Multi-objective optimization and control method for heat transfer characteristics of turbine disk-cavity system during mode switching

Abstract

The present invention discloses a multi-objective optimization and control method for the heat transfer characteristics of a turbine disc cavity system during mode switching. First, based on a numerical simulation model of an aeroengine turbine disc cavity, the heat transfer characteristics and temperature field distribution during mode switching of the turbine disc cavity system under different operating conditions are obtained. This data is used as training samples to establish a database of the heat transfer characteristics of the turbine disc cavity system during mode switching. Then, a mapping relationship between input features and corresponding outputs is established using a neural network method, constructing a rapid prediction model for the heat transfer characteristics of the turbine disc cavity during mode switching. By optimizing and controlling the disc cavity air cooling source input parameters, the present invention achieves the purpose of multi-objective optimization of the heat transfer characteristics of the turbine disc cavity and real-time control of the temperature field during mode switching.

Description

Technical Field

[0001] The present invention belongs to the technical field of aero-engine thermal management, and in particular relates to a multi-objective optimization and control method for heat exchange characteristics of a turbine disc cavity system during mode switching. Background Art

[0002] Aircraft engines are essential power plants for modern aircraft, and their performance and reliability directly impact aircraft safety and flight efficiency. As a key component of aircraft engines, the heat transfer characteristics and temperature distribution of the turbine disc cavity system significantly impact the engine's operational stability and lifespan. In particular, during mode switching, the heat transfer characteristics and temperature field of the turbine disc cavity system change, adversely affecting system performance and lifespan. Therefore, achieving multi-objective optimization of the turbine disc cavity system and real-time control of the temperature field during mode switching has become a key technical challenge.

[0003] Traditional methods for controlling the temperature field of turbine disc cavities rely primarily on numerical simulations, which are time-consuming and difficult to implement in real time. Traditional optimization methods, on the other hand, are limited in their effectiveness when dealing with multi-objective optimization problems. These methods are unable to meet the high-precision and high-efficiency performance requirements of future turbine disc cavity systems during mode switching. Summary of the Invention

[0004] To address the problems existing in the prior art, the present invention aims to provide a multi-objective optimization and control method for the heat transfer characteristics of a turbine disc cavity system during mode switching. This method obtains the heat transfer characteristics and temperature field distribution based on a numerical simulation model of an aircraft engine turbine disc cavity, and establishes a corresponding database. A rapid prediction model is established using a neural network to achieve rapid prediction of the heat transfer characteristics of the turbine disc cavity system. Finally, by optimizing and controlling the disc cavity air cooling source input parameters (cooling air pressure, temperature, etc.), the multi-objective optimization of the turbine disc cavity heat transfer characteristics and real-time control of the temperature field during mode switching are achieved.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching includes the following steps:

[0007] Step 1: Establish a three-dimensional geometric model of the aero-engine turbine disc cavity system; determine the geometric boundary conditions, including input variables and thermal loads, based on the aero-engine mode switching process;

[0008] Step 2: Establish a three-dimensional numerical simulation model of the aircraft engine turbine disc cavity system during the mode switching process, and determine a calculation method for evaluating the heat transfer characteristics of the turbine disc cavity system during the process;

[0009] Step 3: Using Latin hypercube sampling within the range of the input variables, an input data set is generated. After sampling, the corresponding parameters in the input data set are introduced into the three-dimensional numerical simulation model of the turbine disc cavity system to calculate the heat transfer characteristics of the turbine disc cavity system. The heat transfer characteristic evaluation index is used as the output data set.

[0010] Step 4: randomly divide the input data set and output data set obtained in step 3 into a training set, a test set, and a validation set, which are used for fitting, optimizing, and validating the parameters in the neural network model, respectively;

[0011] Step 5: Based on the neural network model, the training set obtained in step 4 is used to train the constructed neural network model. The validation set and test set are used to verify and test the neural network model to improve the generalization ability of the model. The neural network model is used to train and test the data to obtain the mapping relationship between the input features (turbine disc cavity speed, intake pressure, temperature) and the corresponding output (turbine disc cavity heat transfer characteristics). A rapid prediction model for the heat transfer characteristics of the turbine disc cavity system during mode switching under different operating conditions is obtained.

[0012] Step 6: Within the possible variation range of the input variables, the neural network-based rapid prediction model for heat transfer characteristics established in Step 5 is called upon according to the multi-objective optimization algorithm to select the input parameters of the disc cavity, thereby optimizing the heat transfer characteristics of the turbine disc cavity and obtaining a Pareto solution set of the heat transfer characteristics;

[0013] Step 7: Based on the Pareto solution set obtained from the multi-objective optimization and combined with the actual temperature control requirements, a suitable control scheme for the disk cavity temperature field is obtained.

[0014] In step 1, the input variables include the pressure and temperature of each air system flow path and the turbine disk rotor speed.

[0015] In step 2, the turbine disc cavity temperature field numerical simulation model is obtained based on computational fluid dynamics (CFD) through known techniques.

[0016] In step 3, the heat transfer characteristic evaluation indicators include three indicators: turbine disc cavity inlet flow rate, turbine disc temperature uniformity, and turbine disc cavity system maximum temperature.

[0017] Turbine disc cavity temperature uniformity:

[0018]

[0019] in, represents the average temperature of the characteristic point, and S represents the standard deviation between the temperature of the measurement point and the average temperature, which is determined by the following formula:

[0020]

[0021] In the formula, n represents the number of feature points, t i represents the temperature of the i-th point.

[0022] Maximum temperature of turbine disc cavity system:

[0023] T max =max(T1,T2,...,T i ) (3)

[0024] Where, max(T1,T2,...,T i ) is the maximum temperature value of each node of the turbine disk cavity system calculated by the CFD method.

[0025] Turbine disc cavity intake flow:

[0026] Turbine disc cavity system intake flow m all The definition is as follows:

[0027]

[0028] Among them, m i is the intake flow rate of the i-th inlet.

[0029] In step 5, the deep neural network model is used to fit the heat transfer characteristics of the turbine disc cavity during the mode switching process, as shown in the following formula:

[0030]

[0031] Among them, f i (x) is the heat transfer characteristic expression of the turbine disc cavity during the transition state, σ is the activation function, w k is the weight of the kth layer of the neural network, x k is the variable input to the kth layer of the neural network, b is the bias corresponding to each layer of the network, k is the kth neuron, and L is the number of neurons in each layer.

[0032] The multi-objective optimization algorithm mentioned in step 6 is the multi-objective particle swarm optimization algorithm.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] Based on the relationship between different input conditions and the heat transfer characteristics of the turbine disc cavity system during mode switching, this invention constructs a rapid prediction model for the heat transfer characteristics of the turbine disc cavity during mode switching. This allows for rapid estimation of the heat transfer characteristics of the turbine disc cavity within milliseconds. Based on this rapid prediction model, it is possible to evaluate the impact of different cooling strategies on the disc cavity heat transfer characteristics and temperature field to meet practical needs. Furthermore, by optimizing and controlling the disc cavity air cooling source input parameters (cooling air pressure, temperature, etc.), a multi-objective optimization of the turbine disc cavity heat transfer characteristics and real-time temperature field control during mode switching is achieved using a multi-objective intelligent optimization algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1a This is a three-dimensional geometric model of a typical turbine disk cavity system.

[0036] Figure 1b This is the computational domain mesh model of a typical turbine disk cavity system.

[0037] Figure 2 A rapid prediction model for turbine disk cavity heat transfer characteristics based on neural network.

[0038] Figure 3 Accuracy verification of the rapid estimation model of heat transfer characteristics.

[0039] Figure 4 It is the Pareto solution set obtained by the multi-objective particle swarm optimization algorithm. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0041] Example

[0042] Taking a turbine disc cavity structure as an example, the multi-objective optimization and temperature field control method in the process of turbine disc cavity system mode switching is explained. Figure 1a Figure 2 shows a two-dimensional geometric model of a turbine disc cavity system, consisting of a fluid region and a solid turbine disc region. This system includes two air cooling inlets, Inlet 1 and Inlet 2, and four outlets, Outlet 1 through Outlet 4. A heat source, q, is located at the top of the turbine disc, depending on the operating conditions of the turbine disc cavity mode switching process. Based on these two models, the pressure (P*1) and temperature (T*1) of the cooling inlet Inlet 1, the pressure (P*2) and temperature (T*2) of the cooling inlet Inlet 2, and the range of the turbine disc rotor (N*) can be determined. Figure 1b is the computational domain mesh model obtained based on the three-dimensional model.

[0043] Based on the above model, a numerical calculation model of the turbine disc cavity system modal conversion process can be established. The calculation method is the CFD method. After the calculation is completed, the evaluation indicators of the heat transfer characteristics of the turbine disc cavity system are calculated, including the turbine disc cavity inlet flow rate, turbine disc temperature uniformity, and the maximum temperature of the turbine disc cavity system.

[0044] Turbine disc cavity temperature uniformity:

[0045]

[0046] in, represents the average temperature of the characteristic point, and S represents the standard deviation between the temperature of the measurement point and the average temperature, which is determined by the following formula:

[0047]

[0048] In the formula, n represents the number of feature points, t i represents the temperature of the i-th point.

[0049] Maximum temperature of turbine disc cavity system:

[0050] T max =max(T1,T2,...,T i ) (3)

[0051] Where, max(T1,T2,...,T i ) is the maximum temperature value of each node of the turbine disk cavity system calculated by the CFD method.

[0052] Turbine disc cavity intake flow:

[0053] Turbine disc cavity system intake flow m all The definition is as follows:

[0054]

[0055] Among them, m i is the intake flow rate of the i-th inlet.

[0056] Based on a review of the input variable range and the establishment of a numerical model for the turbine disc cavity heat transfer characteristics, Latin hypercube sampling was used to generate 100 feature data sets within the variable study range. The turbine disc cavity heat transfer characteristics were calculated, resulting in a corresponding data set. These data sets were divided into training, validation, and test sets for training and testing the rapid prediction model for the disc cavity temperature field. The training set, test set, and validation set accounted for 80%, 10%, and 10%, respectively.

[0057] Based on the above dataset, we establish Figure 2The neural network-based rapid prediction model for turbine disc cavity heat transfer characteristics is shown. The neural network model consists of an input layer, three hidden layers, and an output layer, which implements a nonlinear mapping from input parameters to output parameters. The neural network output satisfies the following equation:

[0058]

[0059] Among them, f i (x) is the heat transfer characteristic expression of the turbine disc cavity during the transition state, σ is the activation function, w k is the weight of the kth layer of the neural network, x k is the variable input to the kth layer of the neural network, b is the bias corresponding to each layer of the network, k is the kth neuron, and L is the number of neurons in each layer.

[0060] After obtaining the fast estimation model, it is necessary to test the accuracy and generalization ability of the estimation model. Figure 3 As shown in the figure, this prediction model can obtain the heat transfer characteristics of the turbine disc cavity system during the mode switching process with high accuracy, and the data set determination coefficient R 2 At 0.9984, the mean relative error (MRE) is 0.75%. The model is highly accurate, meeting the needs of engineering calculations and providing support for subsequent performance analysis and regulation.

[0061] Based on the above prediction model, with the help of multi-objective particle swarm optimization algorithm, by optimizing and controlling the input parameters of the disk cavity air cooling source (cooling air pressure, temperature, etc.), the multi-objective optimization of the turbine disk cavity heat transfer characteristics during mode switching can be achieved. Figure 4 This is the Pareto solution set obtained by the multi-objective particle swarm optimization algorithm. The three coordinates in the figure represent the three evaluation indicators mentioned above: turbine disc cavity inlet flow, turbine disc temperature uniformity, and turbine disc cavity system maximum temperature. According to the Pareto solution results and the actual temperature control requirements, a suitable control scheme for the disc cavity temperature field can be obtained.

[0062] The present invention provides a method for rapidly estimating the turbine disc cavity temperature field based on an improved deconvolutional neural network. While there are numerous methods and approaches for implementing this technical solution, the aforementioned are merely preferred embodiments of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Any components not specified in this embodiment may be implemented using existing technologies.

Claims

1. A multi-objective optimization and control method for the heat transfer characteristics of a turbine disc cavity system during mode switching, characterized in that: The following steps are involved: Step 1: Establish a three-dimensional geometric model of the aircraft engine turbine disc cavity system; Determine geometric boundary conditions, including input variables and thermal loads, based on the aero-engine mode switching process; Step 2: Establish a three-dimensional numerical simulation model of the aircraft engine turbine disc cavity system during the mode switching process, and determine a calculation method for evaluating the heat transfer characteristics of the turbine disc cavity system during the process; Step 3: Using Latin hypercube sampling within the range of the input variables, an input data set is generated. After sampling, the corresponding parameters in the input data set are introduced into the three-dimensional numerical simulation model of the turbine disc cavity system to calculate the heat transfer characteristics of the turbine disc cavity system. The heat transfer characteristic evaluation index is used as the output data set. Step 4: randomly divide the input data set and output data set obtained in step 3 into a training set, a test set, and a validation set, which are used for fitting, optimizing, and validating the parameters in the neural network model, respectively; Step 5: Based on the neural network model, the training set obtained in step 4 is used to train the constructed neural network model, and the validation set and test set are used to verify and test the neural network model to improve the generalization ability of the model. The neural network model is used to train and test the data to obtain the mapping relationship between input features and corresponding outputs, and a rapid prediction model for the heat transfer characteristics of the turbine disc cavity system during the mode switching process under different operating conditions is obtained. Step 6: Within the range of the input variables, the neural network-based rapid prediction model for heat transfer characteristics established in Step 5 is called upon to select the input parameters of the disc cavity according to the multi-objective intelligent optimization algorithm, thereby optimizing the heat transfer characteristics of the turbine disc cavity and obtaining a Pareto solution set of the heat transfer characteristics. Step 7: Based on the Pareto solution set obtained from the multi-objective optimization and combined with the actual temperature control requirements, a suitable control scheme for the disk cavity temperature field is obtained.

2. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1 is characterized in that: In step 1, the input variables include the pressure and temperature of each air system flow path and the turbine disk rotor speed.

3. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1 is characterized in that: In step 2, the numerical simulation model of the turbine disc cavity temperature field is obtained based on the computational fluid dynamics method.

4. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1 is characterized in that: In step 3, the heat transfer characteristic evaluation indicators include three indicators: turbine disc cavity inlet flow rate, turbine disc temperature uniformity, and turbine disc cavity system maximum temperature.

5. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 4 is characterized in that: The turbine disc cavity inlet flow, turbine disc temperature uniformity, and turbine disc cavity system maximum temperature are expressed using the following formulas: Turbine disc cavity temperature uniformity: Where t represents the average temperature of the characteristic point, and S represents the standard deviation between the temperature of the measurement point and the average temperature, which is determined by the following formula: In the formula, n represents the number of feature points, t i represents the temperature of the i-th point; Maximum temperature of turbine disc cavity system: T max =max(T1,T2,...,T i ) (3) Where, max(T1,T2,...,T i ) is the maximum temperature value of each node of the turbine disk cavity system calculated by CFD method; Turbine disc cavity intake flow: Turbine disc cavity system intake flow m all The definition is as follows: Among them, m i is the intake flow rate of the i-th inlet.

6. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1 is characterized in that: In step 5, the input features are: turbine disc cavity speed, intake pressure, and temperature, and the corresponding output is: turbine disc cavity heat transfer characteristics.

7. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1, characterized in that: In step 5, the deep neural network model is used to fit the heat transfer characteristics of the turbine disc cavity during the mode switching process, as shown in the following formula: Among them, f i (x) is the heat transfer characteristic expression of the turbine disc cavity during the transition state, σ is the activation function, w k is the weight of the kth layer of the neural network, x k is the variable input to the kth layer of the neural network, b is the bias corresponding to each layer of the network, k is the kth neuron, and L is the number of neurons in each layer.

8. The multi-objective optimization and control method for heat transfer characteristics of a turbine disk cavity system during mode switching according to claim 1 is characterized in that: The multi-objective optimization algorithm mentioned in step 6 is the multi-objective particle swarm optimization algorithm.

Images