Heat exchanger optimization method and device based on multi-source data

Abstract

The application discloses a heat exchanger optimization method and device based on multi-source data, which is used for accurately identifying local overheating areas of a heat exchanger and effectively optimizing the heat exchanger, so that the problem of heat exchanger damage caused by local overheating can be reduced, and the heat transfer efficiency and operation stability of the heat exchanger are improved. The method comprises the following steps: collecting multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operation data comprises fluid motion trajectory data, centrifugal force distribution gradient data and flow velocity radial layering data; a flow simulation model of the heat exchanger is built based on the multi-source operation data to generate a centrifugal force-flow velocity correlation matrix based on a spatial position; a random forest algorithm is used to extract coupling features between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix; the local overheating area is identified according to the coupling features; and a preset heat compensation mechanism is used to optimize the local overheating area.

Description

Technical Field

[0001] The present application relates to the technical field of heat exchanger optimization, and in particular to a heat exchanger optimization method and device based on multi-source data. Background Art

[0002] Heat exchangers are widely used in numerous fields, including chemical engineering, energy, and refrigeration. Their performance directly impacts the efficiency and energy consumption of the entire system. During operation, localized overheating often occurs, which not only reduces the heat transfer efficiency but can also damage the equipment, shorten its service life, and increase maintenance costs.

[0003] Currently, to reduce the occurrence of localized overheating, fluid flow adjustments within heat exchangers are typically made based on real-time temperature or flow rate data during operation. However, the heat exchange process within a heat exchanger is a complex physical process, involving the interaction of multiple factors such as fluid movement, force distribution, and temperature changes. Using only a single data type makes it difficult to fully and accurately reflect the actual operating conditions within the heat exchanger, and thus unable to deeply explore the inherent connections between various factors. This makes it difficult to accurately identify localized overheating areas, making it difficult to meet the requirements for efficient and stable operation of heat exchangers in actual production. Summary of the Invention

[0004] The present application provides a heat exchanger optimization method and device based on multi-source data, which can accurately identify the local overheating area of ​​the heat exchanger and effectively optimize it, thereby reducing the problem of heat exchanger damage caused by local overheating and improving the heat transfer efficiency and operating stability of the heat exchanger.

[0005] In a first aspect, the present application provides a heat exchanger optimization method based on multi-source data, comprising:

[0006] Collecting multi-source operating data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operating data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data;

[0007] Building a flow simulation model of the heat exchanger based on the multi-source operating data;

[0008] Obtaining centrifugal force and flow velocity in different regions of the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position;

[0009] A random forest algorithm is used to extract coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix;

[0010] identifying a local overheating area according to the coupling characteristics;

[0011] The local overheating area is optimized using a preset thermal compensation mechanism.

[0012] Optionally, building a flow simulation model of the heat exchanger based on the multi-source operating data includes:

[0013] constructing a three-dimensional physical model using the dimensional parameters of the heat exchanger;

[0014] Using a hybrid grid strategy to grid the three-dimensional physical model;

[0015] Constructing a turbulence model using the centrifugal force distribution gradient data as the input of the turbulence generation term, the flow velocity radial stratification data as the initial condition, and the vorticity information in the fluid motion trajectory data as the initial disturbance;

[0016] The three-dimensional physical model, mesh division results and turbulence model are organically integrated to obtain a flow simulation model.

[0017] Optionally, obtaining the centrifugal force and flow velocity in different areas of the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position includes:

[0018] Running the flow simulation model to obtain simulation data under different working conditions;

[0019] extracting centrifugal forces and flow velocities corresponding to different regions in the heat exchanger based on the simulation data;

[0020] Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different areas;

[0021] A centrifugal force-flow velocity correlation matrix based on spatial position is constructed according to the centrifugal force, flow velocity and correlation coefficient corresponding to different areas. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different areas.

[0022] Optionally, the extracting coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix using a random forest algorithm includes:

[0023] generating a time series of temperature gradient changes over time according to the temperature gradient fluctuation data;

[0024] Using a Kalman filter algorithm to fuse the time series and the centrifugal force-flow velocity correlation matrix to obtain a coupling matrix;

[0025] The coupling features in the coupling matrix are extracted using a random forest algorithm.

[0026] Optionally, identifying the local overheating area according to the coupling feature includes:

[0027] Determining coupling features that do not meet preset conditions as abnormal features;

[0028] Extracting spatial location information from the abnormal features;

[0029] A local overheating area is determined according to the spatial position information.

[0030] Optionally, optimizing the local overheating area by using a preset thermal compensation mechanism includes:

[0031] Extracting an influencing factor affecting the local overheating area from the coupling characteristics;

[0032] Obtaining a preset thermal compensation mechanism matching the influencing factor from a preset database;

[0033] The local overheating area is optimized according to the preset thermal compensation mechanism.

[0034] A second aspect of the present application provides a heat exchanger optimization device based on multi-source data, comprising:

[0035] An acquisition unit is used to acquire multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operation data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data;

[0036] A building unit, configured to build a flow simulation model of the heat exchanger based on the multi-source operating data;

[0037] a generating unit, configured to obtain the centrifugal force and flow velocity of different regions in the heat exchanger according to the flow simulation model, so as to generate a centrifugal force-flow velocity correlation matrix based on spatial position;

[0038] an extraction unit, configured to extract coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix using a random forest algorithm;

[0039] an identification unit, configured to identify a local overheating area according to the coupling feature;

[0040] The optimization unit is used to optimize the local overheating area by using a preset thermal compensation mechanism.

[0041] Optionally, the construction unit is specifically used for:

[0042] constructing a three-dimensional physical model using the dimensional parameters of the heat exchanger;

[0043] Using a hybrid grid strategy to grid the three-dimensional physical model;

[0044] Constructing a turbulence model using the centrifugal force distribution gradient data as the input of the turbulence generation term, the flow velocity radial stratification data as the initial condition, and the vorticity information in the fluid motion trajectory data as the initial disturbance;

[0045] The three-dimensional physical model, mesh division results and turbulence model are organically integrated to obtain a flow simulation model.

[0046] Optionally, the generating unit is specifically configured to:

[0047] Running the flow simulation model to obtain simulation data under different working conditions;

[0048] extracting centrifugal forces and flow velocities corresponding to different regions in the heat exchanger based on the simulation data;

[0049] Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different areas;

[0050] A centrifugal force-flow velocity correlation matrix based on spatial position is constructed according to the centrifugal force, flow velocity and correlation coefficient corresponding to different areas. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different areas.

[0051] Optionally, the extraction unit is specifically configured to:

[0052] generating a time series of temperature gradient changes over time according to the temperature gradient fluctuation data;

[0053] Using a Kalman filter algorithm to fuse the time series and the centrifugal force-flow velocity correlation matrix to obtain a coupling matrix;

[0054] The coupling features in the coupling matrix are extracted using a randomized Morinda algorithm.

[0055] It can be seen from the above technical solutions that this application has the following effects:

[0056] First, multi-source operational data and temperature gradient fluctuation data are collected during the operation of the heat exchanger. This multi-source operational data includes fluid trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data. A flow simulation model of the heat exchanger is then constructed based on this multi-source operational data. The centrifugal force and flow velocity in different regions of the heat exchanger are then derived from the flow simulation model to generate a spatially determined centrifugal force-velocity correlation matrix. A random forest algorithm is then used to extract the coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-velocity correlation matrix. Local overheating regions are then identified based on these coupling characteristics. Finally, a pre-set thermal compensation mechanism is used to optimize these local overheating regions. In this way, a flow simulation model can be constructed using multi-source operational data, including fluid trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data, to map the actual internal operating state of the heat exchanger. Furthermore, the temperature gradient fluctuation data is combined to identify the precise local overheating regions, allowing for targeted optimization of these regions. This approach can reduce the risk of heat exchanger damage caused by local overheating and improve the heat transfer efficiency and operational stability of the heat exchanger. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of an embodiment of a heat exchanger optimization method based on multi-source data in this application;

[0058] Figure 2-1 、 Figure 2-2 and Figure 2-3 This is a schematic diagram of another embodiment of a heat exchanger optimization method based on multi-source data in this application;

[0059] Figure 3 This is a schematic diagram of an embodiment of a heat exchanger optimization device based on multi-source data in this application. DETAILED DESCRIPTION

[0060] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0061] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0062] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0063] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0064] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0065] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0066] To reduce localized overheating, existing technologies typically adjust the fluid flow within a heat exchanger based on real-time temperature or flow rate data during operation. However, the heat exchange process within a heat exchanger is a complex physical process, involving the interaction of multiple factors, such as fluid motion, force distribution, and temperature changes. Using only a single data type makes it difficult to fully and accurately reflect the actual operating conditions within the heat exchanger, and thus, unable to deeply explore the inherent connections between these factors. This makes it difficult to accurately identify localized overheating areas, making it difficult to meet the requirements for efficient and stable heat exchanger operation in actual production.

[0067] Based on this, the present application discloses a heat exchanger optimization method and device based on multi-source data, which can accurately identify the local overheating areas of the heat exchanger and perform effective optimization, thereby reducing the problem of heat exchanger damage caused by local overheating and improving the heat transfer efficiency and operating stability of the heat exchanger.

[0068] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0069] The application described in this application is implemented in a system, terminal or other device with logical analysis and processing capabilities. Figure 1 As shown, an embodiment of the heat exchanger optimization method based on multi-source data in the present application includes:

[0070] 101. Collect multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operation data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data;

[0071] In this embodiment, multi-source operation data is used to reflect the motion characteristics of the fluid in the heat exchanger from different dimensions, and temperature gradient fluctuation data is used to reflect the thermal state of the heat exchanger. In the process of collecting multi-source operation data, for fluid motion trajectory data: multiple high-precision tracer particle sensors are installed inside the heat exchanger, and tracer particles with specific optical or electromagnetic properties are introduced into the fluid. The motion trajectory of the tracer particles is captured in real time by the sensors, thereby obtaining fluid motion trajectory data. For centrifugal force distribution gradient data: micro pressure sensors and acceleration sensors are installed on the key components of the heat exchanger, and the pressure and acceleration changes generated by the fluid on these key components during the movement are measured, combined with the density and motion parameters of the fluid, to calculate the centrifugal force distribution gradient data. For radial stratification data of flow velocity: an ultrasonic Doppler flowmeter is used to measure the fluid in the heat exchanger, and measurement points are set at different radial positions of the heat exchanger to obtain flow velocity data at different radial positions, thereby obtaining radial stratification data of flow velocity. During the collection of temperature gradient fluctuation data, high-precision temperature sensors are installed in various areas of the heat exchanger to monitor the temperature changes at different locations of the heat exchanger in real time, record the temperature fluctuations over time, and obtain temperature gradient fluctuation data.

[0072] 102. Build a flow simulation model of a heat exchanger based on multi-source operating data;

[0073] In this example, computational fluid dynamics software was used to input fluid trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data as boundary conditions and initial conditions into the simulation model. For example, in ANSYS Fluent software, the fluid inlet velocity and direction were determined based on the collected fluid trajectory data, making the simulation more realistic when the fluid enters the heat exchanger. The centrifugal force distribution gradient data was converted into corresponding force boundary conditions and applied to the model to simulate the effects of centrifugal force on the fluid. Flow velocity parameters for different regions were set based on the radial velocity stratification data to accurately reflect the stratified flow characteristics of the fluid within the heat exchanger. This resulted in a flow simulation model that accurately reflects the flow state of the fluid within the heat exchanger.

[0074] 103. Obtain the centrifugal force and flow velocity in different areas of the heat exchanger based on the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position;

[0075] In this embodiment, a centrifugal force-flow velocity correlation matrix is ​​used to quantify and structure the centrifugal force and flow velocity information within the heat exchanger's interior, facilitating subsequent analysis of the relationship between the two and their association with temperature gradient fluctuation data, providing a data foundation for identifying localized overheating areas. After the flow simulation model is constructed, it is run, and the centrifugal force and flow velocity for each grid created during the model construction process are calculated using fluid mechanics equations. During the calculation process, the effect of grid size and shape on the results must be considered. A grid that is too large will result in rough results and fail to accurately reflect the characteristics of the local area; a grid that is too small will increase the computational effort and time. After the calculated data is obtained, it is arranged according to the spatial position of the grid cells to form a matrix, with each element corresponding to the centrifugal force and flow velocity information for a grid cell. By analyzing the matrix, specific distribution patterns of centrifugal force and flow velocity in certain areas can be identified. For example, in a spirally wound heat exchanger in a chemical production facility, the centrifugal force-flow velocity correlation matrix reveals that the centrifugal force is greater at the curved portion of the spiral tube with a smaller radius of curvature, resulting in higher flow velocity on the outside of the tube and lower flow velocity on the inside, creating a distinct flow velocity stratification phenomenon.

[0076] 104. Use the random forest algorithm to extract the coupling characteristics between temperature gradient fluctuation data and centrifugal force-flow velocity correlation matrix;

[0077] In this embodiment, there is a complex nonlinear relationship between the temperature gradient fluctuation data and the centrifugal force-flow velocity association matrix. The random forest algorithm, as an integrated learning algorithm, can process high-dimensional data. By constructing multiple decision trees and comprehensively predicting the results, it has strong generalization ability and robustness, and can extract the coupling characteristics between the two. The temperature gradient fluctuation data and the centrifugal force-flow velocity association matrix are used as input data of the random forest algorithm. A plurality of decision trees are set, and each decision tree is trained based on the input data. During the training process, the potential relationship in the data can be mined through random sampling and feature selection of the input data. For example, after training, the random forest algorithm finds that when the centrifugal force in a certain area is lower than a certain threshold and the flow velocity is also low, the temperature gradient will increase abnormally. This relationship is the coupling feature extracted by the random forest algorithm. It can be understood that by adjusting parameters such as the number of decision trees, the sample selection ratio and the feature selection ratio, the algorithm performance can be optimized and the accuracy and reliability of the coupling feature extraction can be improved.

[0078] 105. Identify local overheating areas based on coupling characteristics;

[0079] In this embodiment, when a region in the centrifugal force-flow velocity correlation matrix data exhibits a specific distribution pattern based on the coupling characteristics, and the corresponding temperature gradient fluctuation data increases abnormally, the region is determined to be a local overheating region. For example, in the centrifugal force-flow velocity correlation matrix, if a grid region has low centrifugal force and low flow velocity, and the temperature in this region continues to rise above a set threshold, the region can be identified as a local overheating region. It should be noted that in actual applications, it is necessary to set a reasonable temperature threshold. If the threshold is too high, the local overheating region may not be identified in a timely manner, while if the threshold is too low, it may lead to misidentification.

[0080] 106. Use the preset thermal compensation mechanism to optimize the local overheating area.

[0081] It is understood that during the actual operation of a heat exchanger, local overheating can be caused by a variety of factors, and different measures may be required to address these factors. In this embodiment, a preset thermal compensation mechanism is utilized to address these factors and dynamically optimize the local overheating area.

[0082] In this embodiment, multi-source operational data and temperature gradient fluctuation data are first collected during the operation of the heat exchanger. This multi-source operational data includes fluid motion trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data. A flow simulation model of the heat exchanger is then constructed based on the multi-source operational data. The centrifugal force and flow velocity in different regions of the heat exchanger are then obtained based on the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial location. A random forest algorithm is further used to extract the coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix. Local overheating regions are further identified based on the coupling characteristics. Finally, a preset thermal compensation mechanism is used to optimize the local overheating regions. In this way, a flow simulation model can be constructed using multi-source operational data such as fluid motion trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data during the operation of the heat exchanger. The flow simulation model can then be used to map the actual internal operating state of the heat exchanger. Furthermore, the temperature gradient fluctuation data is combined to accurately identify the local overheating regions, allowing for targeted optimization of these local overheating regions. This can reduce the problem of heat exchanger damage caused by local overheating and improve the heat transfer efficiency and operating stability of the heat exchanger.

[0083] See also Figure 2-1 、 Figure 2-2 and Figure 2-3 As shown, another embodiment of the heat exchanger optimization method based on multi-source data in the present application includes:

[0084] 201. Collect multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operation data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data;

[0085] Step 201 in this embodiment is the same as the aforementioned Figure 1 Step 101 in the illustrated embodiment is similar and will not be described again here.

[0086] 202. Use the dimensional parameters of the heat exchanger to construct a three-dimensional physical model;

[0087] 203. Use hybrid grid strategy to mesh the 3D physical model;

[0088] 204. Construct a turbulence model using centrifugal force distribution gradient data as the input of turbulence generation term, velocity radial stratification data as initial condition, and vorticity information in fluid motion trajectory data as initial disturbance;

[0089] 205. Organically integrate the three-dimensional physical model, meshing results and turbulence model to obtain a flow simulation model;

[0090] Optionally, in this embodiment, the dimensional parameters of the heat exchanger are the basis for constructing a three-dimensional physical model. These parameters include but are not limited to the length, width, height, diameter of the heat exchange tubes, tube spacing, number of tube rows, and diameter of the shell of the heat exchanger. In actual operation, these dimensional parameters can first be obtained from design drawings, product manuals and other materials. For example, for a shell and tube heat exchanger, its shell diameter is 1000mm and its length is 3000mm. There are 200 heat exchange tubes arranged inside, the heat exchange tube diameter is 25mm, the tube spacing is 32mm, and the number of tube rows is 10. Then, using professional three-dimensional modeling software, the shell, heat exchange tubes and other components of the heat exchanger are drawn in sequence according to these dimensional parameters, and finally combined to form a complete three-dimensional physical model of the heat exchanger. At the same time, for some complex structures, such as baffles and distributors, it is also necessary to accurately model according to the actual dimensions to truly reflect the internal geometric structure of the heat exchanger.

[0091] The mesh is used to discretize the three-dimensional physical model into computational units. As the discretization carrier for the governing equations, the mesh quality directly determines the accuracy of the numerical solution. When meshing, a hybrid meshing strategy can be used. This strategy combines the advantages of structured and unstructured meshes. Structured meshing is used for regions of the heat exchanger with regular geometry and simple flow, such as the interior of straight heat exchange tubes and regular shell sections. Structured meshing offers the advantages of high mesh quality, high computational accuracy, and minimal data storage, enabling efficient computation in these areas. Conversely, unstructured meshing is used for regions of the heat exchanger with complex geometry and drastic flow variations, such as those near baffles and at inlet and outlet pipes. Unstructured meshing better adapts to complex geometries and accurately captures flow details. Alternatively, based on the velocity gradient distribution in the radially layered flow data, the mesh can be appropriately refined in areas with large velocity gradients to improve computational accuracy. This not only more accurately captures flow details but also avoids unnecessary mesh refinement, improving computational efficiency.

[0092] Based on the Reynolds number and flow complexity, the Realizable k-ε model or the LES model is selected as the turbulence model. When setting the turbulence model parameters, because centrifugal force can induce turbulence in the fluid, the centrifugal force distribution gradient data is used as the input for the turbulence generation term. The influence of centrifugal force on turbulence intensity is reflected by modifying the turbulent viscosity calculation formula. The radial stratification of flow velocity data describes the velocity distribution of the fluid in the radial direction of the heat exchanger. This radial stratification of flow velocity data can be used as the initial condition to set the velocity distribution on the pipe cross section, making the turbulence model more consistent with the actual fluid flow state. Vorticity information is obtained by analyzing the fluid motion trajectory through image analysis and fluid dynamics calculations. Vortices play a key role in the formation and development of turbulence. Inputting vorticity information into the turbulence model as an initial perturbation can stimulate the generation of turbulence in the model, enabling the model to more accurately simulate the initiation and development of turbulence and enhance the ability to capture flow instabilities.

[0093] Finally, the inlet boundary condition is set to mass flow inlet, the outlet boundary condition to pressure outlet, and the wall condition to no-slip wall. The appropriate wall roughness is set based on the plate material and surface treatment. The constructed turbulence model is then associated with the 3D physical model and mesh to create a complete flow simulation model. This fully accounts for various factors affecting turbulence, enabling the model to accurately simulate the turbulent flow and heat transfer processes within the heat exchanger, significantly improving the accuracy and reliability of the simulation model.

[0094] 206. Run the flow simulation model to obtain simulation data under different working conditions;

[0095] 207. Extract the centrifugal force and flow rate corresponding to different areas in the heat exchanger based on simulation data;

[0096] 208. Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different regions;

[0097] 209. Constructing a centrifugal force-flow velocity correlation matrix based on spatial position according to the centrifugal force, flow velocity and correlation coefficient corresponding to different regions. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different regions.

[0098] Optionally, in this embodiment, the operating conditions refer to various conditions that affect the flow and heat transfer of the fluid inside the heat exchanger, such as: the inlet flow rate, temperature, pressure of the fluid, and the working environment temperature of the heat exchanger. In actual operation, it is first necessary to set a series of different operating parameters according to the actual working scenario and research purpose of the heat exchanger. These operating parameters are input into the flow simulation model, and by running the flow model, simulation data of the fluid flow inside the heat exchanger under different operating conditions are obtained. These simulation data contain various physical quantity information of each area of ​​the heat exchanger under different operating conditions, such as velocity vector, pressure distribution, temperature distribution, etc., which are not specifically limited here. After obtaining the simulation data under different working conditions, the centrifugal force and flow rate of each grid are calculated. It should be noted that for the calculation of centrifugal force, in the grid area where the fluid performs circular motion or curvilinear motion, the centrifugal force formula can be used according to parameters such as the velocity, density and motion radius of the fluid. Calculate, where is the centrifugal force, is the mass of the fluid element, is the velocity of the fluid element, is the motion radius. For the calculation of flow rate, the velocity vector information of the fluid in each area can be directly obtained from the simulation data, and then the magnitude and direction of the flow rate can be obtained. The correlation coefficient is an indicator to measure the degree of linear correlation between two variables. The Pearson correlation coefficient can be used to calculate the correlation coefficient between centrifugal force and flow rate. After obtaining the centrifugal force, flow rate and the correlation coefficient between them in different areas, a centrifugal force-flow rate correlation matrix based on spatial position can be constructed. The matrix is ​​based on the regional division of the heat exchanger, and the centrifugal force, flow rate and correlation coefficient of each area are arranged as elements of the matrix. Through this centrifugal force-flow rate correlation matrix, the correlation between centrifugal force and flow rate in different areas can be clearly mapped, which provides a comprehensive and effective data basis for the optimal design of the heat exchanger, helps to improve the performance of the heat exchanger, and thus improves energy utilization efficiency and reduces production costs.

[0099] 210. Generate a time series of temperature gradient changes over time based on temperature gradient fluctuation data;

[0100] 211. The Kalman filter algorithm is used to fuse the time series and the centrifugal force-velocity correlation matrix to obtain the coupling matrix;

[0101] 212. Use random forest algorithm to extract coupling features from coupling matrix;

[0102] Optionally, in this embodiment, the temperature gradient fluctuation data reflects the dynamic changes in the temperature gradient distribution of the heat exchanger during operation. By arranging the temperature gradient data in the temperature gradient fluctuation data in chronological order, a time series of temperature gradient changes over time can be generated. The Kalman filter algorithm is used to fuse the temperature gradient time series and the centrifugal force-flow velocity correlation matrix to output a coupling matrix based on space and time. The coupling matrix is ​​used to reflect the relationship between temperature gradient, centrifugal force and flow velocity. The coupling matrix is ​​used as the input of the random forest algorithm. For each data point in the coupling matrix, the random forest algorithm divides it into different decision tree nodes based on the characteristics of the data. Through the construction and training of a large number of decision trees, the random forest algorithm can automatically learn the complex relationship between different variables in the data, thereby extracting the coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix. For example, in a plate heat exchanger, the coupling matrix contains the temperature gradient, centrifugal force, flow velocity and the correlation information between each area. When processing this data, the random forest algorithm may discover that in certain areas, when the temperature gradient fluctuates within a certain range, the centrifugal force and flow velocity exhibit specific trends. This trend is a coupling feature. This coupling feature reflects the inherent connection between temperature changes and fluid flow conditions, providing a deeper understanding of the interaction between various physical quantities within the heat exchanger.

[0103] 213. Determine that a coupling feature that does not meet the preset conditions is an abnormal feature;

[0104] 214. Extracting spatial location information from abnormal features;

[0105] 215. Determine the local overheating area based on the spatial location information;

[0106] Optionally, in this embodiment, after obtaining the coupling feature, it is determined whether the coupling feature meets preset conditions, such as whether the temperature gradient fluctuation meets a threshold, whether the centrifugal force distribution uniformity meets a threshold, or whether the radial velocity distribution uniformity meets a threshold. When the coupling feature does not meet the preset conditions, it can be determined that the coupling feature is an abnormal feature. Since each coupling feature contains the spatial position information of the corresponding grid, after obtaining the abnormal feature, it can be determined that the area where the grid corresponding to the abnormal feature is located is a local overheating area. It is understandable that the operating status of the heat exchanger can be intuitively presented by drawing a temperature distribution cloud map, marking the local overheating area, etc., so that the operator can take timely measures to optimize it.

[0107] 216. Extract the influencing factors affecting the local overheating area from the coupling characteristics;

[0108] 217. Obtaining a preset thermal compensation mechanism that matches the impact factor from a preset database;

[0109] 218. Optimize local overheating areas based on the preset thermal compensation mechanism.

[0110] Optionally, in this embodiment, factors significantly influencing localized overheating are extracted through methods such as correlation analysis. For example, if low flow velocity in a certain area leads to insufficient heat exchange, causing heat accumulation and resulting in localized overheating, flow velocity is a significant influencing factor. Alternatively, if the heat flux in a certain area is too high, exceeding the normal heat dissipation capacity of the heat exchanger, heat flux also becomes an influencing factor. Furthermore, factors such as changes in fluid physical properties (such as viscosity and specific heat capacity) and heat exchanger structural factors (such as channel blockage and fin damage) may also affect localized overheating. After extracting the influencing factors affecting the localized overheating area, these factors are used as search criteria to perform a search and match against a pre-set database. For example, if the influencing factor is determined to be low flow velocity in a certain localized area, searching the pre-set database for the keyword "low flow velocity" will reveal multiple matching pre-set thermal compensation mechanisms. These may include adjusting the opening of the heat exchanger's inlet and outlet valves to increase fluid flow in that area, or optimizing the internal flow channel structure of the heat exchanger to reduce fluid flow resistance and increase flow velocity. If the influencing factor is excessive heat flux, the pre-set thermal compensation mechanism matched in the database may increase the heat transfer area in that area, such as installing heat sink fins, or introduce an auxiliary cooling medium to remove excess heat. By accurately extracting the influencing factors affecting the local overheating area from the coupling characteristics, we can deeply analyze the root cause of the local overheating and provide targeted solutions for different overheating causes. This effectively solves the local overheating problem, improves the stability and heat transfer efficiency of the heat exchanger, and provides reliable guarantees for the efficient operation of the heat exchanger.

[0111] See also Figure 3 As shown, an embodiment of the heat exchanger optimization device based on multi-source data in the present application includes:

[0112] The acquisition unit 301 is used to acquire multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation. The multi-source operation data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data.

[0113] A building unit 302 is used to build a flow simulation model of the heat exchanger based on multi-source operating data;

[0114] A generating unit 303 is configured to obtain the centrifugal force and flow velocity of different regions in the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position;

[0115] An extraction unit 304 is configured to extract coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix using a random forest algorithm;

[0116] an identification unit 305 for identifying a local overheating area based on the coupling characteristics;

[0117] The optimization unit 306 is configured to optimize the local overheating area by using a preset thermal compensation mechanism.

[0118] In this embodiment, a collection unit 301 collects multi-source operating data and temperature gradient fluctuation data during the operation of the heat exchanger. The multi-source operating data includes fluid motion trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data. A construction unit 302 constructs a flow simulation model for the heat exchanger based on the multi-source operating data. A generation unit 303 obtains the centrifugal force and flow velocity in different regions of the heat exchanger based on the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial location. An extraction unit 304 uses a random forest algorithm to extract coupling features between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix. An identification unit 305 identifies local overheating regions based on the coupling features. An optimization unit 306 optimizes the local overheating regions using a preset thermal compensation mechanism. In this way, a flow simulation model can be constructed using multi-source operating data, such as fluid motion trajectory data, centrifugal force distribution gradient data, and radial velocity stratification data, to map the actual internal operating state of the heat exchanger. Furthermore, the flow simulation model is combined with the temperature gradient fluctuation data to identify precise local overheating regions for targeted optimization. This can reduce the problem of heat exchanger damage caused by local overheating and improve the heat transfer efficiency and operating stability of the heat exchanger.

[0119] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0122] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk, and other media that can store program code.

Claims

1. A heat exchanger optimization method based on multi-source data, characterized in that: include: Collecting multi-source operating data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operating data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data; Building a flow simulation model of the heat exchanger based on the multi-source operating data; Obtaining centrifugal force and flow velocity in different regions of the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position; A random forest algorithm is used to extract coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix; identifying a local overheating area according to the coupling characteristics; Optimizing the local overheating area using a preset thermal compensation mechanism; Wherein, building the flow simulation model of the heat exchanger based on the multi-source operation data includes: constructing a three-dimensional physical model using the dimensional parameters of the heat exchanger; Using a hybrid grid strategy to grid the three-dimensional physical model; Constructing a turbulence model using the centrifugal force distribution gradient data as the input of the turbulence generation term, the flow velocity radial stratification data as the initial condition, and the vorticity information in the fluid motion trajectory data as the initial disturbance; Organically integrating the three-dimensional physical model, the meshing result, and the turbulence model to obtain a flow simulation model; The step of obtaining the centrifugal force and flow velocity in different regions of the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position includes: Running the flow simulation model to obtain simulation data under different working conditions; extracting centrifugal forces and flow velocities corresponding to different regions in the heat exchanger based on the simulation data; Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different areas; A centrifugal force-flow velocity correlation matrix based on spatial position is constructed according to the centrifugal force, flow velocity and correlation coefficient corresponding to different areas. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different areas.

2. The heat exchanger optimization method based on multi-source data according to claim 1, characterized in that: The method of extracting the coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix using a random forest algorithm includes: generating a time series of temperature gradient changes over time based on the temperature gradient fluctuation data; Using a Kalman filter algorithm to fuse the time series and the centrifugal force-flow velocity correlation matrix to obtain a coupling matrix; The coupling features in the coupling matrix are extracted using a random forest algorithm.

3. The heat exchanger optimization method based on multi-source data according to claim 1, characterized in that: The identifying of the local overheating area according to the coupling characteristics includes: Determining coupling features that do not meet preset conditions as abnormal features; Extracting spatial location information from the abnormal features; A local overheating area is determined according to the spatial position information.

4. The heat exchanger optimization method based on multi-source data according to any one of claims 1 to 3, characterized in that: The optimizing the local overheating area by using a preset thermal compensation mechanism includes: Extracting an influencing factor affecting the local overheating area from the coupling characteristics; Obtaining a preset thermal compensation mechanism matching the influencing factor from a preset database; The local overheating area is optimized according to the preset thermal compensation mechanism.

5. A heat exchanger optimization device based on multi-source data, characterized in that: include: An acquisition unit is used to acquire multi-source operation data and temperature gradient fluctuation data of the heat exchanger during operation, wherein the multi-source operation data includes fluid motion trajectory data, centrifugal force distribution gradient data, and flow velocity radial stratification data; A building unit, configured to build a flow simulation model of the heat exchanger based on the multi-source operating data; a generating unit, configured to obtain the centrifugal force and flow velocity of different regions in the heat exchanger according to the flow simulation model, so as to generate a centrifugal force-flow velocity correlation matrix based on spatial position; an extraction unit, configured to extract coupling characteristics between the temperature gradient fluctuation data and the centrifugal force-flow velocity correlation matrix using a random forest algorithm; an identification unit, configured to identify a local overheating area according to the coupling feature; An optimization unit for optimizing the local overheating area using a preset thermal compensation mechanism; Wherein, building the flow simulation model of the heat exchanger based on the multi-source operation data includes: constructing a three-dimensional physical model using the dimensional parameters of the heat exchanger; Using a hybrid grid strategy to grid the three-dimensional physical model; Constructing a turbulence model using the centrifugal force distribution gradient data as the input of the turbulence generation term, the flow velocity radial stratification data as the initial condition, and the vorticity information in the fluid motion trajectory data as the initial disturbance; Organically integrating the three-dimensional physical model, the meshing result, and the turbulence model to obtain a flow simulation model; The step of obtaining the centrifugal force and flow velocity in different regions of the heat exchanger according to the flow simulation model to generate a centrifugal force-flow velocity correlation matrix based on spatial position includes: Running the flow simulation model to obtain simulation data under different working conditions; extracting centrifugal forces and flow velocities corresponding to different regions in the heat exchanger based on the simulation data; Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different areas; A centrifugal force-flow velocity correlation matrix based on spatial position is constructed according to the centrifugal force, flow velocity and correlation coefficient corresponding to different areas. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different areas.

6. The heat exchanger optimization device based on multi-source data according to claim 5, characterized in that: The construction unit is specifically used for: constructing a three-dimensional physical model using the dimensional parameters of the heat exchanger; Using a hybrid grid strategy to grid the three-dimensional physical model; Constructing a turbulence model using the centrifugal force distribution gradient data as the input of the turbulence generation term, the flow velocity radial stratification data as the initial condition, and the vorticity information in the fluid motion trajectory data as the initial disturbance; The three-dimensional physical model, mesh division results and turbulence model are organically integrated to obtain a flow simulation model.

7. The heat exchanger optimization device based on multi-source data according to claim 5, characterized in that: The generating unit is specifically configured to: Running the flow simulation model to obtain simulation data under different working conditions; extracting centrifugal forces and flow velocities corresponding to different regions in the heat exchanger based on the simulation data; Calculate the correlation coefficient between the centrifugal force and flow velocity corresponding to different areas; A centrifugal force-flow velocity correlation matrix based on spatial position is constructed according to the centrifugal force, flow velocity and correlation coefficient corresponding to different areas. The centrifugal force-flow velocity correlation matrix is ​​used to map the correlation between centrifugal force and flow velocity in different areas.

8. The heat exchanger optimization device based on multi-source data according to claim 5, characterized in that: The extraction unit is specifically used for: generating a time series of temperature gradient changes over time based on the temperature gradient fluctuation data; Using a Kalman filter algorithm to fuse the time series and the centrifugal force-flow velocity correlation matrix to obtain a coupling matrix; The coupling features in the coupling matrix are extracted using a randomized Morinda algorithm.

Images