A method, apparatus, device, and medium for predicting a maximum temperature of a cable channel

Abstract

A cable channel maximum temperature prediction method, device, equipment and medium are disclosed. The features include: obtaining input parameters of electromagnetic heat flow multi-physics field coupling of the cable channel, constructing a sample database based on the input parameters; determining training data based on the sample database, training a structure prediction sub-model, an environment prediction sub-model and a material prediction sub-model based on the training data and the input parameters; determining a maximum temperature prediction model based on the structure prediction sub-model, the environment prediction sub-model and the material prediction sub-model; collecting cable channel parameters of a target cable channel, performing maximum temperature prediction on the cable channel parameters through the maximum temperature prediction model, and determining a maximum temperature prediction result of the target cable channel. The present application significantly reduces the calculation cost of the maximum temperature prediction of the cable channel, effectively improves the evaluation efficiency, prediction accuracy and stability.

Description

Technical Field

[0001] This invention relates to the field of cable channel technology, and in particular to a method, apparatus, equipment and medium for predicting the maximum temperature of a cable channel. Background Technology

[0002] With the advancement of urban power grid construction and capacity expansion, cable tunnels and prefabricated cable channels are widely used. During operation, cables generate conductor losses, and eddy current losses occur in the cable armor and the metal structure of the channel. These losses, acting as heat sources, are superimposed on the internal heat transfer of the channel and external boundary conditions, determining the temperature rise level and maximum temperature of the cable channel. To ensure the safety margin of cable insulation and sheath materials, the engineering design phase requires a rapid and reliable assessment of the maximum channel temperature under different cable structural dimensions and material properties, channel structure and materials, and environmental conditions. Existing cable channel thermal assessments typically rely on electromagnetic-thermal-fluid multiphysics finite element simulations. This method can obtain high-precision temperature distribution and maximum temperature, but it has shortcomings in multivariate parameterized analysis and rapid scheme iteration: on the one hand, the coupling of factors such as structural size, material parameters, channel geometry and layout, environmental medium and thermal boundary leads to the need for a large number of operating condition simulations to cover typical combinations, resulting in high sample acquisition costs and long cycles; on the other hand, the combined effect of flow heat transfer and loss heat source distribution in the channel further increases the consumption of simulation modeling and computing resources, making it difficult to meet the needs of rapid evaluation and parameter optimization in the design stage. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and medium for predicting the maximum temperature of a cable channel, in order to solve the technical problem that the maximum temperature of a cable channel cannot be accurately and quickly predicted in the prior art.

[0004] According to one aspect of the present invention, a method for predicting the maximum temperature of a cable channel is provided, comprising: The input parameters of electromagnetic, thermal, and current multiphysics coupling in the cable channel are obtained, and a sample database is constructed based on the input parameters. Training data is determined based on the sample database, and structural prediction sub-model, environmental prediction sub-model, and material prediction model are trained based on the training data and the input parameters, respectively. The highest temperature prediction model is determined based on the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model. The cable channel parameters of the target cable channel are collected, and the maximum temperature of the cable channel parameters is predicted using the maximum temperature prediction model to determine the maximum temperature prediction result of the target cable channel.

[0005] According to another aspect of the present invention, a maximum temperature prediction device for a cable channel is provided, comprising: The parameter processing module is used to acquire the input parameters of electromagnetic, thermal and current multi-physics coupling in the cable channel, and to construct a sample database based on the input parameters. The model training module is used to determine training data based on the sample database, and to train the structure prediction sub-model, environment prediction sub-model, and material prediction model based on the training data and the input parameters, respectively. The model fusion module is used to determine the maximum temperature prediction model based on the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model. The prediction module is used to collect cable channel parameters of the target cable channel, predict the maximum temperature of the cable channel parameters using the maximum temperature prediction model, and determine the maximum temperature prediction result of the target cable channel.

[0006] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the maximum temperature prediction method for cable channels according to any embodiment of the present invention.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the maximum temperature prediction method for cable channels according to any embodiment of the present invention.

[0008] The technical solution of this invention obtains the input parameters of the electromagnetic-thermal-fluid multi-physics coupling of the cable channel, constructs a sample database based on the input parameters, and significantly reduces computational costs and improves evaluation efficiency by establishing the sample database. Training data is determined based on the sample database, and structural prediction sub-models, environmental prediction sub-models, and material prediction sub-models are trained based on the training data and the input parameters, respectively. A maximum temperature prediction model is determined based on the structural, environmental, and material prediction sub-models, and the input parameters are modeled according to structural, material, and environmental branches, effectively improving prediction accuracy and stability, and enhancing prediction accuracy and generalization ability. Cable channel parameters of the target cable channel are collected, and the maximum temperature is predicted using the maximum temperature prediction model. The maximum temperature prediction model can quickly predict the maximum temperature of the target operating condition without repeating the full-process finite element simulation, solving the technical problem of inaccurate and rapid prediction of the maximum temperature of cable channels in the prior art, reducing the workload of repetitive simulation, and effectively improving the efficiency of multi-condition thermal evaluation.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart of a method for predicting the maximum temperature of a cable channel is provided as an embodiment of the present invention; Figure 2 A waveform diagram of eddy current loss in the wall of a cable channel provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a steady-state temperature distribution cloud map of a cable channel provided in an embodiment of the present invention; Figure 4 A schematic diagram of fluid velocity distribution cloud map in a cable channel provided in an embodiment of the present invention; Figure 5 A flowchart of another method for predicting the maximum temperature of a cable channel provided in an embodiment of the present invention; Figure 6 A schematic diagram of the structure of a maximum temperature prediction device for a cable channel provided in an embodiment of the present invention; Figure 7 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. Detailed Implementation

[0012] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0013] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0014] Figure 1 This invention provides a flowchart of a method for predicting the maximum temperature of a cable channel, applicable to situations requiring prediction and identification of the maximum temperature of a cable channel. This method can be executed by a maximum temperature prediction device for the cable channel, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes: S110. Obtain the input parameters of electromagnetic-thermal-fluid multi-physics coupling of the cable channel, and construct a sample database based on the input parameters.

[0015] Optionally, the input parameters consist of cable parameters, channel parameters, and environmental parameters. Cable parameters include the cable's structural dimensions and material properties. Structural dimensions include cable type, conductor outer diameter, insulation outer diameter, inner sheath outer diameter, armor outer diameter, and outer sheath outer diameter. The cable type can be a single-core or three-core cable. Material properties include electrical conductivity, relative permeability, density, specific heat capacity, and thermal conductivity. Channel parameters include the channel's structural dimensions and material properties. Structural dimensions include the channel's inner diameter, outer diameter, length, and cable arrangement parameters within the channel. Arrangement parameters include the eccentricity between the cable and the channel center, cable spacing, and arrangement method. Material properties include electrical conductivity, relative permeability, density, specific heat capacity, and thermal conductivity. Environmental parameters include the environmental medium type, ambient temperature, thermal boundary parameters, and fluid boundary conditions. The environmental medium type can be air, soil, or seawater. The ambient temperature is the temperature of the medium within the channel. The thermal boundary parameter is the thermal resistivity, and the fluid boundary condition is the axial velocity of the medium within the channel.

[0016] For example, the cable channel of the present invention is a single-core cable, and the outer diameter and material properties of each structural layer of the cable are shown in the table below: The cable channel's structural dimensions are as follows: inner diameter 2453mm, outer diameter 2495mm, and length 3m. Cables are arranged symmetrically in groups within the channel cross-section, with the channel center set as the origin (0,0). Under this arrangement, the center coordinates (in mm) of each single-core cable conductor are: 440,70), (800,70), (620,520), (440,-380), (800,-380), (620,-780), (-440,70), (-800,70), (-620,520), (-440,-380), (-800,-380), (-620,-780). These coordinates determine the cable spacing, inter-group spacing, and arrangement, which are then input as channel layout parameters for subsequent calculations. The cable channel material properties are as follows: electrical conductivity 1.118×10⁶ S / m, relative magnetic permeability 285, density 7000 kg / m³, specific heat capacity 500 J / (kg·K), and thermal conductivity 35 W / (m·K). Environmental parameters are as follows: ambient medium type is air, medium temperature within the channel is 30℃; thermal boundary parameters are expressed using the equivalent heat transfer coefficient, taken as h = 200 W / (m²·K), with the corresponding equivalent thermal resistance per unit area being R = 1 / h (unit: (m²·K) / W); the axial velocity of the air within the channel is 1 m / s.

[0017] Optionally, the sample database can be a training database used to train the maximum temperature prediction model. It should be noted that the input parameters used when constructing the sample database include cable channel length, relative permeability of the channel material, axial air velocity within the channel, cable channel material properties, and environmental parameters.

[0018] Specifically, the input parameters of the electromagnetic-thermal-fluid multiphysics coupling of the cable channel are obtained, and a sample database is constructed based on the input parameters.

[0019] Optionally, in another optional embodiment of the present invention, the step of obtaining the input parameters of the electromagnetic-thermal-fluid multiphysics coupling of the cable channel and constructing a sample database based on the input parameters includes: At least one sample parameter combination is constructed based on the cable channel length, the relative permeability of the channel material, and the axial air velocity within the channel. For each sample parameter combination, electromagnetic field simulation and thermal-fluid coupling simulation are performed on the sample parameter combination based on the cable channel material property parameters and the environmental parameters to determine the maximum temperature of the cable channel. Data association is performed between each sample parameter combination and the maximum temperature of the cable channel corresponding to the sample parameter combination to construct the sample database.

[0020] The sample parameter combination can be a dataset consisting of cable channel length, relative permeability of channel material, and axial air velocity within the channel. A single sample parameter combination includes three data points: cable channel length, relative permeability of channel material, and axial air velocity within the channel. Different sample parameter combinations correspond to different cable channel lengths, relative permeability of channel material, and axial air velocity within the channel.

[0021] The highest temperature of the cable channel can be the highest temperature value obtained by performing electromagnetic-thermal-fluid multiphysics finite element simulation on a cable channel corresponding to a sample parameter combination.

[0022] Optionally, after obtaining the highest temperature of the cable channel corresponding to the sample parameter combination, the sample parameter combination is associated with the highest temperature of the cable channel to form a separate sample data. All the sample data corresponding to the sample parameter combinations are stored in the database to form a sample database.

[0023] Specifically, at least one sample parameter combination is constructed based on the cable channel length, the relative permeability of the channel material, and the axial air velocity within the channel. For each sample parameter combination, electromagnetic field simulation and thermal-fluid coupling simulation are performed on the sample parameter combination based on the cable channel material properties and environmental parameters to determine the maximum temperature of the cable channel. Data association is performed between each sample parameter combination and the corresponding maximum temperature of the cable channel to construct a sample database.

[0024] Optionally, in another optional embodiment of the present invention, the step of constructing at least one sample parameter combination based on the cable channel length, the relative permeability of the channel material, and the axial air velocity within the channel includes: An initial Latin hypercube sample matrix is ​​constructed based on the cable channel length, the relative permeability of the channel material, and the axial velocity of the air within the channel; the column permutations of the initial Latin hypercube sample matrix are iteratively exchanged to determine an optimized sample matrix; at least one combination of sample parameters is constructed based on the optimized sample matrix.

[0025] The initial Latin hypercube sample matrix can be a data matrix generated by initial sampling of the Latin hypercube.

[0026] Optionally, in this invention, the cable channel length can be represented by L, and the discrete value set of the cable channel length L is L∈{0.5,1,2,3}; the relative permeability of the channel material can be represented by μ. r The relative permeability μ of the channel material is represented as follows. r The discrete set of values ​​is μ r ∈{100,200,285,300,400,500}; The axial velocity of the air in the channel is calculated by V, and the discrete set of values ​​for the axial velocity of the air in the channel V is V∈{0.2,0.4,0.6,0.8,1}.

[0027] Optionally, in this invention, the initial sampling of the Latin hypercube in the continuous space is set to a sample size N of 120, in the normalized space [0,1]. 3 The three variables are each divided into N equal intervals. For each variable dimension, one sample value is drawn from each interval, and the sample values ​​for the three dimensions are randomly permuted and combined to obtain the initial Latin hypercube sample matrix. The initial Latin hypercube sample matrix U is denoted as U∈[0,1]. N×3 .

[0028] Optionally, the optimized sample matrix can be obtained by adjusting and optimizing the minimum distance between each sample point. During the initial Latin hypercube sample matrix sampling process, the minimum distances of the variable values ​​of each sample point were too close, leading to a certain correlation between the sample points. By iteratively swapping the column arrangements of the sample matrix, the minimum distance between each pair of sample points is increased, resulting in the optimized sample matrix. The optimized sample matrix is ​​obtained through U... * To express.

[0029] Optionally, after obtaining the optimized sample matrix, since sampling is performed in the normalized space, the optimized sample matrix U is obtained. *To obtain the normalized sample matrix, it is necessary to map the optimized sample matrix to actual discrete values. This is done by mapping each dimension of each sample point to the corresponding discrete set index according to its position in the [0,1] interval, and taking the corresponding discrete value as the final parameter value. For example, assuming the size of the discrete set of the cable channel length L is 4, then according to k... L =[u L * ·4]+1 to get the sequence number, and take L=L kL ; where k L L is the ordinal number of the variable. kL The kth discrete set of cable channel lengths L L Several variable values. Assuming the discrete set size of the axial air velocity V within the channel is 5, then according to k... V =[u V * ·5]+1 to get the sequence number, and take V=V kV ; where k V V is the ordinal number of the variable. kV The kth discrete set of air axial velocity V within the channel V Several variable values. Let the relative permeability μ of the channel material be... r If the size of the discrete set is 6, then according to k μ =[u μ * ·6]+1 to get the sequence number, and take μ r =μ r,kμ Where, k μ μ is the ordinal number of the variable. r,kμ The relative permeability μ of the channel material r discrete set k-th μ Each variable has a value. This results in a small sample input parameter set containing N sets of sample parameter combinations, where each set of sample parameter combinations is (L, V, μ). r This, together with the material properties and environmental parameters of the cable channel, forms the initial sample set.

[0030] Specifically, an initial Latin hypercube sample matrix is ​​constructed based on the cable channel length, the relative permeability of the channel material, and the axial velocity of the air in the channel; the column permutations of the initial Latin hypercube sample matrix are iteratively exchanged to determine the optimized sample matrix; and at least one combination of sample parameters is constructed based on the optimized sample matrix.

[0031] Optionally, in another optional embodiment of the present invention, the step of performing electromagnetic field simulation and thermal-fluid coupling simulation on the sample parameter combination based on the cable channel material property parameters and the environmental parameters to determine the maximum temperature of the cable channel includes: Electromagnetic field simulation is performed on the sample parameter combination based on the cable channel material properties and environmental parameters to determine conductor loss, armor eddy current loss, and channel structure eddy current loss. The conductor loss, armor eddy current loss, and channel structure eddy current loss are then converted to determine the electromagnetic loss temperature. Thermal-fluid coupling simulation is performed on the sample parameter combination based on the cable channel material properties, electromagnetic loss temperature, and environmental parameters to determine the steady-state temperature distribution cloud map of the cable channel. The highest temperature of the cable channel is determined based on the steady-state temperature distribution cloud map.

[0032] Among them, conductor loss can be the heat generated due to the resistance of the conductor material when the current flows inside the conductive core of the cable; armor eddy current loss can be the heat generated when the alternating magnetic field generated by the current induces eddy currents in the metal armor layer of the cable and the eddy currents flow through the resistance of the armor material; channel structure eddy current loss can be the heat generated when the alternating magnetic field generated by the current induces eddy currents in the metal channel structure around the cable.

[0033] Optionally, for each combination of sample parameters, under the conditions of channel length L and relative permeability of the channel material determined by the sample parameter combination, the present invention applies a 2500A three-phase AC current to the cable for electromagnetic field simulation, and obtains the conductor loss, armor eddy current loss, and channel structure eddy current loss through simulation calculation. For example, Figure 2 A waveform diagram of eddy current loss in the wall of a cable channel provided in an embodiment of the present invention is shown below. Figure 2 As shown, the sample parameter combination is at L=2m, μ r Under operating conditions of 285V and V=1m / s, the eddy current loss of the cable channel structure fluctuates between 200W and 1000W.

[0034] Optionally, the electromagnetic loss temperature can be the temperature in the corresponding area of ​​the cable channel where the conductor loss, armor eddy current loss, and channel structure eddy current loss are applied.

[0035] Among them, the steady-state temperature distribution cloud map of the cable channel can be a three-dimensional image describing the temperature distribution at various points in the cable channel.

[0036] Optionally, conductor loss, armor eddy current loss, and channel structure eddy current loss are applied to the corresponding regions of the cable channel. Under given cable channel material property parameters and environmental parameters, a corresponding axial flow velocity V is applied at the cable channel inlet, and a steady-state temperature field-fluid field coupled simulation is performed to obtain the simulated steady-state temperature distribution cloud map of the cable channel. For example, Figure 3 This is a schematic diagram of a steady-state temperature distribution cloud map of a cable channel provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a fluid velocity distribution cloud map within a cable channel, provided as an embodiment of the present invention. Figure 3 and Figure 4 As shown: Sample parameter combinations at L=2m, μ r Under operating conditions of 285°C and V=1 m / s, the temperature distribution of the cable channel can be referenced from the steady-state temperature distribution cloud map of the cable channel, and the fluid velocity distribution can be referenced from the fluid velocity distribution cloud map.

[0037] Optionally, after obtaining the steady-state temperature distribution cloud map of the cable channel, the highest temperature in the steady-state temperature distribution cloud map of the cable channel can be read to obtain the highest temperature of the cable channel.

[0038] Specifically, electromagnetic field simulation is performed on the sample parameter combination based on the cable channel material property parameters and environmental parameters to determine conductor loss, armor eddy current loss, and channel structure eddy current loss; the conductor loss, armor eddy current loss, and channel structure eddy current loss are converted to determine the electromagnetic loss temperature; thermal-fluid coupling simulation is performed on the sample parameter combination based on the cable channel material property parameters, electromagnetic loss temperature, and environmental parameters to determine the steady-state temperature distribution cloud map of the cable channel; and the highest temperature of the cable channel is determined based on the steady-state temperature distribution cloud map of the cable channel.

[0039] Optionally, after simulation based on each sample parameter combination, the highest cable channel temperature corresponding to each sample parameter combination is obtained. Each sample parameter combination is then associated with other sample parameter combinations to establish a sample database. For example, the highest cable channel temperature is determined by T... max The highest temperature T in the cable channel is represented as follows: max Combined with sample parameters (L,V,μ) r To perform the association, i.e., {(L,V,μ)} r →T max}, thus obtaining the sample database.

[0040] Optionally, the sample database can visually demonstrate the variation of the highest temperature of the cable channel with the cable channel length, the relative permeability of the channel material, and the axial velocity of the air in the channel in the sample parameter combination.

[0041] S120. Based on the sample database, determine the training data, and based on the training data and the input parameters, train the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model respectively.

[0042] Optionally, the training data can be used to train the maximum temperature prediction model, taking {(L,V,μ)} from the sample database. r, →T max The environmental parameters and other parameters are used as the training dataset, which is divided into training data and validation data in an 8:2 ratio. The training data is used to train the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model, respectively. The validation data can be used to validate the structure prediction sub-model, the material prediction model, and the maximum temperature prediction model.

[0043] Optionally, during training, the cable channel length L and the highest cable channel temperature T are... max The input parameters corresponding to the branch of the structural prediction sub-model are used as the input parameters of the cable channel length L as vector inputs to train the structural prediction sub-model. The input parameters for training the structural prediction sub-model include: conductor outer diameter, insulation outer diameter, inner sheath outer diameter, armor layer outer diameter, outer sheath outer diameter, channel inner diameter, channel outer diameter, channel length, eccentricity, cable spacing, and arrangement.

[0044] Optionally, the relative magnetic permeability V of the channel material and the axial velocity μ of the air within the channel can be used as parameters. r and the highest temperature T of the cable channel max The relative permeability V of the channel material and the axial velocity μ of the air within the channel are used as branch inputs for the material prediction sub-model. r The corresponding input parameters are used as vector inputs to train the material prediction sub-model. The input parameters for training the material prediction sub-model include: the electrical conductivity, relative permeability, density, specific heat capacity, and thermal conductivity of the cable material; and the electrical conductivity, relative permeability, density, specific heat capacity, and thermal conductivity of the channel material. Optionally, environmental parameters and the highest temperature T in the cable channel can be included. max As the corresponding branch inputs to the environmental prediction sub-model, the input parameters corresponding to the environmental medium type, environmental temperature, thermal resistivity, and axial flow velocity of the medium within the channel are used as vector inputs to train the environmental prediction sub-model. For example, the environmental parameters are represented by T0.

[0045] Optionally, the structural prediction sub-model can predict the maximum temperature of the cable channel based on cable channel length analysis. The structural prediction sub-model consists of a radial basis regression model.

[0046] Optionally, the material prediction sub-model can predict the maximum temperature of the cable channel based on cable channel length analysis. The structural prediction sub-model consists of a radial basis regression model.

[0047] Optionally, the environmental prediction sub-model can predict the maximum temperature of the cable channel based on environmental parameter analysis. The environmental prediction sub-model consists of a radial basis regression model.

[0048] Specifically, training data is determined based on the sample database, and structural prediction sub-models, material prediction sub-models, and environmental prediction sub-models are trained based on the training data.

[0049] Optionally, in another optional embodiment of the present invention, training the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model based on the training data and the input parameters respectively includes: The first radial basis regression model is trained based on the training data and input parameters to determine the first predictive sub-model. The pre-established second radial basis regression model is trained based on training data and input parameters to determine the second predictive sub-model. The pre-established third radial basis regression model is trained based on training data and input parameters to determine the third predictive sub-model; The radial basis width parameter of the first, second, and third prediction sub-models was optimized by particle swarm optimization to determine the structure prediction sub-model, environment prediction sub-model, and material prediction sub-model.

[0050] The first radial basis regression model can be a pre-established radial basis regression model used to train the first predictive sub-model. The first predictive sub-model can be a predictive model of structural branches trained from the pre-established first radial basis regression model. It should be noted that the first radial basis regression model can consist of an input layer, a radial base layer, and an output layer. The radial base layer contains 12 radial basis nodes. K-means clustering is used on the channel cable lengths in the training data, with 12 clusters. The 12 cluster centers are used as the 12 radial basis centers. The radial basis width is taken as the average distance between adjacent centers, and the minimum width threshold is set to 0.1 times the average value. After the centers and widths are determined, regularized least squares regression is used to solve for the model weights. The regularization coefficient is a fixed value of 1×10⁻⁶. -6 This allows the input layer to take in training data and the output layer to output the predicted temperature value.

[0051] The second radial basis regression model can be a pre-established radial basis regression model used to train the second predictive sub-model. The second predictive sub-model can be a predictive model for the material branch obtained by training the pre-established second radial basis regression model. It should be noted that the second radial basis regression model can consist of an input layer, a radial base layer, and an output layer. The radial base layer contains 12 radial basis nodes. K-means clustering is used on the relative permeability of the channel material and the axial velocity of the air within the channel in the training data, with a cluster size of 12. The 12 cluster centers are used as the 12 radial basis centers. The width of the radial basis center is taken as the average of the distances between adjacent centers, and the minimum width threshold is set to 0.1 times the average value. After the center and width are determined, regularized least squares regression is used to solve for the model weights. The regularization coefficient is a fixed value of 1×10⁻⁶. -6 This allows the input layer to take in training data and the output layer to output the predicted temperature value.

[0052] The third radial basis regression model can be a pre-established radial basis regression model used to train the third predictive sub-model. The third predictive sub-model can be a predictive model of the environmental branch trained from the pre-established third radial basis regression model. It should be noted that the third radial basis regression model can consist of an input layer, a radial base layer, and an output layer. The radial base layer contains 12 radial basis nodes. K-means clustering is used on the environmental parameters in the training data, with 12 clusters. The 12 cluster centers are used as the 12 radial basis centers. The width of the radial basis is taken as the average distance between adjacent centers, and the minimum width threshold is set to 0.1 times the average value. After the centers and width are determined, regularized least squares regression is used to solve for the model weights. The regularization coefficient is a fixed value of 1×10⁻⁶. This allows the input layer to input training data and the output layer to output the predicted temperature value.

[0053] Optionally, after training the first, second, and third predictive sub-models, particle swarm optimization is performed on them. The specific steps are as follows: For the first, second, and third predictive sub-models, the scaling factor of the radial basis width is set as the optimization variable for the particle swarm. The base width determined by the average distance between adjacent centers is multiplied by the scaling factor to obtain the actual radial basis width used for training. Particle swarm parameters are pre-set, i.e., the number of particles is 20, and the maximum number of iterations is 30. The search range of the scaling factor is limited to [0.5, 2]. For each particle with a given scaling factor, the prediction results of the first, second, and third predictive sub-models are verified using each validation data point. The prediction error between each prediction result and the validation data is identified, and the root mean square error (RMSE) of the prediction error is calculated. Error is used as the fitness index; the particle position is iteratively updated according to the particle swarm optimization rules, and the individual optimum and global optimum are updated. The training-verification-update cycle is repeated until the maximum number of iterations is reached or the fitness converges. Finally, the scaling factor corresponding to the global optimum particle is taken as the optimization result. The scaling factor corresponding to the global optimum particle is used to optimize the radial basis width parameters of the first prediction sub-model, the second prediction sub-model and the third prediction sub-model to determine the final radial basis width parameters of the first prediction sub-model, the second prediction sub-model and the third prediction sub-model, and obtain the structural prediction sub-model, the environmental prediction sub-model and the material prediction sub-model.

[0054] Specifically, a pre-established first radial basis regression model is trained based on training data and input parameters to determine the first predictive sub-model; a pre-established second radial basis regression model is trained based on training data and input parameters to determine the second predictive sub-model; a pre-established third radial basis regression model is trained based on training data and input parameters to determine the third predictive sub-model; and the radial basis width parameters of the first, second, and third predictive sub-models are optimized through particle swarm optimization to determine the structural predictive sub-model, environmental predictive sub-model, and material predictive sub-model.

[0055] S130. The highest temperature prediction model is determined based on the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model.

[0056] Among them, the maximum temperature prediction model can be trained based on training data and used to predict the maximum temperature of the cable channel under the current operating conditions.

[0057] Optionally, after obtaining the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model, the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model are weighted and fused to obtain the maximum temperature prediction model.

[0058] Specifically, the maximum temperature prediction model is determined based on the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model.

[0059] S140. Collect the cable channel parameters of the target cable channel, predict the maximum temperature of the cable channel parameters using the maximum temperature prediction model, and determine the maximum temperature prediction result of the target cable channel.

[0060] The target cable channel is the cable channel where the highest temperature of the cable to be predicted is located. It should be noted that the target cable channel can be a cable channel for which a maximum temperature prediction model has been established, allowing the model to be used to predict the maximum temperature of the target cable; alternatively, a maximum temperature prediction model can be pre-established for other cable channels, and this model can then be used to predict the maximum temperature of the target cable.

[0061] Among them, the cable channel parameters can be the cable channel length, the relative magnetic permeability of the channel material, and the axial velocity of the air in the channel under the current operating conditions.

[0062] Among them, the highest temperature prediction result can be the highest temperature predicted by the highest temperature prediction model based on the cable channel parameters of the target cable channel.

[0063] Specifically, the cable channel parameters of the target cable channel are collected, and the highest temperature prediction result of the cable channel parameters is determined by using the highest temperature prediction model.

[0064] The technical solution of this invention obtains the input parameters of the electromagnetic-thermal-fluid multi-physics coupling of the cable channel, constructs a sample database based on the input parameters, and significantly reduces computational costs and improves evaluation efficiency by establishing the sample database. Training data is determined based on the sample database, and structural prediction sub-models, environmental prediction sub-models, and material prediction sub-models are trained based on the training data and the input parameters, respectively. A maximum temperature prediction model is determined based on the structural, environmental, and material prediction sub-models, and the input parameters are modeled according to structural, material, and environmental branches, effectively improving prediction accuracy and stability, and enhancing prediction accuracy and generalization ability. Cable channel parameters of the target cable channel are collected, and the maximum temperature is predicted using the maximum temperature prediction model. The maximum temperature prediction model can quickly predict the maximum temperature of the target operating condition without repeating the full-process finite element simulation, solving the technical problem of inaccurate and rapid prediction of the maximum temperature of cable channels in the prior art, reducing the workload of repetitive simulation, and effectively improving the efficiency of multi-condition thermal evaluation.

[0065] Figure 5 This is a flowchart illustrating another method for predicting the maximum temperature of a cable channel according to an embodiment of the present invention. The relationship between this embodiment and the above embodiments is that this describes the detailed process of training a model for predicting the maximum temperature of a cable channel. Figure 5 As shown, the method includes: S510. Obtain the input parameters of electromagnetic-thermal-fluid multi-physics coupling of the cable channel, and construct a sample database based on the input parameters.

[0066] S520. Based on the sample database, determine the training data, and based on the training data and the input parameters, train the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model respectively.

[0067] S530. Based on the validation data of the sample database, perform model validation on the structure prediction sub-model, the environment prediction sub-model and the material prediction sub-model respectively, and determine at least one prediction error.

[0068] Optionally, the {(L,V,μ)} in the sample database can be used as an example. r →T max The data, including the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model, was divided into training and validation data in an 8:2 ratio. The validation data was used to validate the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model to obtain the prediction error. The prediction error can be the error dataset between the maximum temperature prediction results from the structural prediction sub-model, environmental prediction model, and material prediction sub-model and the validation data.

[0069] Optionally, each set of validation data is input into the structural prediction sub-model to obtain the predicted maximum temperature output by the structural prediction sub-model. The prediction result for each set of validation data is then compared with the validation data to obtain the prediction error of the structural prediction sub-model. Similarly, each set of validation data is input into the environmental prediction sub-model to obtain the predicted maximum temperature output by the environmental prediction sub-model. The prediction result for each set of validation data is then compared with the validation data to obtain the prediction error of the environmental prediction sub-model. Likewise, each set of validation data is input into the material prediction sub-model to obtain the predicted maximum temperature output by the material prediction sub-model. The prediction result for each set of validation data is then compared with the validation data to obtain the prediction error of the material prediction sub-model. For example, T... L T is used to represent the prediction results of the structural prediction sub-model. MV Used to represent the prediction results of the material prediction sub-model; T n Used to represent the prediction results of the environmental prediction sub-model.

[0070] Specifically, the structural prediction sub-model and the material prediction sub-model are validated based on the validation data from the sample database to determine at least one prediction error.

[0071] S540. Based on all the prediction errors, determine the weighting coefficients, and use the weighting coefficients to perform weighted fusion of the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model to determine the maximum temperature prediction model.

[0072] The weighting coefficients can be calculated from the total prediction errors of the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model.

[0073] Optionally, after obtaining the prediction results for the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model based on the obtained validation data, the prediction results of the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model are weighted and summed to obtain the final temperature prediction result. Then, the final temperature prediction result is compared with the validation data to determine the difference between the final temperature prediction result and the highest temperature in the validation data. A set of weighting coefficients is found such that the difference between the final temperature prediction result and the highest temperature in the validation data for each validation data point is minimized, i.e., the prediction error is minimized. These weighting coefficients are then used as the weighting coefficients for the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model. The selection of these weighting coefficients can employ at least one of the following methods: minimizing the maximum error method, minimizing the mean absolute error method, minimizing the mean squared error method, and cross-validation.

[0074] For example, the weighting coefficients are represented by α, β, and γ, where α∈[0,1], β∈[0,1], γ∈[0,1], and α+β+γ=1; through T max As a way to minimize the prediction error, the weighted method is expressed as follows: T max =α×T L +β×T MV +γ×T n The weight coefficients for the branches corresponding to the structural prediction sub-model are α, β, and γ, respectively.

[0075] Specifically, weighting coefficients are determined based on all prediction errors. These weighting coefficients are then used to weight and fuse the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model to determine the highest temperature prediction model.

[0076] Optionally, in another optional embodiment of the present invention, the method further includes: performing model validation on the maximum temperature prediction model based on the validation data; if the validation result of the maximum temperature prediction model is greater than a preset error threshold, generating at least one new training data based on all the prediction errors; updating the training data with the new training data; returning to the step of training the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model based on the training data and the input parameters respectively; and updating the maximum temperature prediction model until the validation result of the maximum temperature prediction model is not greater than the preset error threshold.

[0077] The preset error threshold can be a pre-set data value used to judge the performance of the highest temperature prediction model.

[0078] Optionally, after training the maximum temperature prediction model, the validation data is input into the maximum temperature prediction model for model validation, and the predicted temperature value output by the maximum temperature prediction model is obtained. The difference between the predicted temperature value and the maximum temperature of the validation data is calculated to obtain the temperature difference value, and the absolute value of the temperature difference value is used as the validation result.

[0079] Optionally, if the verification result is greater than the preset error threshold, it indicates that the performance of the maximum temperature prediction model is poor, and the maximum temperature prediction model is iteratively updated; if the verification result is not greater than the preset error threshold, it indicates that the performance of the maximum temperature prediction model meets the requirements, and the maximum temperature prediction model is not iteratively updated.

[0080] The new training data can be a combination of new sample parameters generated based on the prediction error.

[0081] Optionally, when iterating and updating the maximum temperature prediction model, the validation data can be input into the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model of the maximum temperature prediction model, respectively, to obtain the corresponding prediction output T. L T MV and T n T respectively L T MV and T n The absolute errors of the structural, environmental, and material prediction sub-models are calculated by comparing them with the highest temperature in the validation data. After obtaining the absolute errors of these sub-models, the magnitudes of their respective errors are compared to identify the sources of error. When the absolute error of the structural prediction sub-model dominates, the validation data μ is fixed. r In conjunction with V, at least one new cable channel length is generated within the neighborhood of the cable channel length L, to form a combination of multiple new cable channel lengths and fixed verification data μ. r Multiple sample parameter combinations, along with V, are used to perform electromagnetic field simulations and thermal-fluid coupling simulations on each sample parameter combination based on cable channel material properties and environmental parameters. The maximum temperature of the cable channel is determined, and the sample parameter combinations are correlated with the maximum cable channel temperature as new training data. When the absolute error of the material prediction sub-model dominates, the validation data L is fixed, and the parameters are determined by the relative permeability of the channel material and the axial air velocity within the channel (μ). r Generate at least one new value (μ) within the neighborhood of V). r To form multiple sample parameter combinations consisting of the relative magnetic permeability of multiple new channel materials, the axial velocity of air in the channel, and fixed verification data L, electromagnetic field simulation and thermal-fluid coupling simulation are performed on each sample parameter combination based on the cable channel material property parameters and environmental parameters to determine the maximum temperature of the newly added cable channel. The sample parameter combinations, environmental parameters, and the maximum temperature of the newly added cable channel are correlated as new training data.

[0082] Optionally, the newly added training data is incorporated into the sample database and updated in the training data. The structure prediction sub-model, environment prediction sub-model, and material prediction sub-model are retrained separately with the new training data. The weight parameters are then re-determined based on the updated structure prediction sub-model, environment prediction sub-model, and material prediction sub-model to update the maximum temperature prediction model.

[0083] Optionally, after obtaining the updated maximum temperature prediction model, the maximum temperature prediction model is validated again based on the validation data. If the validation result of the maximum temperature prediction model is still greater than the preset error threshold, the model is iterated again in the same way and updated until the validation result of the maximum temperature prediction model is no greater than the preset error threshold.

[0084] Specifically, the maximum temperature prediction model is validated based on the validation data. If the validation result of the maximum temperature prediction model is greater than the preset error threshold, at least one new training data is generated based on all the prediction errors. The new training data is then added to the training data, and the process of training the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model based on the training data and input parameters is resumed. The maximum temperature prediction model is then updated until the validation result of the maximum temperature prediction model is no greater than the preset error threshold.

[0085] S550. Collect the cable channel parameters of the target cable channel, predict the maximum temperature of the cable channel parameters using the maximum temperature prediction model, and determine the maximum temperature prediction result of the target cable channel.

[0086] This invention establishes a sample database based on the highest temperature obtained from finite element multiphysics simulation, and replaces a large number of repetitive simulations with a trained highest temperature prediction model to achieve rapid prediction of the highest temperature of cable channels under multiple operating conditions, reducing computational resource consumption and engineering cycle. By evaluating the error contributions of the structural sub-model and the material sub-model respectively, the source of error is determined, and local densification and supplementation of samples are performed in the corresponding parameter dimensions and the model is iteratively updated to make the newly added finite element samples more targeted, thereby improving model performance with fewer samples.

[0087] Optionally, this embodiment of the invention discloses another implementation step for the method of predicting the maximum temperature of a cable channel, specifically: S1: Determine the input parameters required for electromagnetic-thermal-fluid multiphysics coupling calculations of the cable channel.

[0088] S2: Organize the input parameters and set the value range or optional category, generate small sample input parameter groups according to the preset sampling rules, and form at least one sample parameter combination.

[0089] S3: Perform finite element multiphysics simulation for each sample parameter combination to obtain the highest temperature of the cable channel corresponding to each set of input parameters and establish a sample database.

[0090] S4: Based on the sample database and combined with the training parameters for missing environmental parameters, the input parameters are divided into structural input sub-vectors, material input sub-vectors, and environmental input sub-vectors. The structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model are trained respectively. Based on the output of the three sub-models, a fusion model is constructed to obtain the maximum temperature prediction model.

[0091] S5: Calculate the prediction errors of the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model on the validation data and determine the source of error. Generate new training data according to the source of error and obtain the highest temperature of the new cable channel through finite element simulation to obtain new samples. After merging the new samples into the sample database, retrain the structural prediction sub-model, material prediction sub-model, and fusion model to update the highest temperature prediction model.

[0092] S6: Input the target operating condition input parameter group into the updated maximum temperature prediction model, and output the maximum temperature prediction value under the target operating condition.

[0093] In step S1, the input parameters consist of cable parameters, channel parameters, and environmental parameters.

[0094] The structural dimensions of a cable include its structural type, conductor outer diameter, insulation outer diameter, inner sheath outer diameter, armor outer diameter, and outer sheath outer diameter; the cable structural type is either a single-core cable or a three-core cable.

[0095] The material properties of cables include electrical conductivity, relative permeability, density, specific heat capacity, and thermal conductivity.

[0096] The structural dimensions of the channel include the inner diameter, outer diameter, length, and cable arrangement parameters within the channel; the arrangement parameters include the eccentricity between the cable and the center of the channel, the cable spacing, and the arrangement method.

[0097] The material properties of the channel include electrical conductivity, relative magnetic permeability, density, specific heat capacity, and thermal conductivity.

[0098] Environmental parameters include the type of environmental medium, ambient temperature, thermal boundary parameters, and fluid boundary conditions; the type of environmental medium is air, soil, or seawater; the ambient temperature is the temperature of the medium inside the channel; the thermal boundary parameter is the thermal resistance coefficient; and the fluid boundary condition is the axial velocity of the medium inside the channel.

[0099] In step S2, the preset sampling rule is optimized Latin hypercube sampling; and a small sample input parameter group is generated according to the value range of each input parameter or the optional category.

[0100] In step S3, the finite element multiphysics simulation includes electromagnetic field simulation and temperature field-fluid field coupling simulation.

[0101] Electromagnetic field simulation is used to obtain conductor loss, armored eddy current loss, and channel eddy current loss, and the loss is applied as a volumetric heat source to the corresponding region.

[0102] The temperature field-fluid field coupled simulation solves the temperature distribution and extracts the highest temperature under the thermal boundary conditions corresponding to the heat source and thermal resistance coefficient.

[0103] In step S4, the three branches are organized as follows: the outer diameter of the conductor, the outer diameter of the insulation layer, the outer diameter of the inner sheath, the outer diameter of the armor layer, the outer diameter of the outer sheath, the inner diameter of the channel, the outer diameter of the channel, the channel length, the eccentricity, the cable spacing, and the arrangement are included in the structure input sub-vector; the conductivity, relative permeability, density, specific heat capacity, and thermal conductivity of the cable material, and the conductivity, relative permeability, density, specific heat capacity, and thermal conductivity of the channel material are included in the material input sub-vector; the environmental medium type, environmental temperature, thermal resistance coefficient, and axial flow velocity of the medium in the channel are included in the environmental input sub-vector.

[0104] The structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model are all radial basis regression models.

[0105] The fusion model is a weighted fusion model, which combines the outputs of the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model according to weights to obtain the predicted maximum temperature. The weights in the weighted fusion model are determined by minimizing the prediction error of the fusion output on validation data.

[0106] In step S4, the key parameters of the radial basis regression model are optimized using a particle swarm optimization algorithm. Each particle represents a set of radial basis regression model parameters. The particle swarm uses the prediction error on the validation data as the fitness and iteratively updates to obtain the optimal parameters of the radial basis regression model.

[0107] In step S5, the method for determining the source of error is as follows: calculate the absolute errors of the structural prediction sub-model and the material prediction sub-model on the verification data respectively, and determine the source of error based on the relationship between the magnitudes of their absolute errors.

[0108] In step S5, the method for generating new training data is as follows: taking the input parameter group corresponding to the error as the center, perturb the continuous parameters in the parameter dimension corresponding to the error source to generate candidate supplementary sampling points, and after removing the candidate supplementary sampling points that are duplicated with the existing samples, select a number of supplementary sampling points as new training data.

[0109] In step S5, retraining is performed by incorporating new samples into the sample database and then retraining the structural prediction sub-model, material prediction sub-model, and fusion model based on the updated sample data and environmental parameters to update the maximum temperature prediction model.

[0110] This invention establishes a sample database based on the highest temperature obtained from finite element multiphysics simulation, and replaces a large number of repetitive simulations with a trained highest temperature prediction model to achieve rapid prediction of the highest temperature of cable channels under multiple operating conditions, reducing computational resource consumption and engineering cycle. By evaluating the error contributions of the structural sub-model and the material sub-model respectively, the source of error is determined, and local densification and supplementation of samples are performed in the corresponding parameter dimensions and the model is iteratively updated to make the newly added finite element samples more targeted, thereby improving model performance with fewer samples.

[0111] Figure 6 This is a schematic diagram of a maximum temperature prediction device for a cable channel provided in an embodiment of the present invention. Figure 6 As shown, the device includes: a parameter processing module 610, a model training module 620, a model fusion module 630, and a prediction module 640; wherein, The parameter processing module 610 is used to acquire the input parameters of the electromagnetic-thermal-fluid multi-physics coupling of the cable channel and to construct a sample database based on the input parameters. The model training module 620 is used to determine training data based on the sample database, and to train the structure prediction sub-model, the environment prediction sub-model, and the material prediction model based on the training data and the input parameters, respectively. The model fusion module 630 is used to determine the maximum temperature prediction model based on the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model. The prediction module 640 is used to collect cable channel parameters of the target cable channel, predict the maximum temperature of the cable channel parameters using the maximum temperature prediction model, and determine the maximum temperature prediction result of the target cable channel.

[0112] Optionally, the model training module 620 is specifically used for: The first radial basis regression model is trained based on the training data and the input parameters to determine the first prediction sub-model. The pre-established second radial basis regression model is trained based on the training data and the input parameters to determine the second prediction sub-model. The pre-established third radial basis regression model is trained based on the training data and the input parameters to determine the third prediction sub-model; The radial basis width parameter of the first, second, and third prediction sub-models is optimized by particle swarm optimization to determine the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model.

[0113] Optionally, the model fusion module 630 is specifically used for: Based on the validation data from the sample database, the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model were validated to determine at least one prediction error. Based on all the prediction errors, weighting coefficients are determined, and the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model are weighted and fused using the weighting coefficients to determine the maximum temperature prediction model.

[0114] Optionally, the model fusion module 630 is also specifically used for: The maximum temperature prediction model is validated based on the validation data. If the validation result of the maximum temperature prediction model is greater than a preset error threshold, at least one new training data is generated based on all the prediction errors. The newly added training data is updated into the training data, and the process of training the structure prediction sub-model, environment prediction sub-model and material prediction sub-model based on the training data and the input parameters is returned; and the maximum temperature prediction model is updated until the verification result of the maximum temperature prediction model is not greater than the preset error threshold.

[0115] Optionally, the parameter processing module 610 is specifically used for: Construct at least one combination of sample parameters based on cable channel length, relative magnetic permeability of channel material, and axial air velocity within the channel; For each sample parameter combination, electromagnetic field simulation and thermal flux coupling simulation are performed on the sample parameter combination based on the material property parameters of the cable channel and the environmental parameters to determine the maximum temperature of the cable channel; The sample database is constructed by associating each sample parameter combination with the highest temperature of the cable channel corresponding to that sample parameter combination.

[0116] Optionally, the parameter processing module 610 is also specifically used for: An initial Latin hypercube sample matrix is ​​constructed based on the cable channel length, the relative permeability of the channel material, and the axial velocity of the airflow within the channel. The column permutations of the initial Latin hypercube sample matrix are iteratively exchanged to determine the optimized sample matrix; At least one combination of sample parameters is constructed based on the optimized sample matrix.

[0117] Optionally, the parameter processing module 610 is also specifically used for: Electromagnetic field simulation is performed on the sample parameter combination based on the material property parameters of the cable channel and the environmental parameters to determine conductor loss, armor eddy current loss and channel structure eddy current loss. The electromagnetic loss temperature is determined by converting the conductor loss, armor eddy current loss, and channel structure eddy current loss. Based on the material properties of the cable channel, the electromagnetic loss temperature, and the environmental parameters, a thermal-fluid coupling simulation is performed on the sample parameter combination to determine the steady-state temperature distribution cloud map of the cable channel. The highest temperature of the cable channel is determined based on the steady-state temperature distribution cloud map of the cable channel.

[0118] The maximum temperature prediction device for cable channels provided in this embodiment of the invention can execute the maximum temperature prediction method for cable channels provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0119] Figure 7 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their patterns are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0120] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0121] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer grids such as the Internet and / or various telecommunications grids.

[0122] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for predicting the maximum temperature of a cable channel.

[0123] In some embodiments, the maximum temperature prediction method for the cable channel can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the maximum temperature prediction method for the cable channel described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the maximum temperature prediction method for the cable channel by any other suitable means (e.g., by means of firmware).

[0124] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0125] Computer programs used to implement the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the patterns / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0126] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0127] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0128] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or grid browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication grid). Examples of communication grids include local area networks (LANs), wide area networks (WANs), blockchain grids, and the Internet.

[0129] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS servers, such as high management difficulty and weak business scalability.

[0130] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0131] This embodiment provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the cable channel maximum temperature prediction method provided in any embodiment of the present invention, the method comprising: The input parameters of electromagnetic, thermal, and current multiphysics coupling in the cable channel are obtained, and a sample database is constructed based on the input parameters. Training data is determined based on the sample database, and structural prediction sub-model, environmental prediction sub-model, and material prediction model are trained based on the training data and the input parameters, respectively. The highest temperature prediction model is determined based on the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model. The cable channel parameters of the target cable channel are collected, and the maximum temperature of the cable channel parameters is predicted using the maximum temperature prediction model to determine the maximum temperature prediction result of the target cable channel.

[0132] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0133] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0134] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0135] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of mesh, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0136] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a grid of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computing device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0137] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0138] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for predicting the maximum temperature of a cable channel, characterized in that, include: The input parameters of electromagnetic, thermal, and current multiphysics coupling in the cable channel are obtained, and a sample database is constructed based on the input parameters. Training data is determined based on the sample database, and structural prediction sub-model, environmental prediction sub-model, and material prediction model are trained based on the training data and the input parameters, respectively. The highest temperature prediction model is determined based on the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model. The cable channel parameters of the target cable channel are collected, and the maximum temperature of the cable channel parameters is predicted using the maximum temperature prediction model to determine the maximum temperature prediction result of the target cable channel.

2. The method according to claim 1, characterized in that, The step of training the structure prediction sub-model, environment prediction sub-model, and material prediction model based on the training data and the input parameters includes: The first radial basis regression model is trained based on the training data and the input parameters to determine the first prediction sub-model. The pre-established second radial basis regression model is trained based on the training data and the input parameters to determine the second prediction sub-model. The pre-established third radial basis regression model is trained based on the training data and the input parameters to determine the third prediction sub-model; The radial basis width parameter of the first, second, and third prediction sub-models is optimized by particle swarm optimization to determine the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model.

3. The method according to claim 1, characterized in that, The determination of the maximum temperature prediction model based on the structural prediction sub-model, environmental prediction sub-model, and material prediction sub-model includes: Based on the validation data from the sample database, the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model were validated to determine at least one prediction error. Based on all the prediction errors, weighting coefficients are determined, and the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model are weighted and fused using the weighting coefficients to determine the maximum temperature prediction model.

4. The method according to claim 3, characterized in that, Also includes: The maximum temperature prediction model is validated based on the validation data. If the validation result of the maximum temperature prediction model is greater than a preset error threshold, at least one new training data is generated based on all the prediction errors. The newly added training data is updated into the training data, and the process of training the structure prediction sub-model, the environment prediction sub-model, and the material prediction sub-model based on the training data and the input parameters is returned; and the maximum temperature prediction model is updated until the verification result of the maximum temperature prediction model is not greater than the preset error threshold.

5. The method according to claim 1, characterized in that, The process of obtaining input parameters for the electromagnetic-thermal-fluid multiphysics coupling of the cable channel and constructing a sample database based on these input parameters includes: Construct at least one combination of sample parameters based on cable channel length, relative magnetic permeability of channel material, and axial air velocity within the channel; For each sample parameter combination, electromagnetic field simulation and thermal flux coupling simulation are performed on the sample parameter combination based on the cable channel material property parameters and environmental parameters to determine the maximum temperature of the cable channel; The sample database is constructed by associating each sample parameter combination with the highest temperature of the cable channel corresponding to that sample parameter combination.

6. The method according to claim 5, characterized in that, The construction of at least one sample parameter combination based on the cable channel length, the relative permeability of the channel material, and the axial air velocity within the channel includes: An initial Latin hypercube sample matrix is ​​constructed based on the cable channel length, the relative permeability of the channel material, and the axial velocity of the airflow within the channel. The column permutations of the initial Latin hypercube sample matrix are iteratively exchanged to determine the optimized sample matrix; At least one combination of sample parameters is constructed based on the optimized sample matrix.

7. The method according to claim 5, characterized in that, The process of performing electromagnetic field simulation and thermal-fluid coupling simulation on the sample parameter combination based on the material property parameters of the cable channel and the environmental parameters to determine the maximum temperature of the cable channel includes: Electromagnetic field simulation is performed on the sample parameter combination based on the material property parameters of the cable channel and the environmental parameters to determine conductor loss, armor eddy current loss and channel structure eddy current loss. The electromagnetic loss temperature is determined by converting the conductor loss, armor eddy current loss, and channel structure eddy current loss. Based on the material properties of the cable channel, the electromagnetic loss temperature, and the environmental parameters, a thermal-fluid coupling simulation is performed on the sample parameter combination to determine the steady-state temperature distribution cloud map of the cable channel. The highest temperature of the cable channel is determined based on the steady-state temperature distribution cloud map of the cable channel.

8. A device for predicting the maximum temperature of a cable channel, characterized in that, include: The parameter processing module is used to acquire the input parameters of electromagnetic, thermal and current multi-physics coupling in the cable channel, and to construct a sample database based on the input parameters. The model training module is used to determine training data based on the sample database, and to train the structure prediction sub-model, environment prediction sub-model, and material prediction model based on the training data and the input parameters, respectively. The model fusion module is used to determine the maximum temperature prediction model based on the structural prediction sub-model, the environmental prediction sub-model, and the material prediction sub-model. The prediction module is used to collect cable channel parameters of the target cable channel, predict the maximum temperature of the cable channel parameters using the maximum temperature prediction model, and determine the maximum temperature prediction result of the target cable channel.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the maximum temperature prediction method for the cable channel according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for predicting the maximum temperature of a cable channel as described in any one of claims 1-7.

Images