Three-dimensional temperature field construction method, device, equipment and medium

Abstract

The embodiment of the invention relates to a three-dimensional temperature field construction method and device, equipment and a medium, and the method comprises the steps: obtaining training data determined based on a simulation heat transfer model, and obtaining a temperature field construction network through training based on the training data; obtaining a current measurement temperature of a preset measurement point corresponding to the target object at a current moment, a preorder three-dimensional temperature field constructed by the temperature field construction network for the target object at a preorder moment, and environmental parameters of an environment where the target object is located; and inputting the current measured temperature, the preorder three-dimensional temperature field and the environmental parameters into the temperature field construction network to obtain a current three-dimensional temperature field of the target object at the current moment. Therefore, on the basis of the predicted historical three-dimensional temperature field through the neural network model, the current actual measured temperature and the environmental parameters are combined, iterative calculation of the three-dimensional temperature field is carried out, and the actual physical condition and the historical predicted three-dimensional temperature field are combined; and the rapidity and the accuracy of the whole process of carrying out temperature field prediction on the target object are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of temperature measurement technology, and in particular to a three-dimensional temperature field construction method, device, equipment and medium. Background Art

[0002] Contact temperature measurement is a common temperature measurement technology. It requires contact with the object being measured and provides temperature feedback after thermal equilibrium is reached. Without penetrating or damaging the object, this method can only provide surface temperature information. This method has limitations in its spatial measurement range.

[0003] Non-contact temperature measurement techniques, which don't require direct contact with the object being measured, can measure the temperature inside the object. However, some non-contact temperature measurement methods are only suitable for superficial temperature measurements, some can measure the temperature inside an object but are time-consuming, and some have low resolution. Currently, there is an urgent need for a method that can quickly and accurately measure the entire temperature of an object. Summary of the Invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present disclosure provides a three-dimensional temperature field construction method, device, equipment and medium.

[0005] The present disclosure provides a method for constructing a three-dimensional temperature field, the method comprising:

[0006] Acquire a sample temperature field determined based on a simulated heat transfer model, and train a temperature field construction network based on the sample temperature field;

[0007] Obtaining the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field constructed by the temperature field construction network for the target object at the previous moment, and the environmental parameters of the environment in which the target object is located;

[0008] The current measured temperature, the previous three-dimensional temperature field, and the environmental parameters are input into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

[0009] The present disclosure also provides a three-dimensional temperature field construction device, the device comprising:

[0010] A training module, configured to obtain a sample temperature field determined based on a simulated heat transfer model, and to construct a temperature field network based on the sample temperature field training;

[0011] An acquisition module is used to obtain the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field constructed by the temperature field construction network for the target object at the previous moment, and the environmental parameters of the environment in which the target object is located;

[0012] A construction module is used to input the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

[0013] An embodiment of the present disclosure also provides an electronic device, which includes: a processor; a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the three-dimensional temperature field construction method provided in the embodiment of the present disclosure.

[0014] An embodiment of the present disclosure further provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to execute the three-dimensional temperature field construction method provided in the embodiment of the present disclosure.

[0015] The technical solution provided by the embodiments of the present disclosure has the following advantages over the existing technology: the three-dimensional temperature field construction solution provided by the embodiments of the present disclosure obtains a sample temperature field determined based on a simulation heat transfer model, and obtains a temperature field construction network based on sample temperature field training; obtains the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field at the previous moment constructed by the temperature field construction network for the target object, and the environmental parameters of the environment in which the target object is located; the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters are input into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment. By adopting the above technical solution, a sample temperature field determined by a simulation heat transfer model is obtained in a simulated manner, which saves the time for obtaining the sample temperature field compared to obtaining the sample temperature field simply through experiments. A temperature field construction network is obtained based on the sample temperature field. The temperature field construction network predicts the current three-dimensional temperature field of the object based on the three-dimensional temperature field predicted for the previous moment of the object, the current measured temperature of the measurement point in the object, and the environmental parameters of the object. The neural network model is used to iteratively calculate the three-dimensional temperature field on the basis of the predicted historical three-dimensional temperature field, combined with the current actual measured temperature and environmental parameters. The combination of the actual physical situation and the historically predicted three-dimensional temperature field improves the speed and accuracy of the overall process of temperature field prediction for the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0017] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 A schematic diagram of a process for constructing a three-dimensional temperature field according to an embodiment of the present disclosure;

[0019] Figure 2 A schematic diagram of a process for constructing a temperature field network operation according to an embodiment of the present disclosure;

[0020] Figure 3 A schematic diagram of a temperature field construction network provided in an embodiment of the present disclosure;

[0021] Figure 4 A schematic diagram of a process for determining a sample temperature field according to an embodiment of the present disclosure;

[0022] Figure 5 A schematic diagram of another three-dimensional temperature field construction method provided in an embodiment of the present disclosure;

[0023] Figure 6 A schematic diagram of determining a simulation heat transfer model provided in an embodiment of the present disclosure;

[0024] Figure 7 A schematic structural diagram of a three-dimensional temperature field construction device provided in an embodiment of the present disclosure;

[0025] Figure 8 A schematic structural diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features therein can be combined with each other in the absence of conflict.

[0027] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0028] In order to solve the above problems, the embodiments of the present disclosure provide a three-dimensional temperature field construction method, which is introduced below in conjunction with specific embodiments.

[0029] Figure 1This is a flow chart of a three-dimensional temperature field construction method provided by an embodiment of the present disclosure. The method can be executed by a three-dimensional temperature field construction device, wherein the device can be implemented using software and / or hardware and can generally be integrated into an electronic device. Figure 1 As shown, the method includes:

[0030] Step 101: Obtain a sample temperature field determined based on a simulated heat transfer model, and construct a temperature field network based on training of the sample temperature field.

[0031] Wherein, the simulation heat transfer model is a three-dimensional model for performing numerical simulation calculations on the heat transfer process inside the target object. The sample temperature field can be used as the temperature field of part of the training data of the temperature field construction network. The temperature field construction network can be a neural network model for constructing a three-dimensional temperature field, and the temperature field construction network can be a temperature inversion proxy model established based on a neural network. The present embodiment does not limit the network architecture of the temperature field construction network. For example, the network architecture of the temperature field construction network includes but is not limited to a fully connected neural network (Fully Connected Neural Network, FCNN), a convolutional neural network (Convolutional Neural Network, CNN), a recurrent neural network (Recurrent Neural Network, RNN), an attention mechanism (Attention) neural network, and an operator learning neural network. Wherein, the operator learning neural network can be a deep operator neural network (Deep Operator Network, DeepONet) or a Fourier neural operator network (FNO).

[0032] In an embodiment of the present disclosure, the three-dimensional temperature field construction device can obtain a simulated heat transfer model to determine a sample temperature field in a simulation calculation manner, and use the sample temperature field as part of the training data to train a temperature field construction network.

[0033] Step 102 : obtaining the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field constructed by the temperature field construction network for the target object at the previous moment, and the environmental parameters of the environment in which the target object is located.

[0034] The target object may be an object whose three-dimensional temperature field is currently to be determined, and the target object may be a three-dimensional object having length, width, and height. This embodiment does not limit the type of the target object. In some embodiments of the present disclosure, the type of the target object includes one of an organ type and a component type. The organ type may be understood as a biological organ type. The component type may be understood as an electronic component type. The target object of the component type may include at least one of an integrated circuit, a resistor, an inductor, a capacitor, a diode, and a transistor.

[0035] The preset measurement point may be a pre-set temperature measurement location point of the target object. The preset measurement point may be set according to the type of the target object, the needs of the user, etc., and is not limited in this embodiment. In some embodiments of the present disclosure, the preset measurement point includes a measurement point inside the target object and / or a measurement point on the surface of the target object. The number of measurement points inside the target object may be one or more, and the local actual temperature measurement inside the target object may be achieved through the multiple measurement points inside the target object. The number of measurement points on the surface of the target object may be one or more, and the local actual temperature measurement on the surface of the target object may be achieved through the multiple measurement points on the surface of the target object.

[0036] The currently measured temperature may be the temperature actually measured at a preset measurement point on the target object at the current moment. This embodiment does not limit the measurement method of the currently measured temperature. In some embodiments of the present disclosure, the currently measured temperature is determined by contact temperature measurement and / or non-contact temperature measurement.

[0037] The preceding moment may be a moment before the current moment, and the number of the preceding moments may be one or more. Optionally, the preceding moment may be a moment before the current moment during the periodic iterative calculation of the three-dimensional temperature field. The preceding three-dimensional temperature field may be the three-dimensional temperature field of the target object at the preceding moment predicted by the temperature field construction network. The three-dimensional temperature field can be used to characterize the temperature distribution inside the target object.

[0038] The environment may be the physical environment in which the target object resides. Environmental parameters may be used to describe specific indicators of the characteristics and conditions of the environment at the current moment or between a previous moment and the current moment. This embodiment does not limit the type of environmental parameters. For example, the type of environmental parameters may include, but is not limited to, at least one of ambient temperature, ambient temperature variation, external flow field flow conditions, and external fluid medium type.

[0039] In the embodiments of the present disclosure, the three-dimensional temperature field construction method can be applicable to different ambient temperatures, and the present embodiment does not impose any restrictions on the ambient temperature. For example, the three-dimensional temperature field construction method can be applicable to ambient temperatures ranging from -196 degrees Celsius to 200 degrees Celsius. Alternatively, the ambient temperature can be as low as the liquid nitrogen temperature zone and as high as the temperature zone used for hyperthermia. In some embodiments of the present disclosure, the three-dimensional temperature field construction method can be applied to predicting the three-dimensional temperature field of a target object during a heating or cooling process after the target object is moved from an initial environment to a new environment.

[0040] In the disclosed embodiments, a temperature field construction network can be used to perform periodic iterative calculations of the target object's three-dimensional temperature field. Specifically, at the current moment, the three-dimensional temperature field device can obtain the current measured temperature of the target object at the current moment, obtained through actual measurement at a preset measurement point. Furthermore, the temperature field construction network can be used to obtain the previous three-dimensional temperature field predicted for the target object at the previous moment. Furthermore, the environmental parameters of the target object's environment between the previous moment and the current moment can be obtained.

[0041] Step 103 : Input the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

[0042] The current three-dimensional temperature field may be the three-dimensional temperature field of the target object at the current moment predicted by the temperature field construction network.

[0043] In an embodiment of the present disclosure, the three-dimensional temperature field construction device can input the current measured temperature of the preset measurement point of the target object at the current moment obtained by actual measurement, the previous three-dimensional temperature field of the target object at the previous moment obtained by prediction, and the environmental parameters of the target object into the temperature field construction network, and the temperature field construction network outputs the predicted current three-dimensional temperature field of the target object at the current moment.

[0044] In some embodiments of the present disclosure, the network architecture of the temperature field construction network is a deep operator neural network (DeepOperator Network, DeepONet). The temperature field construction network includes a temperature field branch network (Branch Net) and a temperature field trunk network (Trunk Net). The temperature field branch network can be used to extract features of the preceding three-dimensional temperature field. The temperature field trunk network can be used to extract features of the current measured temperature and environmental parameters.

[0045] Figure 2 A schematic diagram of a process for constructing a network operation for a temperature field provided in an embodiment of the present disclosure is shown as follows: Figure 2 As shown, in some embodiments of the present disclosure, the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters are input into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment, including:

[0046] Step 201: Input the previous three-dimensional temperature field into the temperature field branch network to obtain an initial condition vector, and input the current measured temperature and environmental parameters into the temperature field main network to obtain a boundary condition vector.

[0047] The initial condition vector may be a vector obtained by extracting features using the preceding three-dimensional temperature field as the initial condition. The boundary condition vector may be a vector obtained by extracting features using the current measured temperature and environmental parameters as boundary conditions. The dimensions of the initial condition vector and the boundary condition vector may be the same.

[0048] In this embodiment, the temperature field construction network is based on a deep operator neural network architecture. The network comprises two subnetworks: a temperature field branch network and a temperature field backbone network. The input data for the temperature field branch network includes the three-dimensional temperature field predicted for the previous moment. The temperature field branch network maps the three-dimensional temperature field to a high-dimensional vector and outputs an initial condition vector.

[0049] The input data of the temperature field backbone network include the current measured temperature actually measured at the current moment and the environmental parameters of the target object. The temperature field backbone network maps the current measured temperature and environmental parameters into high-dimensional vectors and outputs a boundary condition vector.

[0050] Step 202: The boundary condition vector and the initial condition vector are fused to obtain a comprehensive condition vector, and the current three-dimensional temperature field is determined according to the comprehensive condition vector.

[0051] The comprehensive condition vector can be a vector obtained by performing a dot product operation on the boundary condition vector and the initial condition vector. The comprehensive condition vector can be used to comprehensively represent the high-dimensional features of the preceding three-dimensional temperature field, the current measured temperature, and the environmental parameters.

[0052] In this embodiment, the temperature field construction network can perform dot product calculation on the boundary condition vector and the initial condition vector to obtain a comprehensive condition vector, and perform further neural network calculation based on the comprehensive condition vector to obtain the current three-dimensional temperature field as the output.

[0053] Figure 3 A schematic diagram of a temperature field construction network provided by an embodiment of the present disclosure, such as Figure 3 As shown, after obtaining the temperature field construction network, the initial conditions are input into the temperature field branch network to obtain the initial condition vector, the boundary conditions are input into the temperature field trunk network to obtain the boundary condition vector, the initial condition vector and the boundary condition vector are dot-producted to obtain the comprehensive condition vector, and the current three-dimensional temperature field is determined based on the comprehensive condition vector.

[0054] In the above scheme, the temperature field branch network is used to process the previous three-dimensional temperature field as the initial condition, so that the processing of the previous three-dimensional temperature field conforms to the conditional constraints played by the previous three-dimensional temperature field in determining the current three-dimensional temperature field. The temperature field backbone network is used to process the current measured temperature and environmental parameters as boundary conditions, so that the processing of the current measured temperature and environmental parameters conforms to the conditional constraints played by the current measured temperature and environmental parameters in determining the current three-dimensional temperature field. This improves the accuracy of the final determined current three-dimensional temperature field.

[0055] The three-dimensional temperature field construction scheme provided by the embodiment of the present disclosure obtains a sample temperature field determined based on a simulation heat transfer model, and obtains a temperature field construction network based on sample temperature field training; obtains the current measured temperature of a preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field at the previous moment constructed by the temperature field construction network for the target object, and the environmental parameters of the environment in which the target object is located; inputs the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment. By adopting the above technical solution, a sample temperature field determined by a simulation heat transfer model is obtained in a simulated manner, which saves the time of obtaining the sample temperature field compared with the method of obtaining the sample temperature field simply through experiments. A temperature field construction network is obtained based on the sample temperature field. The temperature field construction network predicts the current three-dimensional temperature field of the object based on the three-dimensional temperature field predicted for the previous moment of the object, the current measured temperature of the measurement point in the object, and the environmental parameters of the object. The neural network model is used to iteratively calculate the three-dimensional temperature field on the basis of the predicted historical three-dimensional temperature field, combined with the current actual measured temperature and environmental parameters. The combination of the actual physical situation and the historically predicted three-dimensional temperature field improves the speed and accuracy of the overall process of temperature field prediction for the target object.

[0056] In addition, when the type of the target object is an organ type, high-precision temperature field prediction inside the organ is an important basis for achieving long-term low-temperature preservation of the organ. In related technologies, the temperature measurement method for organs is mainly contact temperature measurement. However, contact temperature measurement will cause certain damage to the organ, and the temperature measurement depth of non-contact temperature measurement is limited in a short time, making it difficult to achieve accurate measurement of the temperature field inside the organ. By adopting the three-dimensional temperature field construction method provided by the embodiment of the present disclosure, the temperature inside the organ can be accurately determined in real time in a non-contact manner, revealing the temperature evolution law inside the object during long-term low-temperature preservation, and constructing a high-precision three-dimensional temperature field of the organ.

[0057] In the case where the type of the target object is a component type, a large amount of heat will be generated during the use of the component. If the heat is not dissipated in time, it will cause damage to the component. Therefore, the temperature measurement of the component is very important. In the related art, for more complex components, the heat generated during use may be focused on a certain part of the component, that is, the temperature distribution of the component is not uniform, so it is difficult to measure the overall temperature distribution of the component through contact temperature measurement. Through the three-dimensional temperature field construction method provided by the embodiment of the present disclosure, the overall temperature distribution of the component can be accurately and quickly measured through the non-contact temperature measurement method.

[0058] In the embodiment of the present disclosure, the sample temperature field in the training data corresponding to the temperature field construction network may be determined through experimental measurement and / or determined through heat transfer simulation based on a simulation heat transfer model.

[0059] Taking the sample temperature field as an example, including the temperature field determined by heat transfer simulation based on the simulation heat transfer model, the sample temperature field includes the current sample temperature field of the target object at the sample moment and the historical sample temperature field of the target object at the historical moment; wherein the time sequence of the historical moment is before the sample moment.

[0060] The sample moment may be a moment serving as a sample, and the sample moment may be the moment corresponding to the current sample temperature field. The current sample temperature field may be the three-dimensional temperature field of the target object at the sample moment. The historical moment may be a moment preceding the sample moment, and the number of such historical moments may be one or more. The historical sample temperature field may be the three-dimensional temperature field of the target object at the historical moment.

[0061] Figure 4 A schematic diagram of a process for determining a sample temperature field according to an embodiment of the present disclosure is provided. Figure 4 As shown, the sample temperature field determined based on the simulation heat transfer model is obtained, including:

[0062] Step 401: Construct a three-dimensional simulation model of the target object according to the components of the target object.

[0063] The component may be a part that constitutes the target object. For example, if the target object is a component, the component may include one or more of ceramic, metal, plastic, epoxy resin, and silicon.

[0064] In the disclosed embodiment, the components of the target object are determined in advance by measurement. The three-dimensional temperature field construction device can obtain the components of the target object and, based on the components, invoke numerical simulation software to perform three-dimensional modeling based on the components, thereby obtaining a three-dimensional simulation model of the target object.

[0065] Step 402: Configure the physical field and physical property parameters of each component for the three-dimensional simulation model to obtain a simulation heat transfer model.

[0066] The physical field may be a spatial condition and / or a temporal condition used to implement a simulated display of an environment in a simulation calculation, and this embodiment does not limit the physical field. For example, the physical field may include a heat transfer physical field, a flow physical field, and the like.

[0067] Physical property parameters can characterize the properties of the target object itself and can correspond to the type of target object. For example, if the target object is an organ, the physical property parameters may include, but are not limited to, one or more of thermal conductivity, specific heat capacity, density, viscosity, protective liquid melting point, and latent heat of phase change. If the target object is a component, the physical property parameters may include, but are not limited to, one or more of thermal conductivity, specific heat capacity, and density. The physical property parameters may also include size and / or shape.

[0068] In this embodiment, the three-dimensional temperature field construction device configures the physical field and the physical property parameters of each component in the three-dimensional simulation model for the three-dimensional simulation model to obtain a simulated heat transfer model.

[0069] Step 403: Based on the preset sample environment parameters and the simulation heat transfer model, simulation calculation is performed on the target object at the sample moment to obtain the current sample temperature field, and simulation calculation is performed on the target object at the historical moment to obtain the historical sample temperature field.

[0070] In this embodiment, the three-dimensional temperature field construction device can use the sample environment parameters as specific data for the physical field parameters in the simulated heat transfer model to simulate the environment of the target object. The meshing software or numerical calculation software is called to perform simulation calculations such as heat flux density processing, key parameter derivation, and key parameter integration on the simulated heat transfer model, thereby obtaining the temperature distribution change inside the target object. The temperature distribution change at the sample time is determined to obtain the current sample temperature field, and the temperature distribution change at the historical time is determined to obtain the historical sample temperature field.

[0071] like Figure 3 As shown, the heat transfer simulation model can generate partial training data for the initial network, such as the current sample temperature field and historical sample temperature fields. The initial network is trained based on this training data to obtain a temperature field construction network. Subsequently, the current three-dimensional temperature field is determined based on this temperature field construction network.

[0072] Figure 5 A schematic diagram of another three-dimensional temperature field construction method provided in an embodiment of the present disclosure is shown as follows: Figure 5As shown, after determining the target object, the geometric parameters and physical parameters of each component of the target object are measured, and a three-dimensional simulation model of the target object is constructed based on the geometric parameters of each component. The physical field and the physical parameters of each component are configured for the three-dimensional simulation model to obtain a heat transfer simulation model of the target object. A temperature database of the target object is constructed based on the heat transfer simulation model. The temperature database includes the current sample temperature field, the historical sample temperature field, etc. The initial model is trained based on the temperature database to obtain a temperature field construction network. By performing actual real-time measurement on a part or surface of the target object, the current measured temperature is obtained. Combined with the previous three-dimensional temperature field predicted by the temperature field construction network at the previous moment and the current measured temperature and the environmental parameters of the target object, the current three-dimensional temperature field of the target object at the current moment is constructed in real time.

[0073] In the above scheme, a large amount of sample temperature field data determined based on the simulation heat transfer model is learned based on the deep learning network to obtain a temperature field construction network. The current three-dimensional temperature field is constructed based on the actual measured physical data, the physical environment of the target object, and the previously predicted three-dimensional temperature field through the temperature field construction network, realizing the embedded physics combined with data-driven temperature inversion calculation.

[0074] In some embodiments of the present disclosure, after obtaining the current sample temperature field, the three-dimensional temperature field construction method further includes:

[0075] Obtain a verification temperature, where the verification temperature is the temperature obtained by collecting the temperature of the target object at the sample time under the sample environmental parameters; determine the temperature to be verified corresponding to the verification temperature in the current sample temperature field; if the temperature difference between the verification temperature and the temperature to be verified is greater than a preset difference threshold, generate a model adjustment prompt information for the simulation heat transfer model.

[0076] The temperature to be verified may be the temperature at the same sampling position as the verification temperature in the current sample temperature field. The temperature difference may be the temperature difference between the verification temperature and the temperature to be verified. The preset difference threshold may be the maximum value of the temperature difference when the simulation heat transfer model is not adjusted. The model adjustment prompt information may be prompt information for prompting parameter adjustment of the simulation heat transfer model. This embodiment does not limit the type of the model adjustment prompt information. For example, the model adjustment prompt information may include voice prompt information, pop-up prompt information, etc.

[0077] In this embodiment, the actual temperature of the target object at the sampling time under the sample environmental parameters is collected through contact or non-contact temperature measurement to obtain a verification temperature. A temperature to be verified is determined at a sampling location in the current sample temperature field that is the same as the verification temperature. The temperature difference between the verification temperature and the temperature to be verified is calculated, and a determination is made as to whether the temperature difference is greater than a preset difference threshold.

[0078] If so, a model adjustment prompt for the heat transfer simulation model is generated, prompting the user to adjust one or more of the meshing method, physical parameters, and flow condition equations involved in the heat transfer simulation model to obtain an adjusted heat transfer simulation model. The adjusted heat transfer simulation model is then returned to determine a new temperature to be verified and a new temperature difference. If the new temperature difference is no greater than a preset difference threshold, the new heat transfer simulation model is determined as the final heat transfer simulation model. Training data for training the initial network is subsequently constructed based on the heat transfer simulation model.

[0079] Figure 6 A schematic diagram of determining a simulation heat transfer model provided by an embodiment of the present disclosure is shown in FIG. Figure 6 As shown, a three-dimensional simulation model is constructed based on the components of the target object. The physical field and physical property parameters are configured for the three-dimensional simulation model to obtain a heat transfer simulation model, and the temperature to be verified is determined using the heat transfer simulation model. Furthermore, the temperature of the target object is measured through actual experiments to obtain a verification temperature. If the temperature difference between the verification temperature and the temperature to be verified is too large, feedback is provided that the heat transfer simulation model needs to be corrected until the temperature difference is no greater than a preset difference threshold. Training data for the initial network is generated based on the corrected simulation heat transfer model.

[0080] In this approach, the need for iterative correction of the heat transfer simulation model is determined based on the validation data generated by the heat transfer simulation model and the validation data obtained from actual measurements. This iterative correction improves the simulation accuracy of the heat transfer simulation model, laying the foundation for subsequently obtaining a large amount of relatively accurate training data.

[0081] In some embodiments of the present disclosure, a temperature field construction network is obtained based on sample temperature field training, including: training a pre-set initial network according to the sample measurement temperature of the sample measurement point corresponding to the target object at the sample time, the historical sample temperature field, the environmental sample parameters of the environment in which the target object is located, and the current sample temperature field to obtain a temperature field construction network.

[0082] Among them, the sample measurement point can be a temperature measurement position of the target object set in advance as a sample, and the sample measurement point can be the same as or different from the target measurement point. The sample measurement temperature can be the temperature actually measured at the sample measurement point of the target object at the sample moment. The environmental sample parameters can be used to describe specific indicators of the characteristics and conditions of the environment at the sample moment or between the historical moment and the sample moment. This embodiment does not limit the type of the sample environmental parameters. For example, the type of the sample environmental parameters may include at least one of ambient temperature, ambient temperature change, external flow field flow conditions, and external fluid medium type. The initial network can be an untrained neural network model with the same network architecture as the temperature field construction network.

[0083] In this embodiment, for a sample moment, the three-dimensional temperature field construction apparatus can obtain the sample measurement temperature determined through actual measurement at the sample measurement point corresponding to the target object at that sample moment. A historical moment prior to the sample moment is determined, and the historical sample temperature field of the target object at that historical moment is obtained. The environmental sample parameters of the target object's environment corresponding to the sample moment are determined, and the current sample temperature field of the target object at that sample moment is obtained.

[0084] Furthermore, the three-dimensional temperature field construction device can use the sample predicted temperature, historical sample temperature field, and environmental sample parameters as input data for the initial network, and the current sample temperature field as output data for the initial network. Based on this input and output data, the device optimizes and adjusts the hyperparameters in the initial network to obtain a temperature field construction network. The hyperparameters may include at least one of the network type, layer size, layer depth, and hidden vector dimension of each layer. Subsequent iterative calculations can be performed based on this temperature field construction network to obtain the three-dimensional temperature field of the target object during the temperature increase or decrease process in real time.

[0085] In the above scheme, the construction of the temperature field construction network is realized based on the training data, which creates the basic conditions for the subsequent construction of the three-dimensional temperature field.

[0086] The three-dimensional temperature field construction method provided by the embodiments of the present disclosure collects the local temperature of the surface or the interior of an object through a non-contact temperature measurement method, and then uses a temperature field construction network based on the temperature to predict the three-dimensional temperature field inside the object in real time, thereby achieving rapid and accurate measurement of the internal temperature of the object and solving the problem of difficulty in timely acquisition of the three-dimensional temperature field.

[0087] Figure 7 This is a schematic diagram of the structure of a three-dimensional temperature field construction device provided by an embodiment of the present disclosure. The device can be implemented by software and / or hardware and can generally be integrated into electronic devices. Figure 7 As shown, the device includes:

[0088] A training module 701 is configured to obtain a sample temperature field determined based on a simulated heat transfer model, and to train a temperature field based on the sample temperature field to construct a network;

[0089] An acquisition module 702 is configured to acquire a current measured temperature of a preset measurement point corresponding to a target object at a current moment, a previous three-dimensional temperature field constructed by a temperature field construction network for the target object at a previous moment, and environmental parameters of the environment in which the target object is located;

[0090] The construction module 703 is used to input the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

[0091] In some embodiments of the present disclosure, the type of the target object includes one of an organ type and a component type, and the component type includes at least one of an integrated circuit, a resistor, an inductor, a capacitor, a diode, and a transistor, and the current measured temperature is determined by contact temperature measurement and / or non-contact temperature measurement.

[0092] In some embodiments of the present disclosure, the preset measurement points include measurement points inside the target object and / or measurement points on the surface of the target object.

[0093] In some embodiments of the present disclosure, the temperature field construction network includes a temperature field branch network and a temperature field trunk network;

[0094] The building block 702 is configured to:

[0095] Inputting the preceding three-dimensional temperature field into the temperature field branch network to obtain an initial condition vector, and inputting the current measured temperature and the environmental parameters into the temperature field main network to obtain a boundary condition vector;

[0096] The boundary condition vector and the initial condition vector are fused to obtain a comprehensive condition vector, and the current three-dimensional temperature field is determined according to the comprehensive condition vector.

[0097] In some embodiments of the present disclosure, the sample temperature field includes the current sample temperature field of the target object at the sample moment and the historical sample temperature field of the target object at the historical moment; wherein the time sequence of the historical moment is before the sample moment;

[0098] The obtaining of the sample temperature field determined based on the simulation heat transfer model includes:

[0099] Constructing a three-dimensional simulation model of the target object according to each component of the target object;

[0100] Configuring the physical field and the physical property parameters of each component for the three-dimensional simulation model to obtain a simulation heat transfer model;

[0101] Based on the preset sample environment parameters and the simulation heat transfer model, the target object at the sample moment is simulated and calculated to obtain the current sample temperature field, and the target object at the historical moment is simulated and calculated to obtain the historical sample temperature field.

[0102] In some embodiments of the present disclosure, the device further includes an adjustment prompt module, wherein the adjustment prompt module is configured to:

[0103] After obtaining the current sample temperature field, obtaining a verification temperature, wherein the verification temperature is a temperature obtained by collecting the temperature of the target object at the sample moment under the sample environmental parameters;

[0104] Determining a temperature to be verified corresponding to the verification temperature in the current sample temperature field;

[0105] If the temperature difference between the verification temperature and the temperature to be verified is greater than a preset difference threshold, model adjustment prompt information of the simulation heat transfer model is generated.

[0106] In some embodiments of the present disclosure, the step of constructing a network based on a temperature field obtained through training of the sample temperature field includes:

[0107] The preset initial network is trained according to the sample measurement temperature of the sample measurement point corresponding to the target object at the sample moment, the historical sample temperature field, the environmental sample parameters of the environment in which the target object is located, and the current sample temperature field to obtain the temperature field construction network.

[0108] The three-dimensional temperature field construction device provided in the embodiments of the present disclosure can execute the three-dimensional temperature field construction method provided in any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0109] Figure 8 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 8 As shown, electronic device 800 includes one or more processors 801 and memory 802 .

[0110] The processor 801 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 800 to perform desired functions.

[0111] The memory 802 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 801 may run the program instructions to implement the three-dimensional temperature field construction method of the embodiment of the present disclosure described above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0112] In one example, the electronic device 800 may further include an input device 803 and an output device 804 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0113] In addition, the input device 803 may also include, for example, a keyboard, a mouse, and the like.

[0114] The output device 804 can output various information to the outside, including determined distance information, direction information, etc. The output device 804 can include, for example, a display, a speaker, a printer, a communication network and its connected remote output device, etc.

[0115] Of course, to simplify, Figure 8 Only some of the components related to the present disclosure in the electronic device 800 are shown, and components such as buses, input / output interfaces, etc. are omitted. In addition, according to specific application scenarios, the electronic device 800 may further include any other appropriate components.

[0116] In addition to the above methods and devices, the embodiments of the present disclosure may also be a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, the processor executes the three-dimensional temperature field construction method provided by the embodiments of the present disclosure.

[0117] The computer program product may be written in any combination of one or more programming languages to implement the operations of the disclosed embodiments, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0118] In addition, the embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor executes the three-dimensional temperature field construction method provided by the embodiment of the present disclosure.

[0119] The computer-readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0120] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0121] The foregoing description is intended only to provide specific embodiments of the present disclosure, intended to enable those skilled in the art to understand and implement the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the embodiments described herein, but rather to be construed in the broadest manner consistent with the principles and novel features disclosed herein.

Claims

1. A three-dimensional temperature field construction method, characterized in that: include: Acquire a sample temperature field determined based on a simulated heat transfer model, and train a temperature field construction network based on the sample temperature field; Obtaining the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field constructed by the temperature field construction network for the target object at the previous moment, and the environmental parameters of the environment in which the target object is located; The current measured temperature, the previous three-dimensional temperature field, and the environmental parameters are input into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

2. The method according to claim 1, characterized in that The type of the target object includes one of an organ type and a component type, the component type includes at least one of an integrated circuit, a resistor, an inductor, a capacitor, a diode, and a triode, and the current measured temperature is determined by contact temperature measurement and / or non-contact temperature measurement.

3. The method according to claim 1, characterized in that The preset measurement points include measurement points inside the target object and / or measurement points on the surface of the target object.

4. The method according to claim 1, wherein The temperature field construction network includes a temperature field branch network and a temperature field trunk network; The step of inputting the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment includes: Inputting the preceding three-dimensional temperature field into the temperature field branch network to obtain an initial condition vector, and inputting the current measured temperature and the environmental parameters into the temperature field main network to obtain a boundary condition vector; The boundary condition vector and the initial condition vector are fused to obtain a comprehensive condition vector, and the current three-dimensional temperature field is determined according to the comprehensive condition vector.

5. The method according to claim 1, wherein The sample temperature field includes the current sample temperature field of the target object at the sample moment and the historical sample temperature field of the target object at the historical moment; wherein the time sequence of the historical moment is before the sample moment; The obtaining of the sample temperature field determined based on the simulation heat transfer model includes: Constructing a three-dimensional simulation model of the target object according to each component of the target object; Configuring the physical field and the physical property parameters of each component for the three-dimensional simulation model to obtain a simulation heat transfer model; Based on the preset sample environment parameters and the simulation heat transfer model, the target object at the sample moment is simulated and calculated to obtain the current sample temperature field, and the target object at the historical moment is simulated and calculated to obtain the historical sample temperature field.

6. The method according to claim 5, characterized in that After obtaining the current sample temperature field, the method further includes: Acquiring a verification temperature, wherein the verification temperature is a temperature obtained by collecting the temperature of the target object at the sample time under the sample environmental parameters; Determining a temperature to be verified corresponding to the verification temperature in the current sample temperature field; If the temperature difference between the verification temperature and the temperature to be verified is greater than a preset difference threshold, model adjustment prompt information of the simulation heat transfer model is generated.

7. The method according to claim 5, characterized in that The training of the sample temperature field to obtain a temperature field and constructing a network includes: The preset initial network is trained according to the sample measurement temperature of the sample measurement point corresponding to the target object at the sample moment, the historical sample temperature field, the environmental sample parameters of the environment in which the target object is located, and the current sample temperature field to obtain the temperature field construction network.

8. A three-dimensional temperature field construction device, characterized in that: include: A training module, configured to obtain a sample temperature field determined based on a simulated heat transfer model, and to construct a temperature field network based on the sample temperature field training; An acquisition module is used to obtain the current measured temperature of the preset measurement point corresponding to the target object at the current moment, the previous three-dimensional temperature field constructed by the temperature field construction network for the target object at the previous moment, and the environmental parameters of the environment in which the target object is located; A construction module is used to input the current measured temperature, the previous three-dimensional temperature field, and the environmental parameters into the temperature field construction network to obtain the current three-dimensional temperature field of the target object at the current moment.

9. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the three-dimensional temperature field construction method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and the computer program is used to execute the three-dimensional temperature field construction method according to any one of claims 1 to 7.

Images