Heat pipe heat exchanger fault real-time monitoring system and method based on multi-source data fusion

Abstract

The invention discloses a heat pipe heat exchanger fault real-time monitoring system and method based on multi-source data fusion, and the system comprises a multi-source data collection module which is used for collecting temperature data, pressure data, flow data, vibration data and image data in the operation process of a heat pipe heat exchanger; the invention relates to the technical field of heat pipe heat exchanger fault monitoring. According to the heat pipe heat exchanger fault real-time monitoring system and method based on multi-source data fusion, multi-source data such as temperature, pressure, flow, vibration and images are creatively fused, and compared with traditional single sensor monitoring, operation state information of the heat pipe heat exchanger can be captured in an all-around mode. Various sensors are arranged at key positions, such as temperature sensors at the inlet, the outlet, the pipe wall and the like, so that the internal temperature distribution of the heat exchanger can be comprehensively mastered; the pressure sensor and the flow sensor can reflect the flowing state of fluid; visual basis is provided for diagnosis of mechanical faults and external states by vibration and image data.

Description

Technical Field

[0001] The present invention relates to the technical field of heat pipe heat exchanger fault monitoring, and in particular to a real-time heat pipe heat exchanger fault monitoring system and method based on multi-source data fusion. Background Art

[0002] Heat pipe heat exchangers, due to their efficient heat transfer performance, are widely used in numerous industrial fields, including chemical, electric power, and metallurgy. However, over long-term operation, heat pipe heat exchangers may experience failures such as heat pipe leakage, blockage, and fin damage. These failures can reduce the heat transfer efficiency of the heat exchanger, affect the stable operation of the entire production system, and even cause safety accidents.

[0003] Currently, fault monitoring methods for heat pipe exchangers primarily rely on single-sensor monitoring, such as temperature sensors monitoring the heat exchanger's inlet and outlet temperatures and pressure sensors monitoring internal pressure. However, the data captured by a single sensor is limited and cannot fully reflect the heat exchanger's operating status, making it prone to missed detections and false positives. Furthermore, most existing monitoring systems lack real-time and intelligent diagnostic capabilities, making it difficult to detect and accurately determine the fault type and location in the early stages of a fault. This leads to delayed fault handling, increased equipment maintenance costs, and increased production downtime. To address this issue, we propose a real-time heat pipe exchanger fault monitoring system and method based on multi-source data fusion. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides a real-time monitoring system and method for heat pipe and heat exchanger faults based on multi-source data fusion, which solves the problems raised in the background art.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a real-time monitoring system for heat pipe heat exchanger faults based on multi-source data fusion, comprising: Multi-source data acquisition module, used to collect temperature data, pressure data, flow data, vibration data and image data during the operation of the heat pipe heat exchanger; A data preprocessing module, connected to the multi-source data acquisition module, is used to perform data filtering, data normalization, missing value processing, and outlier removal on the collected data; A data fusion analysis module, connected to the data preprocessing module, for fusing the preprocessed data and extracting data features; A fault diagnosis and early warning module, connected to the data fusion analysis module, is used to determine whether the heat pipe heat exchanger has a fault based on the analysis results and issue an early warning; The human-computer interaction module is connected to the fault diagnosis and early warning module and is used to provide an operation interface to realize data display, parameter setting and system control functions.

[0006] Preferably, the multi-source data acquisition module includes a temperature data acquisition unit, a pressure data acquisition unit, a flow data acquisition unit, a vibration data acquisition unit and an image data acquisition unit; the temperature data acquisition unit is used to arrange temperature sensors at key positions such as the inlet and outlet of the heat pipe exchanger, the heat pipe wall, and the fins to collect temperature information; the pressure data acquisition unit is used to install pressure sensors at the inlet and outlet pipes of the heat exchanger and inside the heat pipe to measure pressure changes; the flow data acquisition unit is used to install flow measurement equipment at the inlet and outlet pipes of the heat exchanger to monitor fluid flow; the vibration data acquisition unit is used to install vibration sensors on the outer casing, bracket and other parts of the heat exchanger to collect vibration signals; and the image data acquisition unit is used to install high-definition cameras around the heat exchanger to capture appearance images.

[0007] Preferably, the data preprocessing module includes a data filtering unit, a data normalization unit, a missing value processing unit and an outlier elimination unit; the data filtering unit adopts a digital filtering algorithm to remove data noise; the data normalization unit normalizes the data to the range of [0, 1] or [-1, 1]; the missing value processing unit adopts an interpolation method, a mean filling method or a model-based prediction method to fill in the missing data; the outlier elimination unit eliminates outliers through a statistical analysis method or an outlier detection algorithm based on machine learning.

[0008] Preferably, the data fusion analysis module includes a data fusion unit, a feature extraction unit and a data analysis unit; the data fusion unit adopts a feature-level fusion method to fuse the features of temperature, pressure, flow, vibration and image data; the feature extraction unit uses principal component analysis, independent component analysis and other dimensionality reduction algorithms to extract key features; the data analysis unit adopts a machine learning algorithm or a deep learning algorithm to analyze the extracted features and establish a heat pipe exchanger operating status model.

[0009] Preferably, the fault diagnosis and warning module includes a fault judgment unit, a fault warning unit and a fault recording unit; the fault judgment unit compares the analysis result with the preset fault threshold to determine the fault and type and location; the fault warning unit issues a warning message through sound and light alarms, SMS notifications, email reminders, etc. after determining the fault; the fault recording unit records and stores the fault information.

[0010] Preferably, the human-computer interaction module includes a data display unit, a parameter setting unit and a system control unit; the data display unit displays operating data and diagnostic results in the form of charts, curves and images; the parameter setting unit is used to set parameters such as sensor acquisition frequency, fault threshold, early warning mode, etc.; the system control unit realizes control functions such as starting, stopping and calibration of the monitoring system.

[0011] Preferably, when the data fusion unit fuses data features, it extracts temperature change rate and temperature gradient features from temperature data, and extracts vibration frequency and amplitude features from vibration data; when the fault judgment unit judges a fault, if the temperature rises abnormally and the pressure drops abnormally, it is judged as a heat pipe leakage fault in combination with the fault model; the data display unit of the human-computer interaction module is also used to display the statistical analysis results of fault history data.

[0012] The present invention also discloses a real-time monitoring method for heat pipe heat exchanger faults based on multi-source data fusion, based on the monitoring system according to any one of claims 1 to 7, the method comprising the following steps: Multi-source data acquisition: The temperature, pressure, flow, vibration and image data of the heat pipe heat exchanger during operation are collected through the multi-source data acquisition module and transmitted to the data preprocessing module; Data preprocessing: Use the data preprocessing module to filter, normalize, process missing values, and remove outliers on the original data to obtain preprocessed data; Data fusion analysis: The data fusion analysis module fuses the preprocessed data features, extracts key features, and uses machine learning or deep learning algorithms to analyze the features and calculate the probability of the current device being in different operating states; Fault diagnosis and early warning: The fault diagnosis and early warning module compares the analysis results with the preset fault threshold to determine whether the heat pipe heat exchanger has a fault and the type and location of the fault. If a fault occurs, it will promptly issue an early warning and record the fault information; Human-computer interaction: Operators can view equipment operation data and fault diagnosis results, set system parameters, and operate and control the monitoring system through the human-computer interaction module.

[0013] Preferably, in the data preprocessing step, the 3σ principle is used to identify abnormal values in the vibration data.

[0014] Preferably, in the data fusion analysis step, when the principal component analysis algorithm is used to reduce the dimension of the fused feature vector, the principal components with a cumulative contribution rate of more than 85% are retained.

[0015] The present invention provides a real-time heat pipe and heat exchanger fault monitoring system and method based on multi-source data fusion. Compared with the existing technology, it has the following advantages: (1) This invention innovatively integrates multi-source data such as temperature, pressure, flow, vibration, and image data. Compared with traditional single-sensor monitoring, it can capture the operating status information of the heat pipe heat exchanger in all directions. By arranging multiple sensors at key locations, such as temperature sensors installed at the inlet and outlet, pipe wall, etc., the internal temperature distribution of the heat exchanger can be fully grasped; pressure and flow sensors can reflect the fluid flow state; vibration and image data provide an intuitive basis for mechanical fault and external status diagnosis. After pre-processing and fusion analysis, multi-source data effectively overcomes the limitations of single data information, greatly reduces missed detection and false detection, greatly improves the accuracy and reliability of fault monitoring, and provides a solid guarantee for the stable operation of the equipment.

[0016] (2) The system has powerful real-time data acquisition and processing capabilities, combined with advanced machine learning and deep learning algorithms, which can quickly analyze and determine the operating status of the equipment. Once a fault is detected, the fault diagnosis and warning module can immediately issue a warning through various means such as sound and light alarms and SMS notifications, allowing operators to respond and handle the fault in a timely manner at the early stage, effectively avoiding the escalation of the fault. By establishing an accurate fault diagnosis model, the system can not only quickly identify the fault, but also accurately determine the fault type and location, greatly improving the efficiency and accuracy of fault diagnosis, significantly reducing equipment maintenance costs and production downtime, and effectively ensuring production continuity and corporate economic benefits.

[0017] (3) The human-computer interaction module is designed with an intuitive and friendly operation interface, which allows operators to view equipment operation data and fault diagnosis results in real time, and flexibly set system parameters to achieve convenient control of the monitoring system, significantly improving the system's usability and operability. At the same time, the system records and stores collected data and fault information in detail, providing rich data support for equipment maintenance, fault analysis, and system optimization. With the help of this data, enterprises can deeply analyze the operation patterns and causes of equipment failures, optimize equipment management strategies in a targeted manner, improve equipment management levels and operating performance, promote intelligent upgrades to production processes, and enhance the core competitiveness of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a system block diagram of the present invention; Figure 2 Schematic diagram of a multi-source data acquisition module of the present invention; Figure 3 Schematic diagram of the data preprocessing module of the present invention; Figure 4 Schematic diagram of the data fusion analysis module of the present invention; Figure 5 This is a schematic diagram of the fault diagnosis and early warning module of the present invention; Figure 6 Schematic diagram of the human-computer interaction module of the present invention.

[0019] In the figure: 01, multi-source data acquisition module; 02, data preprocessing module; 03, data fusion analysis module; 04, fault diagnosis and early warning module; 05, human-computer interaction module; 011, temperature data acquisition unit; 012, pressure data acquisition unit; 013, flow data acquisition unit; 014, vibration data acquisition unit; 015, image data acquisition unit; 021, data filtering unit; 022, data normalization unit; 023, missing value processing unit; 024, outlier elimination unit; 031, data fusion unit; 032, feature extraction unit; 033, data analysis unit; 041, fault judgment unit; 042, fault early warning unit; 043, fault recording unit; 051, data display unit; 052, parameter setting unit; 053, system control unit. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] The present invention provides three technical solutions, including the following embodiments: Example 1, see Figure 1 , a real-time monitoring system for heat pipe heat exchanger faults based on multi-source data fusion, including: Multi-source data acquisition module 01 is used to collect temperature data, pressure data, flow data, vibration data and image data during the operation of the heat pipe heat exchanger; The data preprocessing module 02 is connected to the multi-source data acquisition module 01 and is used to perform data filtering, data normalization, missing value processing and outlier removal on the collected data; The data fusion analysis module 03 is connected to the data preprocessing module 02 and is used to fuse the preprocessed data and extract data features; The fault diagnosis and early warning module 04 is connected to the data fusion analysis module 03 and is used to determine whether the heat pipe heat exchanger has a fault based on the analysis results and issue an early warning; The human-computer interaction module 05 is connected to the fault diagnosis and early warning module 04 and is used to provide an operation interface to realize data display, parameter setting and system control functions.

[0022] On the heat pipe exchanger of this chemical enterprise, various sensors are precisely installed in accordance with the design requirements of the multi-source data acquisition module 01. K-type thermocouple sensors are arranged at 12 key temperature monitoring points such as the heat exchanger inlet and outlet, heat pipe wall, and fins to collect temperature data at different locations in real time; high-precision diffused silicon pressure sensors are installed in the inlet and outlet pipes and inside the heat pipes to ensure accurate measurement of fluid pressure changes; an electromagnetic flowmeter is installed in each of the heat exchanger inlet and outlet pipes to achieve accurate monitoring of fluid flow; three-axis acceleration vibration sensors are installed at the four corners of the heat exchanger casing and key parts of the bracket to capture vibration signals during equipment operation; at the same time, two high-definition industrial cameras are installed in positions with good visibility around the heat exchanger to ensure clear images of the heat exchanger's appearance.

[0023] All sensors were connected to the Multi-Source Data Acquisition Module 01 via dedicated data transmission cables to ensure stable signal transmission. Next, the Multi-Source Data Acquisition Module 01, Data Preprocessing Module 02, Data Fusion Analysis Module 03, Fault Diagnosis and Warning Module 04, and Human-Computer Interaction Module 05 were system integrated and debugged. In the Human-Computer Interaction Module 05, based on the heat pipe exchanger's historical operating data and the company's production process requirements, the sensor acquisition frequency was set to once every 30 seconds. The fault threshold for abnormal temperature rise was set to 15°C above the normal operating temperature, and the threshold for abnormal pressure drop was set to 0.8 MPa below the normal operating pressure. Audible and visual alarms and SMS notifications were selected as the primary fault warning methods.

[0024] Example 2, based on Example 1, see Figure 2-Figure 6 As shown, the multi-source data acquisition module 01 includes a temperature data acquisition unit 011, a pressure data acquisition unit 012, a flow data acquisition unit 013, a vibration data acquisition unit 014 and an image data acquisition unit 015; the temperature data acquisition unit 011 is used to arrange temperature sensors at key locations such as the inlet and outlet of the heat pipe exchanger, the heat pipe wall, and the fins to collect temperature information; the pressure data acquisition unit 012 is used to install pressure sensors at the inlet and outlet pipes of the heat exchanger and inside the heat pipe to measure pressure changes; the flow data acquisition unit 013 is used to install flow measurement equipment at the inlet and outlet pipes of the heat exchanger to monitor fluid flow; the vibration data acquisition unit 014 is used to install vibration sensors at the shell, bracket and other parts of the heat exchanger to collect vibration signals; the image data acquisition unit 015 is used to install high-definition cameras around the heat exchanger to capture appearance images.

[0025] The data preprocessing module 02 includes a data filtering unit 021, a data normalization unit 022, a missing value processing unit 023 and an outlier elimination unit 024; the data filtering unit 021 uses a digital filtering algorithm to remove data noise; the data normalization unit 022 normalizes the data to the range of [0, 1] or [-1, 1]; the missing value processing unit 023 uses interpolation, mean filling or model-based prediction methods to fill in missing data; the outlier elimination unit 024 eliminates outliers through statistical analysis methods or machine learning-based anomaly detection algorithms.

[0026] The multi-source data acquisition module 01 continuously collects temperature, pressure, flow, vibration and image data during the operation of the heat pipe heat exchanger at a set frequency, and transmits these raw data to the data preprocessing module in real time. The data preprocessing module 02 starts working. The data filtering unit 021 uses the Kalman filtering algorithm for the collected temperature data to effectively remove the random noise caused by environmental interference, making the temperature data curve smoother and more accurate; for the pressure data, the median filtering algorithm is used to eliminate the interference caused by instantaneous pressure fluctuations; the data normalization unit 022 uses the minimum-maximum normalization method to normalize data of different dimensions such as temperature, pressure, flow, vibration, etc. to the range of [0,1] to facilitate subsequent data fusion and analysis; when the missing value processing unit 023 detects that the flow data at a certain moment is missing, it uses the linear interpolation method to fill in the data based on the flow data at the previous and next time points to ensure the integrity of the data; the outlier elimination unit 024 analyzes the vibration data according to the 3σ principle, and successfully identifies and eliminates abnormal vibration signals caused by temporary poor contact of the sensor.

[0027] The data fusion analysis module 03 includes a data fusion unit 031, a feature extraction unit 032 and a data analysis unit 033; the data fusion unit 031 adopts a feature-level fusion method to fuse the features of temperature, pressure, flow, vibration and image data; the feature extraction unit 032 uses principal component analysis, independent component analysis and other dimensionality reduction algorithms to extract key features; the data analysis unit 033 adopts a machine learning algorithm or a deep learning algorithm to analyze the extracted features and establish a heat pipe exchanger operating status model.

[0028] The pre-processed data enters the data fusion analysis module 03. The data fusion unit 031 first extracts the temperature change rate, temperature gradient and other features from the temperature data, extracts the pressure fluctuation amplitude, pressure change trend and other features from the pressure data, extracts the flow stability, flow change rate and other features from the flow data, extracts the vibration frequency, amplitude, vibration energy and other features from the vibration data, extracts the fin integrity, whether there are leakage signs and other features from the image data, and then combines these features to form a feature vector containing multi-source information. Feature extraction unit 032 uses the principal component analysis (PCA) algorithm to reduce the dimensionality of the fused feature vectors, retaining the top eight principal components with cumulative contributions exceeding 90%, effectively removing redundant information and reducing data dimensionality. Data analysis unit 033 employs a convolutional neural network (CNN) algorithm, a deep learning framework, to train a model using 5,000 sets of historically accumulated data from the company regarding normal operation and 1,000 sets of data from various fault types. After training, the real-time collected and processed feature vectors are input into the trained CNN model to calculate the probability of the equipment being in different operating states, including normal operation, heat pipe leakage, blockage, and fin damage.

[0029] The fault diagnosis and warning module 04 includes a fault judgment unit 041, a fault warning unit 042 and a fault recording unit 043; the fault judgment unit 041 compares the analysis results with the preset fault threshold to determine the fault and type and location; the fault warning unit 042 issues a warning message through sound and light alarms, SMS notifications, email reminders, etc. after determining the fault; the fault recording unit 043 records and stores the fault information.

[0030] Fault Diagnosis and Warning Module 04 receives the results output by Data Analysis Unit 033. At a certain moment, the system detects that the heat exchanger outlet temperature has increased by 18°C and the internal pressure has dropped by 1.2 MPa within 10 minutes. Data Analysis Unit 033 outputs a 92% probability of a heat pipe leakage fault, exceeding the preset fault threshold. Fault Judgment Unit 041 immediately determines that a heat pipe leakage fault has occurred and locates the fault in a heat pipe in the middle of the heat exchanger. Fault warning unit 042 activated quickly, emitting a piercing siren and flashing red light from the on-site audible and visual alarm. Simultaneously, a text message containing detailed information, including the fault type (heat pipe leak), fault location (center of the heat exchanger), and fault severity (high), was sent to the equipment maintenance supervisor and relevant operators. Fault recording unit 043 recorded the exact time of the fault (XX / XX / XX / XX / XX / XX hour / XX minute, 202X), fault type, and the basis for the judgment in a database, providing a basis for subsequent fault analysis and resolution.

[0031] The human-computer interaction module 05 includes a data display unit 051, a parameter setting unit 052 and a system control unit 053; the data display unit 051 displays operating data and diagnostic results in the form of charts, curves and images; the parameter setting unit 052 is used to set parameters such as sensor acquisition frequency, fault threshold, early warning mode, etc.; the system control unit 053 realizes control functions such as starting, stopping and calibration of the monitoring system.

[0032] The operator uses the data display unit 051 of the human-computer interaction module 05 to view the heat pipe exchanger's real-time temperature curve, pressure trend, flow fluctuation, and vibration data spectrum in intuitive graphical form. They can also view images of the heat exchanger's appearance captured by a high-definition camera, quickly understanding the equipment's operating status. Upon learning of a heat pipe leak, the operator uses the data display unit 051 to further view detailed data before and after the failure to assist in determining the severity of the fault. Subsequently, the operator used parameter setting unit 052 to fine-tune the fault thresholds based on the fault situation, adjusting the threshold for abnormal temperature increase to 12°C and the threshold for abnormal pressure decrease to 0.6MPa, thereby increasing the system's sensitivity to similar faults. After the fault was resolved, the monitoring system was calibrated using system control unit 053 to ensure continued accurate and stable operation. At the same time, technicians regularly conduct in-depth analysis of the system's collected data and fault records, continuously optimizing data processing algorithms and fault diagnosis models to further enhance the system's monitoring performance and reliability.

[0033] When the data fusion unit 031 fuses data features, it extracts the temperature change rate and temperature gradient features from the temperature data, and extracts the vibration frequency and amplitude features from the vibration data; when the fault judgment unit 041 judges a fault, if the temperature rises abnormally and the pressure drops abnormally, it is judged as a heat pipe leakage fault in combination with the fault model; the data display unit 051 of the human-computer interaction module 05 is also used to display the statistical analysis results of the fault history data.

[0034] Example 3, based on Example 2, the present invention further discloses a method for real-time monitoring of heat pipe heat exchanger faults based on multi-source data fusion, comprising the following steps: Multi-source data acquisition: The temperature, pressure, flow, vibration and image data of the heat pipe heat exchanger during operation are collected through the multi-source data acquisition module 01 and transmitted to the data preprocessing module; Data preprocessing: Use the data preprocessing module 02 to filter, normalize, process missing values, and remove outliers on the original data to obtain preprocessed data; Data fusion analysis: Data fusion analysis module 03 fuses the pre-processed data features, extracts key features, and uses machine learning or deep learning algorithms to analyze the features and calculate the probability of the current device being in different operating states; Fault diagnosis and early warning: The fault diagnosis and early warning module 04 compares the analysis results with the preset fault threshold to determine whether the heat pipe heat exchanger has a fault and the type and location of the fault. If a fault occurs, it will promptly issue an early warning message and record the fault information; Human-computer interaction: The operator can view equipment operation data and fault diagnosis results, set system parameters, and operate and control the monitoring system through the human-computer interaction module 05.

[0035] In the data preprocessing step, the 3σ principle is used to identify outliers in the vibration data.

[0036] In the data fusion analysis step, when the principal component analysis algorithm is used to reduce the dimension of the fused feature vector, the principal components with a cumulative contribution rate of more than 85% are retained.

[0037] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0038] The above embodiments of the invention are described in detail, but the contents are only preferred embodiments of the invention and should not be considered to limit the scope of the invention. All equivalent changes and improvements made within the scope of the invention should still fall within the scope of the invention.

Claims

1. A real-time monitoring system for heat pipe and heat exchanger faults based on multi-source data fusion, characterized by: include: A multi-source data acquisition module (01) is used to acquire temperature data, pressure data, flow data, vibration data and image data during the operation of the heat pipe heat exchanger; A data preprocessing module (02) is connected to the multi-source data acquisition module (01) and is used to perform data filtering, data normalization, missing value processing, and outlier removal on the collected data; A data fusion analysis module (03), connected to the data preprocessing module (02), is used to fuse the preprocessed data and extract data features; A fault diagnosis and early warning module (04) is connected to the data fusion analysis module (03) and is used to determine whether a fault occurs in the heat pipe heat exchanger based on the analysis results and issue an early warning; The human-computer interaction module (05) is connected to the fault diagnosis and early warning module (04) and is used to provide an operation interface to realize data display, parameter setting and system control functions.

2. The heat pipe and heat exchanger fault real-time monitoring system based on multi-source data fusion according to claim 1 is characterized by: The multi-source data acquisition module (01) comprises a temperature data acquisition unit (011), a pressure data acquisition unit (012), a flow data acquisition unit (013), a vibration data acquisition unit (014) and an image data acquisition unit (015); the temperature data acquisition unit (011) is used to arrange temperature sensors at key positions such as the inlet and outlet of the heat pipe exchanger, the heat pipe wall, and the fins to collect temperature information; the pressure data acquisition unit (012) is used to install pressure sensors at the inlet and outlet pipes of the heat exchanger and inside the heat pipe to measure pressure changes; the flow data acquisition unit (013) is used to install flow measurement equipment at the inlet and outlet pipes of the heat exchanger to monitor fluid flow; the vibration data acquisition unit (014) is used to install vibration sensors at the outer shell, bracket, and other parts of the heat exchanger to collect vibration signals; and the image data acquisition unit (015) is used to install high-definition cameras around the heat exchanger to capture appearance images.

3. The real-time monitoring system for heat pipe and heat exchanger failures based on multi-source data fusion according to claim 1 is characterized by: The data preprocessing module (02) includes a data filtering unit (021), a data normalization unit (022), a missing value processing unit (023) and an outlier elimination unit (024); the data filtering unit (021) uses a digital filtering algorithm to remove data noise; the data normalization unit (022) normalizes the data to the range of [0, 1] or [-1, 1]; the missing value processing unit (023) uses an interpolation method, a mean filling method or a model-based prediction method to fill in the missing data; the outlier elimination unit (024) eliminates outliers through a statistical analysis method or an outlier detection algorithm based on machine learning.

4. The real-time monitoring system for heat pipe and heat exchanger failures based on multi-source data fusion according to claim 1 is characterized by: The data fusion analysis module (03) includes a data fusion unit (031), a feature extraction unit (032) and a data analysis unit (033); the data fusion unit (031) fuses the features of temperature, pressure, flow, vibration and image data using a feature-level fusion method; the feature extraction unit (032) extracts key features using a dimensionality reduction algorithm such as principal component analysis and independent component analysis; the data analysis unit (033) analyzes the extracted features using a machine learning algorithm or a deep learning algorithm to establish an operating state model of a heat pipe exchanger.

5. The real-time monitoring system for heat pipe and heat exchanger failures based on multi-source data fusion according to claim 4 is characterized in that: The fault diagnosis and warning module (04) comprises a fault judgment unit (041), a fault warning unit (042) and a fault recording unit (043); the fault judgment unit (041) compares the analysis result with a preset fault threshold to judge the fault and its type and location; The fault warning unit (042) issues warning information through sound and light alarms, SMS notifications, email reminders, etc. after determining a fault; and the fault recording unit (043) records and stores the fault information.

6. The real-time monitoring system for heat pipe and heat exchanger failures based on multi-source data fusion according to claim 5 is characterized by: The human-computer interaction module (05) comprises a data display unit (051), a parameter setting unit (052) and a system control unit (053); the data display unit (051) displays operating data and diagnostic results in the form of charts, curves and images; the parameter setting unit (052) is used to set parameters such as sensor acquisition frequency, fault threshold, early warning mode, etc.; the system control unit (053) realizes control functions such as starting, stopping and calibrating the monitoring system.

7. The real-time monitoring system for heat pipe and heat exchanger failures based on multi-source data fusion according to claim 6 is characterized by: When the data fusion unit (031) fuses data features, it extracts temperature change rate and temperature gradient features from the temperature data, and extracts vibration frequency and amplitude features from the vibration data; when the fault judgment unit (041) judges a fault, if the temperature rises abnormally and the pressure drops abnormally, it is judged as a heat pipe leakage fault in combination with the fault model; the data display unit (051) of the human-computer interaction module (05) is also used to display statistical analysis results of fault history data.

8. A real-time monitoring method for heat pipe and heat exchanger failures based on multi-source data fusion, characterized by: Based on the monitoring system according to any one of claims 1 to 7, the method comprises the following steps: Multi-source data acquisition: The temperature, pressure, flow, vibration and image data of the heat pipe heat exchanger during operation are collected through the multi-source data acquisition module (01) and transmitted to the data pre-processing module; Data preprocessing: Use the data preprocessing module (02) to filter, normalize, process missing values, and remove outliers on the original data to obtain preprocessed data; Data fusion analysis: The data fusion analysis module (03) fuses the pre-processed data features, extracts key features, and uses machine learning or deep learning algorithms to analyze the features and calculate the probability of the current device being in different operating states; Fault diagnosis and early warning: The fault diagnosis and early warning module (04) compares the analysis results with the preset fault threshold to determine whether the heat pipe heat exchanger has a fault and the type and location of the fault. If a fault occurs, it will issue a warning message in time and record the fault information; Human-computer interaction: The operator can view the equipment operation data and fault diagnosis results through the human-computer interaction module (05), set system parameters and operate and control the monitoring system.

9. The method for real-time monitoring of heat pipe and heat exchanger faults based on multi-source data fusion according to claim 8, characterized in that: In the data preprocessing step, the 3σ principle is used to identify abnormal values in the vibration data.

10. The method for real-time monitoring of heat pipe and heat exchanger faults based on multi-source data fusion according to claim 8, characterized in that: In the data fusion analysis step, when the principal component analysis algorithm is used to reduce the dimension of the fused feature vector, the principal components with a cumulative contribution rate of more than 85% are retained.

Images