Fuel control feedback adjusting method and control system of micro turbojet engine

By obtaining a variety of measurement data in a micro-turbine jet engine and adjusting flow using intelligent prediction models, the problem of measuring the internal temperature and pressure of the engine is solved, real-time monitoring and performance prediction of the internal parts of the engine is achieved, and the test efficiency is improved and reliability.

CN118997930APending Publication Date: 2024-11-22HANGXING AEROSPACE (TAICANG) KINETIC ENERGY TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411264191.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

In micro-turbine jet engines, it is difficult to measure the temperature and pressure of the main components inside the engine without interfering with the normal circulation of the airflow and not affecting the performance of the engine, which hinders a detailed understanding of the working conditions of the internal components during engine test runs.

Method used

A fuel control feedback adjustment method is adopted to obtain a variety of measurement data, and use intelligent prediction models (such as BP neural networks) to analyze and adjust the ignition channel flow, main oil channel flow and cooling channel flow to achieve the temperature of key components inside the engine. Real-time monitoring and performance prediction, including using measurement components to obtain data, the control center outputs control signals, and the oil supply valve group to regulate flow.

Benefits of technology

Real-time adjustment of the fuel supply system and real-time monitoring of the engine working status are realized, and engine test efficiency and working reliability are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118997930A_ABST
    Figure CN118997930A_ABST
Patent Text Reader

Abstract

The invention relates to the field of fuel supply of jet propulsion devices, in particular to a fuel control feedback regulation method and control system for a microminiature turbojet engine, and the fuel control feedback regulation method for the microminiature turbojet engine comprises the following steps: obtaining measurement data, the measured data comprises impeller rotating speed, ignition path flow, main oil path flow, cooling path flow, pressure and temperature at an engine inlet, pressure and temperature at a gas outlet end of a gas compressor diffuser, pressure and temperature at a gas inlet end of a turbine, and pressure and temperature at the tail end of an exhaust nozzle; the intelligent prediction model obtains measurement data and outputs key data; and the ignition path flow, the main oil path flow and the cooling path flow are adjusted according to the key data. The device has the effect of obtaining the temperature and the pressure of main components in the engine on the premise that normal circulation of airflow in the engine is not interfered, the working performance of the engine is not influenced, and no measuring device is arranged in the engine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of fuel supply of a jet propulsion device, and in particular to a fuel control feedback adjustment method and a control system for a micro-sized turbojet engine. Background Art

[0002] Micro turbojet engines have been vigorously developed due to their advantages of small size, low cost, and high thrust-to-weight ratio, and have entered the stage of cooperation between academia and industry. Products for various application targets are developed, including small unmanned (tactical) reconnaissance aircraft, small decoy aircraft, small multi-purpose cruise aircraft, small high-speed target aircraft, precision guided weapons, model aircraft, etc.

[0003] As the main power of light unmanned aerial vehicles and model aircraft, micro turbojet engines have difficulties that cannot be ignored and necessary optimization design solutions in terms of structure, overall performance design, compressor aerodynamics design, high-speed rotor dynamics design, combustion chamber design, use of high-temperature materials, bearing service life, engine thermal protection, etc.

[0004] Due to the influence of the rotating parts inside the micro-turbojet engine and the layout of the engine's own flame tube and bearings, it is relatively difficult to measure the temperature and pressure of the main components inside the engine without interfering with the normal flow of air and affecting the engine performance. This hinders researchers from understanding the detailed working conditions of internal components during engine testing. Summary of the invention

[0005] In order to overcome the problem that it is difficult to measure the temperature and pressure of the main components inside the engine without interfering with the normal circulation of the airflow inside the engine, affecting the working performance of the engine, and without setting a measuring device inside the engine, the present invention provides a fuel control feedback adjustment method and control system for a micro-turbojet engine.

[0006] The present invention provides a fuel control feedback adjustment method for a micro-sized turbojet engine, which adopts the following technical solution:

[0007] A fuel control feedback adjustment method for a micro-sized turbojet engine comprises the following steps:

[0008] Obtaining measurement data, including impeller speed, ignition path flow, main oil path flow, cooling path flow, pressure and temperature at the engine inlet, pressure and temperature at the compressor diffuser outlet, pressure and temperature at the turbine inlet, and pressure and temperature at the tail nozzle end;

[0009] The intelligent prediction model obtains the measured data and outputs key data;

[0010] Analyze the engine conditions according to key data and adjust the ignition path flow, main oil path flow, and cooling path flow.

[0011] In a specific implementation scheme, based on the key data being greater than or equal to a preset critical value, a shutdown signal is output to shut down the engine.

[0012] In a specific implementation scheme, the intelligent prediction model is a BP neural network.

[0013] In a specific embodiment, the method for adjusting the cooling path flow rate includes:

[0014] If the front bearing temperature and the rear bearing temperature are lower than a first temperature threshold, reducing the cooling path flow rate;

[0015] If the front bearing temperature and the rear bearing temperature are higher than the second temperature threshold, the cooling path flow rate is increased.

[0016] In a specific feasible implementation scheme, after the BP neural network outputs the key data, a feature subset is constructed based on the key data and the measured data, and each feature subset is trained through a binary classification model to output the predicted probability. The probability output by each feature subset is used as the input of a multi-classification model to train the multi-classification model and output the corresponding situation between the measured data and the key data.

[0017] In a specific feasible implementation scheme, based on the correspondence between the measured data and the key data, an optimal strategy is generated through an inverse reinforcement learning algorithm, and the ignition path flow rate, the main oil path flow rate, and the cooling path flow rate are adjusted according to the optimal strategy.

[0018] In a specific feasible implementation scheme, the sample library used for model training of the intelligent prediction model is the statistical results obtained after a large number of simulations of the same model engine after grid independence verification using different engine inlet pressure and outlet flow boundary conditions.

[0019] The present invention provides a fuel control system for a micro-sized turbojet engine, which adopts the following technical solution:

[0020] A fuel control system for a micro-sized turbojet engine, used in conjunction with the above-mentioned fuel control feedback adjustment method for a micro-sized turbojet engine, comprises:

[0021] A measuring element, used for measuring and obtaining measuring data;

[0022] The control center obtains the key data output by the intelligent prediction model and outputs control signals to the oil supply valve group based on the key data;

[0023] The oil supply valve group obtains the control signal output by the control center and adjusts the flow of the ignition path, the main oil path, and the cooling path;

[0024] The oil pump obtains the control signal from the control center and controls the oil pressure and flow of the fuel output from the oil tank.

[0025] In a specific possible implementation scheme, the measuring elements include a total pressure sensor and a total temperature sensor at the engine inlet, a total pressure sensor and a total temperature sensor at the compressor diffuser outlet, a total pressure sensor and a total temperature sensor at the turbine outlet, a total pressure sensor and a total temperature sensor installed on the outer wall of the tail nozzle, a speed sensor and a flow meter.

[0026] In a specific feasible implementation scheme, the oil supply valve group includes a main oil supply valve, a first oil supply valve, a second oil supply valve and a third oil supply valve. The first oil supply valve is connected to the ignition oil circuit, the second oil supply valve is connected to the main oil circuit, the third oil supply valve is connected to the cooling circuit, and the main oil supply valve is connected to the outlet end of the oil pump.

[0027] In summary, the beneficial effects of the present invention compared with the prior art are:

[0028] By acquiring relatively easy-to-measure measurement data and calculating key data, real-time adjustment of the fuel supply system, real-time monitoring of the compressor operating status and performance, and real-time prediction of the performance of key components inside the engine can be achieved, thereby improving engine test efficiency and engine operating reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is an overall schematic diagram of the fuel control system of a micro turbojet engine.

[0030] Figure 2 It is a schematic diagram showing the data flow of a micro turbojet engine.

[0031] Figure 3 It is a schematic diagram of the BP neural network control and regulation process of a micro-turbojet engine.

[0032] Figure 4 It is the algorithm diagram of the BP neural network training process of a micro turbojet engine.

[0033] Explanation of the accompanying drawings: 1. Oil storage tank; 2. Filter; 3. Main oil supply valve; 4. First oil supply valve; 5. Second oil supply valve; 6. Third oil supply valve; 7. Flow meter; 8. Speed ​​sensor; 9. Control center; 10. Oil pump. DETAILED DESCRIPTION

[0034] The following is combined with Figure 1-4 The present invention is described in further detail.

[0035] Embodiment 1:

[0036] Reference Figure 1The fuel control feedback adjustment method of a micro-turbojet engine comprises the following steps:

[0037] S100, obtaining measurement data.

[0038] The measured data include impeller speed, ignition path flow, main oil path flow, cooling path flow, pressure p1 and temperature t1 at the engine inlet, pressure p2 and temperature t2 at the compressor diffuser outlet, pressure p3 and temperature t3 at the turbine inlet, and pressure p4 and temperature t4 at the tail nozzle end.

[0039] The impeller speed is obtained through the speed sensor. The ignition path flow, main oil path flow, and cooling path flow are obtained through the flow meter.

[0040] The pressure p1 and temperature t1 at the engine inlet are obtained through the total pressure sensor P1 and the total temperature sensor T1 at the engine inlet.

[0041] The pressure p2 and temperature t2 at the compressor diffuser outlet are obtained through the total pressure sensor P2 and the total temperature sensor T2 at the compressor diffuser outlet.

[0042] The pressure p3 and temperature t3 at the turbine outlet are obtained through the total pressure sensor P3 and the total temperature sensor T3 at the turbine outlet.

[0043] The pressure p4 and temperature t4 at the end of the tail nozzle are obtained by the total pressure sensor P4 and the total temperature sensor T4 installed on the outer wall of the tail nozzle.

[0044] S200, calculating key data based on the measured data.

[0045] A BP neural network with a 3-layer network structure consisting of 12-10-8 is pre-built. The BP neural network includes an input layer, a hidden layer, and an output layer. The input layer includes 12 input variables corresponding to 12 measurement data, the output layer includes 8 target outputs corresponding to 8 key data, and the hidden layer includes 10 node neurons. The key data include compressor pressure ratio and working efficiency, combustion chamber gas temperature and pressure, flame tube inner wall temperature, flame tube outer wall temperature, front bearing temperature, and rear bearing temperature.

[0046] The BP neural network has been pre-trained and verified through a sample database. The data in the sample database are statistical results obtained after a large number of simulations of the same model engine after grid independence verification using different engine inlet pressure and outlet flow boundary conditions.

[0047] The measured data is used as the input of the BP neural network. After calculation in the hidden layer, it outputs key data to predict the state of the engine.

[0048] S300, controls the ignition path flow, main oil path flow, and cooling path flow according to key data.

[0049] If the front bearing temperature and the rear bearing temperature are lower than a first temperature threshold, reducing the cooling path flow rate;

[0050] If the front bearing temperature and the rear bearing temperature are higher than a second temperature threshold, increasing the cooling path flow rate;

[0051] If the front bearing temperature and the rear bearing temperature are between the first temperature threshold and the second temperature threshold, it indicates that the front bearing temperature and the rear bearing temperature are normal.

[0052] According to the compressor pressure ratio and working efficiency, the working condition of the compressor and the engine air intake volume can be understood in real time.

[0053] According to the gas conditions in the combustion chamber, the flow of the main oil circuit is adjusted in real time to adjust the combustion chamber performance to a suitable working state.

[0054] When the bearing temperature exceeds the preset critical value or other extreme conditions occur, a shutdown signal is output to shut down the engine.

[0055] The present invention also discloses a fuel control system for a micro-sized turbojet engine, comprising:

[0056] The measuring elements include a total pressure sensor P1 and a total temperature sensor T1 at the engine inlet, a total pressure sensor P2 and a total temperature sensor T2 at the compressor diffuser outlet, a total pressure sensor P3 and a total temperature sensor T3 at the turbine outlet, a total pressure sensor P4 and a total temperature sensor T4 installed on the outer wall of the tail nozzle, a speed sensor and a flow meter, which are used to measure different measurement data and output them to the control center.

[0057] The control center can store the BP neural network model, obtain the measured data and output it to the BP neural network model, obtain the key data output by the BP neural network model, and output the control signal according to the comparison result between the key data and the preset first temperature threshold, second temperature threshold, etc.

[0058] The oil pump is connected to the oil storage tank, obtains the control signal output by the control center, and controls the oil pressure and flow rate of the fuel output from the oil storage tank.

[0059] The oil supply valve group includes a main oil supply valve, a first oil supply valve, a second oil supply valve and a third oil supply valve. The first oil supply valve is connected to the ignition oil circuit, and the first oil supply valve obtains a control signal output by a control center to control the flow of the ignition oil circuit; the second oil supply valve is connected to the main oil circuit, and the second oil supply valve obtains a control signal output by the control center to control the flow of the main oil circuit; the third oil supply valve is connected to the cooling oil circuit, and the third oil supply valve obtains a control signal output by the control center to control the flow of the cooling circuit; the main oil supply valve is connected to the outlet end of the oil pump, obtains a control signal output by the control center, and controls the total flow of fuel to the engine.

[0060] Embodiment 2:

[0061] Embodiment 2 discloses a fuel control feedback adjustment method for a micro-sized turbojet engine, which differs from Embodiment 1 in that: in step S200, during the model training process, after the BP neural network model outputs key data, a label is created for the possible situations of the engine of the same model, that is, each label corresponds to a certain situation of the engine of the same model. Each group of key data and measured data are grouped under the corresponding label to form a feature subset, and each feature subset contains the parameters that cause the engine to have the situation, as well as normal parameters. A binary classification model is constructed for each feature subset, such as a support vector machine, logistic regression, etc. The binary classification model outputs the predicted probability of the corresponding situation of each feature subset. The predicted probability of each feature subset is used as the input of the multi-classification model, and the model is trained. The multi-classification model can be a decision tree and a random forest. The multi-classification model judges different situations of the same model engine based on the measured data and key data, so as to automatically analyze the current state of the current engine after obtaining the key data.

[0062] Through the adjustment schemes of ignition path flow, main oil path flow, and cooling path flow under the same model and different situations, an expert scheduling strategy set is constructed. Among them, m is the total number of experts in the expert scheduling strategy set D, is the state sequence trajectory of the i-th expert at time t. The multi-classification model outputs the classification results and the corresponding key data and measurement data as feature attributes.

[0063] Constructing the reward function: R(s) = ω T φ(s), where φ(s) is the characteristic attribute vector in state s, ω is the weight vector, and T is the transfer matrix, which represents the state transition probability.

[0064] Randomly generate a strategy π (0) , the characteristic expected value μ(π (0) ). Get the expert strategy π from the expert strategy set D E , calculate the characteristic expected value μ(πE ). Solve t by linear optimization algorithm (i) The optimal weight ω at the moment (i) . Determine whether the model has converged. If it has converged, set the optimal weight ω (i) Substitute into the reward function R(s) to obtain the optimal strategy π (i) , the optimal strategy π (i) That is, under the current state, the optimal ignition flow, main oil flow, and cooling flow configuration based on the model of the engine in this case. If it does not converge, the optimal weight ω (i) Substituting into the reward function R(s), the inverse reinforcement learning algorithm loops the aforementioned steps of calculating feature expectations and linear optimization algorithms.

[0065] S300, automatically controlling the first fuel supply valve, the second fuel supply valve and the third fuel supply valve according to the output ignition path flow rate, the main fuel path flow rate and the cooling path flow rate.

[0066] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A fuel control feedback adjustment method for a micro-sized turbojet engine, characterized in that: The steps include: Obtaining measurement data, including impeller speed, ignition path flow, main oil path flow, cooling path flow, pressure and temperature at the engine inlet, pressure and temperature at the compressor diffuser outlet, pressure and temperature at the turbine inlet, and pressure and temperature at the tail nozzle end; The intelligent prediction model obtains the measured data and outputs key data; Analyze the engine conditions according to key data and adjust the ignition path flow, main oil path flow, and cooling path flow.

2. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 1, characterized in that: When the key data is greater than or equal to a preset critical value, a shutdown signal is output to shut down the engine.

3. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 1, characterized in that: The intelligent prediction model is a BP neural network.

4. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 1, characterized in that: Methods for adjusting cooling flow include: If the front bearing temperature and the rear bearing temperature are lower than a first temperature threshold, reducing the cooling path flow rate; If the front bearing temperature and the rear bearing temperature are higher than the second temperature threshold, the cooling path flow rate is increased.

5. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 1, characterized in that: After the BP neural network outputs the key data, a feature subset is constructed based on the key data and the measured data. Each feature subset is trained through a binary classification model and the predicted probability is output. The probability output by each feature subset is used as the input of the multi-classification model. The multi-classification model is trained and the corresponding situation of the measured data and the key data is output.

6. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 5, characterized in that: According to the correspondence between the measured data and the key data, the optimal strategy is generated through the inverse reinforcement learning algorithm, and the ignition path flow rate, main oil path flow rate and cooling path flow rate are adjusted according to the optimal strategy.

7. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 1, characterized in that: The sample library used for model training of the intelligent prediction model is the statistical results obtained after a large number of simulations of the same model engine after grid independence verification using different engine inlet pressure and outlet flow boundary conditions.

8. A fuel control system for a micro-turbojet engine, characterized in that: Used in conjunction with the fuel control feedback adjustment method for a micro-sized turbojet engine according to any one of claims 1 to 7, comprising: A measuring element, used for measuring and obtaining measuring data; The control center obtains the key data output by the intelligent prediction model and outputs control signals to the oil supply valve group based on the key data; The oil supply valve group obtains the control signal output by the control center and adjusts the flow of the ignition path, the main oil path, and the cooling path; The oil pump obtains the control signal from the control center and controls the oil pressure and flow of the fuel output from the oil tank.

9. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 8, characterized in that: The measuring elements include the total pressure sensor and total temperature sensor at the engine inlet, the total pressure sensor and total temperature sensor at the compressor diffuser outlet, the total pressure sensor and total temperature sensor at the turbine outlet, the total pressure sensor and total temperature sensor installed on the outer wall of the tail nozzle, the speed sensor and the flow meter.

10. The fuel control feedback adjustment method for a micro-sized turbojet engine according to claim 8, characterized in that: The oil supply valve group includes a main oil supply valve, a first oil supply valve, a second oil supply valve and a third oil supply valve. The first oil supply valve is connected to the ignition oil circuit, the second oil supply valve is connected to the main oil circuit, the third oil supply valve is connected to the cooling circuit, and the main oil supply valve is connected to the outlet end of the oil pump.