A heat exchanger tube bundle blockage processing method, device and electronic equipment
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XINJIANG ZHUNENG CHEMICAL CO LTD
- Filing Date
- 2026-04-03
- Publication Date
- 2026-06-09
Smart Images
Figure CN122171671A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of equipment monitoring technology, and specifically to a method, apparatus, and electronic device for treating heat exchanger tube bundle blockage. Background Technology
[0002] Heat exchangers are critical equipment in industries such as petroleum, chemical, and power. Tube blockage is a major cause of decreased heat transfer efficiency, soaring energy consumption, and even unplanned shutdowns. Existing technologies typically predict heat exchanger blockage by inputting a set of blockage variables into a blockage index calculation formula. This prediction method can only predict whether blockage has occurred, but cannot determine the specific type and location of the blockage. Summary of the Invention
[0003] In view of this, embodiments of the present invention provide a method, apparatus and electronic device for treating heat exchanger tube bundle blockage, so as to achieve early detection and cleaning of heat exchanger blockage.
[0004] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:
[0005] A method for treating heat exchanger tube bundle blockage includes:
[0006] Predict whether the heat exchanger is blocked;
[0007] When a blockage is predicted in the heat exchanger, a multi-signal analysis algorithm is used to perform a directional spectrum scan on the sensor signals output by the sensor array to obtain acoustic localization results. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger.
[0008] By extracting features from acoustic localization results and combining them with real-time operational data to achieve spatiotemporal alignment and fusion, a three-dimensional acoustic-thermal coupling confidence distribution map is constructed. Based on this map, blockage hotspots and blockage types are identified.
[0009] Optionally, in the above-mentioned heat exchanger tube bundle blockage treatment method, feature extraction is performed on the acoustic localization results, and spatiotemporal alignment and fusion are completed by combining real-time operating data to construct a three-dimensional acoustic-thermal coupling confidence distribution map. Based on the three-dimensional acoustic-thermal coupling confidence distribution map, blockage hotspots and blockage types are identified, including:
[0010] The acoustic localization results are processed to extract the acoustic feature map of the acoustic localization results. The acoustic feature map includes at least the frequency and amplitude of the sensor signal.
[0011] The acoustic localization results are spatiotemporally aligned and fused with real-time operational data, and the joint confidence level of each grid point in each sensor array is calculated using evidence theory to generate a three-dimensional acoustic-thermal coupling confidence level distribution map.
[0012] The spatial distribution characteristics of congestion hotspots were identified based on the three-dimensional acoustic-thermal coupling confidence distribution map.
[0013] Based on the three-dimensional acoustic-thermal coupling confidence distribution map, the acoustic feature spectrum corresponding to the blockage hotspot is determined;
[0014] Material properties analysis of blockage hotspots is performed based on acoustic feature maps to determine the type of blockage.
[0015] Optionally, in the above-mentioned method for handling heat exchanger tube bundle blockage, predicting whether the heat exchanger is blocked includes:
[0016] Obtain real-time operating data of the heat exchanger;
[0017] Obtain the analysis results of the artificial intelligence model on real-time operating data. The analysis results include the predicted degree of blockage and the predicted blockage area of the heat exchanger.
[0018] Determine if the predicted degree of blockage is greater than the target blockage threshold. If it is greater than the target blockage threshold, determine that the heat exchanger is blocked.
[0019] A multi-signal analysis algorithm is used to perform directional spectral scanning of the sensor signals output by the sensor array, including:
[0020] A multi-signal analysis algorithm is used to perform directional spectral scanning of the sensor signals of the sensor array corresponding to the predicted blockage area.
[0021] Optionally, the above-mentioned method for treating heat exchanger tube bundle blockage, before obtaining the analysis results of the artificial intelligence model on real-time operating data, further includes:
[0022] Acquire real-time data of the heat exchanger under all operating conditions, establish a normal state sound fingerprint database. The real-time data includes ultrasonic signals and thermal parameters, including temperature, pressure and flow rate. Simulate different degrees of blockage by adjusting the valves of the heat exchanger, and establish an abnormal state sound fingerprint database under blockage conditions.
[0023] A dataset with labeled samples was constructed based on the normal state sound fingerprint database and the abnormal state sound fingerprint database;
[0024] The AI model is trained based on a dataset. The input of the AI model is real-time data, and the output is the predicted level of congestion and the predicted congestion area.
[0025] Optionally, in the above heat exchanger tube bundle blockage treatment method, the target blockage threshold includes a first target blockage threshold, a second target blockage threshold, and a third target blockage threshold. The second target blockage threshold is greater than the first target blockage threshold, and the third target blockage threshold is greater than the second target blockage threshold. When the predicted blockage degree is greater than any of the first, second, and third target blockage thresholds, it is determined that the heat exchanger is blocked.
[0026] The method also includes: generating a first-level early warning signal when the predicted congestion level is greater than the first target congestion threshold, as well as abnormal signal characteristics of real-time operating data;
[0027] When the predicted congestion level exceeds the second target congestion threshold, a secondary early warning signal is generated, along with abnormal signal characteristics of real-time operational data.
[0028] When the predicted congestion level exceeds the third target congestion threshold, a level three early warning signal is generated, along with abnormal signal characteristics of real-time operating data and a preliminary diagnostic report.
[0029] Optionally, in the above-mentioned method for handling heat exchanger tube bundle blockage, when the predicted blockage degree is greater than the target blockage threshold, the method further includes:
[0030] The operating parameters of the sensor array corresponding to the predicted congestion area are adjusted based on the target adjustment strategy, and the time delay of the sensor signal of the sensor array corresponding to the predicted congestion area is estimated by using the generalized cross-correlation function.
[0031] Optionally, in the above-mentioned heat exchanger tube bundle blockage treatment method, after determining the blockage hotspots and the type of blockage based on acoustic localization results, it further includes:
[0032] The cleaning strategy for the heat exchanger is determined based on the type of blockage and the location of the blockage hotspot. The cleaning strategy includes at least cleaning pressure, cleaning agent ratio, and operation path planning.
[0033] The heat exchanger is cleaned based on a cleaning strategy.
[0034] Optionally, in the above-mentioned method for treating heat exchanger tube bundle blockage, after determining the type of blockage, it further includes:
[0035] Calculate cleaning parameters for the blocked area based on the spatial distribution characteristics of the blockage hotspots and the type of blockage;
[0036] Cleaning heat exchangers based on cleaning strategies includes cleaning heat exchangers based on cleaning strategies and cleaning parameters.
[0037] A heat exchanger tube bundle blockage early warning device, comprising:
[0038] A blockage prediction unit is used to predict whether a blockage exists in the heat exchanger;
[0039] The acoustic localization unit is used to perform directional spectrum scanning of the sensor signals output by the sensor array using a multi-signal analysis algorithm when a blockage is predicted in the heat exchanger, in order to obtain the acoustic localization result. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger.
[0040] The hotspot and blockage type analysis unit is used to determine the type of blockage hotspots and blockages based on acoustic localization results.
[0041] An electronic device includes: at least one processing device and a storage device connected to the processing device, wherein:
[0042] Storage devices are used to store computer programs;
[0043] The processing device is used to execute a computer program to enable electronic equipment to implement the heat exchanger tube bundle blockage treatment method as described above.
[0044] Based on the above technical solution, the solution provided in this embodiment of the invention relies on acoustic localization technology to obtain the location and distribution information of the sound source, performs targeted feature extraction on the original acoustic localization results, and extracts key acoustic feature information that can reflect the location, intensity, and distribution characteristics of the blockage. Simultaneously, real-time operating data during the heat exchanger's operation is collected, and the acoustic feature information is matched with the multi-dimensional operating data on the time axis and spatial coordinates to achieve unified alignment and data fusion processing in the spatiotemporal dimensions. Based on this, a three-dimensional acoustic-thermal coupling confidence distribution map is constructed, which can simultaneously reflect the coupling relationship between the sound field and the temperature field, by integrating acoustic signal characteristics and thermal field variation patterns. This distribution map can intuitively present the probability and degree of blockage in different areas of the heat exchanger. Finally, based on this distribution map, through feature matching and pattern recognition, the blockage hotspots and blockage types with blockage risk inside the heat exchanger are accurately located.
[0045] Furthermore, once the hotspots and types of blockages are identified, users can configure targeted flushing strategies based on these factors to improve cleaning effectiveness and prevent energy waste during the cleaning process. Attached Figure Description
[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0047] Figure 1This is a schematic flowchart of a heat exchanger tube bundle blockage treatment method disclosed in an embodiment of this application;
[0048] Figure 2 This is a flowchart illustrating the blockage hotspot and type analysis disclosed in an embodiment of this application;
[0049] Figure 3 This is a flowchart illustrating the training process of the artificial intelligence training model disclosed in the embodiments of this application;
[0050] Figure 4 This is a schematic diagram of the structure of a heat exchanger tube bundle blockage early warning device disclosed in an embodiment of this application. Detailed Implementation
[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0052] Before implementing this solution, a non-contact ultrasonic sensor array needs to be uniformly arranged on the surface of the heat exchanger. The sensors selected for the array should have a frequency range of 100kHz to 1MHz, withstand high temperatures up to 300℃, and have strong anti-interference capabilities. The array layout design should be based on the heat exchanger structure, planning a three-dimensional grid of points on the outside of the shell. The grid spacing should be 20cm or other values; the higher the required accuracy, the smaller the grid spacing. At least two sensors should be added at the heat exchanger inlet / outlet and at elbows. In this solution, magnetic bases can be used to fix the sensors. All sensors are connected to a multi-channel synchronous data acquisition card, with the sampling frequency and resolution set to target values, for example, a sampling frequency of 2MHz and a resolution of 24 bits. The card is then connected to an edge computing gateway and an industrial computer to analyze the sensor output results using pre-set algorithms and models.
[0053] Based on the aforementioned sensor array, this application discloses a method for handling heat exchanger tube bundle blockage. This method can be applied to the aforementioned computing gateway or industrial computer. See [link to relevant documentation]. Figure 1 The method may include:
[0054] Step S101: Obtain the operating data of the heat exchanger and predict whether there is blockage in the heat exchanger based on the operating data.
[0055] When the heat exchanger is running, real-time operating data is acquired. This real-time operating data characterizes the heat exchanger's operating status and includes thermal parameters and ultrasonic data. Thermal parameters include the heat exchanger's input temperature, output temperature, input pressure, output pressure, and flow rate. Ultrasonic data includes sensor signals output from the ultrasonic sensors in the sensor array.
[0056] The operational data is then sent to a trained artificial intelligence (AI) model. This model uses the operational data to perform predictive analysis on the heat exchanger's blockage. The analysis results can include the predicted degree of blockage and the predicted blockage area. In this process, real-time operational data is used as input to the AI model, and the model's output is obtained. The AI model is a pre-built predictive model; its input is the heat exchanger's operational data, and its output is the predicted degree of blockage and the predicted blockage area, which is the region of the heat exchanger where blockage is likely to occur.
[0057] Then, based on the predicted degree of blockage, it is determined whether the heat exchanger is blocked. Specifically, this application pre-configures a target blockage threshold. When the blockage level of the heat exchanger reaches the target blockage threshold, the heat exchanger needs to be cleaned; when it does not reach the target blockage threshold, cleaning is not required. The presence of blockage in the heat exchanger can be determined by comparing the predicted blockage level with the target blockage threshold.
[0058] Step S102: When blockage is predicted in the heat exchanger, a multi-signal analysis algorithm is used to perform directional spectrum scanning on the sensor signals output by the sensor array to obtain acoustic localization results. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger.
[0059] When a blockage is predicted in the heat exchanger, an early warning signal can be raised. Upon detection of the warning signal, to facilitate precise location of the blockage, three-dimensional spatial spectrum estimation and acoustic source imaging technology can be used to accurately locate the blockage. Specifically, a high-precision three-dimensional coordinate system is constructed on the surface of the heat exchanger using adaptive mesh generation technology, and the mesh density can be adjusted according to design requirements. Then, the acoustic signals (sensor signals) received by the ultrasonic sensor array are scanned in a directional manner using the Multiple Signal Classification (MUSIC) algorithm to obtain the acoustic location result.
[0060] Step S103: By extracting features from the acoustic localization results, combining them with real-time running data to complete spatiotemporal alignment and fusion, and constructing a three-dimensional acoustic-thermal coupling confidence distribution map, the blockage hotspots and blockage types are identified based on the three-dimensional acoustic-thermal coupling confidence distribution map.
[0061] This process begins by acquiring the location and distribution information of the sound source using acoustic localization technology. Targeted feature extraction is then performed on the original acoustic localization results to extract key acoustic features that reflect the location, intensity, and distribution characteristics of blockages. Simultaneously, real-time operational data such as temperature, pressure, flow rate, and vibration during heat exchanger operation are collected. The acoustic feature information is then matched with the multi-dimensional operational data on both the time and spatial axes, achieving unified alignment and data fusion across the spatiotemporal dimensions. Based on this, a three-dimensional acoustic-thermal coupling confidence distribution map is constructed, integrating acoustic signal characteristics and thermodynamic field variation patterns. This map visually presents the likelihood and degree of blockage in different areas of the heat exchanger. Finally, based on this distribution map, feature matching and pattern recognition are used to locate blockage hotspots within the heat exchanger that pose a risk of blockage, and the type of blockage is further identified.
[0062] The above-mentioned solution provided by the embodiments of the present invention uses ultrasonic sensors that are uniformly distributed on the surface of the heat exchanger to form a sensor array. When a blockage is predicted in the heat exchanger, a multi-signal analysis algorithm is used to perform directional spectrum scanning on the sensor signals output by the sensor array to obtain acoustic localization results. Then, based on the acoustic localization results, the blockage hotspots and the type of blockage are determined.
[0063] This embodiment provides a specific process for identifying congestion hotspots and types of congestion. (See [link to documentation]) Figure 2 Step S103 may specifically include:
[0064] Step S1031: Process the acoustic positioning results and extract the acoustic feature maps of each sensor signal in the acoustic positioning results. The acoustic feature maps include at least the frequency and amplitude of the sensor signals.
[0065] When processing the acoustic localization results, the signal data collected by each sensor is first obtained from the results. Then, signal analysis algorithms, such as Fast Fourier Transform (FFT), are used to convert the time-domain signal into a frequency-domain signal to extract the frequency components of the signal. At the same time, the amplitude is determined by measuring parameters such as the peak value and RMS value of the signal. Finally, the acoustic feature map is obtained by combining key information such as frequency and amplitude.
[0066] Step S1032: The acoustic localization results are spatiotemporally aligned and fused with the real-time running data, and the joint confidence of each grid point in each sensor array is calculated through evidence theory to generate a three-dimensional acoustic-thermal coupling confidence distribution map.
[0067] This step involves spatiotemporally aligning and fusing the acoustic localization results with real-time operational data, matching them to a unified time axis coordinate system. Then, using evidence theory, the acoustic localization results and real-time operational data at the same moment are calculated and analyzed. Based on the analysis results, the joint confidence of the corresponding grid points of each ultrasonic sensor signal in the sensor array is calculated, and a three-dimensional acoustic-thermal coupling confidence distribution map is generated. Thus, a more accurate three-dimensional view can be constructed through multi-dimensional combination.
[0068] Step S1033: Identify the spatial distribution characteristics of the blockage hotspots based on the three-dimensional acoustic-thermal coupling confidence distribution map.
[0069] In this step, based on the three-dimensional acoustic-thermal coupling confidence distribution map, the high-confidence abnormal areas are analyzed and quantified. Key information such as the coordinate position, geometric shape, extension direction, aggregation density, and confidence gradient change of each abnormal area in three-dimensional space are comprehensively extracted. Based on this information, the blockage hotspots are identified, and the spatial arrangement, concentration, and diffusion trend of the blockage hotspots inside the heat exchanger are analyzed, thereby fully depicting the spatial distribution characteristics of the blockage hotspots.
[0070] Step S1034: Determine the acoustic feature spectrum corresponding to the blockage hotspot based on the three-dimensional acoustic-thermal coupling confidence distribution map.
[0071] Once the blockage hotspot is identified, the acoustic spectrum of the ultrasonic signal output by the ultrasonic sensor corresponding to the blockage hotspot in the sensor array is obtained, along with the confidence level corresponding to the blockage point. Specifically, this process first identifies high-confidence blockage hotspot areas, then traces back and extracts the original acoustic positioning signals collected by the sensors corresponding to that area. Key acoustic parameters such as sound source intensity, spectral distribution, reflection characteristics, and attenuation characteristics in the original acoustic positioning signals are extracted, normalized, and structured. Environmental noise and irrelevant interference components are removed, ultimately forming an acoustic feature spectrum that accurately corresponds to the spatial location of the blockage hotspot. Alternatively, the acoustic feature spectrum corresponding to the blockage hotspot can be directly obtained from the acoustic feature spectrum created in step S1031.
[0072] Step S1035: Analyze the material properties of the blockage hotspots based on acoustic feature maps to determine the type of blockage.
[0073] Based on the acoustic feature map of the output signal of the ultrasonic sensor corresponding to the blockage hotspot and the confidence level of the blockage hotspot, the type of blockage in the blockage hotspot can be determined by performing feature analysis and comparison on the acoustic feature map. Specifically, the acoustic feature map corresponding to each blockage hotspot can be analyzed by a support vector machine classifier to accurately identify the type of blockage in each area.
[0074] Specifically, firstly, by utilizing the abnormal signal characteristics of real-time operating data (such as specific frequencies, amplitude peaks, or phase distortions), the predicted blockage area in the three-dimensional space of the heat exchanger is accurately located. Then, a high-precision three-dimensional coordinate system is constructed in the predicted blockage area using adaptive mesh generation technology, and the mesh density can be adjusted automatically according to design requirements. Next, the acoustic signal (sensor signal) received by the ultrasonic sensor array corresponding to the predicted blockage area is scanned in a directional manner using the Multiple Signal Classification (MUSIC) algorithm to obtain the acoustic localization result.
[0075] In this embodiment, whether the heat exchanger is blocked can be determined by judging whether the blockage degree output by the model is greater than the target blockage threshold. When the predicted blockage degree is greater than the target blockage threshold, the heat exchanger is predicted to be blocked, and an early warning signal and abnormal signal characteristics of real-time operating data can be generated. In this scheme, the target blockage threshold includes a first target blockage threshold, a second target blockage threshold, and a third target blockage threshold. The second target blockage threshold is greater than the first target blockage threshold, and the third target blockage threshold is greater than the second target blockage threshold. A graded early warning mechanism can be set according to the magnitude of the blockage degree output by the artificial intelligence model. Specifically, when the predicted blockage degree output by the artificial intelligence model is found to be greater than the first target blockage threshold (e.g., 5%), a first-level early warning signal is triggered; when the predicted blockage degree output by the artificial intelligence model is found to be greater than the second target blockage threshold (e.g., 10%), a second-level early warning signal is triggered; and when the predicted blockage degree output by the artificial intelligence model is found to be greater than the third target blockage threshold (e.g., 15%), a third-level early warning signal is triggered. While triggering the first, second, or third-level early warning signals, abnormal signal characteristics of real-time operating data can also be output to allow users to view the operating status of the heat exchanger. When a Level 3 warning signal is triggered, in addition to outputting the abnormal signal characteristics of real-time operating data, a preliminary diagnostic report can also be generated. The preliminary diagnostic report includes at least the predicted degree of blockage and the predicted blockage area output by the artificial intelligence model, so as to help users control the blockage status of the heat exchanger and estimate the blockage area.
[0076] In this embodiment, after identifying the hotspots and types of blockages, a cleaning strategy can be automatically configured based on the location of the hotspots and the type of blockage. The cleaning strategy includes at least cleaning pressure, cleaning agent ratio, and workflow planning. Specifically, after determining the location of the hotspots and the type of blockage, the cleaning pressure, cleaning agent ratio, and workflow planning are configured based on experience. The workflow is the flow path of the cleaning fluid, which can be controlled by adjusting the opening status of various valves in the heat exchanger. The hotspots are located on the workflow planning path, allowing the cleaning fluid to flow through them. The cleaning pressure and cleaning agent ratio are determined based on the type of blockage; different types of blockages correspond to different cleaning pressures and cleaning agent ratios. Once the cleaning strategy is determined, it is executed to unclog the heat exchanger, thereby achieving automatic cleaning of the heat exchanger tubes.
[0077] In this embodiment, before obtaining the analysis results of the artificial intelligence model on the real-time running data, it is necessary to pre-build and train the artificial intelligence model, see [link to relevant documentation]. Figure 3 The training process for an artificial intelligence training model can include:
[0078] Step S301: Obtain real-time data of the heat exchanger under all operating conditions, establish a normal state sound fingerprint database, the real-time data includes ultrasonic signals and thermal parameters, the thermal parameters include temperature, pressure and flow rate, and simulate different degrees of blockage by adjusting the valves of the heat exchanger to establish an abnormal state sound fingerprint database under blockage conditions.
[0079] In this step, when the heat exchanger is in a clean state, real-time operating data is continuously collected for at least 72 hours or other set durations under all operating conditions. This real-time operating data includes ultrasonic data and thermal parameters, including temperature, pressure, and flow rate, to establish a normal-state acoustic fingerprint database. Simultaneously, under various operating conditions, the valve opening of the heat exchanger is adjusted to simulate different degrees of blockage (e.g., 5%, 10%, 20%), and real-time operating data under each blockage level is obtained to establish an abnormal-state acoustic fingerprint database under blockage conditions.
[0080] Based on the thermal parameters contained in each data sample in the normal state sound fingerprint database and the abnormal state sound fingerprint database, the heat transfer efficiency of the heat exchanger under various operating conditions is calculated.
[0081] Then, the ultrasonic signals and thermal parameters in the normal state sound fingerprint database and the abnormal state sound fingerprint database are time-aligned and fused, and a thermal acoustic correlation model is established using a multiple linear regression algorithm.
[0082] Variational mode decomposition algorithm is used to denoise data samples in the normal state sound fingerprint database and the abnormal state sound fingerprint database. Butterworth bandpass filter is used to retain the key frequency band from 100kHz to 1MHz. FastICA algorithm is used to perform independent component analysis to remove external interference sources in the data samples.
[0083] The denoised multi-channel acoustic signal (the sensor signal output by the ultrasonic sensor) is used to construct a three-dimensional data volume of the time-frequency sensor position, i.e., an acoustic image. A convolutional neural network is used to automatically extract a 1024-dimensional feature vector from the acoustic image. At the same time, 128 traditional features such as Mel frequency cepstral coefficients and spectral centroids are extracted as supplements to form a fused feature vector as the sound fingerprint at that moment.
[0084] For each normal operating condition and different degree of blockage, a corresponding voiceprint fingerprint template is established and stored in the feature database. The voiceprint template is dynamically optimized based on the new verified data.
[0085] Step S302: Construct a dataset with labeled samples based on the normal state sound fingerprint database and the abnormal state sound fingerprint database.
[0086] The feature vectors corresponding to each data sample in the normal state sound fingerprint database and the abnormal state sound fingerprint database, as well as the heat exchanger status label and blockage location information corresponding to the feature vectors, are divided into a training set, a validation set, and a test set with labeled samples in a ratio of 7:2:1.
[0087] Step S303: Train the artificial intelligence model based on the dataset. The input of the artificial intelligence model is real-time data, and the output is the predicted degree of congestion and the predicted congestion area.
[0088] An initial AI model was constructed, employing an attention-enhanced cross-scale temporal convolutional network as the core algorithm. This network comprises four convolutional layers, two LSTM layers, and one attention layer, automatically focusing on key features and capturing long-term dependencies. The initial AI model was pre-trained using labeled data (training, validation, and test sets with labeled samples) using the Adam optimizer with an initial learning rate of 0.001, fine-tuned for a specific heat exchanger. After system deployment, an online learning mechanism was introduced to continuously optimize the model based on operator feedback and cleaned validation results, thus obtaining a satisfactory AI model.
[0089] In this embodiment, when the predicted congestion level exceeds the target congestion threshold, the method further includes: adjusting the operating parameters of the ultrasonic sensors in the sensor array corresponding to the predicted congestion area based on a target adjustment strategy, and using a generalized cross-correlation function to estimate the time delay of the sensor signals in the sensor array corresponding to the predicted congestion area. High-precision time delay estimation can reduce errors in location calculation, making the positioning results more accurate and reliable. Furthermore, high-precision time delay estimation helps improve the spatial resolution of the system, i.e., the ability to distinguish adjacent targets. When multiple targets exist, accurate time delay estimation can better distinguish the signals of different targets, avoiding mutual interference and confusion between targets.
[0090] In this embodiment, after determining the type of blockage, the method further includes: calculating cleaning parameters for the blockage area based on the spatial distribution characteristics of the blockage hotspots and the type of blockage. Specifically, based on the spatial distribution characteristics of the blockage hotspots and the type of blockage, the optimal cleaning parameters for each blockage area are calculated using a material property database and a cleaning kinetic model to generate a complete personalized cleaning plan. The cleaning parameters may include cleaning duration, impact frequency, etc. At this time, the heat exchanger is cleaned based on the cleaning strategy, including: cleaning the heat exchanger based on the cleaning strategy and cleaning parameters.
[0091] In this embodiment, to facilitate user observation of the cleaning progress, after determining the heat exchanger cleaning strategy, the method further includes: registering the three-dimensional acoustic-thermal coupling confidence distribution map and the operation path plan with the spatial coordinates of the heat exchanger, and constructing an augmented reality operation scene. After cleaning the heat exchanger based on the cleaning strategy, the method further includes: displaying the current operation position in real time through the augmented reality operation scene using augmented reality devices. That is, during the cleaning operation, the current operation position and the process parameter requirements for the next operation specified in the cleaning strategy are displayed in real time through augmented reality devices, so that the user can adjust the cleaning operation process based on the process parameter requirements to ensure the accurate execution of the cleaning plan.
[0092] Furthermore, in this embodiment, during the execution of the cleaning strategy, real-time operating data of the heat exchanger needs to be continuously collected via a sensor array to monitor the cleaning effect and operational status in real time. Based on the real-time monitoring results (cleaning effect and operational status), the cleaning parameters and operational path are dynamically adjusted. After cleaning is completed, acceptance data is collected and compared with pre-configured cleaning quality standards to generate a cleaning effect evaluation report.
[0093] This embodiment discloses a heat exchanger tube bundle blockage early warning device. For the specific operation of each unit in the device, please refer to the above method embodiment. The heat exchanger tube bundle blockage early warning device provided in this embodiment is described below. The heat exchanger tube bundle blockage early warning device described below can be referred to in correspondence with the heat exchanger tube bundle blockage handling method described above. See [link to documentation] Figure 4The device may include:
[0094] Blockage prediction unit 10 is used to predict whether there is blockage in the heat exchanger;
[0095] The acoustic localization unit 20 is used to perform directional spectrum scanning of the sensor signals output by the sensor array using a multi-signal analysis algorithm when a blockage is predicted in the heat exchanger, so as to obtain the acoustic localization result. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger.
[0096] Hotspot and blockage type analysis unit 30 is used to determine the type of blockage hotspot and blockage based on acoustic localization results.
[0097] Corresponding to the above method, this application also discloses an electronic device, which includes at least one processing device and a storage device connected to the processing device, wherein: the storage device is used to store a computer program; the processing device is used to execute the computer program so that the electronic device can implement the heat exchanger tube bundle blockage treatment method shown in any of the above embodiments.
[0098] For example, the electronic device is used to: acquire real-time operating data of the heat exchanger; acquire the analysis results of the real-time operating data by an artificial intelligence model, the analysis results including the predicted degree of blockage and the predicted blockage area of the heat exchanger; determine whether the predicted blockage degree is greater than a target blockage threshold; when it is greater than the target blockage threshold, generate a warning signal and abnormal signal features of the real-time operating data; when the warning signal is detected, use a multi-signal analysis algorithm to perform directional spectrum scanning on the sensor signals of the sensor array corresponding to the predicted blockage area to obtain acoustic positioning results, the sensor array being composed of ultrasonic sensors uniformly distributed on the surface of the heat exchanger; process the acoustic positioning results, and extract the acoustic feature spectrum of the acoustic positioning results, the acoustic feature spectrum being at least The process includes: analyzing the frequency and amplitude of sensor signals; spatiotemporally aligning and fusing the acoustic localization results with the real-time operational data; calculating the joint confidence level of each grid point in each sensor array using evidence theory to generate a three-dimensional acoustic-thermal coupling confidence distribution map; identifying the spatial distribution characteristics of blockage hotspots based on the three-dimensional acoustic-thermal coupling confidence distribution map; determining the acoustic feature spectrum corresponding to the blockage hotspots based on the three-dimensional acoustic-thermal coupling confidence distribution map; performing material property analysis on the blockage hotspots based on the acoustic feature spectrum to determine the type of blockage; determining the cleaning strategy for the heat exchanger based on the type of blockage and the location of the blockage hotspots, the cleaning strategy including at least cleaning pressure, cleaning agent ratio, and operation path planning; and cleaning the heat exchanger based on the cleaning strategy.
[0099] Corresponding to the above method, this application also discloses a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the heat exchanger tube bundle blockage treatment method as described in any of the above embodiments.
[0100] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0101] For ease of description, the above system is described by dividing it into various modules based on their functions. Of course, in implementing this invention, the functions of each module can be implemented in one or more software and / or hardware components.
[0102] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0103] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0104] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0105] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0106] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for treating heat exchanger tube bundle blockage, characterized in that, include: Predict whether the heat exchanger is blocked; When a blockage is predicted in the heat exchanger, a multi-signal analysis algorithm is used to perform a directional spectrum scan on the sensor signals output by the sensor array to obtain acoustic localization results. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger. By extracting features from the acoustic localization results and combining them with real-time operational data to achieve spatiotemporal alignment and fusion, a three-dimensional acoustic-thermal coupling confidence distribution map is constructed. Based on the three-dimensional acoustic-thermal coupling confidence distribution map, blockage hotspots and blockage types are identified.
2. The method for treating heat exchanger tube bundle blockage according to claim 1, characterized in that, By extracting features from acoustic localization results and combining them with real-time operational data to achieve spatiotemporal alignment and fusion, a three-dimensional acoustic-thermal coupling confidence distribution map is constructed. Based on this three-dimensional acoustic-thermal coupling confidence distribution map, blockage hotspots and blockage types are identified, including: The acoustic localization results are processed to extract an acoustic feature map, which includes at least the frequency and amplitude of the sensor signal. The acoustic localization results are spatiotemporally aligned and fused with the real-time operating data, and the joint confidence level of each grid point in each sensor array is calculated using evidence theory to generate a three-dimensional acoustic-thermal coupling confidence level distribution map. The spatial distribution characteristics of the congestion hotspots were identified based on the three-dimensional acoustic-thermal coupling confidence distribution map. Based on the three-dimensional acoustic-thermal coupling confidence distribution map, the acoustic feature map corresponding to the blockage hotspot is determined; Based on the acoustic feature spectrum, the material properties of the blockage hotspots are analyzed to determine the type of blockage.
3. The method for treating heat exchanger tube bundle blockage according to claim 1, characterized in that, Predicting whether a heat exchanger is clogged includes: Obtain real-time operating data of the heat exchanger; The analysis results of the artificial intelligence model on the real-time operating data are obtained, and the analysis results include the predicted degree of blockage and the predicted blockage area of the heat exchanger. Determine whether the predicted degree of blockage is greater than the target blockage threshold. If it is greater than the target blockage threshold, determine that the heat exchanger is blocked. The step of performing directional spectrum scanning of the sensor signals output by the sensor array using a multi-signal analysis algorithm includes: A multi-signal analysis algorithm is used to perform directional spectral scanning of the sensor signals of the sensor array corresponding to the predicted blockage area.
4. The method for treating heat exchanger tube bundle blockage according to claim 3, characterized in that, Before obtaining the analysis results of the artificial intelligence model on the real-time running data, the following steps are also included: Real-time data of the heat exchanger under all operating conditions is acquired, and a normal state sound fingerprint database is established. The real-time data includes ultrasonic signals and thermal parameters, including temperature, pressure and flow rate. Different degrees of blockage are simulated by adjusting the valves of the heat exchanger, and an abnormal state sound fingerprint database under blockage conditions is established. A dataset with labeled samples was constructed based on the normal state sound fingerprint database and the abnormal state sound fingerprint database; The artificial intelligence model is trained based on the dataset. The input of the artificial intelligence model is real-time data, and the output is the predicted degree of congestion and the predicted congestion area.
5. The method for treating heat exchanger tube bundle blockage according to claim 3, characterized in that, The target blockage threshold includes a first target blockage threshold, a second target blockage threshold, and a third target blockage threshold. The second target blockage threshold is greater than the first target blockage threshold, and the third target blockage threshold is greater than the second target blockage threshold. When the predicted blockage degree is greater than any of the first, second, and third target blockage thresholds, it is determined that the heat exchanger is blocked. The method also includes: generating a first-level early warning signal and abnormal signal characteristics of the real-time operating data when the predicted congestion level is greater than the first target congestion threshold; When the predicted congestion level is greater than the second target congestion threshold, a secondary early warning signal is generated, along with abnormal signal characteristics of the real-time operating data; When the predicted congestion level is greater than the third target congestion threshold, a level three early warning signal is generated, along with the abnormal signal characteristics of the real-time operating data and a preliminary diagnostic report.
6. The method for treating heat exchanger tube bundle blockage according to any one of claims 3-6, characterized in that, When the predicted congestion level is greater than the target congestion threshold, the method further includes: The operating parameters of the sensor array corresponding to the predicted congestion area are adjusted based on the target adjustment strategy, and the time delay of the sensor signal of the sensor array corresponding to the predicted congestion area is estimated using a generalized cross-correlation function.
7. The method for treating heat exchanger tube bundle blockage according to claim 1, characterized in that, After determining the blockage hotspots and the type of blockage based on the acoustic localization results, the process also includes: The cleaning strategy for the heat exchanger is determined based on the type of blockage and the location of the blockage hotspot. The cleaning strategy includes at least cleaning pressure, cleaning agent ratio, and operation path planning. The heat exchanger is cleaned based on the cleaning strategy described above.
8. The method for treating heat exchanger tube bundle blockage according to claim 7, characterized in that, After determining the type of blockage, the following is also included: Calculate the cleaning parameters for the blocked area based on the spatial distribution characteristics of the blockage hotspots and the type of blockage. Cleaning the heat exchanger based on the cleaning strategy includes cleaning the heat exchanger based on the cleaning strategy and cleaning parameters.
9. A heat exchanger tube bundle blockage early warning device, characterized in that, include: A blockage prediction unit is used to predict whether a blockage exists in the heat exchanger; An acoustic localization unit is used to perform directional spectrum scanning of the sensor signals output by the sensor array using a multi-signal analysis algorithm when a blockage is predicted in the heat exchanger, in order to obtain an acoustic localization result. The sensor array consists of ultrasonic sensors uniformly distributed on the surface of the heat exchanger. The hotspot and blockage type analysis unit is used to determine the type of blockage hotspot and blockage based on the acoustic localization results.
10. An electronic device, characterized in that, include: At least one processing device and a storage device connected to the processing device, wherein: The storage device is used to store computer programs; The processing device is used to execute the computer program so that the electronic device can implement the heat exchanger tube bundle blockage treatment method as described in any one of claims 1 to 8.