Aeroengine Surge Pressure Simulation System and Method Based on Gaussian Prediction Model

Abstract

The present invention relates to a surge pressure simulation system and method for an aeroengine based on a Gaussian prediction model, including a data acquisition module, a mapping module and a control module; the data acquisition module obtains the relationship between the output flow rate of the switching valve and the on-duty ratio of the high-speed switching valve; the control module calculates the real-time error based on the measured pressure value of the cylinder and the real-time on-duty ratio of the high-speed switching valve, and further performs predictive calculation through the real-time error to obtain a control quantity, which, after passing through the mapping module, controls the switching valve group to change the pressure in the cylinder, where the pressure in the cylinder is the surge simulation pressure of the aeroengine. The present invention can timely compensate for the uncertainty caused by internal or external disturbances of the switching valve, has strong robustness and good dynamic performance, and improves the control accuracy and response speed of the real output pressure value of the surge pressure simulation of the aeroengine.

Description

Technical Field

[0001] The present invention belongs to the technical field of pneumatic servo control, and particularly relates to an aero-engine surge pressure simulation system and method based on a Gaussian prediction model. Background Art

[0002] In industrial automation, robotics, and aero-engine hardware-in-the-loop simulation, precise control of the pressure in a fixed-volume container is very important. In previous studies, electromagnetic switching valves and proportional valves were the two most commonly used pneumatic actuators for pressure regulation. Although the proportional valve has the advantage of such proportional output, its high cost and complex structure reduce its versatility. In recent years, various pneumatic manufacturers have produced various electromagnetic switching valves with accelerated opening characteristics, shortening the opening and closing times to 4 ms or more, which has also led to the gradual use of switching valves instead of proportional valves in precise pressure control applications. In an increasingly complex industrial process, traditional mathematical models have been unable to accurately model important variables, and important processes cannot be effectively optimized and diagnosed. The Gaussian process model is a machine learning method based on Bayesian theory and is a hot topic in current international machine learning research, providing a new idea for the current challenges. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide an aero-engine surge pressure simulation system and method based on a Gaussian prediction model, which can timely compensate for the uncertainty caused by internal or external disturbances of the switching valve, has strong robustness and good dynamic performance, and improves the control accuracy and response speed of the real output pressure value of the aero-engine surge pressure simulation.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions:

[0005] An aero-engine surge pressure simulation system based on a Gaussian prediction model includes a data acquisition module, a mapping module, and a control module; the data acquisition module obtains the relationship between the output flow rate of the switching valve and the on-duty ratio of the high-speed switching valve; the control module calculates the real-time error based on the measured pressure value of the cylinder and the real-time on-duty ratio of the high-speed switching valve, and performs predictive calculation through the real-time error to obtain a control quantity, which is passed through the mapping module and then controls the switching valve group to change the pressure in the cylinder, where the pressure in the cylinder is the aero-engine surge simulation pressure.

[0006] Further, the data acquisition module includes an air compressor, a pressure regulating valve, a high-speed switching valve, and a flowmeter; the air inlet of the air compressor is connected to the atmosphere, the air inlet of the pressure regulating valve is connected to the exhaust port of the air compressor to access a high-pressure air source, the exhaust port of the pressure regulating valve is connected to the air inlet of the flowmeter through the high-speed switching valve, and the exhaust port of the flowmeter is connected to the atmosphere; the pressure regulating valve is used to reduce the pressure of the high-pressure air source and output a given pressure; the flowmeter is used to measure the output flow of the high-speed switching valve under different pressure conditions.

[0007] Further, the high-speed switching valve is a two-position two-way high-speed switching valve, and the opening and closing of the switching valve can be controlled by energizing and de-energizing the coil, and the maximum pressure it can withstand is 0.6 MPa; among them, the switching valve group consists of 2 high-speed switching valves, which are respectively connected to the air inlet and exhaust port of the cylinder, and are used as the air inlet valve and exhaust valve of the cylinder to adjust the pressure in the cylinder.

[0008] Further, the control module includes a controller; the controller is built-in with a switching valve flow Gaussian model trained based on the switching valve volume flow characteristic data set; the simulation system is also provided with a switching valve group, a cylinder, and a pressure sensor; the switching valve group consists of 2 high-speed switching valves, which are respectively connected to the air inlet and exhaust port of the cylinder; the pressure sensor is connected to the inner cavity air hole of the cylinder; the switching valve group and the pressure sensor are respectively connected to the controller through cables, and the controller controls the opening and closing of the switching valve group and measures and processes the values fed back by the pressure sensor.

[0009] Further, the pressure sensor is a pressure sensor with a 0-10V electrical signal, which can transmit a pressure of -0.1 MPa to 0.6 MPa, and is used to measure the real-time pressure in the cylinder to simulate the real pressure of the aero-engine surge, and output the measured pressure to the control module for calculation.

[0010] A control method for an aero-engine surge pressure simulation system based on a Gaussian prediction model includes the following steps:

[0011] Step S1: Adjust the input pressure of the high-speed switching valve connected thereto by adjusting the pressure regulating valve, and adjust the opening duty ratio of the high-speed switching valve at the preset control frequency, record the average volume flow output by the flowmeter, and finally integrate the collected data to obtain a switching valve volume flow characteristic data set;

[0012] Step S2: Extract the minimum opening duty ratio D min and the maximum opening duty ratio D max and the maximum flow data in the switching valve volume flow data set, and respectively fit them to obtain the minimum duty ratio D min and the maximum opening duty ratio D max and the maximum flow curve data, and combine them with the data collected in Step S1 into a complete data set;

[0013] Step S3: Based on the complete data set, construct and train a Gaussian model for the flow rate of the on-off valve;

[0014] Step S4: The Gaussian prediction model for the flow rate of the on-off valve takes the measured pressure value of the cylinder measured by the current pressure sensor and the real-time opening duty ratio of the high-speed on-off valve obtained by the mapping module as inputs, predicts the current volume flow rate, and calculates the predicted pressure of the cylinder at the next moment through pressure conversion;

[0015] Step S5: Based on the difference between the predicted value at the current time obtained in Step S4 and the input of the internally set simulation signal of the surge pressure of the aero-engine, it is used as the pressure error;

[0016] Step S6: The control module calculates and processes the minimum control amount u through the rolling optimization performance index function according to the pressure error, and transmits it to the mapping module;

[0017] Step S7: Calculate the opening duty ratios of the 2 high-speed on-off valves in the on-off valve group through the mapping module, output the control matrix, and the control module outputs digital signals based on the control matrix through the digital output card to control the opening and closing of the on-off valve group. The on-off valve group changes its duty ratio to make the measured pressure output by the cylinder, that is, the output value of the real surge pressure signal of the aero-engine, and the error between the given pressure, that is, the input value of the simulation signal of the surge pressure of the aero-engine, is minimized and kept stable.

[0018] Further, the specific content of Step S1 is as follows: First, set the control frequency to 100 Hz. After adjusting the opening of the pressure regulating valve to change the input-output pressure difference of the high-speed on-off valve, use an automatic acquisition system to automatically adjust the duty ratio of the on-off valve, and record the outlet flow rate and inlet flow rate of the on-off valve in real time, where the pressure difference range is 0 - 3 MPa; Through multiple acquisitions, obtain the flow characteristics of the high-speed on-off valve at different pressure differences under the same control frequency of 100 Hz, as well as its minimum opening duty ratio D min and maximum opening duty ratio D max ; According to the collected data, remove the data that does not meet the preset requirements to obtain the data set of the volume flow rate characteristics of the on-off valve.

[0019] Further, the specific content of Step S2 is as follows: Extract the minimum opening duty ratio D min and maximum opening duty ratio D max corresponding to the high-speed on-off valve under the collected pressure difference, and the maximum flow rate, and respectively perform fitting to obtain the predicted curves of the minimum opening duty ratio D min and maximum opening duty ratio D max , and maximum flow rate corresponding to 0.3 - 0.7 MPa; Among them, the fitting function of the minimum opening duty ratio D min is: D min =-7.745*ΔP -0.06504+11.52, the maximum opening duty cycle D max The fitting function is: D max = -2.728 * ΔP -0.4997 +96.15, the maximum flow rate fitting function is: V max = 114.7 * ΔP 0.5843 -2.697; where the minimum duty cycle curve is the duty cycle at which the acceleration switch valve critically opens at the corresponding pressure difference, and the corresponding flow rate is 0; the final combination is a dataset of the volume flow rate characteristics of the switch valve including the three parameters of pressure difference, duty cycle, and flow rate.

[0020] Further, the step S3 is specifically: using the switch valve volume flow rate dataset M = {(x i , y i ) | i = 1, 2,..., n} obtained from the switch valve volume flow rate characteristic data acquisition as the dataset input for the Gaussian process modeling of the switch valve flow rate; and determining the kernel function as the exponential kernel function, that is where γ 2 represents the amplitude of the Gaussian kernel function, and α represents the characteristic length scale; using the maximum likelihood method to obtain the extreme value of the likelihood function to estimate the hyperparameters; obtaining the Gaussian prediction model of the switch valve flow rate through Gaussian process modeling.

[0021] Further, the pressure conversion formula is:

[0022]

[0023] where V is the volume flow rate, P in is the absolute intake air pressure of the switch valve, P out is the absolute exhaust air pressure of the switch valve, C d is the discharge coefficient of the switch valve, T0 is the stagnation temperature, and T in is the air temperature at the intake port of the switch valve. P c = P out / P in is the intake and exhaust pressure ratio of the switch valve, and P cr is the critical pressure ratio.

[0024] The present invention has the following beneficial effects compared with the prior art:

[0025] The present invention can quickly obtain an accurate system model through the actual system test data, avoiding the modeling errors caused by the simplified conditions of the mathematical model; adopting the rolling optimization strategy of the switch valve flow rate prediction control method can timely compensate for the uncertainties caused by internal or external disturbances of the switch valve, has strong robustness and good dynamic performance, and improves the control accuracy and response speed of the aeroengine surge pressure to simulate the real output pressure value. Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0027] Figure 2 It is the modeling result of the high-speed switch valve flow Gaussian process in an embodiment of the present invention. Specific embodiments

[0028] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0029] Please refer to Figure 1 , the present invention provides an aero-engine surge pressure simulation system based on a Gaussian prediction model, including a data acquisition module, a mapping module and a control module; the data acquisition module obtains the relationship between the output flow of the switch valve and the opening duty ratio of the high-speed switch valve; the control module calculates the real-time error based on the measured pressure value of the cylinder and the real-time opening duty ratio of the high-speed switch valve, and performs predictive calculation through the real-time error to obtain a control quantity, which is passed through the mapping module and then controls the switch valve group to change the pressure in the cylinder, where the pressure in the cylinder is the aero-engine surge simulation pressure.

[0030] In this embodiment, the data acquisition module is composed of an air compressor, a pressure regulating valve, a high-speed switch valve and a flowmeter, the control module is composed of a controller, and the systems are connected by air pipes and cables; the switch valve group is composed of 2 high-speed switch valves, and the 2 high-speed switch valves are respectively connected to the cylinder through air pipes; one end of the air compressor in the data acquisition module is connected to the atmosphere, and the other end is connected to the intake end of the pressure regulating valve. One end of the high-speed switch valve is connected to the exhaust end of the pressure regulating valve, and the other end is connected to the flowmeter. The other end of the flowmeter is connected to the atmosphere. By adjusting the intake pressure, opening duty ratio and control frequency of the high-speed switch valve, the volume flow of the flowmeter is collected, and finally the data set is input to the training set of the switch valve flow Gaussian prediction module through processing.

[0031] In this embodiment, for the switch valve flow Gaussian prediction module, after the training set input, kernel function and hyperparameters are determined, the switch valve flow Gaussian model is trained, and finally the switch valve flow Gaussian prediction model is obtained; the first input of the control module is the output model of the switch valve flow Gaussian prediction module, the second input is the output duty ratio D of the mapping module x , and the third input is the measured value P(k) of the pressure sensor. In this module, the pressure error is calculated minJ(k) is calculated under the constraint conditions of the pneumatic system, and finally the control quantity u is obtained and output to the mapping module, where the constraint conditions of the pneumatic system are related to the switch valve group and the cylinder; the mapping module calculates the opening duty ratio D of the switch valve group according to the control quantity u x, and output it to the switching valve flow Gaussian prediction model of the switching valve group and the control module; the pressure sensor is connected to the cylinder, measures the real-time pressure in the cylinder, that is, the output value P(k) of the real aero-engine surge pressure signal, and outputs the actual measured value to the control module.

[0032] In this embodiment, the control module, the mapping module, and the pressure sensor are operated by an embedded controller NI cRIO-9074, and are connected with a digital output card (model NI 9401) for digital signal output, and an analog acquisition card (model NI 9205) is connected to the pressure sensor for analog signal acquisition.

[0033] Preferably, in this embodiment, the high-speed switching valve is a two-position two-way high-speed switching valve (FESTO MHJ10-S-0,35-QS-4MF), and the opening and closing of the switching valve can be controlled by energizing and de-energizing the coil, and the maximum pressure it can withstand is 0.6 MPa; among them, the switching valve group is composed of 2 high-speed switching valves, which are respectively connected to the air inlet and exhaust port of the cylinder, and are used as the air inlet valve and exhaust valve of the cylinder to adjust the pressure in the cylinder. The cylinder is a standard cylinder with a maximum pressure resistance of 0.7 MPa.

[0034] Preferably, in this embodiment, the pressure sensor adopts a high-performance silicon piezoresistive pressure oil-filled core pressure sensor with 0-10V electrical signal (model MIK-P300), which can transmit a pressure of -0.1 MPa to 0.6 MPa, and is used to measure the real-time pressure in the cylinder to simulate the real air pressure of aero-engine surge, and output the measured pressure to the switching valve prediction model for calculation.

[0035] Preferably, in this embodiment, the flowmeter adopts a micro gas thermal flowmeter that outputs 0-5V electrical signal, and is used to measure the output flow of the high-speed switching valve under different pressure differences, and combines and outputs the measured flow to the switching valve flow process modeling module as a training set. The pressure regulating valve can reduce the high-pressure gas source pressure and output a given pressure.

[0036] In this embodiment, the method of the aero-engine surge pressure simulation system based on the high-speed switching valve flow Gaussian prediction model refers to Figure 1 the overall model system, including a data acquisition module, a switching valve flow Gaussian prediction module, a control module, and a mapping module; the output end of the data acquisition module is connected to the input end of the switching valve flow Gaussian prediction module; the output end of the switching valve flow Gaussian prediction module is connected to the first input end of the control module; the output end of the control module is connected to the input end of the mapping module; the first output end of the mapping module is connected to the input end of the switching valve group, and the second output end is connected to the second input end of the control module; the output end of the pressure sensor is connected to the third input end of the control module;

[0037] The data acquisition module is used to acquire the flow rate characteristic data set of the high-speed on-off valve; the on-off valve flow Gaussian prediction module is used to perform Gaussian modeling on the input training set; the control module is used to calculate the control quantity to reduce the error between the given pressure and the measured pressure; the mapping module is used for the conversion between the control quantity and the duty cycle; specifically, it includes the following steps:

[0038] Step S1: First, set the control frequency to 100 Hz. After adjusting the opening of the pressure regulating valve to change the input-output pressure difference of the high-speed on-off valve, use the automatic acquisition system to automatically adjust the duty cycle of the on-off valve and record the average volume flow rate at the outlet of the on-off valve in real time. Through continuous experiments, obtain the flow rate characteristics of the high-speed on-off valve at different pressure differences under the same control frequency of 100 Hz, as well as its minimum opening duty cycle D min and the maximum opening duty cycle D max ; according to the collected data, remove some inaccurate data, and finally obtain the volume flow rate characteristic data set of the on-off valve.

[0039] Step S2: Extract the minimum opening duty cycle D min and the maximum opening duty cycle D max corresponding to the collected pressure difference of the high-speed on-off valve, as well as the maximum flow rate, and perform fitting respectively to obtain the prediction curves of the minimum opening duty cycle D min and the maximum opening duty cycle and the maximum flow rate corresponding to 0.3 - 0.7, and combine them with the data collected in step 1 to form a complete data set; among them, the formula for the minimum opening duty cycle D min is: D min =-7.745*ΔP -0.06504 +11.52, where the formula for the maximum opening duty cycle D max is: D max =-2.728*ΔP -0.4997 +96.15; where the formula for the maximum flow rate is: V max =114.7*ΔP 0.5843 -2.697.

[0040] Step S3: Gaussian process modeling of the on-off valve flow: Use the on-off valve volume flow rate data set D={(x i ,y i )|i = 1,2,...,n} obtained from the on-off valve volume flow rate characteristic data acquisition as the data set input for the Gaussian process modeling of the on-off valve flow; and determine the kernel function as the exponential kernel function, that is where γ 2 represents the amplitude of the Gaussian kernel function, and α represents the characteristic length scale; use the maximum likelihood method to find the extreme value of the likelihood function to estimate the hyperparameters; obtain the on-off valve flow Gaussian prediction model through Gaussian process modeling, as shown in Figure 2 .

[0041] Step S4: The switching valve flow Gaussian prediction model takes the measured pressure value of the cylinder measured by the current pressure sensor and the real-time opening duty ratio of the high-speed switching valve obtained by the mapping module as inputs, predicts the current volume flow rate, and obtains the predicted pressure of the cylinder at the next moment through pressure conversion; the pressure conversion formula is:

[0042]

[0043] where V is the volume flow rate, P in is the absolute intake pressure of the switching valve, P out is the absolute exhaust pressure of the switching valve, C d is the discharge coefficient of the switching valve, T0 is the stagnation temperature, T in is the air temperature at the intake port of the switching valve. P c = P out / P in is the intake-exhaust pressure ratio of the switching valve, P cr = 0.38 is the critical pressure ratio;

[0044] Step S5: By calculating the difference between the real-time predicted value of the switching valve flow prediction module and the input of the internally set aero-engine surge pressure simulation signal, it is used as the pressure error

[0045] Step S6: The control module, according to the pressure error, calculates and processes through rolling optimization of the performance index function

[0046] to obtain the minimum control amount u, and transmits it to the mapping module.

[0047] Step S7: Calculate the opening duty ratios of the 2 high-speed switching valves in the switching valve group through the mapping module, and output the control matrix where D L represents the opening duty ratio of the intake valve of the switching valve group, D R represents the opening duty ratio of the exhaust valve of the switching valve group, f L and f R respectively represent the control frequencies of the intake valve and the exhaust valve of the switching valve group; and then output digital signals through the digital output card to control the opening and closing of the switching valve group. The switching valve group changes its duty ratio to make the error between the measured pressure output by the cylinder, that is, the output value of the real aero-engine surge pressure signal, and the given pressure, that is, the input value of the aero-engine surge pressure simulation signal, remain the smallest and stable. The calculation strategy of the mapping module is shown in the following table.

[0048]

[0049]

[0050] The above are only the preferred embodiments of the present invention, and all equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope of the present invention.

Claims

1. A control method for a surge pressure simulation system of an aero-engine based on a Gaussian prediction model, characterized in that, The described control method is implemented based on a surge pressure simulation system of an aero-engine using a Gaussian prediction model, and includes a data acquisition module, a mapping module, and a control module; the data acquisition module obtains the relationship between the output flow rate of the switching valve and the on-duty ratio of the high-speed switching valve; the control module calculates the real-time error based on the measured pressure value of the cylinder and the real-time on-duty ratio of the high-speed switching valve, and performs predictive calculation through the real-time error to obtain the control quantity. After passing through the mapping module, the control quantity is used to control the switching valve group to change the pressure in the cylinder, where the pressure in the cylinder is the surge simulation pressure of the aero-engine. The control method includes the following steps: Step S1: Adjust the input pressure of the high-speed switching valve connected thereto by adjusting the pressure regulating valve, and adjust the on-duty ratio of the high-speed switching valve under the preset control frequency. Record the average volume flow rate output by the flow meter, and finally integrate the collected data to obtain the switching valve volume flow rate characteristic data set. Step S2: Extract the minimum opening duty ratio D in the switching valve volume flow rate dataset min and the maximum opening duty ratio D max , the maximum flow rate data, and respectively perform fitting on these to obtain the minimum duty ratio D min and the maximum opening duty ratio D max , the maximum flow rate curve data, and combine them with the data collected in Step S1 to form a complete dataset; Step S3: Based on the complete data set, construct and train the switching valve flow Gaussian model. Step S4: The switching valve flow Gaussian prediction model takes the measured pressure value of the cylinder measured by the current pressure sensor and the real-time on-duty ratio of the high-speed switching valve obtained by the mapping module as inputs, predicts the current volume flow rate, and calculates the predicted pressure of the cylinder at the next moment through pressure conversion. Step S5: Based on the difference between the time prediction value obtained in Step S4 and the input of the internal set aero-engine surge pressure simulation signal, it is used as the pressure error. Step S6: The control module calculates and processes the minimum control quantity u through the rolling optimization performance index function according to the pressure error, and transmits it to the mapping module. Step S7: Calculate the on-duty ratio of each of the 2 high-speed switching valves in the switching valve group through the mapping module, output the control matrix, and the control module outputs digital signals through the digital quantity output card based on the control matrix to control the opening and closing of the switching valve group. The switching valve group changes its on-duty ratio to make the error between the measured pressure output by the cylinder, that is, the output value of the real aero-engine surge pressure signal, and the given pressure, that is, the input value of the aero-engine surge pressure simulation signal, remain the smallest and stable.

2. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 1, characterized in that, The data acquisition module includes an air compressor, a pressure regulating valve, a high-speed switching valve, and a flow meter; the air inlet of the air compressor is connected to the atmosphere, the air inlet of the pressure regulating valve is connected to the exhaust port of the air compressor to access the high-pressure air source, the exhaust port of the pressure regulating valve is connected to the air inlet of the flow meter through the high-speed switching valve, and the exhaust port of the flow meter is connected to the atmosphere; the pressure regulating valve is used to reduce the pressure of the high-pressure air source and output a given pressure; the flow meter is used to measure the output flow rate of the high-speed switching valve under different pressure difference conditions.

3. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 2, characterized in that, The high-speed switching valve is a two-position two-way high-speed switching valve, and the opening and closing of the switching valve can be controlled by energizing and de-energizing the coil, and the maximum bearing pressure is 0.6 MPa; among them, the switching valve group is composed of 2 high-speed switching valves, which are respectively connected to the air inlet and exhaust port of the cylinder, and are used as the cylinder inlet valve and exhaust valve to adjust the pressure in the cylinder.

4. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 1, characterized in that, The control module includes a controller, and a switching valve flow Gaussian model trained based on a switching valve volume flow characteristic data set is built in the controller; the simulation system is further provided with a switching valve group, a cylinder and a pressure sensor; the switching valve group is composed of two high-speed switching valves which are respectively connected to the air inlet and the air outlet of the cylinder; the pressure sensor is connected to the inner cavity air hole of the cylinder; the switching valve group and the pressure sensor are respectively connected to the controller through cables, and the controller controls the opening and closing of the switching valve group and measures and processes the values fed back by the pressure sensor.

5. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 4, characterized in that, The pressure sensor is a pressure sensor with a 0-10V electrical signal, which can transmit a pressure of -0.1MPa to 0.6MPa, and is used to measure the real-time pressure in the cylinder to simulate the real pressure of the aero-engine surge, and output the measured pressure to the control module for calculation.

6. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 1, characterized in that, The specific content of step S1 is as follows: First, set the control frequency to 100Hz. After adjusting the opening of the pressure regulating valve to change the input-output pressure difference of the high-speed switching valve, use an automatic acquisition system to automatically adjust the duty cycle of the switching valve and record the flow rate at the outlet of the switching valve in real time, where the pressure difference range is 0-3MPa. The flow characteristics of the high-speed on-off valve at different pressure differences and its minimum opening duty ratio D are obtained by multiple acquisitions at the same control frequency of 100 Hz. min and the maximum opening duty ratio D max ; According to the collected data, the data that does not meet the preset requirements is removed to obtain a switching valve volume flow characteristic data set.

7. The control method of the surge pressure simulation system of an aero-engine based on the Gaussian prediction model according to claim 1, characterized in that The specific steps of step S2 are as follows: extract the minimum opening duty ratio D corresponding to the high-speed switching valve at the collected pressure difference min and the maximum opening duty ratio D max , the maximum flow rate, and respectively perform fitting to obtain the minimum opening duty ratio D corresponding to 0.3 to 0.7 MPa min and the maximum opening duty ratio D max , the maximum flow rate prediction curves; among them, the fitting function of the minimum opening duty ratio D min is: D min =-7.745*ΔP -0.06504 +11.52, the fitting function of the maximum opening duty ratio D max is: D max =-2.728*ΔP -0.4997 +96.15, the fitting function of the maximum flow rate is: V max =114.7*ΔP 0.5843 -2.697; among them, the minimum duty ratio curve is the duty ratio at which the acceleration switching valve is critically opened at the corresponding pressure difference, and the corresponding flow rate is 0; finally, it is combined into a dataset of the volume flow characteristics of the switching valve including three parameters: pressure difference, duty ratio, and flow rate.

8. The control method of the aeroengine surge pressure simulation system based on the Gaussian prediction model according to claim 1, characterized in that The specific content of step S3 is as follows: Using the dataset M of the volumetric flow rate of the on-off valve obtained from the acquisition of the volumetric flow rate characteristics of the on-off valve, M = {(x i , y i ) | i = 1, 2,..., n} as the dataset input for the Gaussian process modeling of the on-off valve flow rate; And The kernel function is determined to be an exponential kernel function, that is where γ 2 represents the amplitude of the Gaussian kernel function, and α represents the characteristic length scale; Using the maximum likelihood method, find the extreme value of the likelihood function to estimate the hyperparameters; obtain a switching valve flow Gaussian prediction model through Gaussian process modeling.

9. The control method of the surge pressure simulation system of an aero-engine based on a Gaussian prediction model according to claim 1, characterized in that, The pressure conversion formula is: Wherein, V is the volume flow rate, P in is the absolute intake pressure of the on-off valve, P out is the absolute exhaust pressure of the on-off valve, C d is the discharge coefficient of the on-off valve, T0 is the stagnation temperature, T in is the air temperature at the intake port of the on-off valve; P c = P out / P in is the intake-exhaust pressure ratio of the on-off valve, P cr is the critical pressure ratio.

Images