Intelligent control system of miniature turbocharged engine

Abstract

The invention discloses an intelligent control system for a micro turbocharged engine, belongs to the technical field of engines, and solves the technical problems that a turbocharging device and an engine are easy to generate high temperature to cause failure and stop working when working. According to the system, an air inlet pipe is connected with a fuel oil atomization mixing device through a supercharger of a turbocharging device, a heat dissipation air pipe air throttle is installed on the air inlet pipe between the turbocharging device and the fuel oil atomization mixing device, an oil tank is connected with the fuel oil atomization mixing device through an oil supply pipe, and an oil supply valve is installed on the oil supply pipe. The fuel atomization mixing device is connected with the engine through a mixing pipe, and the engine is connected with a turbine of the turbocharging device through an exhaust pipe. The control module calculates and obtains the optimal opening degree of a fuel supply valve, the opening degree of an air throttle of a heat dissipation air pipe, the air-fuel ratio and the fuel flow through an improved SSA-GRU control algorithm according to the fuel temperature, the turbine temperature, the fuel pressure, the rotating speed and a control signal of an upper computer; and the oil supply valve and the engine are controlled to work.

Description

Technical Field

[0001] The present invention relates to the technical field of engines, and more specifically, to an intelligent control system for a micro turbocharged engine. Background Art

[0002] The development of small UAVs has put forward higher requirements on the performance and efficiency of the power system. As the core power source of UAVs, the control system of micro engines directly affects the endurance, fuel efficiency and reliability of the aircraft.

[0003] After the engine is miniaturized, its power will be reduced. Usually, a turbocharger system is equipped to increase the engine's air intake, thereby increasing the engine's power. After an engine is equipped with a turbocharger, its maximum power can be increased by about 40%, and it can improve fuel economy and reduce exhaust emissions. However, when the turbocharger and the engine are working, high temperatures will be generated. When the working temperature of the turbocharger and the micro engine does not match, the working efficiency will be reduced, and even failure and stop working will occur. Summary of the invention

[0004] The technical problem to be solved by the present invention is to address the above-mentioned deficiencies in the prior art, and an object of the present invention is to provide an intelligent control system for a micro turbocharged engine.

[0005] The technical solution of the present invention is: a micro-turbocharged engine intelligent control system, including a turbocharger, an air intake pipe, a heat dissipation air pipe throttle valve, a fuel atomization mixing device, a fuel supply pipe, a fuel supply valve, a fuel tank, a heat dissipation fuel pipe, an engine, a heat dissipation air pipe, a mixing pipe, an exhaust pipe, and a control module;

[0006] The intake pipe is connected to the fuel atomizing and mixing device through the supercharger of the turbocharger, the radiator air pipe throttle valve is installed in the intake pipe between the turbocharger and the fuel atomizing and mixing device, the fuel tank is connected to the fuel atomizing and mixing device through the fuel supply pipe, the fuel supply valve is installed in the fuel supply pipe, the fuel atomizing and mixing device is connected to the engine through the mixing pipe, and the engine is connected to the turbine of the turbocharger through the exhaust pipe;

[0007] The fuel tank is connected to the heat dissipation fuel pipe, a circulation pump is provided on the heat dissipation fuel pipe, and the heat dissipation fuel pipe is sequentially wound around the outer wall of the engine and the turbocharger; the heat dissipation air pipe throttle valve is connected to the heat dissipation air pipe, and the heat dissipation air pipe is sequentially wound around the outer wall of the engine and the turbocharger;

[0008] The control module is electrically connected to the radiator air pipe throttle valve, the fuel supply valve, the circulation pump, and the temperature sensor module, the fuel pressure sensor, and the speed sensor of the engine. The temperature sensor module includes a fuel temperature sensor and a turbine temperature sensor.

[0009] The control module is communicatively connected to the host computer;

[0010] The control module calculates the optimal fuel supply valve opening, cooling air pipe throttle valve opening, air-fuel ratio, and fuel flow rate through an improved SSA-GRU control algorithm according to the fuel temperature, turbine temperature, fuel pressure, rotation speed, and control signal of the host computer; and controls the fuel supply valve and engine operation according to the fuel supply valve opening, air-fuel ratio, and fuel flow rate to achieve the optimal air-fuel ratio in the current state to improve fuel efficiency; and controls the operation of the cooling air pipe throttle valve according to the cooling air pipe throttle valve opening to regulate the engine and the turbocharger to always be in the optimal operating temperature state.

[0011] As a further improvement, the traditional formula of the SSA algorithm in the improved SSA-GRU control algorithm is:

[0012] x i,d =L d +(U d -L d )×r(1)

[0013] Where d is the spatial dimension, x i,d represents the position of the i-th sparrow in the d-th dimension, U d and L d They represent the maximum and minimum values ​​in the d dimension respectively, and r is a random number uniformly distributed in the range [0,1].

[0014] Furthermore, the SSA algorithm uses chaotic mapping when initializing the algorithm. The principle of chaotic mapping is:

[0015]

[0016] Among them, m is an empirical parameter that can be obtained through experiments. The algorithm formula after the chaos mapping improvement is:

[0017] x i ′ ,d =L d +(U d -L d )×x n (3)

[0018] Among them, x n is the nth chaotic variable obtained by chaotic mapping.

[0019] Furthermore, the following improvements are made to the sparrow position formula of the SSA algorithm:

[0020]

[0021] in, represents the spatial position of the i-th sparrow in the n-th generation in the j-th dimension; α is a control parameter; epmax is the maximum number of iterations; μ is a random number between 0 and 1; R2 is the warning threshold; ST is the safety threshold; Q is a random number that obeys the standard normal distribution; L is a 1xj unit matrix; W is a random number that obeys the Weibull distribution, and its shape parameter k is dynamically adjusted with the number of iterations t; σ is a random number with a value of -1 or 1, which is used to change the search direction.

[0022] Furthermore, when using the SSA-GRU control algorithm, first use Formula 3 to initialize the position of the sparrow. The position of the discoverer sparrow represents the position of the food, which may be the desired solution. The position of the discoverer sparrow is updated using Formula 4. Whether to update the current sparrow's position depends on the loss function of the GRU. When the loss function of the GRU is at the minimum, stop updating, otherwise continue to look for the optimal hyperparameters of the GRU.

[0023] Furthermore, the optimal hyperparameters include at least one of a learning rate, a dimension of a hidden vector, and a number of iterations.

[0024] Furthermore, the GRU algorithm uses the sigmoid function in the reset gate and the update gate.

[0025] Furthermore, the GRU algorithm also uses the PRELU function.

[0026] Beneficial Effects

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] 1. The present invention can regulate the engine operating temperature so that the micro engine is always in an optimal operating temperature state.

[0029] 2. The present invention uses an intelligent control algorithm to calculate the optimal air-fuel ratio of the engine, thereby improving the fuel efficiency of the engine and preventing cylinder explosion.

[0030] 3. The present invention uses part of the air from the air inlet to cool the turbocharger, thereby reducing the volume and weight of the cooling device and improving the load-bearing performance and endurance of the UAV.

[0031] 4. The present invention utilizes the high temperature of the turbocharger to heat the fuel, which is beneficial to the atomization and combustion of the fuel, thereby improving the utilization rate of the fuel and reducing emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural schematic diagram of the present invention;

[0033] Figure 2 This is the wiring schematic diagram of the control module in the present invention;

[0034] Figure 3 is a control flow chart of the present invention;

[0035] Figure 4 It is a control flow chart of the valve in the present invention.

[0036] Among them: 1-turbocharger, 2-engine intake pipe, 3-cooling air pipe throttle valve, 4-fuel atomization mixing device, 5-fuel supply pipe, 6-fuel supply valve, 7-fuel tank, 8-cooling fuel pipe, 9-engine, 10-cooling air pipe, 11-mixing pipe, 12-exhaust pipe. DETAILED DESCRIPTION

[0037] The present invention will be further described below with reference to the specific embodiments in the accompanying drawings.

[0038] See also Figure 1 to Figure 4 A micro-turbocharged engine intelligent control system, characterized in that it includes a turbocharger 1, an intake pipe 2, a heat dissipation air pipe throttle valve 3, a fuel atomization mixing device 4, a fuel supply pipe 5, a fuel supply valve 6, a fuel tank 7, a heat dissipation fuel pipe 8, an engine 9, a heat dissipation air pipe 10, a mixing pipe 11, an exhaust pipe 12, and a control module.

[0039] The intake pipe 2 is connected to the fuel atomizing mixing device 4 through the supercharger of the turbocharger 1, the heat dissipation air pipe throttle valve 3 is installed in the intake pipe 2 between the turbocharger 1 and the fuel atomizing mixing device 4, the fuel tank 7 is connected to the fuel atomizing mixing device 4 through the fuel supply pipe 5, the fuel supply valve 6 is installed in the fuel supply pipe 5, the fuel atomizing mixing device 4 is connected to the engine 9 through the mixing pipe 11, and the engine 9 is connected to the turbine of the turbocharger 1 through the exhaust pipe 12.

[0040] The fuel tank 7 is connected to the heat dissipation fuel pipe 8, and a circulation pump (not shown) is provided on the heat dissipation fuel pipe 8. The heat dissipation fuel pipe 8 is sequentially wound around the outer wall of the engine 9 and the turbocharger 1. Under the action of the circulation pump, the heat dissipation fuel pipe 8 in the fuel tank 7 circulates between the fuel tank 7 and the fuel tank 7, and the engine 9 and the turbocharger 1 can heat the fuel.

[0041] The heat dissipation pipe throttle valve 3 is connected to the heat dissipation pipe 10, and the heat dissipation pipe 10 is wound around the outer wall of the engine 9 and the turbocharger 1 in sequence. When the engine is working, it drives the turbocharger to inhale air. When the engine temperature rises, the heat dissipation pipe throttle valve will open, and part of the compressed air of the supercharger will be discharged after passing through the heat dissipation pipe 10. The relatively low temperature air takes away the heat, which can cool down the engine 9 and the turbocharger 1.

[0042] It can suppress the excessive temperature of the engine and turbocharger, and avoid the installation of cooling fans to reduce the weight of the drone and save space. The heated fuel can be more easily atomized and fully mixed with the air, thereby improving the combustion efficiency of the fuel.

[0043] The control module is electrically connected to the heat dissipation pipe throttle valve 3, the fuel supply valve 6, the circulation pump, and the temperature sensor module, the fuel pressure sensor, and the speed sensor of the engine 9. The temperature sensor module includes a fuel temperature sensor and a turbine temperature sensor.

[0044] The control module is connected to the host computer through communication, and the host computer controls the operation of the engine 9, such as acceleration and deceleration, by sending control signals.

[0045] The control module calculates the optimal fuel supply valve opening, radiator air pipe throttle valve opening, air-fuel ratio, and fuel flow rate according to the fuel temperature, turbine temperature, fuel pressure, speed, and control signal of the host computer through the improved SSA-GRU (Sparrow Algorithm Optimized Gated Cycle Unit) control algorithm; and controls the fuel supply valve 6 and engine 9 to achieve the optimal air-fuel ratio in the current state to improve fuel efficiency; controls the radiator air pipe throttle valve 3 to operate according to the radiator air pipe throttle valve opening to regulate the engine 9 and the turbocharger 1 to always be in the optimal operating temperature state.

[0046] The traditional formula of the SSA algorithm in the improved SSA-GRU control algorithm is:

[0047] x i,d =L d +(U d -L d )×r(1)

[0048] Where d is the spatial dimension, x i,d represents the position of the i-th sparrow in the d-th dimension, U d and L d They represent the maximum and minimum values ​​in the d dimension respectively, and r is a random number uniformly distributed in the range [0,1].

[0049] The SSA (Sparrow Algorithm Optimization) algorithm uses chaotic mapping when initializing the algorithm. The principle of chaotic mapping is:

[0050]

[0051] Among them, m is an empirical parameter that can be obtained through experiments. The algorithm formula after the chaos mapping improvement is:

[0052] x i ′ ,d=L d +(U d -L d )×x n (3)

[0053] Among them, x n is the nth chaotic variable obtained by chaotic mapping.

[0054] The traditional SSA algorithm is prone to fall into a local optimal solution. In order to solve this problem, the following improvements are made to the sparrow position formula of the SSA algorithm:

[0055] The improved formula is:

[0056] (The original formula is: )

[0057] in, represents the spatial position of the i-th sparrow in the n-th generation in the j-th dimension; α is a control parameter; epmax is the maximum number of iterations; μ is a random number between 0 and 1; R2 is the warning threshold; ST is the safety threshold; Q is a random number that obeys the standard normal distribution; L is a 1xj unit matrix; W is a random number that obeys the Weibull distribution, and its shape parameter k is dynamically adjusted with the number of iterations t; σ is a random number with a value of -1 or 1, which is used to change the search direction.

[0058] When using the SSA-GRU control algorithm, first use Formula 3 to initialize the position of the sparrow. The position of the discoverer sparrow represents the position of the food, which may be the desired solution. The position of the discoverer sparrow is updated using Formula 4. Whether to update the current sparrow's position depends on the loss function of the GRU. When the loss function of the GRU is at the minimum, stop updating. Otherwise, continue to find the best hyperparameters of the GRU. The best hyperparameters include at least one of the learning rate, the dimension of the hidden vector, and the number of iterations.

[0059] GRU (Gated Recurrent Unit) is an improved recurrent neural network, which has the advantages of low computational complexity and saving computational resources, and can solve the problems of gradient vanishing and gradient exploding in traditional algorithms. The hyperparameters of the algorithm are found by the SSA optimization algorithm. Compared with the traditional manual experience, the hyperparameters found by SSA are more suitable for the GRU algorithm running under various working conditions, ensuring its optimal performance. The algorithm uses the sigmoid function in the reset gate and the update gate. In order to further optimize its computational efficiency, the present disclosure uses the PRELU function, which simplifies the calculation while retaining the advantages of the sigmoid function, and does not require the use of exponential operations, making the calculation more efficient.

[0060] In the GRU algorithm part, Xt is the input parameter, including temperature, pressure, speed data and the control signal of the host computer. h0 can be 0 by default. After GRU operation, the optimal valve opening, air-fuel ratio, and fuel flow are output.

[0061] The above are only preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the structure of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A micro turbocharger engine intelligent control system, characterized in that: It comprises a turbocharger (1), an air intake pipe (2), a heat dissipation pipe throttle valve (3), a fuel atomization mixing device (4), a fuel supply pipe (5), a fuel supply valve (6), a fuel tank (7), a heat dissipation fuel pipe (8), an engine (9), a heat dissipation pipe (10), a mixing pipe (11), an exhaust pipe (12), and a control module; The intake pipe (2) is connected to the fuel atomizing mixing device (4) through the supercharger of the turbocharger (1); the heat dissipation pipe throttle valve (3) is installed on the intake pipe (2) between the turbocharger (1) and the fuel atomizing mixing device (4); the fuel tank (7) is connected to the fuel atomizing mixing device (4) through the fuel supply pipe (5); the fuel supply valve (6) is installed on the fuel supply pipe (5); the fuel atomizing mixing device (4) is connected to the engine (9) through the mixing pipe (11); and the engine (9) is connected to the turbine of the turbocharger (1) through the exhaust pipe (12); The oil tank (7) is connected to the heat dissipation fuel pipe (8), a circulation pump is provided on the heat dissipation fuel pipe (8), and the heat dissipation fuel pipe (8) is sequentially wound around the outer walls of the engine (9) and the turbocharger (1); the heat dissipation air pipe throttle valve (3) is connected to the heat dissipation air pipe (10), and the heat dissipation air pipe (10) is sequentially wound around the outer walls of the engine (9) and the turbocharger (1); The control module is electrically connected to the heat dissipation air pipe throttle valve (3), the fuel supply valve (6), the circulation pump, and the temperature sensor module, the fuel pressure sensor, and the rotation speed sensor of the engine (9), wherein the temperature sensor module includes a fuel temperature sensor and a turbine temperature sensor; The control module is communicatively connected to the host computer; The control module calculates the optimal fuel supply valve opening, heat dissipation pipe throttle valve opening, air-fuel ratio and fuel flow rate through an improved SSA-GRU control algorithm according to the fuel temperature, turbine temperature, fuel pressure, rotation speed and control signal of the upper computer; and controls the fuel supply valve (6) and the engine (9) to achieve the optimal air-fuel ratio in the current state to improve fuel efficiency; and controls the operation of the heat dissipation pipe throttle valve (3) according to the heat dissipation pipe throttle valve opening to regulate the engine (9) and the turbocharger (1) to always be in the optimal operating temperature state.

2. The micro-turbocharged engine intelligent control system according to claim 1, characterized in that: The traditional formula of the SSA algorithm in the improved SSA-GRU control algorithm is: x i,d =L d +(U d -L d )×r(1) Where d is the spatial dimension, x i,d represents the position of the i-th sparrow in the d-th dimension, U d and L d They represent the maximum and minimum values ​​in the d dimension respectively, and r is a random number uniformly distributed in the range [0,1].

3. The micro-turbocharged engine intelligent control system according to claim 2, characterized in that: The SSA algorithm uses chaotic mapping when initializing the algorithm. The principle of chaotic mapping is: Among them, m is an empirical parameter that can be obtained through experiments. The algorithm formula after the chaos mapping improvement is: x i ′ ,d =L d +(U d -L d )×x n (3) Among them, x n is the nth chaotic variable obtained by chaotic mapping.

4. The micro-turbocharged engine intelligent control system according to claim 3, characterized in that: The following improvements are made to the sparrow position formula of the SSA algorithm: in, represents the spatial position of the i-th sparrow in the n-th generation in the j-th dimension; α is a control parameter; epmax is the maximum number of iterations; μ is a random number between 0 and 1; R2 is the warning threshold; ST is the safety threshold; Q is a random number that obeys the standard normal distribution; L is a 1xj unit matrix; W is a random number that obeys the Weibull distribution, and its shape parameter k is dynamically adjusted with the number of iterations t; σ is a random number with a value of -1 or 1, which is used to change the search direction.

5. The micro-turbocharged engine intelligent control system according to claim 4, characterized in that: When using the SSA-GRU control algorithm, first use Formula 3 to initialize the position of the sparrow. The position of the discoverer sparrow represents the position of the food, which may be the desired solution. The position of the discoverer sparrow is updated using Formula 4. Whether to update the current sparrow's position depends on the loss function of the GRU. When the loss function of the GRU is at the minimum, stop updating, otherwise continue to look for the best hyperparameters of the GRU.

6. The micro-turbocharged engine intelligent control system according to claim 5, characterized in that: The optimal hyperparameters include at least one of a learning rate, a dimension of a hidden vector, and a number of iterations.

7. The micro-turbocharged engine intelligent control system according to claim 5, characterized in that: The GRU algorithm uses the sigmoid function in the reset gate and the update gate.

8. The micro-turbocharged engine intelligent control system according to claim 7, characterized in that: The GRU algorithm also uses the PRELU function.

Images