Explosion range prediction method, device and storage medium

Abstract

This application provides an explosion range prediction method, device, and storage medium, relating to the field of pipeline design technology, to address the problem of how to quickly and accurately predict explosion ranges. The method comprises: obtaining data to be predicted from a detection pipeline; using the data to be predicted to predict the explosion range in the event of an explosion in the detection pipeline; inputting the data to be predicted into a pre-trained explosion range prediction model to obtain a prediction result for the detection pipeline's explosion range; the explosion range prediction model is trained using a target loss function; the target loss function comprises: an explosion range prediction loss, a physical residual loss determined by a relevant physical formula, and an entropy loss. This application can quickly and accurately predict the explosion range, obtaining an explosion range prediction result that conforms to physical constraints.

Description

Technical Field

[0001] The present application relates to the technical field of pipeline design, and in particular to an explosion range prediction method, device, and storage medium. Background Art

[0002] Hydrogen storage and transportation, a key link in the hydrogen energy industry chain, currently relies primarily on hydrogen pipelines for transportation. However, hydrogen is flammable and explosive, and an explosion can cause severe damage. Therefore, the safe design and operation of hydrogen pipelines are crucial.

[0003] Accurately predicting the explosion range of hydrogen pipelines can provide an important basis for the safe design of hydrogen pipelines. For example, accurately predicting the explosion range of hydrogen pipelines can not only provide accurate data support for the design of safe distances between hydrogen pipelines and surrounding facilities, but also provide a reference for the evaluation and modification of safe distances of underground hydrogen pipelines constructed in the early stage.

[0004] Therefore, how to quickly and accurately predict the explosion range of hydrogen pipelines is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The purpose of this application is to provide an explosion range prediction method, device and storage medium, aiming to solve the problem of how to quickly and accurately predict the explosion range.

[0006] To achieve the above objectives, this application adopts the following technical solutions:

[0007] In a first aspect, a method for predicting an explosion range is provided, comprising: obtaining data to be predicted from a detection pipeline; the data to be predicted is used to predict the explosion range when the detection pipeline explodes; the data to be predicted is input into a pre-trained explosion range prediction model to obtain an explosion range prediction result of the detection pipeline; the explosion range prediction model is trained according to a target loss function; the target loss function comprises: a first loss function for determining an explosion range prediction loss, a second loss function for determining a physical residual loss, and a third loss function for determining an entropy loss; the explosion range prediction loss is used to represent the difference between an explosion range prediction result predicted by the explosion range prediction model and an actual explosion range result; and the physical residual loss is used to ensure that the explosion range prediction result predicted by the explosion range prediction model conforms to physical constraints.

[0008] Optionally, the explosion range prediction model is trained by:

[0009] A training sample set is obtained; the training sample set includes multiple training samples and a label for each training sample; the training sample includes training data of a sample pipeline; and the label of the training sample includes a true result of the explosion range of the sample pipeline.

[0010] Based on the training sample set, the original explosion range prediction model is trained to obtain a trained explosion range prediction model.

[0011] Optionally, the original explosion range prediction model is trained based on the training sample set to obtain a trained explosion range prediction model, including:

[0012] Input the training sample set into the explosion range prediction model to obtain the loss value of the current training round, and train the explosion range prediction model based on the loss value of the current training round until the loss value meets the convergence condition to obtain a trained explosion range prediction model; the loss value of the current training round is obtained by the following method:

[0013] The training samples are input into the explosion range prediction model of the current training time to obtain the current explosion range prediction result corresponding to the training samples.

[0014] The current explosion range prediction loss is determined based on the current explosion range prediction result, the label of the training sample and the first loss function.

[0015] Determine the current physical residual loss based on the training sample, the physical residual equation and the second loss function.

[0016] Determine the current entropy loss based on the training samples, the entropy regularization term equation and the third loss function.

[0017] The loss value of the current training round is determined based on the current explosion range prediction loss, current physical residual loss, current entropy loss and target loss function.

[0018] Optionally, the loss value for the current training round is determined based on the current explosion range prediction loss, the current physical residual loss, the current entropy loss, and the target loss function, including:

[0019] The objective loss function satisfies the following formula:

[0020]

[0021] in, is the loss value of the current training time, λ1 is the weight of the first loss function, W is the current explosion range prediction result, W Real is the true result of the explosion range of the sample pipeline, λ2 is the weight of the second loss function, f i is the i-th physical residual equation, λ3 is the weight of the third loss function, is the entropy regularization equation.

[0022] Optionally, obtain a training sample set, including:

[0023] Get the raw data corresponding to the training data of the sample pipeline.

[0024] The raw data is preprocessed to obtain data features corresponding to the raw data; the preprocessing includes: at least one of: outlier removal, standardization, and missing value filling.

[0025] Determine the degree of correlation between any two variable features in the data features.

[0026] According to the degree of correlation between any two variable features, the data features are screened to determine the training sample set.

[0027] Optionally, determine the degree of correlation between any two variable features in the data features, including:

[0028] The degree of correlation between any two variable features satisfies the following formula:

[0029]

[0030] Among them, r s is the degree of association between any two variable features, d i is the rank difference of the i-th rank data value in any two variable features, and m is the number of training samples.

[0031] Optionally, data features are screened based on the degree of correlation between any two variable features to determine a training sample set, including:

[0032] When the correlation degree between any two variable features is greater than or equal to a first preset threshold, the variable features with smaller variable feature values ​​are eliminated; and when the correlation degree between any two variable features is less than the first preset threshold, any two variable features are retained to obtain a training sample set.

[0033] Optionally, the original data includes: chemical property data of hydrogen transmitted in the sample pipeline, physical property data of the sample pipeline, mechanical property data of the soil in the area where the sample pipeline is located, and at least one of energy release data generated when the sample pipeline explodes.

[0034] The data to be predicted includes at least one of: chemical property data of hydrogen transmitted in the detection pipeline, physical property data of the detection pipeline, mechanical property data of soil in the area where the detection pipeline is located, and energy release data generated when the detection pipeline explodes.

[0035] In a second aspect, an explosion range prediction device is provided, comprising: a communication unit and a processing unit.

[0036] The communication unit is used to obtain the data to be predicted of the detection pipeline; the data to be predicted is used to predict the explosion range when the detection pipeline explodes.

[0037] A processing unit is used to input the data to be predicted into a pre-trained explosion range prediction model to obtain the explosion range prediction result of the detection pipeline; the explosion range prediction model is trained according to the target loss function; the target loss function includes: a first loss function for determining the explosion range prediction loss, a second loss function for determining the physical residual loss and a third loss function for determining the entropy loss; the explosion range prediction loss is used to represent the difference between the explosion range prediction result predicted by the explosion range prediction model and the actual explosion range result; the physical residual loss is used to make the explosion range prediction result predicted by the explosion range prediction model conform to the physical constraints.

[0038] In a third aspect, an explosion range prediction device is provided, comprising a memory and a processor; the memory is used to store computer-executable instructions, and the processor is connected to the memory via a bus; when the explosion range prediction device is running, the processor executes the computer-executable instructions stored in the memory, so that the explosion range prediction device performs the explosion range prediction method of the first aspect.

[0039] The explosion range prediction device can be an electronic device or a component within an electronic device, such as a system-on-chip within the electronic device. The system-on-chip is configured to support the electronic device in implementing the functions described in the first aspect and any possible implementation thereof, such as acquiring, determining, and transmitting data and / or information used in the explosion range prediction method. The system-on-chip includes a chip and may also include other discrete components or circuit structures.

[0040] According to a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium includes computer-executable instructions. When the computer-executable instructions are executed on a computer, the computer is caused to execute the explosion range prediction method according to the first aspect.

[0041] In a fifth aspect, a computer program product is further provided. The computer program product includes computer instructions. When the computer instructions are executed on an explosion range prediction device, the explosion range prediction device executes the explosion range prediction method according to the first aspect.

[0042] It should be noted that the aforementioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer-readable storage medium may be packaged together with the processor of the explosion range prediction device, or may be packaged separately from the processor of the explosion range prediction device, and this is not limited in the present embodiment.

[0043] The description of the second, third, fourth and fifth aspects of this application can refer to the detailed description of the first aspect.

[0044] In the embodiments of this application, the name of the explosion range prediction device does not limit the device or functional modules themselves. In actual implementation, these devices or functional modules may appear with other names. For example, the communication unit may also be called a communication module. As long as the functions of each device or functional module are similar to those of this application, they are within the scope of the claims of this application and their equivalents.

[0045] The technical solution provided by this application brings at least the following beneficial effects:

[0046] Based on any of the aforementioned aspects, embodiments of the present application provide an explosion range prediction method, comprising: first, obtaining data to be predicted from a detection pipeline (the data to be predicted is used to predict the explosion range in the event of an explosion in the detection pipeline); then, inputting the data to be predicted into a pre-trained explosion range prediction model to obtain a prediction result for the explosion range of the detection pipeline. The pre-trained explosion range prediction model is trained using a target loss function; the target loss function includes a first loss function for determining explosion range prediction loss, a second loss function for determining physical residual loss, and a third loss function for determining entropy loss.

[0047] As can be seen from the above, firstly, the present application can obtain the explosion range prediction result of the detection pipeline by inputting the data to be predicted into the pre-trained explosion range prediction model, thereby improving the prediction efficiency.

[0048] Secondly, since the above-mentioned explosion range prediction model introduces the first loss function, it can determine the difference between the explosion range prediction result predicted by the explosion range prediction model and the actual explosion range result, thereby improving the prediction accuracy of the explosion range prediction model.

[0049] Then, because the above explosion range prediction model introduces the second loss function, the explosion range prediction results predicted by the explosion range prediction model conform to physical constraints, further improving the prediction accuracy of the explosion range prediction model. Thirdly, because the above explosion range prediction model introduces the third loss function, the entropy loss can improve the generalization ability of the explosion range prediction model and prevent the overfitting problem of the explosion range prediction model.

[0050] The beneficial effects of the first, second, third, fourth and fifth aspects of this application can all be referred to the analysis of the above beneficial effects, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0052] Figure 1 A schematic diagram of the structure of an explosion range prediction system provided in an embodiment of the present application;

[0053] Figure 2 A schematic diagram of the hardware structure of an explosion range prediction device provided in an embodiment of the present application;

[0054] Figure 3 A flow chart of a method for predicting explosion range provided in an embodiment of the present application;

[0055] Figure 4 A schematic flow chart of another explosion range prediction method provided in an embodiment of the present application;

[0056] Figure 5 A schematic flow chart of another explosion range prediction method provided in an embodiment of the present application;

[0057] Figure 6 A schematic flow chart of another explosion range prediction method provided in an embodiment of the present application;

[0058] Figure 7 A schematic diagram of the structure of an explosion range prediction model provided in an embodiment of the present application;

[0059] Figure 8 A schematic structural diagram of an explosion range prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0061] In the description of this application, it should be understood that the terms "upper," "lower," "left," "right," "front," "back," "inner," "outer," and the like, indicating directions or positional relationships, are based on the directions or relative positional relationships shown in the accompanying drawings and are intended solely to facilitate the description of this application and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific direction, be constructed, or operate in a specific direction. Therefore, they should not be construed as limitations on this application. Unless otherwise specified, the above-mentioned directionality descriptions may be flexibly set in actual application, provided that the relative positional relationships shown in the accompanying drawings are met.

[0062] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0063] In the embodiments of the present application, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, 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, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of other identical elements in the process, article, or device comprising the element.

[0064] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0065] In the description of this specification, specific features, structures, materials or characteristics may be combined in an appropriate manner in any one or more embodiments or examples.

[0066] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0067] Before giving a detailed introduction to the explosion range prediction method of this application, a brief introduction to the application scenario and implementation environment of the design of this application is first given.

[0068] First, a brief introduction to the application scenarios of this application design is given.

[0069] As described in the background, hydrogen energy plays a crucial role in global energy resources, making its storage and transportation crucial. Currently, hydrogen energy is primarily transported via hydrogen pipelines. However, due to the properties of hydrogen, an explosion could cause severe damage, making the safe design and operation of hydrogen pipelines crucial. Predicting the explosion range of a pipeline can provide an important basis for its safe design. Therefore, how to quickly and accurately predict the explosion range of a hydrogen pipeline is a current technical problem that needs to be addressed.

[0070] To address the above issues, embodiments of the present application provide an explosion range prediction method, comprising: first, obtaining data to be predicted from a detection pipeline (the data to be predicted is used to predict the explosion range in the event of an explosion in the detection pipeline); then, inputting the data to be predicted into a pre-trained explosion range prediction model to obtain a prediction result for the detection pipeline's explosion range. The pre-trained explosion range prediction model is trained using a target loss function; the target loss function includes a first loss function for determining the explosion range prediction loss, a second loss function for determining the physical residual loss, and a third loss function for determining the entropy loss.

[0071] As can be seen from the above, firstly, the present application can obtain the explosion range prediction result of the detection pipeline by inputting the data to be predicted into the pre-trained explosion range prediction model, thereby improving the prediction efficiency.

[0072] Secondly, since the above-mentioned explosion range prediction model introduces the first loss function, it can determine the difference between the explosion range prediction result predicted by the explosion range prediction model and the actual explosion range result, thereby improving the prediction accuracy of the explosion range prediction model.

[0073] Then, because the above explosion range prediction model introduces the second loss function, the explosion range prediction results predicted by the explosion range prediction model conform to physical constraints, further improving the prediction accuracy of the explosion range prediction model. Thirdly, because the above explosion range prediction model introduces the third loss function, the entropy loss can improve the generalization ability of the explosion range prediction model and prevent the overfitting problem of the explosion range prediction model.

[0074] The implementation environment of the above-mentioned explosion range prediction method can be the explosion range prediction system provided in the embodiment of the present application.

[0075] Figure 1 FIG. 1 shows a schematic diagram of the structure of an explosion range prediction system provided by an embodiment of the present application. Figure 1As shown, the explosion range prediction system includes: an explosion range prediction device 101 and a data providing device 102.

[0076] The explosion range prediction device 101 and the data providing device 102 are communicatively connected with each other.

[0077] In practical applications, the explosion range prediction device 101 can be connected to any number of data providing devices 102. Figure 1 An example of an explosion range prediction device 101 connected to a data providing device 102 is used for description.

[0078] In the application embodiment, the data providing device 102 is used to provide the detection pipeline's predicted data to the explosion range prediction device 101, so that the explosion range prediction device 101 predicts the explosion range when the detection pipeline explodes based on the detection pipeline's predicted data.

[0079] Optionally, the data providing device 102 may also be a data acquisition device for collecting data on hydrogen, pipelines, soil, explosions, etc. The collected data sources include, but are not limited to, experimental field monitoring data, historical failure database data, and combustion and explosion consequence data obtained from simulations.

[0080] Optionally, the explosion range prediction device 101 may be a server, a terminal, or other types of electronic devices, which is not limited in this embodiment of the present application.

[0081] Optionally, the terminal may be a device that provides voice and / or data connectivity to a user, a handheld device with wireless connection capabilities, or other processing devices connected to a wireless modem. A wireless terminal may communicate with one or more core networks via a radio access network (RAN). A wireless terminal may be a mobile terminal, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal, or a portable, pocket-sized, handheld, computer-built-in, or vehicle-mounted mobile device that exchanges voice and / or data with a radio access network, such as a mobile phone, tablet computer, laptop computer, netbook, or personal digital assistant (PDA).

[0082] Optionally, the above-mentioned server can be a server in a server cluster (consisting of multiple servers), or a chip in the server, or a system on a chip in the server, or can be implemented by a virtual machine (VM) deployed on a physical machine, which is not limited in this embodiment of the present application.

[0083] Optionally, explosion range prediction device 101 and data providing device 102 may be two independent devices or integrated into the same device. When explosion range prediction device 101 and data providing device 102 are integrated into the same device, data providing device 102 may be a storage module (e.g., a database) of explosion range prediction device 101.

[0084] It will be readily understood that when explosion-range prediction device 101 and data providing device 102 are integrated into the same device, the communication between them is carried out between the internal modules of the device. In this case, the communication process between them is the same as when explosion-range prediction device 101 and data providing device 102 are independent.

[0085] For ease of understanding, this application is described by taking the explosion range prediction device 101 and the data providing device 102 as an example in which they are independent of each other.

[0086] The explosion range prediction device 101 in the explosion range prediction system includes: Figure 2 The components included. Figure 2 As an example, the hardware structure of the explosion range prediction device 101 is introduced.

[0087] Figure 2 This is a schematic diagram of the hardware structure of an explosion range prediction device provided in an embodiment of the present application. The explosion range prediction device includes a processor 21, a memory 22, a communication interface 23, and a bus 24. The processor 21, the memory 22, and the communication interface 23 can be connected via the bus 24.

[0088] Processor 21 is the control center of the explosion range prediction device and can be a single processor or a collective term for multiple processing elements. For example, processor 21 can be a general-purpose central processing unit (CPU) or other general-purpose processor. A general-purpose processor can be a microprocessor or any other conventional processor.

[0089] As an embodiment, the processor 21 may include one or more CPUs, such as Figure 2 CPU0 and CPU1 are shown in the figure.

[0090] The memory 22 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0091] In one possible implementation, memory 22 can exist independently of processor 21 and can be connected to processor 21 via bus 24 to store instructions or program code. When processor 21 calls and executes the instructions or program code stored in memory 22, the explosion range prediction method provided in the following embodiments of this application can be implemented.

[0092] In the embodiment of the present application, the explosion range prediction device 101 and the data providing device 102 have different software programs stored in the memory 22, so the explosion range prediction device 101 and the data providing device 102 implement different functions. The functions performed by each device will be described in conjunction with the following flowchart.

[0093] In another possible implementation, the memory 22 may also be integrated with the processor 21 .

[0094] The communication interface 23 is used to connect the explosion range prediction device to other devices via a communication network. The communication network can be Ethernet, a wireless access network, a wireless local area network (WLAN), etc. The communication interface 23 can include a receiving unit for receiving data and a sending unit for sending data.

[0095] The bus 24 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, Figure 2Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0096] It should be pointed out that Figure 2 The structure shown in the figure does not constitute a limitation on the explosion range prediction device, except Figure 2 In addition to the components shown, the explosion range prediction device may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0097] The explosion range prediction method provided in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0098] The explosion range prediction method provided in the embodiment of the present application is applied to Figure 1 The explosion range prediction device 101 in the explosion range prediction system shown is as follows: Figure 3 As shown, the explosion range prediction method provided by the embodiment of the present application includes:

[0099] S301: Obtain data to be predicted from a detection pipeline.

[0100] The data to be predicted is used to predict the explosion range when detecting pipeline explosion.

[0101] In some embodiments, the data to be predicted for the detection pipeline may include: chemical property data of the hydrogen transmitted in the detection pipeline, physical property data of the detection pipeline, mechanical property data of the soil in the area where the detection pipeline is located, and at least one of energy release data generated when the detection pipeline explodes.

[0102] Optionally, the chemical property data of the hydrogen transmitted in the detection pipeline may include: hydrogen combustion heat , the diffusion coefficient of hydrogen in air D H , hydrogen mass concentration C H wait.

[0103] Optionally, the physical characteristic data of the detected pipeline may include: pipeline pressure , pipe diameter , Pipeline burial depth wait.

[0104] Optionally, the mechanical property data of the soil in the area where the detection pipeline is located may include: soil density ρ soil , Poisson's ratio wait.

[0105] Optionally, the energy release data generated when detecting a pipeline explosion may include: temperature And the real results of the explosion range, etc.

[0106] Optionally, the explosion range when detecting a pipeline explosion may be explosion-related range data such as the width of an explosion pit.

[0107] S302: Input the data to be predicted into a pre-trained explosion range prediction model to obtain a prediction result of the explosion range of the detection pipeline.

[0108] Specifically, the pre-trained explosion range prediction model can predict the explosion range prediction result of the detection pipeline based on the to-be-predicted data of the detection pipeline. Therefore, the explosion range prediction result of the detection pipeline can be obtained by inputting the to-be-predicted data into the pre-trained explosion range prediction model.

[0109] The explosion range prediction model is trained using a target loss function. This target loss function includes a first loss function for determining the explosion range prediction loss, a second loss function for determining the physical residual loss, and a third loss function for determining the entropy loss. The explosion range prediction loss represents the difference between the explosion range prediction model's predicted result and the actual explosion range; the physical residual loss ensures that the explosion range prediction model's prediction conforms to physical constraints.

[0110] Optionally, the entropy loss can be obtained by the entropy regularization term. The introduction of the entropy regularization term can improve the generalization ability of the model and prevent overfitting.

[0111] Optionally, the target loss function may be obtained by weighted summation of the first loss function, the second loss function, and the third loss function.

[0112] Alternatively, the physical residual loss may be determined by the physical equations associated with the explosion. Figure 4 As shown in Figure 2, the explosion range prediction model is trained in the following way:

[0113] S401: Obtain a training sample set.

[0114] The training sample set includes multiple training samples and a label for each training sample. The training sample includes training data of the sample pipeline; and the label of the training sample includes the true result of the explosion range of the sample pipeline.

[0115] Specifically, the explosion range prediction device trains the explosion range prediction model, and firstly requires a training sample set required for the explosion range prediction model training as training data.

[0116] Optionally, the training sample set is a set of model training data.

[0117] S402: Based on the training sample set, the original explosion range prediction model is trained to obtain a trained explosion range prediction model.

[0118] In some embodiments, combined Figure 4 ,like Figure 5 As shown, in the above S402, the explosion range prediction device trains the original explosion range prediction model based on the training sample set to obtain the trained explosion range prediction model, which specifically includes:

[0119] S501. Input the training sample set into the explosion range prediction model to obtain the loss value of the current training time, and train the explosion range prediction model according to the loss value of the current training time until the loss value meets the convergence condition to obtain a trained explosion range prediction model.

[0120] The loss value of the current training round is obtained in the following way:

[0121] The training samples are input into the explosion range prediction model of the current training time to obtain the current explosion range prediction result corresponding to the training samples.

[0122] The current explosion range prediction loss is determined based on the current explosion range prediction result, the label of the training sample and the first loss function.

[0123] Determine the current physical residual loss based on the training sample, the physical residual equation and the second loss function.

[0124] Determine the current entropy loss based on the training samples, the entropy regularization term equation and the third loss function.

[0125] The loss value of the current training round is determined based on the current explosion range prediction loss, current physical residual loss, current entropy loss and target loss function.

[0126] Optionally, the current explosion range prediction loss can be weighted by the squared difference between the current explosion range prediction result and the actual explosion range result of the training sample's labeled sample pipeline. The current physical residual loss can be calculated by solving the physical residual equation using the training sample data and then applying the corresponding weights. The current entropy loss is obtained by weighting the result after entropy regularization. The objective loss function is the sum of the current explosion range prediction loss, the current physical residual loss, and the current entropy loss.

[0127] Optionally, the loss value of the current training round is the result of the target loss function of this training.

[0128] Optionally, the physical equations are substituted into the explosion range prediction model, and the residuals are calculated using automatic differentiation techniques to determine the physical residual equations. The detailed determination process can be found in the relevant description of the general technology and will not be repeated here.

[0129] For example, the physical residual equation mainly includes:

[0130] Conservation equation for the total mass of hydrogen-air mixture:

[0131]

[0132] The conservation equation of hydrogen components during combustion is:

[0133]

[0134] Combustion momentum conservation equation:

[0135]

[0136] Energy conservation equation:

[0137]

[0138] Soil motion equation:

[0139]

[0140] Soil constitutive equation:

[0141]

[0142] Among them, the soil constitutive equation is implemented based on the premise of linear elastic model.

[0143] Explosion range result geometric constraint equation:

[0144]

[0145] in, is the density of the hydrogen-air mixture; is the divergence operator; is the velocity vector of the mixture; C H is the mass concentration of hydrogen; D H is the diffusion coefficient of hydrogen in air, is the thermal conductivity coefficient, is the hydrogen combustion mass consumption rate, is the calorific value of hydrogen combustion; is the pressure of the mixture; is the dynamic viscosity of the mixture; is the acceleration due to gravity; For time; C P is the constant pressure specific heat capacity of the mixture; is the temperature of the mixture; ρ soil is the soil density; is the Cauchy tension; F pipe Pipeline impact loads; is the shear modulus; is the strain tensor, , for The transpose of is the soil Lame constant, and Poisson's ratio Related; is the unit tensor; Predict results for explosion range; is the displacement component of the radial coordinate under polar coordinates; is the velocity component of the r axis; is the radial coordinate; 0 means the coordinate of the z-axis is 0; The moment when the explosion process ends.

[0146] In some embodiments, the explosion range prediction device determines the loss value of the current training round based on the current explosion range prediction loss, the current physical residual loss, the current entropy loss, and the target loss function, including:

[0147] The objective loss function satisfies the following formula:

[0148]

[0149] in, is the loss value of the current training time, λ1 is the weight of the first loss function, W is the current explosion range prediction result, W Real is the true result of the explosion range of the sample pipeline, λ2 is the weight of the second loss function, f i is the i-th physical residual equation, λ3 is the weight of the third loss function, is the entropy regularization equation.

[0150] Specifically, the explosion range prediction device determines whether the loss value of the current explosion range prediction model meets the preset conditions by judging the loss value of the current training time. Therefore, the explosion range prediction device needs to determine the loss value of the current training time.

[0151] For example, assuming the threshold is , the current loss value is less than , stop training and get the trained explosion range prediction model.

[0152] Optionally, entropy loss can be added to the loss calculation to improve the generalization ability of the model and prevent overfitting. The entropy loss result is determined by the entropy regularization term, which is expressed as follows:

[0153] Entropy regularization term:

[0154]

[0155] Entropy production rate:

[0156]

[0157] in is the entropy regularization term; N is the number of training samples; i is the i-th training sample; is the entropy production rate; is the dynamic viscosity of the mixture; is the velocity vector of the mixture; for The transpose of is the thermal conductivity coefficient; is temperature; is the gradient operator.

[0158] In some embodiments, combined Figure 5 ,like Figure 6 As shown, in the above S401, the method for the explosion range prediction device to obtain a training sample set specifically includes:

[0159] S601: Obtain original data corresponding to the training data of the sample pipeline.

[0160] In some embodiments, the raw data includes at least one of: chemical property data of hydrogen transmitted in the sample pipeline, physical property data of the sample pipeline, mechanical property data of the soil in the area where the sample pipeline is located, and energy release data generated when the sample pipeline explodes.

[0161] Optionally, the chemical property data of the hydrogen gas transmitted in the sample pipeline may include the heat of combustion of hydrogen gas. , the diffusion coefficient of hydrogen in air D H , hydrogen mass concentration C H wait.

[0162] Optionally, the physical characteristic data of the sample pipeline may include pipeline pressure , pipe diameter , Pipeline burial depth wait.

[0163] Optionally, the mechanical property data of the soil in the area where the sample pipe is located may include the soil density ρ soil , Poisson's ratio wait.

[0164] Optionally, the energy release data generated when the sample pipe explodes may include temperature , the real results of the explosion range, etc.

[0165] S602: Preprocess the original data to obtain data features corresponding to the original data.

[0166] The preprocessing includes at least one of: outlier elimination, standardization, and missing value filling.

[0167] Specifically, before training the explosion range prediction model, the explosion range prediction device needs to preprocess the acquired raw data to adapt the data to the model's requirements. Therefore, preprocessing is necessary. Secondly, data features are the input data for the explosion range prediction model, so the data features corresponding to the raw data need to be obtained.

[0168] Exemplarily, preprocessing the raw data includes at least one of: outlier removal, standardization, and missing value filling.

[0169] Exemplarily, performing outlier removal processing on the original data to obtain data after outlier removal includes:

[0170] The box plot method is used to identify outliers in the original data. The specific identification rules are as follows:

[0171]

[0172] Where x is the original data, Q1 is the upper quartile, Q3 is the lower quartile, and IQR is the interquartile range.

[0173] From the above formula, we can know that when the original data exceeds the upper quartile Q1, the lower quartile Q3 and the abnormal value range determined by the interquartile range IQR (less than the value Q1-1.5 IQR or greater than Q3+1.5 IQR), the original data will be identified as outliers, otherwise, they will be identified as normal values.

[0174] Exemplarily, the data after outliers are removed is normalized, including:

[0175] After removing outliers, select data without missing values. Use the Z-transformation score (Z-score) method to standardize the data. The formula is as follows:

[0176]

[0177] Among them, X is the data after removing outliers, X ＇ is the data after standardization, μ is the mean, and σ is the standard deviation.

[0178] Exemplarily, the missing value filling process is performed on the standardized data, including:

[0179] For data with missing values, the K nearest neighbor algorithm is used to calculate the distance between the sample point with missing values ​​and other sample points in the complete data set. The formula is as follows:

[0180]

[0181] Select the K sample points closest to each other, and then use the missing feature means corresponding to the selected K nearest sample points to fill the missing values ​​in the data. The formula is as follows:

[0182]

[0183] in, is the distance between the sample point with missing values ​​and other sample points in the complete data set, is the value of sample 1 of feature x, is the value of sample 2 of feature x, is the value of sample 1 of feature y, is the value of sample 2 of feature y, is a missing value, is the total number of the most adjacent sample points selected, For the The value of adjacent sample points, i is the i-th most adjacent sample point.

[0184] It can be seen that the explosion range prediction equipment performs missing value filling processing on the standardized data, which can handle the blank values ​​in the data and ensure the integrity of the data.

[0185] Secondly, when constructing the explosion range prediction model, the explosion range prediction equipment can use a priori physical laws to fill the gaps in sparse experimental data areas by embedding explosion-related physical equations into the explosion range prediction model as physical residual equations, thereby reducing dependence on high-cost and high-risk explosion experimental data.

[0186] S603: Determine the degree of correlation between any two variable features in the data features.

[0187] Specifically, there are multiple variable features in the data features. The explosion range prediction device needs to determine whether the variable features should be removed based on the degree of correlation between any two variable features, thereby achieving the screening of data features. Therefore, the explosion range prediction device needs to determine the degree of correlation between the variable features.

[0188] Alternatively, the degree of association between variable features can also be called correlation.

[0189] Optionally, the data features may include: geometric parameters, experimental operating parameters, soil properties near the pipeline, medium properties, and actual results of explosion range prediction, etc.

[0190] Optionally, the geometric parameters may include variable features such as time t, cylindrical coordinates (r, z), pipe diameter D, and pipe burial depth H.

[0191] Optionally, the experimental operating parameters may include variable characteristics such as pipeline pressure P.

[0192] Optionally, soil properties near the pipeline may include soil density ρ soil and Poisson's ratio ν and other variable characteristics.

[0193] Optionally, the media characteristics may include the heat of combustion of hydrogen , the diffusion coefficient of hydrogen in air D H , hydrogen mass concentration C H Equal variable characteristics.

[0194] S604: Screen the data features according to the degree of correlation between any two variable features to determine a training sample set.

[0195] In some embodiments, the explosion range prediction device determines the degree of correlation between any two variable features in the data features, including:

[0196] The degree of correlation between any two variable features satisfies the following formula:

[0197]

[0198] Among them, r s is the degree of association between any two variable features, d i is the rank difference of the i-th rank data value in the arbitrary two variable features, and m is the number of the training samples.

[0199] Optional, The rank difference of the ranked data values ​​represents the difference between the i-th rank values ​​after sorting the eigenvalues ​​of the two variables. The specific sorting process can refer to the relevant description in the general technology and will not be repeated here.

[0200] For example, assuming that any two variable features are variable features and variable characteristics ,but The rank difference of the ranked data values ​​is the variable characteristic The i-th rank value and variable characteristics The difference between the i-th rank values ​​of .

[0201] In some embodiments, the explosion range prediction device screens data features based on the degree of correlation between any two variable features to determine a training sample set, including:

[0202] When the correlation degree between any two variable features is greater than or equal to a first preset threshold, the variable features with smaller variable feature values ​​are eliminated; and when the correlation degree between any two variable features is less than the first preset threshold, any two variable features are retained to obtain a training sample set.

[0203] It should be noted that when the correlation between any two variable features is less than a first preset threshold, specifically, when the correlation between any two variable features is greater than or equal to a second preset threshold and less than the first preset threshold, a prompt message may be output, and manual screening may be performed based on experience; and when the correlation between any two variable features is less than the second preset threshold, the two variable features are retained. The first preset threshold is greater than the second preset threshold.

[0204] Optionally, the above-mentioned explosion range prediction model can be a physics-informed neural network (PINNs), which is an innovative deep neural network architecture that can solve supervised learning tasks. Its core advantage lies in the ability to seamlessly integrate physical prior knowledge into the network structure, thereby imposing strict physical constraints on the network during training. With the help of automatic differentiation technology, PINNs can accurately incorporate partial differential equations and their boundary conditions into the loss function, ensuring that the model output is strictly consistent with physical laws. Compared with traditional numerical methods, the numerical solutions obtained by PINNs are not only more accurate, but also closer to real physical processes. They can better capture the essential characteristics of physical phenomena and have unique advantages in dealing with complex geometries and multi-physics field coupling problems.

[0205] It's important to understand that explosion range prediction is essentially a nonlinear dynamics problem involving a strong coupling of multiple physical fields. It involves the interactions of four dimensions: the chemical properties of hydrogen, the physical characteristics of the pipeline, the mechanical properties of the soil, and the energy released by the explosion. These complex nonlinear mappings make accurate predictions difficult using traditional deterministic mathematical models. Among the common methods for predicting explosion range outcomes, numerical simulation and machine learning are the most commonly used. Numerical simulation methods primarily rely on dynamics software such as flame acceleration simulation (FLACS) and computational fluid dynamics (CFD) to build detailed physical models to simulate the explosion process and thus predict the explosion range. While this approach theoretically offers high accuracy, it requires significant computing resources and time, and places high demands on the accuracy of model parameters. Machine learning methods can automatically learn patterns and regularities from large amounts of historical data by analyzing them, making them particularly advantageous in handling complex nonlinear relationships. However, most machine learning models require large amounts of labeled data. When data is scarce or labeling is costly, model performance can be severely impacted, and key physical mechanisms may be overlooked.

[0206] It can be seen that the explosion range prediction model constructed by the explosion range prediction device can also be called a physical information neural network model.

[0207] First, the explosion range prediction device constructs an explosion range prediction model and trains the explosion range prediction model based on data collected from hydrogen, pipelines, soil, explosions, etc., so as to predict the explosion range results. This solves the problem that mathematical models are difficult to predict explosion range results, and avoids the cumbersome grid processing process of traditional numerical methods in general technologies, thereby improving the efficiency of calculating explosion range prediction results and the accuracy of explosion range prediction results.

[0208] Secondly, the explosion range prediction device embeds the explosion-related physical equations into the explosion range prediction model through automatic differentiation technology, and fills the data gaps through physical regularization terms (such as mass conservation residuals and energy constraint terms), greatly reducing the dependence on experimental data, further improving the accuracy of the explosion range prediction results, and can also achieve efficient solution of complex physical processes, solving the problem of high data dependence and possible neglect of physical mechanisms in machine learning methods in general technologies.

[0209] For example, Figure 7 A schematic diagram of the structure of an explosion range prediction model provided in an embodiment of the present application is shown as follows: Figure 7 As shown in Figure 2, the explosion range prediction model mainly includes:

[0210] First, we choose a multi-layer perceptron (MLP) as the basic architecture to build a fully connected neural network. The fully connected neural network consists of an input layer, a hidden layer, and an output layer.

[0211] The input layer is used to input the training sample set, including several neurons with data such as geometric parameters, working condition parameters, soil properties, hydrogen energy properties, etc. (e.g. Figure 7 in ).

[0212] The hidden layer (also called multi-layer hidden layer) contains weight parameters, bias parameters, and nonlinear activation functions.

[0213] The output layer is used to output the calculation results of the fully connected neural network, that is, the explosion range prediction results.

[0214] Then, the present application can also embed the physical equations related to pipeline explosion into the fully connected neural network through automatic differentiation technology as physical residual equations, thereby constructing the original explosion range prediction model.

[0215] In the process of training the original explosion range prediction model, the data prediction variance can be determined by the explosion range prediction results output by the fully connected neural network, thereby constructing the first loss function of the target loss function.

[0216] Secondly, the variance of the partial differential equation determined by the physical residual equation can be used to construct a second loss function of the target loss function.

[0217] Again, the third loss function of the target loss function can also be constructed through the entropy regularization term equation.

[0218] Subsequently, the target loss function can be constructed through the first loss function, the second loss function and the third loss function.

[0219] Next, we can judge whether the loss value converges based on the loss value output by the objective loss function.

[0220] If so, the trained explosion range prediction model is obtained.

[0221] Optionally, the trained explosion range prediction model may be the optimal model parameters of the trained explosion range prediction model.

[0222] If not, continue training the explosion range prediction model until the loss value meets the convergence condition.

[0223] In some embodiments, the explosion range prediction device can also test the explosion range prediction model and evaluate the performance of the explosion range prediction model to obtain a test result of the explosion range prediction model. In this case, the explosion range prediction device can use the test set to test the trained explosion range prediction model and evaluate the performance of the model.

[0224] Among them, the test set S contains , N test sample data.

[0225] Optionally, the collected original data can be divided into a training set and a test set in a ratio of 7:3.

[0226] The explosion range prediction device can calculate the prediction error on the test set and evaluate the accuracy and generalization ability of the model by comparing the prediction results of the explosion range prediction model with the actual results of the explosion range.

[0227] Optionally, the explosion range prediction device can evaluate the performance of the model through methods such as receiver operating characteristic (ROC) curves and recall rates, and can evaluate the accuracy and generalization ability of the model by calculating the model's precision and accuracy.

[0228] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of method. In order to realize the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily appreciate that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0229] In the embodiments of the present application, the explosion range prediction device can be divided into functional modules based on the above-described method examples. For example, individual functional modules can be divided according to their respective functions, or two or more functions can be integrated into a single processing module. These integrated modules can be implemented as either hardware or software functional modules. The module division in the embodiments of the present application is illustrative and represents only a logical functional division. In actual implementation, other division methods may be employed.

[0230] like Figure 8 FIG. 1 is a schematic diagram of the structure of an explosion range prediction device provided in an embodiment of the present application. Figure 8The explosion range prediction device shown includes: a communication unit 801 and a processing unit 802 .

[0231] The communication unit 801 is used to obtain the data to be predicted from the detection pipeline.

[0232] The data to be predicted is used to predict the explosion range when detecting pipeline explosion.

[0233] The processing unit 802 is used to input the data to be predicted into a pre-trained explosion range prediction model to obtain the explosion range prediction result of the detection pipeline.

[0234] Among them, the explosion range prediction model is trained according to the target loss function; the target loss function includes: a first loss function for determining the explosion range prediction loss, a second loss function for determining the physical residual loss and a third loss function for determining the entropy loss; the explosion range prediction loss is used to represent the difference between the explosion range prediction result predicted by the explosion range prediction model and the actual result of the explosion range; the physical residual loss is used to make the explosion range prediction result predicted by the explosion range prediction model conform to physical constraints.

[0235] Optionally, the explosion range prediction model is trained in the following way:

[0236] The processing unit 802 is further configured to obtain a training sample set.

[0237] The training sample set includes multiple training samples and a label for each training sample. The training sample includes training data of the sample pipeline; and the label of the training sample includes the true result of the explosion range of the sample pipeline.

[0238] The processing unit 802 is further configured to train the original explosion range prediction model based on the training sample set to obtain a trained explosion range prediction model.

[0239] Optionally, the processing unit 802 is specifically configured to:

[0240] Input the training sample set into the explosion range prediction model to obtain the loss value of the current training round, and train the explosion range prediction model based on the loss value of the current training round until the loss value meets the convergence condition to obtain a trained explosion range prediction model; the loss value of the current training round is obtained by the following method:

[0241] The training samples are input into the explosion range prediction model of the current training time to obtain the current explosion range prediction result corresponding to the training samples.

[0242] According to the current explosion range prediction result, the label of the training sample and the first loss function, the current explosion range prediction loss is determined; according to the training sample, the physical residual equation and the second loss function, the current physical residual loss is determined; according to the training sample, the entropy regularization term equation and the third loss function, the current entropy loss is determined; according to the current explosion range prediction loss, the current physical residual loss, the current entropy loss and the target loss function, the loss value of the current training time is determined.

[0243] Optionally, the processing unit 802 is specifically configured to:

[0244] The objective loss function satisfies the following formula:

[0245]

[0246] in, is the loss value of the current training time, λ1 is the weight of the first loss function, W is the current explosion range prediction result, W Real is the true result of the explosion range of the sample pipeline, λ2 is the weight of the second loss function, f i is the i-th physical residual equation, λ3 is the weight of the third loss function, is the entropy regularization equation.

[0247] Optionally, the processing unit 802 is specifically configured to:

[0248] Get the raw data corresponding to the training data of the sample pipeline.

[0249] The raw data is preprocessed to obtain data features corresponding to the raw data; the preprocessing includes: at least one of: outlier removal, standardization, and missing value filling.

[0250] Determine the degree of correlation between any two variable features in the data features.

[0251] According to the degree of correlation between any two variable features, the data features are screened to determine the training sample set.

[0252] Optionally, the processing unit 802 is specifically configured to determine the degree of correlation between any two variable features in the data features, including:

[0253] The degree of correlation between any two variable features satisfies the following formula:

[0254]

[0255] Among them, r s is the degree of association between any two variable features, d i is the rank difference of the i-th rank data value in any two variable features, and m is the number of training samples.

[0256] Optionally, the processing unit 802 is specifically configured to:

[0257] When the correlation degree between any two variable features is greater than or equal to a first preset threshold, the variable features with smaller variable feature values ​​are eliminated; and when the correlation degree between any two variable features is less than the first preset threshold, any two variable features are retained to obtain a training sample set.

[0258] Optionally, the original data includes: chemical property data of hydrogen transmitted in the sample pipeline, physical property data of the sample pipeline, mechanical property data of the soil in the area where the sample pipeline is located, and at least one of energy release data generated when the sample pipeline explodes.

[0259] The data to be predicted includes at least one of: chemical property data of hydrogen transmitted in the detection pipeline, physical property data of the detection pipeline, mechanical property data of soil in the area where the detection pipeline is located, and energy release data generated when the detection pipeline explodes.

[0260] An embodiment of the present application further provides a computer-readable storage medium, which includes computer-executable instructions. When the computer-executable instructions are executed on a computer, the computer executes the explosion range prediction method provided in the above embodiment.

[0261] The present application also provides a computer program product that can be directly loaded into a memory and contains software code. When loaded and executed by a computer, the computer program product can implement the explosion range prediction method provided in the above-described embodiments. Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art will understand that modifications or equivalent substitutions may be made to the technical solutions of the present invention, and such modifications or equivalent substitutions will not cause the modified technical solutions to depart from the spirit and scope of the present invention.

[0262] The system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiment can be combined into one module or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps and are not to be regarded as improper limitations of the present invention.

[0263] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory, internal memory, read-only memory, electrically erasable programmable read-only memory, registers, hard disks, removable disks, compact disc read-only memory (CD-ROM), or any other form of storage medium known in the art. In order to clearly illustrate the interchangeability of electronic hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed by electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

Claims

1. A method for predicting explosion range, characterized in that: include: Obtain the data to be predicted from the detection pipeline; The data to be predicted is used to predict the explosion range when the detection pipeline explodes; The data to be predicted includes: chemical property data of the hydrogen transmitted in the detection pipeline, physical property data of the detection pipeline, mechanical property data of the soil in the area where the detection pipeline is located, and energy release data generated when the detection pipeline explodes; The data to be predicted is input into a pre-trained explosion range prediction model to obtain a prediction result of the explosion range of the detection pipeline; the explosion range prediction model is trained according to a target loss function; the target loss function includes: a first loss function for determining an explosion range prediction loss, a second loss function for determining a physical residual loss, and a third loss function for determining an entropy loss; the explosion range prediction loss is used to represent the difference between the explosion range prediction result predicted by the explosion range prediction model and the actual explosion range result; the physical residual loss is used to ensure that the explosion range prediction result predicted by the explosion range prediction model conforms to physical constraints; The explosion range prediction model is trained in the following way: Obtain a training sample set; the training sample set includes multiple training samples and a label for each training sample; the training sample includes training data of a sample pipeline; the label of the training sample includes a true result of the explosion range of the sample pipeline; The training sample set is input into the explosion range prediction model to obtain a loss value of the current training time, and the explosion range prediction model is trained according to the loss value of the current training time until the loss value meets the convergence condition to obtain the trained explosion range prediction model; the loss value of the current training time is obtained by: Inputting the training sample into the explosion range prediction model of the current training time to obtain the current explosion range prediction result corresponding to the training sample; Determining a current explosion range prediction loss based on the current explosion range prediction result, the label of the training sample, and the first loss function; Determining a current physical residual loss based on the training sample, the physical residual equation, and the second loss function; the physical residual equation includes: a total mass conservation equation for a hydrogen-air mixture, a hydrogen component conservation equation during combustion, a combustion momentum conservation equation, an energy conservation equation, a soil motion equation, a soil constitutive equation, and a geometric constraint equation for explosion range results; Determining a current entropy loss based on the training sample, the entropy regularization term equation, and the third loss function; Determining the loss value of the current training time according to the current explosion range prediction loss, the current physical residual loss, the current entropy loss, and the target loss function includes: The objective loss function satisfies the following formula: in, is the loss value of the current training time, λ1 is the weight of the first loss function, W is the current explosion range prediction result, W Real is the true result of the explosion range of the sample pipeline, λ2 is the weight of the second loss function, f i is the i-th physical residual equation, λ3 is the weight of the third loss function, is the entropy regularization term equation.

2. The method according to claim 1, characterized in that The obtaining of the training sample set includes: Obtaining original data corresponding to the training data of the sample pipeline; Preprocessing the raw data to obtain data features corresponding to the raw data; the preprocessing includes at least one of: outlier removal, standardization, and missing value filling; Determining the degree of correlation between any two variable features in the data features; According to the degree of correlation between any two variable features, the data features are screened to determine the training sample set.

3. The method according to claim 2, characterized in that Determining the degree of correlation between any two variable features in the data features includes: The degree of correlation between any two variable features satisfies the following formula: Among them, r s is the degree of association between any two variable features, d i is the rank difference of the i-th rank data value in the arbitrary two variable features, and m is the number of the training samples.

4. The method according to claim 2, characterized in that The step of screening the data features according to the degree of correlation between any two variable features to determine the training sample set includes: When the degree of correlation between any two variable features is greater than or equal to a first preset threshold, the variable features with smaller variable feature values ​​are eliminated; and when the degree of correlation between any two variable features is less than the first preset threshold, the any two variable features are retained to obtain the training sample set.

5. The method according to claim 2, characterized in that The original data includes at least one of: chemical property data of the hydrogen transmitted in the sample pipeline, physical property data of the sample pipeline, mechanical property data of the soil in the area where the sample pipeline is located, and energy release data generated when the sample pipeline explodes.

6. An explosion range prediction device, characterized in that: include: A processor and a memory; wherein the memory is used to store one or more programs, and the one or more programs include computer-executable instructions. When the explosion range prediction device is running, the processor executes the computer-executable instructions stored in the memory to enable the explosion range prediction device to perform the method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that When the computer-executable instructions stored in the computer-readable storage medium are executed by a processor of an explosion range prediction device, the explosion range prediction device can perform the method according to any one of claims 1 to 5.

Images