Temperature field and pressure field prediction method and system for small turbojet engine gas compressor
By using a neural network model based on gated cyclic units in the compressor of a small turbojet engine, the problems of gradient vanishing and cumulative error in unsteady flow fields were solved, achieving higher accuracy in temperature and pressure field prediction and improving prediction efficiency.
Patent Information
- Application Number
- CN202510439666.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-09
- Publication Date
- 2026-01-23
AI Technical Summary
Existing technologies suffer from gradient vanishing and cumulative error problems in the unsteady flow field of small turbojet engine compressors, leading to inaccurate predictions of temperature and pressure fields.
A neural network model based on gated recurrent units is adopted, combining convolutional layers and fully connected layers. Through data processing of training and test sets, a prediction model for temperature and pressure fields is established. Mean square error is used to optimize the training. A neural network structure is built using multi-layer gated recurrent units to learn information at key time points and improve prediction accuracy.
It improves the prediction accuracy of temperature and pressure fields, reduces prediction errors over longer time spans, and increases the efficiency of prediction work.
Smart Images
Figure CN121389695A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of small turbojet engine technology, specifically relating to a method and system for predicting the temperature and pressure fields of a small turbojet engine compressor. Background Technology
[0002] Small turbojet engines are engines that use high-speed jet propulsion to generate thrust and propel aircraft. The research, design, and optimization of the compressor are the primary focus of their development. The compressor's main function is to increase the total pressure of the incoming air in the combustion chamber, thereby improving the efficiency of the engine's thermodynamic cycle. With increasingly demanding engine performance requirements, the design load on the compressor is constantly increasing, the flow complexity is growing, and flow stability issues are becoming increasingly prominent. Therefore, in the practical engineering stages of compressor design and evaluation, it is necessary to predict the temperature and pressure fields of the compressor.
[0003] Because compressors involve complex airflow and dynamic load changes during actual operation, their flow fields often exhibit unsteady characteristics. Recurrent Neural Networks (RNNs) excel at processing time-series data, capturing temporal correlations from signals, and can be used to predict unsteady flow fields. Gated Recurrent Units (GRUs), a variant of RNNs, alleviate the gradient vanishing and exploding problems inherent in classic RNNs. Compared to Long Short-Term Memory (LSTM), it requires fewer parameters and is easier to train, while achieving similar results. Therefore, selecting a novel neural network model integrating GRUs, convolutional layers, and fully connected layers to predict the temperature and pressure fields within the compressor flow field can effectively address the shortcomings of traditional prediction methods. Summary of the Invention
[0004] This invention provides a method and system for predicting the temperature and pressure fields of a small turbojet engine compressor, in order to solve the problems of gradient vanishing and long prediction time due to accumulated errors in the application of neural networks in unsteady flow fields.
[0005] This invention is achieved through the following technical solution:
[0006] A method for predicting the temperature and pressure fields of a small turbojet engine compressor, the method comprising:
[0007] S1: Obtain data for different operating conditions and corresponding temperature and pressure fields, match the obtained time series with the temperature and pressure field data, and then divide the data into training and test sets;
[0008] S2: Establish a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0009] S3: Train a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0010] S4: Test the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0011] S5: Enables prediction of temperature and pressure fields in the compressor of a small turbojet engine.
[0012] Furthermore, S1 specifically involves normalizing the original dataset using the following formula:
[0013]
[0014] Where, x * For normalized data; x is the measured data of a certain parameter; x max The maximum value of this parameter term, x min This is the minimum value of this parameter.
[0015] Furthermore, S2 specifically refers to the prediction model comprising an input layer, a convolutional layer, a GRU layer, a transposed convolutional layer, a fully connected layer, and an output layer;
[0016] First, the training data of temperature field and stress field are normalized. The processed training data is then input into convolutional layer, GRU layer and transposed convolutional layer in sequence. The output data of transposed convolutional layer is then input into fully connected layer. After information is integrated by fully connected layer, predicted value with normalized attribute is output. Finally, the actual predicted value is obtained by inverse normalization.
[0017] Furthermore, the calculation method of the gated loop unit is specifically as follows: reset gate r t The calculation formula is:
[0018] r t =σ(W r ·[h t-1 ,x t ]+b r )
[0019] Where σ is the sigmoid function, and W r To reset the weight matrix corresponding to the gate, x t h is the input vector at the current time step. t-1 b is the hidden state from the previous time step. r To reset the gate offset vector;
[0020] Update Gate Z t The calculation formula is:
[0021] z t =σ(W z ·[h t-1 ,x t ]+b z )
[0022] W z b represents the weight matrix corresponding to the update gate. z Update the gate offset vector;
[0023] Reset gate r t The calculation formula is:
[0024]
[0025] Where tanh represents the tanh activation function, W h Let b represent the weight matrix of the candidate hidden layer. h It is the offset vector;
[0026] From the hidden state h of the previous time step t-1 Candidate hidden states at the current time step h was calculated t Calculation formula:
[0027]
[0028] h t This is the output of the current time step, which is then input into the fully connected layer.
[0029] Furthermore, the normalized data values need to be denormalized when obtaining predicted values after training. The formula for denormalization is shown below:
[0030] x real =(x max -x min )·x p +x min
[0031] Where, x real The predicted value is the actual dimensional data after inverse normalization, x p For the normalized evaluation value, x min and x max This indicates the maximum and minimum values of this parameter.
[0032] Furthermore, S3 specifically refers to,
[0033] S3.1: Based on the size and structure of the training set, initially set the model training parameters such as the number of training iterations, learning rate, and required training accuracy.
[0034] The mean squared error is chosen as the loss function and is defined as follows:
[0035]
[0036] Where m is the number of training samples, y(t) is the actual value, and y i (t) represents the predicted value;
[0037] S3.2: Use the training set as input data to train a neural network prediction model based on gated recurrent units for temperature and pressure fields.
[0038] Furthermore, S4 specifically refers to,
[0039] S4.1: Verify the model's performance by using the test set as input data and inputting it into the model to obtain prediction results;
[0040] S4.2: Compare the model's predictions with the acquired temperature and pressure field data. The model's performance is evaluated using four metrics: root mean square error, mean absolute error, mean square error, and coefficient of determination. The specific formulas are as follows:
[0041]
[0042]
[0043] in, and y i These are the predicted value and the actual value, respectively. is the average of the true values; m is the total number of test samples; if the index meets the accuracy requirements, the training ends and the final prediction model is obtained; otherwise, the parameters of the prediction model are adjusted and retraining is performed.
[0044] A temperature and pressure field prediction system for a small turbojet engine compressor, the system employing the temperature and pressure field prediction method for a small turbojet engine compressor as described above, the system comprising:
[0045] Data acquisition module: Acquires data under different operating conditions and corresponding temperature and pressure fields, maps the obtained time series data to the temperature and pressure field data, and then divides the data into training and test sets;
[0046] Prediction model building module: Builds a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0047] Model training module: Trains a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0048] Model performance testing module: Tests the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0049] To enable the prediction of temperature and pressure fields of the compressor of a small turbojet engine.
[0050] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method described above.
[0051] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described above.
[0052] The beneficial effects of this invention are:
[0053] This invention constructs a neural network structure through multi-layer gated recurrent units, learns information at key time points, improves the prediction effect over a longer time span, avoids the prediction error of long-term explanatory variables due to information transmission, and thus improves the prediction accuracy of temperature and pressure fields.
[0054] This invention employs a neural network prediction model based on gated cyclic units for temperature and pressure fields, enabling rapid acquisition of compressor temperature and pressure field data, thus significantly improving the efficiency of temperature and pressure pulsation prediction. Attached Figure Description
[0055] Figure 1 This is a flowchart of the method of the present invention.
[0056] Figure 2 This is a schematic diagram of the prediction model of the present invention.
[0057] Figure 3 This is a schematic diagram of the gated loop unit of the present invention. Detailed Implementation
[0058] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.
[0059] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.
[0060] It should also be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.
[0061] The following is in conjunction with the appendix to this application specification. Figure 1-3 The technical solutions in the embodiments of this application are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0062] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.
[0063] Implementation Method 1
[0064] This embodiment provides a method for predicting the temperature and pressure fields of a small turbojet engine compressor, such as... Figure 1-2 As shown, the method includes:
[0065] S1: Obtain data for different operating conditions and corresponding temperature and pressure fields, match the obtained time series with the temperature and pressure field data, and then divide the data into training and test sets;
[0066] S2: Establish a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0067] S3: Train a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0068] S4: Test the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0069] S5: Enables prediction of temperature and pressure fields in the compressor of a small turbojet engine.
[0070] Furthermore, S1 specifically involves the fact that the parameters in the original dataset have different measurement scales, thus requiring normalization to limit the preprocessed data to a certain range. This reduces the adverse effects of outlier data, facilitating model training. Normalization also reduces the computational cost of the model and improves convergence speed. The calculation formula is as follows:
[0071]
[0072] In the formula: x * For normalized data; x is the measured data of a certain parameter; x max The maximum value of this parameter term, x min This is the minimum value of this parameter.
[0073] Furthermore, S2 specifically refers to the neural network prediction model based on gated recurrent units for temperature and pressure fields, which includes an input layer, a convolutional layer, a GRU layer, a transposed convolutional layer, a fully connected layer, and an output layer.
[0074] First, the training data of temperature field and stress field are normalized. The processed training data is then input into convolutional layer, GRU layer and transposed convolutional layer in sequence. The output data of transposed convolutional layer is then input into fully connected layer. After information is integrated by fully connected layer, predicted value with normalized attribute is output. Finally, the actual predicted value is obtained by inverse normalization.
[0075] First, the training data is normalized and then input into a convolutional layer to extract features. Next, it is input into a GRU layer, which is a neural network structure built from multiple gated recurrent units. The data is trained within the GRU network to learn information at key time points, improving prediction performance over longer time spans. The output of the GRU layer is then re-inputted into a transposed convolutional layer, and its output is concatenated and fused with a fully connected layer. The predicted values from all gated recurrent units are used as input to the fully connected layer. After information synthesis by the fully connected layer, a predicted value with normalized properties is output. Finally, the actual predicted value is obtained through inverse normalization.
[0076] Furthermore, such as Figure 3 As shown, the specific calculation method of the gated loop unit is as follows: reset gate r t The formula for calculating how much information from the previous time step can be passed to the candidate states of the current time step is:
[0077] r t =σ(W r ·[h t-1 ,x t ]+b r )
[0078] Where σ is the sigmoid function, and W r To reset the weight matrix corresponding to the gate, x t h is the input vector at the current time step. t-1 b is the hidden state from the previous time step. r To reset the gate offset vector;
[0079] Update Gate Z t Its function is to filter out some information from the hidden state of the previous time step, retaining useful information. The calculation formula is:
[0080] z t =σ(W z ·[h t-1 ,x t ]+b z )
[0081] W z b represents the weight matrix corresponding to the update gate. z Update the gate offset vector;
[0082] Reset gate r t The hidden state h of the previous time step t-1 Corresponding elements are multiplied to select whether to preserve the hidden state h. t-1 And input x at the current time step t Candidate hidden states for the current time step are generated using the tanh activation function. The calculation formula is:
[0083]
[0084] Where tanh represents the tanh activation function, W h Let b represent the weight matrix of the candidate hidden layer. h It is the offset vector;
[0085] From the hidden state h of the previous time step t-1 Candidate hidden states at the current time step h was calculated t Calculation formula:
[0086]
[0087] h t This is the output of the current time step, which is then input into the fully connected layer.
[0088] Furthermore, the normalized data values need to be denormalized when obtaining predicted values after training. The formula for denormalization is shown below:
[0089] x real =(x max-x min )·x p +x min
[0090] Where, x real The predicted value is the actual dimensional data after inverse normalization, x p For the normalized evaluation value, x min and x max This indicates the maximum and minimum values of this parameter.
[0091] Furthermore, S3 specifically refers to,
[0092] S3.1: Based on the size and structure of the training set, initially set the model training parameters such as the number of training iterations, learning rate, and required training accuracy.
[0093] The mean-squared error (MSE) is chosen as the loss function and is defined as follows:
[0094]
[0095] Where m is the number of training samples, y(t) is the actual value, and y i (t) represents the predicted value. The expected value of the squared difference between elements is calculated and can be used as a measure of the accuracy of a single output matrix. The optimization algorithm uses the Adam optimizer.
[0096] S3.2: Use the training set as input data to train a neural network prediction model based on gated recurrent units for temperature and pressure fields.
[0097] Furthermore, S4 specifically refers to,
[0098] S4.1: Verify the model's performance by using the test set as input data and inputting it into the model to obtain prediction results;
[0099] S4.2: Compare the model predictions with the acquired corresponding temperature and pressure field data, using the root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), and coefficient of determination (R-squared, R0). 2 The model's performance is evaluated using four metrics, as shown in the following formula:
[0100]
[0101] in, and y i These are the predicted value and the actual value, respectively. is the average of the true values; m is the total number of test samples; the closer the root mean square error, mean absolute error, and mean square error are to 0, and the closer the coefficient of determination is to 1, the better the model's performance. If the metrics meet the accuracy requirements, training ends, and the final prediction model is obtained; otherwise, the parameters of the prediction model are adjusted, and training is repeated.
[0102] Implementation Method 2
[0103] This embodiment provides a temperature and pressure field prediction system for a small turbojet engine compressor. The system uses the temperature and pressure field prediction method for a small turbojet engine compressor as described in Embodiment 1. The system includes:
[0104] Data acquisition module: Acquires data under different operating conditions and corresponding temperature and pressure fields, maps the obtained time series data to the temperature and pressure field data, and then divides the data into training and test sets;
[0105] Prediction model building module: Builds a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0106] Model training module: Trains a neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0107] Model performance testing module: Tests the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields;
[0108] To enable the prediction of temperature and pressure fields of the compressor of a small turbojet engine.
[0109] Implementation Method 3
[0110] This invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor. The memory stores software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory and processor are connected via a bus. Specifically, the processor implements any step in Embodiment 1 by running the computer program stored in the memory.
[0111] It should be understood that, in the embodiments of the present invention, the processor may be a Central Processing Unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.
[0112] Memory may include read-only memory, flash memory, and random access memory, and provides instructions and data to the processor. Some or all of the memory may also include non-volatile random access memory.
[0113] As can be seen from the above, the electronic device provided by the embodiments of the present invention can implement the self-supervised group behavior recognition method as described in Embodiment 1 by running a computer program, resulting in a new end-to-end framework called a group feature self-supervised model, which is used for group behavior representation learning. By capturing contextual relationships through context-aware relation prediction encoding, it not only considers spatial interaction contextual relationships but also overall scene temporal change contextual relationships, thereby obtaining a more comprehensive representation of group behavior features.
[0114] It should be understood that if the integrated modules / units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods described above can also be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.
[0115] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the above device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0117] It should be noted that the methods and detailed examples provided in the above embodiments can be incorporated into the apparatus and devices provided in the embodiments for mutual reference, and will not be repeated here.
[0118] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0119] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / device embodiments described above are merely illustrative. For instance, the division of modules or units described above is merely a logical functional division, and in actual implementation, it can be divided in other ways. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.
[0120] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for predicting the temperature and pressure fields of a small turbojet engine compressor, characterized in that, The method includes: S1: Obtain data for different operating conditions and corresponding temperature and pressure fields, match the obtained time series with the temperature and pressure field data, and then divide the data into training and test sets; S2: Establish a neural network prediction model based on gated recurrent units for temperature and pressure fields; S3: Train a neural network prediction model based on gated recurrent units for temperature and pressure fields; S4: Test the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields; S5: Enables prediction of temperature and pressure fields in the compressor of a small turbojet engine.
2. The prediction method according to claim 1, characterized in that, Specifically, S1 involves normalizing the original dataset using the following formula: Where, x * For normalized data; x is the measured data of a certain parameter; x max The maximum value of this parameter term, x min This is the minimum value of this parameter.
3. The prediction method according to claim 1, characterized in that, Specifically, S2 refers to the prediction model comprising an input layer, a convolutional layer, a GRU layer, a transposed convolutional layer, a fully connected layer, and an output layer. First, the training data of the temperature field and stress field are normalized. The processed training data is then input into the convolutional layer, the GRU layer, and the transposed convolutional layer in sequence. The output data of the transposed convolutional layer is then input into the fully connected layer. After the information is synthesized by the fully connected layer, the predicted value with normalized attributes is output. Finally, the actual predicted value is obtained through inverse normalization.
4. The prediction method according to claim 3, characterized in that, The specific calculation method of the gated loop unit is as follows: reset gate r t The calculation formula is: r t =σ(W r ·[h t-1 ,x t ]+b r ) Where σ is the sigmoid function, and W r To reset the weight matrix corresponding to the gate, x t h is the input vector at the current time step. t-1 b is the hidden state from the previous time step. r To reset the gate offset vector; Update Gate Z t The calculation formula is: z t =σ(W z ·[h t-1 ,x t ]+b z ) W z b represents the weight matrix corresponding to the update gate. z Update the gate offset vector; Reset gate r t The calculation formula is: Where tanh represents the tanh activation function, W h Let b represent the weight matrix of the candidate hidden layer. h It is the offset vector; From the hidden state h of the previous time step t-1 Candidate hidden states at the current time step h was calculated t Calculation formula: h t This is the output of the current time step, which is then input into the fully connected layer.
5. The prediction method according to claim 4, characterized in that, The normalized data values still need to be denormalized when obtaining prediction values after training. The formula for denormalization is as follows: x real =(x max -x min )·x p +x min Where, x real The predicted value is the actual dimensional data after inverse normalization, x p For the normalized evaluation value, x min and x max This indicates the maximum and minimum values of this parameter.
6. The prediction method according to claim 2, characterized in that, Specifically, S3 is... S3.1: Based on the size and structure of the training set, initially set the model training parameters such as the number of training iterations, learning rate, and required training accuracy; The mean squared error is chosen as the loss function and is defined as follows: Where m is the number of training samples, y(t) is the actual value, and y i (t) represents the predicted value; S3.2: Use the training set as input data to train a neural network prediction model based on gated recurrent units for temperature and pressure fields.
7. The prediction method according to claim 2, characterized in that, Specifically, S4 is... S4.1: Verify the model's performance by using the test set as input data and inputting it into the model to obtain prediction results; S4.2: Compare the model's predictions with the acquired temperature and pressure field data. The model's performance is evaluated using four metrics: root mean square error, mean absolute error, mean square error, and coefficient of determination. The specific formulas are as follows: in, and y i These are the predicted value and the actual value, respectively. is the average of the true values; m is the total number of test samples; if the index meets the accuracy requirements, the training ends and the final prediction model is obtained; otherwise, the parameters of the prediction model are adjusted and retraining is performed.
8. A system for predicting the temperature and pressure fields of a small turbojet engine compressor, characterized in that, The system uses the temperature and pressure field prediction method for small turbojet engine compressors as described in any one of claims 1-7, and the system includes: Data acquisition module: Acquires data under different operating conditions and corresponding temperature and pressure fields, maps the obtained time series data to the temperature and pressure field data, and then divides the data into training and test sets; Prediction model building module: Builds a neural network prediction model based on gated recurrent units for temperature and pressure fields; Model training module: Trains a neural network prediction model based on gated recurrent units for temperature and pressure fields; Model performance testing module: Tests the performance of the neural network prediction model based on gated recurrent units for temperature and pressure fields; To enable the prediction of temperature and pressure fields of the compressor of a small turbojet engine.
9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1-7.