Crack water seepage prediction model construction method, prediction model, prediction method, electronic equipment and storage medium

Abstract

The invention discloses a gap water seepage prediction model construction method, a prediction model, a prediction method, electronic equipment and a storage medium. The method comprises the following steps: S01, determining key factors influencing gap water seepage; s02, listing a corresponding gap water seepage characteristic orthogonal test table according to the selected factors and levels; s03, simulation software is adopted to conduct simulation analysis on the seepage flow field of the gaps in the airplane under the raining condition; s04, corresponding simulation parameters are set according to the orthogonal test table in the step S02, and simulation results are recorded as gap outlet flow and flow velocity under each simulation condition; and S05, building an adaptive learning algorithm model, and training the adaptive learning algorithm model by using the simulation data obtained in the step S04 as a training sample to obtain a gap water seepage prediction model. The prediction model constructed by the method is high in practicability, and can be suitable for prediction analysis of gap water seepage conditions of most equipment under the raining condition.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer modeling and simulation, and specifically relates to a method for constructing a crack water seepage prediction model, a prediction model, a prediction method, an electronic device, and a storage medium. Background Art

[0002] At present, whether the gaps in aircraft structures are leaking water is mainly verified by tests, but the tests have problems such as large test scale, high manpower and material costs, long cycle, and low efficiency in solving problems.

[0003] With the rapid development of computing and simulation technologies, simulation has played a vital role in process design and R&D. However, predicting water seepage through structural cracks in aircraft exposed to rain using conventional simulation methods or small models typically takes several hours to a day, while large-scale models often require several days to weeks. This results in inefficient and difficult-to-guarantee predictive analysis. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for constructing a gap water seepage prediction model, a prediction model, a prediction method, an electronic device, and a storage medium to solve the problem of low efficiency and low accuracy of existing methods in predicting and analyzing gap water seepage in aircraft structures.

[0005] The present invention is achieved through the following technical solutions: The method for constructing a crack seepage prediction model includes the following steps: S01. Determine the key factors affecting crack seepage, including crack length, crack width, and wall roughness; S02. Select three identified factors as test factors, and select multiple levels corresponding to each factor based on common operating conditions of aircraft internal gaps, including length level, width level, and roughness level. List the corresponding orthogonal test table for gap water seepage characteristics based on the selected factors and levels. S03. Use simulation software to simulate and analyze the flow field of water seepage through gaps inside the aircraft under rain conditions; S04, setting corresponding simulation parameters according to the orthogonal test table in step S02, and obtaining simulation results under various simulation conditions, including the flow rate and flow velocity at the slit outlet; S05. Build an adaptive learning algorithm model, use the simulation data obtained in step S04 as a training sample to train the adaptive learning algorithm model, and obtain a crack seepage prediction model.

[0006] In some embodiments of the present invention, in step S03, a velocity inlet for rainwater entry is set at the top of the gap, and a pressure outlet is set on the irrelevant right wall and the bottom gap outlet wall to establish a simulation model of the gap seepage flow field.

[0007] In some embodiments of the present invention, in step S04, the influence of various factors on the gap seepage flow field is analyzed according to orthogonal experiments, and the range size is calculated. The influence of various factors on the gap seepage flow field is ranked according to the range, and the construction of the gap seepage prediction model is optimized according to the ranking results.

[0008] In some embodiments of the present invention, the constructed adaptive learning algorithm model includes an input layer L1, a hidden layer L2, and an output layer L3; The input layer has n nodes, the hidden layer has p nodes, and the nodes are connected through the weight matrix W. The weight vector is , where i represents the number of input layer nodes and j represents the number of hidden layer nodes; The output of the n processing units in the input layer is represented as X=(x1, …, x i , …,x n ) T ; The threshold of the hidden layer processing unit is θ j Indicates that for any node j in the hidden layer processing unit, the input weighted sum s j Expressed as: ; Where n is the number of input layer nodes, w ij is the weight from the i-th node in the input layer to the j-th node in the hidden layer, x i is the output value of the i-th node in the input layer; The calculated weighted sum s j Enter a transfer function, the activation function of the hidden layer is a sigmoid function, and determine the output y of the jth node in the hidden layer j for: ; The weight vector between the hidden layer and the output layer is expressed as ,in, v jt It represents the connection weight from the jth node in the hidden layer to the tth node in the output layer. t represents the number of nodes in the output layer. There are q processing units in the output layer, and the activation function of the output layer adopts the Sigmoid function.

[0009] In some embodiments of the present invention, the step of using the simulation data obtained in step S04 as a training sample to train the adaptive learning algorithm model includes: S06. Input a set of samples. The input parameters are the gap length, gap width, and wall roughness. The output parameter is the gap outlet flow rate c. t ; S07, repeat step S06, use the error back propagation algorithm, combined with the expected output vector , calculate the corrected error value of all processing units in the output layer, expressed as: ; Where, l t is the weighted sum of the input to the output layer, is the derivative of the transfer function; According to the corrected error of the output layer, the corrected error value of each unit in the hidden layer is calculated, which is expressed as: ; in, is the derivative of the Sigmoid function; Using learning rate Update the weights connecting the hidden layer and the output layer and the bias of the output layer ; Use another learning rate Update the weights connecting the input layer and the hidden layer and the bias of the hidden layer ; Determine the degree of conformity between the overall error E of the adaptive learning algorithm and the required accuracy, and complete the training of the adaptive learning algorithm model.

[0010] In some embodiments of the present invention, in step S07, it is determined whether E satisfies a condition that it is less than or equal to a precision setting value ε. If so, the training is terminated. If not, the training is repeated to obtain a crack water seepage prediction model.

[0011] On the other hand, the present invention also provides a crack water seepage prediction model, which is constructed using the crack water seepage prediction model construction method.

[0012] On the other hand, the present invention also provides a method for predicting crack water seepage, which uses the crack water seepage prediction model to predict the crack water seepage situation.

[0013] In another aspect, the present invention further provides an electronic device, comprising: processor; and, a memory for storing executable instructions of the processor; The processor is configured to execute the method for predicting water seepage due to cracks by executing the executable instructions.

[0014] On the other hand, the present invention also provides a computer storage medium having a computer program stored thereon, which implements the method for predicting crack water seepage when the computer program is executed by a processor.

[0015] Compared with the prior art, the present invention has the following advantages and beneficial effects: The present invention uses orthogonal experiments combined with range analysis to qualitatively analyze the order of the influence of various factors on gap seepage, thereby improving simulation efficiency. A gap seepage prediction model is established based on an adaptive learning algorithm, and the model is trained with simulation data. The mean square error is used as the loss function, combined with a back propagation algorithm that adjusts its own parameters according to the error of the prediction results, thereby improving the accuracy of the prediction results of the prediction model.

[0016] The prediction model constructed by the present invention can quickly predict the water seepage of the gap based on the parameters of the gap, and uses simulation methods to analyze the impact of various factors on the internal gaps of the aircraft under rain conditions. There is no need to conduct rain tests, which reduces the number of sample tests and the test costs, and shortens the product development cycle.

[0017] The prediction model built based on the adaptive learning algorithm can quickly predict the gap seepage flow rate under given parameters. Compared with manual modeling and simulation, it only needs to input the parameters of the gap during design to quickly output the gap seepage flow rate to determine whether the gap will seep water under rain conditions, which greatly improves the prediction efficiency.

[0018] The prediction model constructed by the present invention is highly practical and can provide important reference for engineering design and water resource management in different fields. It has broad practical application prospects and can be applied to the prediction and analysis of gap seepage in most equipment under rain conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings in the embodiments will be briefly introduced below. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 Schematic diagram of the geometric modeling of the gap in an embodiment of the present invention.

[0021] Figure 2 Schematic diagram of the simulation model of the gap seepage flow field in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. It should also be noted that, for ease of description, only parts related to the present application, not all of the contents, are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0023] In some embodiments of the present invention, a method for constructing a crack water seepage prediction model includes the following steps: S01. Based on known experience and laminar flow calculation formula, determine the key factors affecting gap seepage, including gap length, gap width and wall roughness.

[0024] S02. Select the three factors determined in step S01 as test factors, and select three appropriate levels based on common operating conditions of gaps inside aircraft. List corresponding orthogonal test tables of gap water seepage characteristics based on the three factors and the three levels.

[0025] In orthogonal experimental design, levels refer to the specific parameter values selected for each factor. For example, for an aircraft interior gap, the gap length can be selected from three levels: 10mm, 20mm, and 30mm, representing a typical assembly gap length range; the gap width can be selected from three levels: 0.1mm, 0.5mm, and 1.0mm, representing a typical assembly gap width range; and the wall roughness can be selected from three levels: Ra0.8μm, Ra1.6μm, and Ra3.2μm, corresponding to different surface finish levels.

[0026] S03. Use simulation software to simulate and analyze the flow field of water seepage through gaps inside the aircraft under rain conditions; In order to improve the simulation efficiency and simulation mesh quality, the gap seepage flow field is simplified. The multiphase flow model and laminar flow model are used. A velocity inlet for rainwater entry is set at the top of the gap, and a pressure outlet is set on the irrelevant right wall and the bottom gap outlet wall.

[0027] S04. Setting corresponding simulation parameters according to the orthogonal test table in step S02, and recording the simulation results as the slit outlet flow rate and flow velocity under various simulation conditions.

[0028] According to the orthogonal test, the influence of various factors on the gap seepage flow field is analyzed, the range size is calculated, and the influence of various factors on the gap seepage flow field is ranked in order of priority.

[0029] When constructing a model, the ranking results obtained through range analysis are used to assign weights to hidden layer neurons. The connection weights for the width parameter, which has a greater impact on the range, can be initialized to [-1, 1]. The connection weights for the roughness parameter, which has a smaller impact on the range, can be further narrowed during initialization to improve the efficiency of predictive model construction and the accuracy of predictions. Range analysis can also be applied to engineering optimization. For example, during engineering optimization, the width parameter can be prioritized to an appropriate value to effectively reduce penetration.

[0030] The flow rate value output by the simulation can be used for orthogonal test range analysis, in which the flow value can be calculated through the flow rate and cross-sectional area, and the flow value is used as the final output indicator.

[0031] The flow rate values output by the simulation can be used to verify the model. For example, when the error between the predicted flow rate and the actual simulated flow rate is greater than 10%, the model can be verified through the abnormal distribution of the flow rate field.

[0032] S05. Use a three-layer adaptive learning algorithm model to predict the gap seepage flow field. The adaptive learning algorithm model includes an input layer L1, a hidden layer L2 and an output layer L3. The input layer is provided with n processing units, and there is a weight connection between the input layer and the hidden layer.

[0033] The input layer has n nodes, the hidden layer has p nodes, and the nodes are connected through the weight matrix W. The weight vector is , where i represents the number of input layer nodes and j represents the number of hidden layer nodes.

[0034] The output of the n processing units in the input layer is the column vector of all the processing units in the hidden layer, which is expressed as X=(x1, …, x i , …,x n ) T ; The threshold of the hidden layer processing unit is θ j Indicates that for any node j in the hidden layer processing unit, the input weighted sum s j It can be expressed as: ; Where n is the number of input layer nodes, w ij is the weight from the i-th node in the input layer to the j-th node in the L2 hidden layer, x i is the output value of the i-th node in the input layer.

[0035] The calculated weighted sum s jEnter a transfer function, where the activation function of the hidden layer is selected as a sigmoid function, and determine the output y of the jth node in the hidden layer j for: .

[0036] The weight vector between the hidden layer and the output layer is expressed as , where t represents the number of nodes in the output layer. The output layer has a total of q processing units. v jt Represents the connection weight from the jth node in the hidden layer to the tth node in the output layer, j=1,2,3,4, corresponding to 4 hidden neurons, t=1, corresponding to a single output node; the activation function of the output layer also uses the Sigmoid function.

[0037] S06: Based on the adaptive learning algorithm model built in step S05, input a set of samples, select the three variables of gap length, gap width, and wall roughness as input parameters, and output the gap outlet flow rate c. t .

[0038] S07, select the data in step S04 as the training sample data, repeat step S06, use the error back propagation algorithm, combined with the expected output vector , calculate the corrected error value of all processing units in the output layer, expressed as: ; Where, l t is the weighted sum of the input to the output layer, is the derivative of the transfer function.

[0039] The corrected error value is the difference between the desired output and the actual output multiplied by the derivative of the transfer function.

[0040] According to the corrected error of the output layer, the corrected error value of each unit in the hidden layer is calculated, which is expressed as: ; in, It is the derivative of the Sigmoid function, which is used for chain derivation during error back propagation.

[0041] Using learning rate To update the weights connecting the hidden layer and the output layer and the bias of the output layer ; Use another learning rate Update the weights connecting the input layer and the hidden layer and the bias of the hidden layer .

[0042] The degree of conformity between the overall error E of the adaptive learning algorithm and the required accuracy is determined, that is, E is less than or equal to ε. If this condition is met, the training ends; if not, the training is repeated to obtain a crack seepage prediction model.

[0043] On the other hand, some embodiments of the present invention provide a crack water seepage prediction model, which is constructed using the above-mentioned crack water seepage prediction model construction method.

[0044] This gap water seepage prediction model can be applied to the prediction and analysis of gap water seepage in most equipment under rain conditions.

[0045] On the other hand, some embodiments of the present invention provide a method for predicting crack water seepage, which uses the above-mentioned crack water seepage prediction model to predict the crack water seepage situation.

[0046] The gap seepage prediction model obtained by the above training is used to predict the gap seepage flow rate. Inputting the design parameters of the gap into the gap seepage prediction model can quickly output the gap seepage flow rate, so as to determine whether the gap is leaking under rain conditions, which greatly improves the efficiency of prediction analysis.

[0047] On the other hand, some embodiments of the present invention provide an electronic device, including: processor; and, a memory for storing executable instructions of the processor; The processor is configured to execute the method for predicting water seepage through gaps in the above embodiment by executing executable instructions.

[0048] On the other hand, some embodiments of the present invention are a computer storage medium having a computer program stored thereon, which implements the method for predicting crack water seepage in the above embodiment when the computer program is executed by a processor.

[0049] The construction of the crack water seepage prediction model of the present invention and the method for predicting crack water seepage based on the constructed prediction model are described in detail below with reference to specific embodiments.

[0050] Reference Figure 1 , which is a schematic diagram of the geometric modeling of the gap between two parts inside the aircraft, where h is the gap length and d is the gap width. The following describes the method for constructing the gap water seepage prediction model in combination with the schematic diagram of the geometric modeling of the gap.

[0051] S01. In one-dimensional longitudinal flow, the mass conservation equation for gap seepage can be simplified to: ; in, is the liquid density, A is the area of the flow cross section, v is the flow velocity, y is the vertical coordinate, and t is time.

[0052] The momentum conservation equation can be written as: ; Where μ is the liquid dynamic viscosity, v is the flow velocity, g is the acceleration due to gravity, is the wall shear stress.

[0053] According to the above formula, the factors affecting gap seepage in a rainy environment are determined, including gap length, gap width, wall roughness, etc.

[0054] S02. Select the three factors in S01 (gap length, gap width, wall roughness) as test factors, and refer to the common working conditions of internal gaps in aircraft to select 3 appropriate levels respectively, and list the factor level table of the orthogonal test of gap water seepage characteristics, a total of 3 3 =27 groups of experiments.

[0055] S03. Use simulation software to simulate and analyze the flow field of water seepage in the gaps inside the aircraft under rain conditions. In order to improve the simulation efficiency and simulation mesh quality, the flow field of water seepage in the gaps is simplified. Using the multiphase flow model and the laminar flow model, a velocity inlet for rainwater entry is set at the top of the gap, and pressure outlets are set on the irrelevant right wall and the bottom wall of the gap outlet. The simulation model diagram of the water seepage flow field in the gap is established as shown below. Figure 2 shown.

[0056] S04. Setting corresponding simulation parameters according to the orthogonal test table in step S02, and recording the simulation results as the slit outlet flow rate and flow velocity under various simulation conditions.

[0057] According to the orthogonal test, the influence of various factors on the gap seepage flow field is analyzed, the range size is calculated, and the influence of various factors on the gap seepage flow field is ranked in order of priority.

[0058] The steps for performing range analysis on orthogonal test results include: S041. Calculate the total level of each parameter, denoted by T. The jth column shows the total level of the parameter i, denoted by T ji express, T ji The value of can be expressed as: ; It represents the i-th level Result y i sum; Indicates the total number of times the j-th column parameter has the same level.

[0059] S042. Find the average value of the test index corresponding to each level in the jth column , expressed as: ; S043. Calculate the range D j , which is equal to the maximum minus the minimum value among the average values of the test indicators corresponding to each level in the jth column. The calculation formula is as follows: .

[0060] In this embodiment, the influence of factors on the crack seepage flow field is ranked by priority through range analysis, resulting in the following: width range > length range > roughness range. The construction of the crack seepage prediction model is optimized based on the ranking results. For example, when constructing the model, when assigning weights to hidden layer neurons, the connection weights of the width parameter, which has a greater impact on the range, are initialized to [-1, 1], while the connection weights of the roughness parameter, which has a smaller impact on the range, can be further narrowed during initialization. Furthermore, during engineering optimization, engineering parameters can be prioritized based on the ranking results. For example, prioritizing the width to an appropriate value can effectively reduce leakage.

[0061] S05, select a three-layer adaptive learning algorithm model to predict the gap seepage flow field; wherein, the input layer L1 is set with 3 processing units, and the input pattern vector is , corresponding to the three features of the input data (i.e., the relevant parameters of gap seepage), the three nodes of the input layer strictly correspond to: x1: length (normalized to the interval [0,1]), x2: width (normalized to the interval [0,1]), x3: roughness (normalized to the interval [0,1]); where m is the number of learning pattern pairs; The expected output vector corresponding to the input pattern is ,The output layer has only one neuron, which is used to predict the flow rate of gap seepage; The hidden layer L2 sets 4 processing units, and the net input vector of the hidden layer is , the output vector is ; The net input vector in the output layer is , the actual output vector is ; The weight vector between the input layer and the hidden layer is , the weight vector between the hidden layer and the output layer is , the threshold of all units in the hidden layer is , the unit threshold in the output layer is , where i=1,2,3, j=1,2,3,4, t=1; assign connection weights W, V and thresholds 、 Any value between the interval [-1, 1]; v jt Represents the connection weight from the jth node in the hidden layer to the tth node in the output layer, j=1,2,3,4, corresponding to 4 hidden neurons, t=1, corresponding to a single output node.

[0062] The weighted sum of the inputs obtained by the hidden layer processing units is as follows: ; Where j=1,2,3,4.

[0063] The activation function of the hidden layer selects the Sigmoid function, and the output form of the hidden layer processing unit is expressed as: ; The activation function of the output layer adopts the Sigmoid function.

[0064] S06, corresponding to the construction framework of S05, select the data of the orthogonal test as the training sample, and transform the sample pair (X k , Y k ) is assigned to the adaptive learning algorithm, the input training samples are transmitted to the hidden layer units, and the net input vector and output vector of all processing units in the hidden layer are obtained according to the following formula: ; ; Where j=1,2,3,4; is the output vector of the jth unit in the hidden layer; is the net input vector of the jth unit in the hidden layer. The superscript k represents the number of the training sample. The input sample pair (X k , Y k ), X k is the kth group of gap parameters (length, width, roughness), is the normalized value of each group of input in the orthogonal test, and Y k is the flow rate value obtained by simulation, and is the normalized value of the flow rate value obtained by simulation.

[0065] The net input vector and actual output vector of all processing units in the output layer are obtained according to the following formula: ; ; Where t=1; is the net input vector of the output layer, is the actual output vector of the output layer.

[0066] Combined with the expected output vector , calculate the corrected error values of all processing units in the output layer: ; The corrected error values of all processing units in the hidden layer are calculated according to the following formula: ; Change the weight V between the hidden layer and the output layer and the threshold of each neuron in the output layer according to the following formula , expressed as: ; ; Where, is the learning rate value, 0< <1.

[0067] v jt is the connection weight from the jth node in the hidden layer to the tth node in the output layer, forward propagation , update the weights; similarly, according to Threshold value for output layer to update.

[0068] The weight W connecting the input layer and the hidden layer and the threshold value of the hidden layer neurons are calculated according to the following formula: , expressed as: ; ; Where β is the learning rate value, 0<β<1.

[0069] is the connection weight from the i-th node in the input layer to the j-th node in the hidden layer, forward propagation , update the weights; similarly, according to Hidden layer threshold to update.

[0070] S07: Select another set of training sample data and assign it to the adaptive learning algorithm, and repeat step S06 until the training of all training sample data is completed.

[0071] Calculate the degree of conformity between the overall error E of the adaptive learning algorithm and the required accuracy, that is, E is less than or equal to ε; if it is satisfied, end the training; if not, continue to repeat the training.

[0072] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A method for constructing a crack seepage prediction model, characterized in that: The following steps are involved: S01. Determine the key factors affecting crack seepage, including crack length, crack width, and wall roughness; S02. Select three identified factors as test factors, and select multiple levels corresponding to each factor based on common operating conditions of aircraft internal gaps, including length level, width level, and roughness level. List the corresponding orthogonal test table for gap water seepage characteristics based on the selected factors and levels. S03. Use simulation software to simulate and analyze the flow field of water seepage through gaps inside the aircraft under rain conditions; S04, setting corresponding simulation parameters according to the orthogonal test table in step S02, and obtaining simulation results under various simulation conditions, including the flow rate and flow velocity at the slit outlet; S05. Build an adaptive learning algorithm model, use the simulation data obtained in step S04 as a training sample to train the adaptive learning algorithm model, and obtain a crack seepage prediction model.

2. The method for constructing a crack water seepage prediction model according to claim 1, wherein: In step S03, a velocity inlet for rainwater to enter is set at the top of the gap, and a pressure outlet is set on the irrelevant right wall and the bottom gap outlet wall to establish a simulation model of the gap seepage flow field.

3. The method for constructing a crack seepage prediction model according to claim 1, wherein: In step S04, the influence of various factors on the gap seepage flow field is analyzed according to the orthogonal test, and the range size is calculated. The influence of various factors on the gap seepage flow field is sorted according to the range, and the construction of the gap seepage prediction model is optimized according to the sorting results.

4. The method for constructing a crack seepage prediction model according to claim 1, wherein: The adaptive learning algorithm model constructed includes input layer L1, hidden layer L2 and output layer L3; The input layer has n nodes, the hidden layer has p nodes, and the nodes are connected through the weight matrix W. The weight vector is , where i represents the number of input layer nodes and j represents the number of hidden layer nodes; The output of the n processing units in the input layer is represented as X=(x1, …, x i , …,x n ) T ; The threshold of the hidden layer processing unit is θ j Indicates that for any node j in the hidden layer processing unit, the input weighted sum s j Expressed as: ； Where n is the number of input layer nodes, w ij is the weight from the i-th node in the input layer to the j-th node in the hidden layer, x i is the output value of the i-th node in the input layer; The calculated weighted sum s j Enter a transfer function, the activation function of the hidden layer is a sigmoid function, and determine the output y of the jth node in the hidden layer j for: ； The weight vector between the hidden layer and the output layer is expressed as ,in, v jt It represents the connection weight from the jth node in the hidden layer to the tth node in the output layer. t represents the number of nodes in the output layer. There are q processing units in the output layer, and the activation function of the output layer adopts the Sigmoid function.

5. The method for constructing a crack water seepage prediction model according to claim 4, wherein: The steps of using the simulation data obtained in step S04 as training samples to train the adaptive learning algorithm model include: S06. Input a set of samples. The input parameters are the gap length, gap width, and wall roughness. The output parameter is the gap outlet flow rate c. t ; S07, repeat step S06, use the error back propagation algorithm, combined with the expected output vector , calculate the corrected error value of all processing units in the output layer, expressed as: ； Where, l t is the weighted sum of the input to the output layer, is the derivative of the transfer function; According to the corrected error of the output layer, the corrected error value of each unit in the hidden layer is calculated, which is expressed as: ； in, is the derivative of the Sigmoid function; Using learning rate Update the weights connecting the hidden layer and the output layer and the bias of the output layer ; Use another learning rate Update the weights connecting the input layer and the hidden layer and the bias of the hidden layer ; Determine the degree of conformity between the overall error E of the adaptive learning algorithm and the required accuracy, and complete the training of the adaptive learning algorithm model.

6. The method for constructing a crack water seepage prediction model according to claim 5, wherein: In step S07, it is determined whether E satisfies the requirement of being less than or equal to the accuracy setting value ε. If so, the training is terminated. If not, the training is repeated to obtain a crack water seepage prediction model.

7. The crack seepage prediction model is characterized by: The method for constructing a crack water seepage prediction model according to any one of claims 1 to 6 is used to construct the crack water seepage prediction model.

8. A method for predicting crack water seepage, characterized in that: The crack water seepage prediction model described in claim 7 is used to predict the crack water seepage situation.

9. An electronic device, characterized in that include: processor; as well as, a memory for storing executable instructions of the processor; The processor is configured to execute the method for predicting crack water seepage according to claim 8 by executing the executable instructions.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting crack water seepage according to claim 8 is implemented.

Images