrum data, and environmental parameters. Second, the extraction of the cross-modal correlation feature matrix, combined with the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework, enables the hybrid digital twin model to accurately simulate the normal working condition of the temperature transmitter, providing strong support for behavior deviation analysis. Third, the construction of the hierarchical collaborative network realizes the distributed processing and intelligent analysis of data, improves the efficiency and accuracy of fault diagnosis, and the collaborative work of the edge layer, fog node layer, and cloud layer makes the fault analysis results more comprehensive and in-depth, providing a scientific basis for fault prevention and maintenance decision-making. In addition, the fault optimization parameters determined based on the multi-layer fault analysis results effectively guide the fault diagnosis of the intelligent temperature transmitter. The optimization of these parameters helps to reduce the fault occurrence rate, extend the equipment life, and reduce the maintenance cost. Finally, the application of the entire method significantly improves the operation reliability of the temperature transmitter, ensuring the continuity and safety of the production process. Therefore, the present invention can improve the fault analysis efficiency of the temperature transmitter. Description of the Drawings

[0016] Figure 1 It is a schematic flowchart of the remote fault diagnosis method for the intelligent temperature transmitter provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the module for implementing the remote fault diagnosis method for the intelligent temperature transmitter provided by an embodiment of the present invention.

[0017] The implementation, functional features and advantages of the present invention will be further described in conjunction with embodiments with reference to the accompanying drawings. Specific Embodiments

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

[0019] The embodiments of the present application provide a method for remote fault diagnosis of an intelligent temperature transmitter. The execution subject of the method for remote fault diagnosis of the intelligent temperature transmitter includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiments of the present application. In other words, the method for remote fault diagnosis of the intelligent temperature transmitter 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 cloud server cluster, etc.

[0020] Embodiment 1: Referring to Figure 1 As shown, it is a schematic flow chart of a method for remote fault diagnosis of an intelligent temperature transmitter provided by an embodiment of the present invention. In this embodiment, the method for remote fault diagnosis of the intelligent temperature transmitter includes: S1. Collect multi-source sensing data of the temperature transmitter. Among them, the multi-source sensing data includes temperature signal time series data, power supply current harmonic feature data, device vibration spectrum data, and environmental parameters. Define a confidence assignment function for the multi-source sensing data, and eliminate conflicts for the multi-source sensing data through the confidence assignment function to obtain conflict-eliminated data.

[0021] It should be explained that the temperature signal time series data refers to the sequence of temperature readings recorded by the temperature transmitter over time. The power supply current harmonic feature data is the harmonic component data obtained by analyzing the power supply current waveform. The device vibration spectrum data is the spectrum analysis result of the device vibration signal collected by a vibration sensor. The environmental parameters refer to the environmental condition data that affects the performance of the temperature transmitter, and may include, but are not limited to, data such as environmental temperature, humidity, pressure, wind speed, pollution level, etc.

[0022] The present invention defines the confidence assignment function of the multi-source sensing data, which can provide a basis for subsequent data conflict elimination.

[0023] Specifically, the defining of the confidence assignment function for the multi-source sensing data includes: Perform preprocessing on the multi-source sensing data to obtain processed multi-source sensing data; Analyze the confidence influence factors for processing multi-source sensing data, where the confidence influence factors include data source accuracy, data source stability, data source reliability, and environmental influence coefficient; Define the factor weights of the confidence influence factors; Based on the confidence influence factors and the factor weights, use the following formula to construct the confidence assignment function for the multi-source sensing data, where the confidence assignment function includes: ; Where, represents the confidence of the c-th data source of the multi-source sensing data, represents the value of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor, represents the maximum value of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor, represents the factor weight of the c-th data source of the multi-source sensing data corresponding to the confidence influence factor.

[0024] Where, the processing of multi-source sensing data refers to the preprocessed sensor data, the data source accuracy refers to the degree of proximity between the sensor measurement value and the true value, the data source stability refers to the degree of change of the sensor output over a long period of time, the data source reliability refers to the ability of the sensor to work without failure under specific conditions, the environmental influence coefficient refers to the degree of influence of environmental factors (such as temperature, humidity, pressure, etc.) on the sensor readings, the factor weight refers to the relative importance of each confidence influence factor in the overall confidence evaluation, the confidence assignment function refers to the algorithm for calculating the confidence of each data source, and the maximum value of the confidence influence factor refers to the maximum value that each confidence influence factor can reach.

[0025] The present invention effectively eliminates conflicts in the multi-source sensing data through the conflict elimination of the multi-source sensing data by the confidence assignment function, and the obtained conflict elimination data can effectively eliminate conflicts in the multi-source sensing data and improve the reliability of the data and the overall performance of the system.

[0026] Specifically, the conflict elimination of the multi-source sensing data by the confidence assignment function to obtain conflict elimination data includes: Analyze the confidence of the multi-source sensing data through the confidence assignment function; Based on the multi-source sensing data, determine the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term of the temperature transmitter corresponding to the multi-source sensing data; Analyze the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term; Analyze the data conflict of the multi-source sensing data according to the energy deviation value; Eliminate the conflict of the multi-source sensing data through the data conflict and the confidence level to obtain conflict-eliminated data.

[0027] Among them, the confidence level is a quantitative index of the reliability of each data source, reflecting the credibility of the sensor data under the current working conditions. The heat transfer coefficient is a physical quantity representing heat conduction / convection efficiency. The effective heat transfer area is the actual surface area of the temperature transmitter participating in heat exchange. The temperature difference is the temperature gradient between the temperature transmitter and the environment or adjacent components. The radiative heat transfer term is the radiative energy calculated by the Stefan-Boltzmann law, which is proportional to the surface emissivity and the fourth power of the temperature. The energy deviation value is the absolute difference between the measured heat change rate and the predicted value of the theoretical model. The data conflict is the inconsistency between multi-source data that violates physical laws or statistical laws. The conflict-eliminated data is the fused data obtained by removing data conflicts from multi-source sensing data after confidence level weighting and physical verification.

[0028] Further, the analyzing the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term includes: Calculate the heat change rate of the temperature transmitter according to the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term by using the following formula: ; Among them, represents the heat change rate of the temperature transmitter, represents the heat transfer coefficient, represents the effective heat transfer area, represents the temperature difference, represents the radiative heat transfer term, represents the Stefan-Boltzmann constant, represents the surface emissivity, represents the radiative surface area, represents the absolute surface temperature; Calculate the energy deviation value of the temperature transmitter based on the pre-measured actual heat change rate and the heat change rate.

[0029] Among them, the heat conversion rate refers to the rate of change of thermal energy predicted by the theoretical model, the actual heat change rate refers to the heat flow change directly calculated from the measured values of the temperature transmitter, the Stefan-Boltzmann constant refers to the proportionality constant of the blackbody radiation law, usually 5.67×10−8W / m2⋅K4, the surface emissivity refers to the ratio of the radiation ability of the surface of the temperature transmitter material to that of an ideal blackbody, the radiation surface area refers to the effective surface area participating in thermal radiation, and the surface absolute temperature refers to the Kelvin temperature at the measurement point of the temperature transmitter.

[0030] S2. Extract the cross-modal correlation feature matrix of the conflict elimination data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct the hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework.

[0031] The cross-modal correlation feature matrix of the conflict elimination data extracted by the present invention can provide a data basis for later fault analysis.

[0032] Specifically, the extraction of the cross-modal correlation feature matrix of the conflict elimination data includes: Establish a coordinate system transformation matrix for the conflict elimination data; Perform spatial registration on the conflict elimination data according to the coordinate system transformation matrix to obtain registered data; Extract the single-modal features of the registered data; Construct a graph structure of the single-modal features; Construct the cross-modal correlation feature matrix of the conflict elimination data through the graph structure.

[0033] Among them, the coordinate system transformation matrix refers to the transformation matrix describing the spatial relationship of measurement data of different sensors, including rotation and translation parameters, the registered data refers to the multi-source data after spatial alignment, ensuring that all sensor data is in the same coordinate system, the single-modal features refer to the time-domain / frequency-domain / statistical features extracted from single-sensor data, the graph structure refers to the topological structure representing the association relationship between modalities, the nodes are feature vectors, and the edge weights reflect the coupling strength, and the cross-modal correlation feature matrix refers to the matrix quantifying the non-linear relationship between multi-modal features and is used for comprehensive fault diagnosis.

[0034] Optionally, the graph structure of the single-modal features can be constructed through a graph attention network.

[0035] The present invention determines that the thermodynamic transfer sub-equation of the temperature transmitter and the LSTM-PHY joint modeling framework can establish a model that combines deep learning and physical knowledge. This model can more accurately predict the behavior of the temperature transmitter and improve the understanding of complex thermodynamic processes.

[0036] Specifically, the determination of the thermodynamic transfer sub-equation of the temperature transmitter and the LSTM-PHY joint modeling framework includes: Dividing the temperature transmitter into multiple control volumes; Determining the heat balance equations of the multiple control volumes; Constructing the thermodynamic transfer sub-equation of the temperature transmitter according to the heat balance equations; Determining the LSTM-PHY network structure of the temperature transmitter; Integrating the thermodynamic transfer sub-equation into the LSTM-PHY network structure to obtain the LSTM-PHY joint modeling framework of the temperature transmitter.

[0037] Among them, the multiple control volumes refer to dividing the device into multiple control volumes in the temperature transmitter, which means regarding different parts of the transmitter (such as sensor elements, electronic components, housing, etc.) as independent volume units. The heat balance equation refers to the equation that describes how heat is exchanged through conduction, convection, and radiation within a control volume, and how it interacts with other forms of energy within the system. The thermodynamic transfer sub-equation refers to the equation that describes how heat is transferred between multiple control volumes. The LSTM-PHY network structure refers to the neural network structure that combines the long short-term memory network (LSTM) and the physical model (PHY). The LSTM-PHY joint modeling framework refers to the integrated framework that combines the sequence data processing ability of the LSTM network and the physical model.

[0038] Furthermore, the construction of the thermodynamic transfer sub-equation of the temperature transmitter according to the heat balance equations includes: Analyzing the time constant, pure delay time, thermoelastic coupling coefficient, mechanical vibration natural frequency, and damping ratio of the temperature transmitter according to the heat balance equations; Based on the time constant, pure delay time, thermoelastic coupling coefficient, mechanical vibration natural frequency, and damping ratio, constructing the thermodynamic transfer sub-equation of the temperature transmitter using the following formula, where the thermodynamic transfer sub-equation: ; Where represents the thermodynamic transfer sub-equation, represents the time constant, represents the pure delay time, denotes the Laplace variable, denotes the exponential function, denotes the thermo-elastic coupling coefficient, denotes the damping ratio, denotes the natural frequency of mechanical vibration.

[0039] Among them, the time constant refers to the time required for the system response to reach 63.2% of its final value, which reflects the speed of the thermal dynamic characteristics of the temperature transmitter system. The pure delay time refers to the time lag between the input change and the start of the system response change. The thermo-elastic coupling coefficient refers to the degree of change in mechanical stress and strain caused by temperature change. The natural frequency of mechanical vibration refers to the frequency of free vibration of the system without external force. The damping ratio refers to the ratio of the damping force to the critical damping force, which determines the speed of vibration attenuation. The Laplace variable refers to a complex variable used in the Laplace transform to convert differential equations in the time domain into algebraic equations in the frequency domain.

[0040] According to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the present invention, a hybrid digital twin model of the temperature transmitter can be constructed. This model combines the traditional accuracy of the physical model and the prediction ability of the deep learning model, thus providing a powerful analysis and optimization tool for the temperature transmitter. Among them, the hybrid digital twin model refers to a composite model integrating the thermodynamic transfer sub-equation (physical model) and the LSTM-PHY joint modeling framework (data-driven model). Specifically, the hybrid digital twin model is realized by integrating the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework into a pre-constructed digital twin framework.

[0041] S3. Simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix.

[0042] It should be explained that the normal working condition refers to the scenario behavior state of simulating the normal operation of the temperature transmitter by inputting multi-source sensing data into the hybrid digital twin model.

[0043] Based on the normal working condition and the cross-modal correlation feature matrix of the present invention, analyzing the behavior deviation of the temperature transmitter can systematically analyze the deviation of the temperature transmitter, thereby realizing the monitoring and maintenance of the equipment performance.

[0044] Specifically, analyzing the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix includes: Analyze the characteristic distribution of the normal working condition; Normalize the cross-modal correlation feature matrix to obtain a normalized correlation feature matrix; Define the deviation index of the temperature transmitter; Based on the deviation index, the normalized correlation feature matrix, and the feature distribution, analyze the behavior deviation of the temperature transmitter.

[0045] Among them, the feature distribution refers to the statistical characteristics of the feature values extracted from the data, including but not limited to mean, median, standard deviation, variance, distribution shape (such as normal distribution, skewed distribution), range, quartiles, etc. The normalized correlation feature matrix refers to the matrix obtained by performing a mathematical transformation on the cross-modal correlation feature matrix, where the feature values have been adjusted to a common scale so that different features can be directly compared. The deviation index refers to the elements in the feature matrix that are significantly different from the normal operating conditions, including indicators such as abnormal temperature changes, abnormal current harmonics, and abnormal vibration amplitudes. The behavior deviation refers to the difference between the actual behavior of the temperature transmitter and the expected normal behavior.

[0046] Optionally, the feature distribution of the normal operating conditions can be analyzed by statistical methods.

[0047] S4. Construct a hierarchical collaborative network for the temperature transmitter. Among them, the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer. According to the behavior deviation, use the hierarchical collaborative network to analyze the multi-layer fault analysis results of the temperature transmitter.

[0048] The present invention constructs a hierarchical collaborative network for the temperature transmitter. Among them, the hierarchical collaborative network including an edge layer, a fog node layer, and a cloud layer can construct an efficient and reliable hierarchical collaborative network for real-time monitoring, data analysis, and intelligent control of the temperature transmitter.

[0049] Specifically, the construction of the hierarchical collaborative network for the temperature transmitter includes: Construct a hierarchical collaborative structure for the temperature transmitter. Among them, the hierarchical collaborative structure includes an edge layer structure, a fog node layer structure, and a cloud layer structure; Perform network integration on the edge layer structure, the fog node layer structure, and the cloud layer structure to obtain a hierarchical collaborative network structure; Define the collaborative analysis rules of the hierarchical collaborative network structure; Analyze the inter-layer interoperability of the hierarchical collaborative network structure; When the inter-layer interoperability meets the preset inter-layer interoperability threshold, based on the hierarchical collaborative network structure and the collaborative analysis rules, construct the hierarchical collaborative network for the temperature transmitter.

[0050] Among them, the edge layer structure refers to the computing and storage resources near the data source (such as a temperature transmitter). It is responsible for collecting data, performing local processing, and preliminary analysis. The fog node layer structure refers to the intermediate layer between the edge layer and the cloud. It usually includes relatively strong computing and storage capabilities, and is used to process data from the edge layer and perform more complex computing tasks. The cloud layer structure refers to the high-performance computing and storage resources at the top layer of the network. It is responsible for storing a large amount of data, performing complex data analysis, and machine learning model training. The hierarchical collaborative network structure refers to the network architecture that integrates the edge layer, fog node layer, and cloud layer. It can achieve distributed processing and analysis of data, as well as effective management of resources. The collaborative analysis rules refer to the set of rules that define how different network layers cooperate in analysis and decision-making, including data transmission protocols, data processing flows, selection of analysis models, etc. The inter-layer interoperability refers to the ability of different network layers (edge layer, fog node layer, cloud layer) to communicate and cooperate effectively, including data exchange, control instruction transmission, and task coordination. The inter-layer interoperability threshold refers to the interoperability performance standard set to ensure the normal operation of the network, such as the minimum requirements for indicators such as data transmission delay, bandwidth, and reliability. The hierarchical collaborative network refers to a network system that integrates the edge layer, fog node layer, and cloud layer. It can achieve real-time data collection, distributed processing, and intelligent decision-making to improve the monitoring and control efficiency of the temperature transmitter.

[0051] The present invention utilizes the hierarchical collaborative network to analyze the multi-layer fault analysis results of the temperature transmitter. The multi-layer fault analysis results of the temperature transmitter can be effectively analyzed by the hierarchical collaborative network, thereby improving the accuracy and efficiency of fault diagnosis. Among them, the multi-layer fault analysis results refer to the output results of fault diagnosis and analysis of the temperature transmitter at different network levels (edge layer, fog node layer, cloud layer). The multi-layer fault analysis results include fault diagnosis: the fault diagnosis information obtained by each layer after analyzing the data, such as the fault type, occurrence time, duration, etc.; fault mode: the law and mode of fault occurrence identified through data analysis; fault cause: the inference of the root cause of the fault, which may include equipment aging, environmental factors, operation errors, etc.; fault impact: the impact assessment of the fault on the performance of the temperature transmitter and the operation of the entire system.

[0052] S5. Determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

[0053] Based on the above multi-layer fault analysis results, determining the fault optimization parameters of the temperature transmitter can effectively optimize the fault parameters of the temperature transmitter, improving its reliability and operating efficiency. Among them, the fault optimization parameters refer to a series of parameters adjusted to reduce the frequency of faults, reduce the impact of faults, improve the reliability of equipment and maintenance efficiency, including parameters such as threshold setting, warning conditions, and maintenance cycles.

[0054] Compared with the problems described in the background art, first, the confidence assignment function of multi-source sensing data effectively eliminates data conflicts, improves data quality and reliability, and obtains more comprehensive and accurate equipment status information by integrating temperature signal time-series data, power supply current harmonic feature data, equipment vibration spectrum data, and environmental parameters. Second, the extraction of the cross-modal correlation feature matrix, combined with the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework, enables the hybrid digital twin model to accurately simulate the normal working conditions of the temperature transmitter, providing strong support for behavior deviation analysis. Third, the construction of the hierarchical collaborative network realizes the distributed processing and intelligent analysis of data, improves the efficiency and accuracy of fault diagnosis, and the collaborative work of the edge layer, fog node layer, and cloud layer makes the fault analysis results more comprehensive and in-depth, providing a scientific basis for fault prevention and maintenance decision-making. In addition, the fault optimization parameters determined based on the multi-layer fault analysis results effectively guide the fault diagnosis of the intelligent temperature transmitter. The optimization of these parameters helps to reduce the failure rate, extend the equipment life, and reduce the maintenance cost. Finally, the application of the entire method significantly improves the operating reliability of the temperature transmitter, ensuring the continuity and safety of the production process. Therefore, the present invention can improve the fault analysis efficiency of the temperature transmitter.

[0055] Embodiment 2: As Figure 2 shown, it is a functional module diagram of a remote fault diagnosis system for an intelligent temperature transmitter according to the present invention.

[0056] The remote fault diagnosis system 200 for an intelligent temperature transmitter according to the present invention can be installed in an electronic device. According to the functions achieved, the remote fault diagnosis system for an intelligent temperature transmitter can include a data conflict elimination module 201, a twin model construction module 202, a behavior deviation analysis module 203, a multi-layer fault analysis module 204, and a fault parameter optimization module 205. The modules in the present 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 complete fixed functions, and are stored in the memory of the electronic device.

[0057] In the embodiments of the present invention, the functions of each module / unit are as follows: The data conflict elimination module 201 is used to collect multi-source sensing data of the temperature transmitter. Among them, the multi-source sensing data includes temperature signal timing data, power supply current harmonic characteristic data, device vibration spectrum data, and environmental parameters. Define the confidence assignment function of the multi-source sensing data, and eliminate conflicts for the multi-source sensing data through the confidence assignment function to obtain conflict-eliminated data; The twin model construction module 202 is used to extract the cross-modal correlation feature matrix of the conflict-eliminated data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct the hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework; The behavior deviation analysis module 203 is used to simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix; The multi-layer fault analysis module 204 is used to construct a hierarchical collaborative network of the temperature transmitter. Among them, the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer, and analyzes the multi-layer fault analysis results of the temperature transmitter by using the hierarchical collaborative network according to the behavior deviation; The fault parameter optimization module 205 is used to determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

[0058] Specifically, each module in the intelligent temperature transmitter remote fault diagnosis system 200 in the embodiment of the present invention adopts the same technical means as the intelligent temperature transmitter remote fault diagnosis method described above Figure 1 and can produce the same technical effects, which will not be elaborated here.

[0059] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A remote fault diagnosis method for an intelligent temperature transmitter, characterized in that The method includes: Collect multi-source sensing data of a temperature transmitter. Among them, the multi-source sensing data includes temperature signal time-series data, power supply current harmonic characteristic data, device vibration spectrum data, and environmental parameters. Define a confidence assignment function for the multi-source sensing data, and eliminate conflicts in the multi-source sensing data through the confidence assignment function to obtain conflict-eliminated data; Extract the cross-modal correlation feature matrix of the conflict-eliminated data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct a hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework; Simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix; Construct a hierarchical collaborative network of the temperature transmitter. Among them, the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer. Analyze the multi-layer fault analysis results of the temperature transmitter by using the hierarchical collaborative network according to the behavior deviation; Determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

2. The remote fault diagnosis method of the intelligent temperature transmitter according to claim 1, characterized in that, The defining of the confidence assignment function for the multi-source sensing data includes: Preprocess the multi-source sensing data to obtain processed multi-source sensing data; Analyze the confidence influence factors of the processed multi-source sensing data. Among them, the confidence influence factors include data source accuracy, data source stability, data source reliability, and environmental influence coefficient; Define the factor weights of the confidence influence factors; Based on the confidence influence factors and the factor weights, use the following formula to construct the confidence assignment function for the multi-source sensing data. Among them, the confidence assignment function includes: ； Among them, represents the confidence of the c-th data source of the multi-source sensing data, represents the value of the confidence influence factor corresponding to the c-th data source of the multi-source sensing data for the confidence influence factor, represents the maximum value of the confidence influence factor corresponding to the c-th data source of the multi-source sensing data for the confidence influence factor, represents the factor weight of the confidence influence factor corresponding to the c-th data source of the multi-source sensing data for the confidence influence factor.

3. The remote fault diagnosis method for the intelligent temperature transmitter according to claim 2, characterized in that, The eliminating of conflicts in the multi-source sensing data through the confidence assignment function to obtain conflict-eliminated data includes: Analyze the confidence of the multi-source sensing data through the confidence assignment function; Based on the multi-source sensing data, determine the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term of the temperature transmitter corresponding to the multi-source sensing data; Analyze the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term; Analyze the data conflict of the multi-source sensing data according to the energy deviation value; Eliminate conflicts in the multi-source sensing data through the data conflict and the confidence to obtain conflict-eliminated data.

4. The remote fault diagnosis method of the intelligent temperature transmitter according to claim 3, wherein The analyzing of the energy deviation value of the temperature transmitter through the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term includes: According to the heat transfer coefficient, effective heat transfer area, temperature difference, and radiative heat transfer term, use the following formula to calculate the heat change rate of the temperature transmitter: ； Among them, represents the heat conversion rate of the temperature transmitter, represents the heat transfer coefficient, represents the effective heat transfer area, represents the temperature difference, represents the radiative heat transfer term, represents the Stefan - Boltzmann constant, represents the surface emissivity, represents the radiative surface area, represents the absolute surface temperature; Calculate the energy deviation value of the temperature transmitter based on the actually measured heat change rate and the heat change rate measured in advance.

5. The remote fault diagnosis method for the intelligent temperature transmitter according to claim 4, characterized in that, The cross-modal correlation feature matrix for extracting the conflict elimination data includes: Establish a coordinate system transformation matrix for the conflict elimination data; Perform spatial registration on the conflict elimination data according to the coordinate system transformation matrix to obtain registered data; Extract the single-modal features of the registered data; Construct a graph structure for the single-modal features; Construct the cross-modal correlation feature matrix of the conflict elimination data through the graph structure.

6. The remote fault diagnosis method for the intelligent temperature transmitter according to claim 5, characterized in that, The determination of the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter includes: Divide the temperature transmitter into multiple control volumes; Determine the heat balance equations of the multiple control volumes; Construct the thermodynamic transfer sub-equation of the temperature transmitter according to the heat balance equations; Determine the LSTM-PHY network structure of the temperature transmitter; Integrate the thermodynamic transfer sub-equation into the LSTM-PHY network structure to obtain the LSTM-PHY joint modeling framework of the temperature transmitter.

7. The remote fault diagnosis method of the intelligent temperature transmitter according to claim 6, characterized in that The construction of the thermodynamic transfer sub-equation of the temperature transmitter according to the heat balance equations includes: Analyze the time constant, pure delay time, thermo-elastic coupling coefficient, mechanical vibration natural frequency, and damping ratio of the temperature transmitter according to the heat balance equations; Based on the time constant, pure delay time, thermo-elastic coupling coefficient, mechanical vibration natural frequency, and damping ratio, use the following formula to construct the thermodynamic transfer sub-equation of the temperature transmitter, where the thermodynamic transfer sub-equation: ； Among them, represents the thermodynamic transfer sub-equation, represents the time constant, represents the pure delay time, represents the Laplace variable, represents the exponential function, represents the thermo-elastic coupling coefficient, represents the damping ratio, represents the natural frequency of mechanical vibration.

8. The remote fault diagnosis method for the intelligent temperature transmitter according to claim 7, characterized in that, The analysis of the behavior deviation of the temperature transmitter based on the normal condition and the cross-modal correlation feature matrix includes: Analyze the feature distribution of the normal condition; Standardize the cross-modal correlation feature matrix to obtain a standardized correlation feature matrix; Define the deviation index of the temperature transmitter; Analyze the behavior deviation of the temperature transmitter based on the deviation index, the standardized correlation feature matrix, and the feature distribution.

9. The remote fault diagnosis method of the intelligent temperature transmitter according to claim 8, characterized in that The construction of the hierarchical collaborative network of the temperature transmitter includes: Construct the hierarchical collaborative structure of the temperature transmitter, where the hierarchical collaborative structure includes an edge layer structure, a fog node layer structure, and a cloud layer structure; Perform network integration on the edge layer structure, the fog node layer structure, and the cloud layer structure to obtain a hierarchical collaborative network structure; Define the collaborative analysis rules of the hierarchical collaborative network structure; Analyze the inter-layer interoperability of the hierarchical collaborative network structure; When the inter-layer interoperability meets the preset inter-layer interoperability threshold, construct the hierarchical collaborative network of the temperature transmitter based on the hierarchical collaborative network structure and the collaborative analysis rules.

10. An intelligent temperature transmitter remote fault diagnosis system, characterized in that, The system includes: A data conflict elimination module for collecting multi-source sensing data of a temperature transmitter, where the multi-source sensing data includes temperature signal time series data, power supply current harmonic feature data, equipment vibration spectrum data, and environmental parameters, defining a confidence assignment function for the multi-source sensing data, and performing conflict elimination on the multi-source sensing data through the confidence assignment function to obtain conflict elimination data; The twin model construction module is used to extract the cross-modal correlation feature matrix of the conflict elimination data, determine the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework of the temperature transmitter, and construct the hybrid digital twin model of the temperature transmitter according to the thermodynamic transfer sub-equation and the LSTM-PHY joint modeling framework; The behavior deviation analysis module is used to simulate the normal working condition of the temperature transmitter through the hybrid digital twin model, and analyze the behavior deviation of the temperature transmitter based on the normal working condition and the cross-modal correlation feature matrix; The multi-layer fault analysis module is used to construct a hierarchical collaborative network of the temperature transmitter, where the hierarchical collaborative network includes an edge layer, a fog node layer, and a cloud layer, and analyze the multi-layer fault analysis results of the temperature transmitter by using the hierarchical collaborative network according to the behavior deviation; The fault parameter optimization module is used to determine the fault optimization parameters of the temperature transmitter through the multi-layer fault analysis results, and perform the fault diagnosis of the intelligent temperature transmitter based on the fault optimization parameters.

Images