Temperature prediction method and device for digital cable based on digital twinning technology

Abstract

The invention discloses a digital cable temperature prediction method and device based on a digital twinning technology, and relates to the technical field of intelligent power grids. The method comprises the following steps: acquiring position information and current-carrying capacity information of a target point of the digital cable; the position information and the current-carrying capacity information are input into a temperature prediction model, temperature information is determined, and the temperature prediction model is obtained through the following steps that an electric-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established; solving the simulation model to obtain actual temperature information of the sample points under different position information and different current-carrying capacity information; and training a physical information neural network model by taking the position information and the current-carrying capacity information of the sample points as input and taking the actual temperature information as output. According to the temperature prediction method provided by the invention, the mutual influence among various physical fields such as electricity, magnetism and heat can be comprehensively considered through the temperature prediction model, so that the description of the operation state of the digital cable under the actual complex working condition is more accurate and comprehensive, and the accuracy of temperature prediction is improved.

Description

Technical Field

[0001] The present application relates to the technical field of smart grids, and particularly to a temperature prediction method and device for digital cables based on digital twin technology. Background Art

[0002] With the rapid development of smart grids and cyber-physical systems (CPS), as a new type of power cable, digital cables not only need to meet the basic requirements of traditional power transmission, but also need to have additional functions such as communication and monitoring. Therefore, accurately grasping the operating state of digital cables, especially the temperature distribution, is of great significance for ensuring the safe operation of digital cables, extending their service life, and optimizing maintenance plans.

[0003] In related technologies, usually methods such as the thermal network method and the finite volume method are used to study the heat generated during the operation of digital cables and their heat transfer processes such as conduction, convection, and radiation, which helps to understand the temperature rise characteristics of digital cables under different load conditions.

[0004] However, the above technologies are difficult to comprehensively reflect the complex interactions between multiple physical fields such as electricity, magnetism, and heat in the operating environment of digital cables, and thus cannot comprehensively describe the operating state of digital cables under actual complex working conditions, resulting in inaccurate temperature prediction. Summary of the Invention

[0005] Based on the above problems, the present application provides a temperature prediction method and device for digital cables based on digital twin technology, which can improve the accuracy of temperature prediction.

[0006] The embodiments of the present application disclose the following technical solutions:

[0007] In a first aspect, the present application discloses a temperature prediction method for digital cables based on digital twin technology, the method comprising:

[0008] Obtain the position information and current-carrying capacity information of a target point of the digital cable, wherein the position information indicates the distance from the target point to the axis of the digital cable;

[0009] Determine the temperature information of the target point by inputting the position information and current-carrying capacity information of the target point into a temperature prediction model, wherein the temperature prediction model is obtained as follows:

[0010] Establish an electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable according to Maxwell's equations and Fourier's law of heat conduction;

[0011] Solve the simulation model to obtain the actual temperature information of sample points under different position information and different current-carrying capacity information;

[0012] Use the position information and current-carrying capacity information of the sample points as the input of the physical information neural network model, and use the actual temperature information of the sample points as the output of the physical information neural network model to train the physical information neural network model to obtain a temperature prediction model.

[0013] Optionally, when the current-carrying capacity information is the same, the temperature information of the target point is negatively correlated with the position information, and when the position information is the same, the temperature information of the target point is positively correlated with the current-carrying capacity information.

[0014] Optionally, when the current-carrying capacity information is the same, the relationship between the temperature information of the target point and the position information is as follows:

[0015]

[0016] where T j is the temperature information at the axis center of the digital cable, x is the position information of the target point, and d i represents the distance from the outer surface of each layer structure to the axis center.

[0017] Optionally, when the position information is the same, the relationship between the temperature information of the target point and the current-carrying capacity information is as follows:

[0018]

[0019] where T is the temperature information, K is the equivalent thermal conductivity, and I 0 is the current-carrying capacity information.

[0020] Optionally, the step of using the position information and current-carrying capacity information of the sample points as the input of the physical information neural network model, and using the actual temperature information of the sample points as the output of the physical information neural network model to train the physical information neural network model includes:

[0021] Input the position information and current-carrying capacity information of the sample points into the physical information neural network model to obtain the predicted temperature information of the sample points;

[0022] According to the actual temperature information and predicted temperature information of the sample points, train the physical information neural network model by minimizing the total loss function of the physical information neural network model, where the total loss function is the sum of the physical loss function, the initial loss function, and the convenience loss function.

[0023] Optionally, the step of establishing the electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable according to Maxwell's equations and Fourier's law of heat conduction includes:

[0024] Based on Maxwell's equations, the electromagnetic control equation of the digital cable is constructed;

[0025] Based on Fourier's law of heat conduction and the law of conservation of energy, the heat transfer control equation of the digital cable is constructed;

[0026] Using the conductivity-temperature expression, the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation are coupled to establish an electro-magnetic-thermal multi-physics field coupling simulation model of the digital cable.

[0027] In a second aspect, the present application discloses a temperature prediction device for a digital cable based on digital twin technology, and the device includes: an acquisition module and a determination module;

[0028] The acquisition module is used to acquire the position information and current-carrying capacity information of the target point of the digital cable, where the position information indicates the distance from the target point to the axis of the digital cable;

[0029] The determination module is used to determine the temperature information of the target point by inputting the position information and current-carrying capacity information of the target point into a temperature prediction model, where the temperature prediction model is obtained as follows:

[0030] According to Maxwell's equations and Fourier's law of heat conduction, an electro-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established;

[0031] The simulation model is solved to obtain the actual temperature information of the sample points under different position information and different current-carrying capacity information;

[0032] The position information and current-carrying capacity information of the sample points are used as the input of the physics-informed neural network model, and the actual temperature information of the sample points is used as the output of the physics-informed neural network model to train the physics-informed neural network model to obtain a temperature prediction model.

[0033] Optionally, when the current-carrying capacity information is the same, the temperature information of the target point is negatively correlated with the position information, and when the position information is the same, the temperature information of the target point is positively correlated with the current-carrying capacity information.

[0034] Optionally, when the current-carrying capacity information is the same, the relationship between the temperature information of the target point and the position information is as follows:

[0035]

[0036] where T j is the temperature information at the axis of the digital cable, x is the position information of the target point, and d i represents the distance from the outer surface of each layer structure to the axis.

[0037] Optionally, when the position information is consistent, the relationship between the temperature information and the current-carrying capacity information of the target point is as follows:

[0038]

[0039] where T is the temperature information, K is the equivalent thermal conductivity, and I 0 is the current-carrying capacity information.

[0040] Optionally, the determining module includes: a first determining sub-module and a second determining sub-module;

[0041] The first determining sub-module is configured to input the position information and the current-carrying capacity information of the sample point into the physical information neural network model to obtain the predicted temperature information of the sample point;

[0042] The second determining sub-module is configured to train the physical information neural network model by minimizing the total loss function of the physical information neural network model according to the actual temperature information and the predicted temperature information of the sample point, where the total loss function is the sum of a physical loss function, an initial loss function, and a convenience loss function.

[0043] Optionally, the determining module is specifically configured to: based on Maxwell's equations, construct the electromagnetic control equation of the digital cable; based on Fourier's law of heat conduction and the law of conservation of energy, construct the heat transfer control equation of the digital cable; use the conductivity-temperature expression to couple the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation to establish the electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable.

[0044] Compared with the prior art, the present application has the following beneficial effects:

[0045] The embodiments of the present application provide a temperature prediction method and device for a digital cable based on digital twin technology. The method includes: obtaining the position information and current-carrying capacity information of a target point of the digital cable, where the position information indicates the distance from the target point to the axis of the digital cable; by inputting the position information and current-carrying capacity information of the target point into a temperature prediction model, determining the temperature information of the target point, where the temperature prediction model is obtained as follows: according to Maxwell's equations and Fourier's law of heat conduction, establishing an electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable; solving the simulation model to obtain the actual temperature information of sample points under different position information and different current-carrying capacity information; using the position information and current-carrying capacity information of the sample points as the input of a physics-informed neural network model, and using the actual temperature information of the sample points as the output of the physics-informed neural network model, training the physics-informed neural network model to obtain the temperature prediction model. Thus, the temperature prediction method provided by the present application can comprehensively consider the mutual influence between multiple physical fields such as electricity, magnetism, and heat through the temperature prediction model, making the description of the operating state of the digital cable under actual complex working conditions more accurate and comprehensive, and improving the accuracy of temperature prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0047] Figure 1 It is a flowchart of a temperature prediction method for a digital cable based on digital twin technology provided by the embodiments of the present application;

[0048] Figure 2 It is a schematic diagram of a temperature prediction model provided by the embodiments of the present application;

[0049] Figure 3 It is a schematic diagram of a temperature prediction device for a digital cable based on digital twin technology provided by the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] As described above, in the related art, usually methods such as the thermal network method and the finite volume method are used to study the heat generated during the operation of digital cables and its heat transfer processes such as conduction, convection, and radiation, which helps to understand the temperature rise characteristics of digital cables under different load conditions. However, the above technologies are difficult to comprehensively reflect the complex interactions between multiple physical fields such as electricity, magnetism, and heat in the operating environment of digital cables, and thus cannot comprehensively describe the operating state of digital cables under actual complex working conditions, resulting in inaccurate temperature prediction.

[0051] To overcome the limitations of a single physical field model, a multi-physical field coupling model has emerged. The multi-physical field coupling model realizes the joint solution of multiple physical fields such as electricity, magnetism, and heat by combining different physical field equations. However, although the multi-physical field coupling model can jointly solve problems of different physical fields, its computational complexity is high and it requires a large amount of computing resources. In addition, the multi-physical field coupling model often depends on accurate initial conditions and boundary conditions, which are difficult to obtain in practical applications, and this also limits the application scope of the multi-physical field coupling model.

[0052] After research, the inventors proposed a temperature prediction method and device for a digital cable based on digital twin technology. The temperature prediction method provided in this application can comprehensively consider the mutual influence between multiple physical fields such as electricity, magnetism, and heat through a temperature prediction model, making the description of the operating state of the digital cable under actual complex working conditions more accurate and comprehensive, and improving the accuracy of temperature prediction. Further, the temperature prediction model constructed based on the physics-informed neural network is adopted in this application, which can significantly reduce the computational complexity and resource requirements while maintaining high accuracy, making this method more suitable for real-time monitoring and large-scale applications. Moreover, the method of this application reduces the dependence on precise conditions such as initial conditions and boundary conditions by using the physics-informed neural network, enhancing the applicability and flexibility of the model in actual engineering. This means that users can apply this technology in a wider range of scenarios without worrying about the problem of difficult determination of initial conditions and boundary conditions.

[0053] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0054] See Figure 1 , which is a flowchart of a temperature prediction method for a digital cable based on digital twin technology provided by an embodiment of this application. The method includes:

[0055] S101: Obtain the position information and current-carrying capacity information of the target point of the digital cable, where the position information indicates the distance from the target point to the axis of the digital cable.

[0056] First, obtain the position information and current-carrying capacity information of the target point on the digital cable through sensors or measuring devices. Among them, the position information refers to the distance from the target point to the axis of the digital cable (for example, 5 millimeters from the axis). The current-carrying capacity information refers to the current load at the target point (for example, 500 amperes).

[0057] S102: Determine the temperature information of the target point by inputting the position information and current-carrying capacity information of the target point into the temperature prediction model.

[0058] Input the obtained position information and current-carrying capacity information of the target point into the pre-trained temperature prediction model. This temperature prediction model is related to parameters such as the material properties of the digital cable, the ambient temperature, and the heat dissipation conditions, and can accurately predict the temperature information of the target point based on the position information and current-carrying capacity information of the target point. Exemplarily, by inputting the position information (5 mm) and current-carrying capacity information (500 A) into the temperature prediction model, the predicted temperature of the target point output by the temperature prediction model can be 65 °C.

[0059] It can be understood that based on the obtained temperature information of the target point, it is possible to determine whether the operating state of the digital cable is safe. Exemplarily, if the temperature information exceeds the preset temperature threshold (for example, 70 °C), the warning mechanism is triggered to remind the operation and maintenance personnel to take cooling measures or adjust the load.

[0060] In some specific implementation manners, the obtaining manner of the temperature prediction model can be as shown in steps A1 - A5 below:

[0061] A1: Based on Maxwell's equations, construct the electromagnetic control equation of the digital cable.

[0062] First step, by deriving Maxwell's equations, obtain the vector magnetic potential equations of each region in the digital cable. Specifically, the vector magnetic potential equations of each region in the digital cable can be as shown in formula (1) below:

[0063]

[0064] Among them, is the Laplacian operator, A is the magnetic vector potential, J s is the current density, and μ is the magnetic permeability.

[0065] Second step, by solving the vector magnetic potential equations of each region in the digital cable, obtain the magnetic vector potential A under a specific current distribution.

[0066] Third step, based on the magnetic vector potential A under a specific current distribution, determine the electric field strength E caused by the source current. Specifically, determining the electric field strength E caused by the source current can be as shown in formula (2) below:

[0067] E = -jωA (2)

[0068] Among them, E is the electric field strength, j is the imaginary unit, ω is the angular frequency, and A is the magnetic vector potential.

[0069] Step 4: Determine the current density J in each layer of the digital cable according to the electric field strength E i . Specifically, determine the current density J in each layer of the digital cable i . The formula can be shown as the following formula (3):

[0070] J i = σ i E + J s (3)

[0071] where J i is the current density of the i-th layer, σ i is the conductivity of the i-th layer, E is the electric field strength, and J s is the source current density.

[0072] Step 5: According to the current density J in each layer of the digital cable i , determine the power loss P in each layer of the digital cable i . In some specific implementation manners, the formula for determining the power loss P in each layer of the digital cable i can be shown as the following formula (4):

[0073] P i = σ i - 1 ∫J i 2 dv(4)

[0074] where the power loss P i is the power loss of the i-th layer, σ i is the conductivity of the i-th layer, and J i is the current density of the i-th layer.

[0075] Step 6: Based on the above results, construct the electromagnetic control equation of the digital cable

[0076] A2: Based on Fourier's law of heat conduction and the law of conservation of energy, construct the heat transfer control equation of the digital cable

[0077] Step 1: Use the infinitesimal element method in combination with the law of conservation of energy and Fourier's heat conduction equation to construct the internal heat transfer model of the digital cable. Specifically, according to the law of conservation of energy, the sum of the heat Φ in flowing into the infinitesimal element and the heat Q generated inside the infinitesimal element is equal to the heat Φ out flowing out of the infinitesimal element and the internal energy increment U per unit time, that is, dΦ in + dQ = dΦ out + dU. Then, by combining the law of conservation of energy and Fourier's heat conduction equation, the internal heat transfer model of the digital cable can be constructed. Specifically, the internal heat transfer model of the digital cable can be shown as the following formula (5):

[0078]

[0079] Among them, ρ is the material density, C p is the constant-pressure heat capacity of the fluid, u is the fluid velocity field, is the temperature gradient, t is the time, q is the heat flux density, Q is the heat generation power density of the cable, and k is the thermal conductivity of the material.

[0080] In the second step, describe the surface heat transfer model of the digital cable according to the convective heat flux formula and the radiative heat transfer formula.

[0081] Specifically, the convective heat flux formula can be shown as formula (6) below, and the radiative heat transfer formula can be shown as formula (7) below:

[0082] q = h(T ext - T)(6)

[0083]

[0084] Among them, q is the heat flux density, h is the heat transfer coefficient, T ext is the external temperature, T is the temperature, n is the surface normal vector, σ is the Stefan-Boltzmann constant, and ε is the emissivity.

[0085] In the third step, determine the heat transfer control equation according to the internal heat transfer model and the surface heat transfer model of the digital cable.

[0086] A3: Use the conductivity-temperature expression to couple the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation, and establish an electro-magnetic-thermal multi-physics field coupling simulation model of the digital cable.

[0087] For a digital cable, the heat generation power density Q of the cable is related to the current density J and is positively correlated. The specific formula is shown as formula (8) below:

[0088]

[0089] The change in the heat generation amount will in turn affect the magnitude of the material conductivity σ, as shown in formula (9) below:

[0090]

[0091] Among them, σ is the material conductivity, σ 20 is the material conductivity at 20°C, α is the temperature coefficient of conductivity, T is the temperature, and T ref is the reference temperature.

[0092] Thus, the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation can be coupled by using the conductivity-temperature expression to establish an electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable.

[0093] A4: Solve the simulation model to obtain the actual temperature information of the sample points under different position information and different current-carrying capacity information.

[0094] Before solving the simulation model, it is necessary to determine the structure and material parameters of the digital cable (such as conductor radius, insulation layer thickness, etc.), and based on this, use COMSOL Multiphysics software to build a simulation model of the digital cable. Moreover, in the simulation model of the digital cable, it is also necessary to set the coil model of the cable in COMSOL Multiphysics software according to the electromagnetic control equation, define the current density distribution and boundary conditions, and according to the heat transfer control equation, set the heat transfer mode of the digital cable (such as natural convection, forced convection, etc.), the working environment temperature and the boundary conditions. Subsequently, solve the simulation model again.

[0095] When the current-carrying capacity information is the same, the temperature information of the target point is negatively correlated with the position information. When the position information is the same, the temperature information of the target point is positively correlated with the current-carrying capacity information.

[0096] Specifically, when the magnitude of the current-carrying capacity remains unchanged, the temperature inside each layer of the cable is linearly related to the distance from the axis. Generally speaking, the distance-from-axis - temperature relationship formula is a piecewise broken line function, as shown in the following formula (10), and the number of segments depends on the number of layers of the cable physical structure.

[0097]

[0098] Among them, T j is the temperature information at the axis of the digital cable, x is the position information of the target point, and d i represents the distance from the outer surface of each layer structure to the axis.

[0099] When the position remains unchanged (such as only monitoring the temperature on the outer surface of the outer sheath), its temperature is proportional to the square of the current-carrying capacity, as shown in the following formula (11):

[0100]

[0101] Among them, T is the temperature information, K is the equivalent thermal conductivity, and I 0 is the current-carrying capacity information.

[0102] A5: Use the position information and current-carrying capacity information of the sample points as the input of the physics-informed neural network model, and use the actual temperature information of the sample points as the output of the physics-informed neural network model to train the physics-informed neural network model to obtain a temperature prediction model.

[0103] In the temperature prediction method for digital cables based on digital twin technology provided in the embodiments of the present application, the physics-informed neural network model is a six-layer fully connected neural network, and each layer has a number of neurons. The first five layers of the fully connected neural network use the hyperbolic tangent function (tanh) as the activation function, so that the physics-informed neural network receives the position information and current-carrying capacity information (x, I 0 ) as the input and outputs the corresponding temperature information T(x, I 0 ), thus achieving the purpose of deducing the temperature given the current-carrying capacity and position.

[0104] During the model training process, first, input the position information and current-carrying capacity information of the sample points into the physics-informed neural network model to obtain the predicted temperature information of the sample points. Second, according to the actual temperature information and predicted temperature information of the sample points, use the backpropagation algorithm and gradient descent method (or its variants, such as the Adam optimizer) to minimize the total loss function of the physics-informed neural network model and train the physics-informed neural network model (until the model converges). Among them, the total loss function is the sum of the physical loss function, the initial loss function, and the boundary loss function. The total loss function can be shown as the following formula (12):

[0105] L = L PDE + L IC + L BC (12)

[0106] Where L is the total loss function, L PDE is the physical loss function, L IC is the initial loss function, and L BC is the boundary loss function. Specifically, the physical loss function ensures that the predictions of the model conform to known physical laws. For example, for the cable temperature prediction problem, the physical loss function can be constructed based on Maxwell's equations and Fourier's law of heat conduction. The initial loss function can ensure that the predictions of the model are accurate under the initial conditions. The boundary loss function ensures that the predictions of the model are accurate under the boundary conditions.

[0107] It should be noted that the physics-informed neural network model can include two sub-models. Specifically, for the first sub-model, first, the temperature data at the center of the digital cable under different current-carrying capacities are used to train the first sub-model, so that the first sub-model can predict the center temperature according to the given current-carrying capacity information. Then, the obtained center temperature is added to the relationship between the center distance and temperature to construct a temperature distribution model of the cable. Subsequently, with the relationship between the center distance and temperature as the physical loss function, the boundary loss function is set according to the geometric structure of the digital cable, and the hyperbolic tangent function is used as the activation function to construct the second sub-model. For the second sub-model, first, the temperature data at each position of the digital cable are used to train the second sub-model, so that under the selected current-carrying capacity, the temperature T(x) can be calculated given the position information x. Finally, the above two neural networks are combined to construct a unified model, so that given the position information x and the current-carrying capacity information I of the observation point, the temperature T(x, I) at this point can be calculated.

[0108] It should also be noted that if the physics-informed neural network model still does not converge after N iterations, it is necessary to re-execute step A4.

[0109] It should also be noted that if the position information, current-carrying capacity information, and actual temperature information of M sample points are obtained, the position information, current-carrying capacity information, and actual temperature information of M1 sample points can be selected to train the machine learning model. After obtaining the temperature prediction model, the position information, current-carrying capacity information, and actual temperature information of M - M1 sample points (i.e., the test set) are used to verify the trained temperature prediction model to ensure that the temperature prediction model can accurately predict the temperature on unseen data. If the performance of the temperature prediction model on the test set is good, it means that the temperature prediction model is successfully trained; otherwise, it may be necessary to adjust the model structure of the temperature prediction model or retrain the temperature prediction model.

[0110] In summary, the present application discloses a temperature prediction method for digital cables based on digital twin technology. The temperature prediction method provided by the present application can comprehensively consider the mutual influence between multiple physical fields such as electricity, magnetism, and heat through the temperature prediction model, making the description of the operating state of digital cables under actual complex working conditions more accurate and comprehensive, and improving the accuracy of temperature prediction. Further, the temperature prediction model constructed based on the physics-informed neural network adopted by the present application can significantly reduce the computational complexity and resource requirements while maintaining high accuracy, which makes this method more suitable for real-time monitoring and large-scale applications. Moreover, the method of the present application reduces the dependence on precise conditions such as initial conditions and boundary conditions by using the physics-informed neural network, enhancing the applicability and flexibility of the model in actual engineering. This means that users can apply this technology in a wider range of scenarios without worrying about the difficulty of determining the initial conditions and boundary conditions.

[0111] SeeFigure 3 , this figure is a schematic diagram of a temperature prediction device for a digital cable based on digital twin technology provided by an embodiment of the present application. The temperature prediction device 300 for a digital cable based on digital twin technology includes: an acquisition module 301 and a determination module 302.

[0112] The acquisition module 301 is configured to acquire the position information and current-carrying capacity information of the target point of the digital cable, wherein the position information indicates the distance from the target point to the axis of the digital cable;

[0113] The determination module 302 is configured to determine the temperature information of the target point by inputting the position information and current-carrying capacity information of the target point into a temperature prediction model, wherein the temperature prediction model is obtained as follows:

[0114] According to Maxwell's equations and Fourier's law of heat conduction, establish an electro-magnetic-thermal multi-physical field coupling simulation model of the digital cable;

[0115] Solve the simulation model to obtain the actual temperature information of the sample points under different position information and different current-carrying capacity information;

[0116] Use the position information and current-carrying capacity information of the sample points as the input of the physics-informed neural network model, and use the actual temperature information of the sample points as the output of the physics-informed neural network model to train the physics-informed neural network model to obtain the temperature prediction model.

[0117] Optionally, when the current-carrying capacity information is the same, the temperature information of the target point is negatively correlated with the position information, and when the position information is the same, the temperature information of the target point is positively correlated with the current-carrying capacity information.

[0118] Optionally, when the current-carrying capacity information is the same, the relationship between the temperature information of the target point and the position information is shown in the following formula (13):

[0119]

[0120] where, T j is the temperature information at the axis of the digital cable, x is the position information of the target point, and d i represents the distance from the outer surface of each layer structure to the axis.

[0121] Optionally, when the position information is the same, the relationship between the temperature information of the target point and the current-carrying capacity information is shown in the following formula (14):

[0122]

[0123] where, T is the temperature information, K is the equivalent thermal conductivity, and I 0 is the current-carrying capacity information.

[0124] Optionally, the determination module 302 includes: a first determination sub-module and a second determination sub-module;

[0125] The first determination sub-module is configured to input the position information and current-carrying capacity information of the sample points into the physical information neural network model to obtain the predicted temperature information of the sample points;

[0126] The second determination sub-module is configured to train the physical information neural network model by minimizing the total loss function of the physical information neural network model according to the actual temperature information and predicted temperature information of the sample points, where the total loss function is the sum of the physical loss function, the initial loss function, and the convenience loss function.

[0127] Optionally, the determination module 302 is specifically configured to: based on Maxwell's equations, construct an electromagnetic control equation for the digital cable; based on Fourier's law of heat conduction and the law of conservation of energy, construct a heat transfer control equation for the digital cable; use the conductivity-temperature expression to couple the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation to establish an electro-magnetic-thermal multi-physical field coupling simulation model for the digital cable.

[0128] In summary, the present application discloses a temperature prediction device for a digital cable based on digital twin technology. The temperature prediction device provided by the present application can comprehensively consider the mutual influence between multiple physical fields such as electricity, magnetism, and heat through the temperature prediction model, so that the description of the operating state of the digital cable under actual complex working conditions is more accurate and comprehensive, and the accuracy of temperature prediction is improved. Further, the temperature prediction model constructed by the present application based on the physical information neural network can significantly reduce the computational complexity and resource requirements while maintaining high accuracy, which makes the device more suitable for real-time monitoring and large-scale applications. Moreover, the device of the present application reduces the dependence on precise conditions such as initial conditions and boundary conditions by using the physical information neural network, and enhances the applicability and flexibility of the model in actual engineering. This means that users can apply this technology in a wider range of scenarios without worrying about the difficulty of determining the initial conditions and boundary conditions.

[0129] It should be noted that the embodiments in this specification are all described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the embodiments of the device and system, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content. The device and system embodiments described above are only illustrative. The units described as separate components may or may not be physically separated, and the components referred to as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0130] As described above, this is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A temperature prediction method for digital cables based on digital twin technology, characterized in that: The method comprises: Acquire position information and current carrying capacity information of a target point of the digital cable, wherein the position information indicates the distance from the target point to the axis of the digital cable; The temperature information of the target point is determined by inputting the position information and the current carrying capacity information of the target point into a temperature prediction model, wherein the temperature prediction model is obtained as follows: According to Maxwell's equations and Fourier's heat transfer law, an electric-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established; Solving the simulation model to obtain actual temperature information of sample points under different position information and different current carrying capacity information; The position information and current carrying capacity information of the sample points are used as inputs of the physical information neural network model, and the actual temperature information of the sample points is used as outputs of the physical information neural network model. The physical information neural network model is trained to obtain a temperature prediction model.

2. The method according to claim 1, characterized in that When the current carrying capacity information is consistent, the temperature information of the target point is negatively correlated with the position information, and when the position information is consistent, the temperature information of the target point is positively correlated with the current carrying capacity information.

3. The method according to claim 2, characterized in that When the current carrying capacity information is consistent, the relationship between the temperature information of the target point and the position information is as follows: Among them, T j is the temperature information at the axis of the digital cable, x is the position information of the target point, d i Indicates the distance from the outer surface of each layer structure to the axis.

4. The method according to claim 2, characterized in that: When the position information is consistent, the relationship between the temperature information of the target point and the current carrying capacity information is as follows: Among them, T is temperature information, K is equivalent thermal conductivity, and I0 is current carrying capacity information.

5. The method according to claim 1, characterized in that The method of using the position information and the current carrying capacity information of the sample points as inputs of the physical information neural network model, using the actual temperature information of the sample points as outputs of the physical information neural network model, and training the physical information neural network model includes: Inputting the position information and the current carrying capacity information of the sample point into the physical information neural network model to obtain the predicted temperature information of the sample point; According to the actual temperature information and the predicted temperature information of the sample point, the physical information neural network model is trained by minimizing the total loss function of the physical information neural network model, wherein the total loss function is the sum of the physical loss function, the initial loss function and the convenient loss function.

6. The method according to claim 1, characterized in that The electric-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established according to Maxwell's equations and Fourier's heat transfer law, including: Based on Maxwell's equations, constructing electromagnetic control equations of the digital cable; Based on Fourier's heat transfer law and the law of conservation of energy, construct the heat transfer control equation of the digital cable; The conductivity-temperature expression is used to couple the electromagnetic field corresponding to the electromagnetic control equation and the thermal field corresponding to the heat transfer control equation, and an electric-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established.

7. A temperature prediction device for a digital cable based on digital twin technology, characterized in that: The device comprises: an acquisition module and a determination module; The acquisition module is used to acquire the position information and current carrying capacity information of the target point of the digital cable, wherein the position information indicates the distance from the target point to the axis of the digital cable; The determination module is used to determine the temperature information of the target point by inputting the position information and the current carrying capacity information of the target point into a temperature prediction model, wherein the temperature prediction model is obtained as follows: According to Maxwell's equations and Fourier's heat transfer law, an electric-magnetic-thermal multi-physics field coupling simulation model of the digital cable is established; Solving the simulation model to obtain actual temperature information of sample points under different position information and different current carrying capacity information; The position information and current carrying capacity information of the sample points are used as inputs of the physical information neural network model, and the actual temperature information of the sample points is used as outputs of the physical information neural network model. The physical information neural network model is trained to obtain a temperature prediction model.

8. The device according to claim 7, characterized in that When the current carrying capacity information is consistent, the temperature information of the target point is negatively correlated with the position information, and when the position information is consistent, the temperature information of the target point is positively correlated with the current carrying capacity information.

9. The device according to claim 8, characterized in that When the current carrying capacity information is consistent, the relationship between the temperature information of the target point and the position information is as follows: Among them, T j is the temperature information at the axis of the digital cable, x is the position information of the target point, d i Indicates the distance from the outer surface of each layer structure to the axis.

10. The device according to claim 8, characterized in that When the position information is consistent, the relationship between the temperature information of the target point and the current carrying capacity information is as follows: Among them, T is temperature information, K is equivalent thermal conductivity, and I0 is current carrying capacity information.

Images