An aircraft electric propulsion system

Abstract

The present invention discloses an electric propulsion system for an aircraft, and relates to the field of aerospace technology. Compared with previous electric propulsion systems for aircraft, the present invention solves the problem that the distributed electric propulsion systems of existing aircraft basically use radial magnetic motors without reasonable motor matching, resulting in an increase in the weight of the electric power system; and the additional packaging structure and the use of longer wires to transmit electrical energy further increase the weight of the power system; it provides an electric aircraft distributed electric power system layout, in which the electric power system that provides propulsion power uses radial magnetic motors, and the electric power system that provides lift uses axial magnetic motors. By utilizing the advantages of different types of motors, the total weight of the power system is reduced. The integrated design of the motor controller and the motor further improves the integration of the aircraft and the optimization of the internal structure of the aircraft, thereby improving the endurance performance and reliability of the aircraft, and providing strong support for applications in the aircraft field.

Description

Technical Field

[0001] The present invention relates to the field of aerospace technology, and in particular to an aircraft electric propulsion system. Background Art

[0002] As the energy density of power batteries increases, electric propulsion systems are increasingly being used in transportation. The automotive industry has seen the most rapid and successful adoption. Similarly, the use of electric propulsion systems in aircraft is also increasing. Currently, electric aircraft are rapidly emerging and developing both domestically and internationally.

[0003] Currently, electric propulsion systems are typically deployed in a distributed configuration on aircraft, with a typical application being eVTOL (evolved vertically-outlined vehicles). These aircraft typically use a longitudinal arrangement of electric propulsion systems, typically fore and aft, to provide propulsion. These systems typically have the propellers' rotational plane perpendicular to the aircraft's flight direction. Furthermore, multiple electric propulsion systems are arranged in a regular pattern within the plane of the aircraft's wing to provide lift. These systems typically have the propellers' rotational plane parallel to the wing or perpendicular to the direction of gravity.

[0004] All distributed electric propulsion systems for aircraft models disclosed in prior art essentially utilize radial magnet motors, and most utilize independent motor controller designs. Independent motor controllers require additional packaging, increasing the weight of the power system. Furthermore, the motor controllers are located far from the motors, requiring long wires to transmit power, further increasing the weight of the power system. All existing distributed electric propulsion systems for aircraft utilize radial magnet motors, without proper motor coordination, resulting in increased weight.

[0005] In order to solve the above problems, the present invention proposes an aircraft electric propulsion system. Summary of the Invention

[0006] The purpose of the present invention is to provide an aircraft electric propulsion system to solve the problems raised in the background technology:

[0007] The distributed electric propulsion systems of existing aircraft basically use radial magnetic motors, which are not reasonably matched with the motors, resulting in an increase in the weight of the electric power system; and the motor controllers that mostly use independent design solutions require additional packaging structures, which increases the gravity of the power system; and the motor controllers are far away from the motors, requiring longer wires to transmit electrical energy, further increasing the weight of the power system.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] An aircraft electric propulsion system includes: a DC battery, a power distribution module, a flight control system, lift propellers arranged axially at four vertices in a rectangular shape, and a propeller arranged radially at the top. The DC battery is intelligently managed by a BMS system and also reduces the output voltage through DC / DC conversion to provide power for the flight control system, which is used to control the operation of the lift propellers and propellers. The DC battery also provides operating voltage for the four lift propellers and the propeller through the power distribution module.

[0010] Preferably, the voltage range of the DC battery is 600-800V.

[0011] Preferably, the four lift propellers all adopt electrically controlled integrated axial magnetic motors, and the propulsion propeller adopts an electrically controlled integrated radial magnetic motor; the magnetic flux direction of the electrically controlled integrated axial magnetic motor is parallel to the motor rotation axis, and the magnetic flux direction of the electrically controlled integrated radial magnetic motor is distributed along the radial direction of the motor.

[0012] Preferably, the electrically controlled integrated axial magnetic motor and the electrically controlled integrated radial magnetic motor both adopt an integrated motor controller and motor design, and are cooled by water cooling or air cooling.

[0013] Preferably, the motor controller and motor integrated design solution also integrates a wireless communication model to enable the system to communicate wirelessly with a ground monitoring center.

[0014] Preferably, the integrated design of the motor controller and the motor uses carbon fiber composite materials to make an integrated housing, and is also provided with heat dissipation channels and heat dissipation fins made of phase change materials or nanomaterials; and electromagnetic shielding technology is also used to reduce the impact of electromagnetic interference on the equipment.

[0015] Preferably, the structure of the heat dissipation fins includes wavy heat dissipation fins, honeycomb heat dissipation fins, spiral heat dissipation fins, hollow heat dissipation fins, multi-layer three-dimensional heat dissipation fins, curved three-dimensional heat dissipation fins, composite structure heat dissipation fins, deformable heat dissipation fins, adjustable heat dissipation fins, bionic heat dissipation fins, and modular heat dissipation fins; the heat dissipation fins are also prepared in combination with 3D printing technology.

[0016] Preferably, the motor controller and motor integrated design solution also integrates sensors, power modules and energy storage systems.

[0017] Preferably, the motor controller and motor integrated design solution further provides an automatic intelligent control method according to the operating state of the motor and the external environmental conditions, and the automatic intelligent control method automatically adjusts the control parameters according to the real-time operating state of the motor and the real-time external environmental conditions;

[0018] The automatic intelligent control method is specifically as follows:

[0019] S1: Real-time collection of the operating status and external environmental conditions of the motor controller and motor, and pre-processing and storage of data based on big data technology;

[0020] S2: Control simulation of the motor controller and motor integrated structure based on neural networks, as well as optimization and adjustment of layout and wiring;

[0021] The neural network takes the pre-processed motor controller and motor operating status and external environmental conditions as input to build a BiLSTM-FCN network. The BiLSTM-FCN network builds a parallel improved BiLSTM network and FCN network respectively. Finally, the outputs of the improved BiLSTM network and the FCN network are connected through a connection function, and the control results are output through the connection layer and the output layer.

[0022] The output of the improved BiLSTM network is as follows:

[0023]

[0024] in, and are the output results of the forward layer at time t and time t-1 respectively; and are the output results of the backward layer at time t and time t-1 respectively; and are the weight coefficients between the input layer and the forward layer and the backward layer respectively. is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t; is the weight coefficient between the backward propagation unit at time t-1 and the backward propagation unit at time t; and are the weight coefficients between the forward layer, backward layer and output layer respectively; f(·) is the internal calculation function of the neuron; g(·) is the function for calculating the total result of the forward layer and backward layer; y t is the output result of the BiLSTM network at time t;

[0025] The improved BiLSTM network also incorporates an attention mechanism into each propagation unit, as follows:

[0026] Atten t =αtanh(Wx t +B)

[0027]

[0028] Among them, Atten t is the value of the attention probability distribution at time t; α, W and b are model learning parameters; wt is the attention weight; τ is the dimension of the model input vector; h t is the output at time t;

[0029] The FCN network contains three temporal convolution blocks, each with 128, 256, and 128 filters respectively. Each convolution layer is activated by a ReLU activation function, and then connected to a global pooling layer for output. The convolution kernel of the FCN network is a sliding window on the time series to extract the short-term features of the time series. The update is as follows:

[0030]

[0031] Among them, z i and z j are the i-th and j-th feature maps respectively; σ is the activation function; ω ij and b are model learning parameters;

[0032] The BiLSTM-FCN network also learns model parameters based on the harmony search algorithm, as follows:

[0033] The control balance degree U={u1,u2,Λ,u M} is used as the optimization target, M is the total number of control signal combinations of the motor and the motor controller, and the relevant parameters are initialized;

[0034] u ab =u min +(u max -u min )×r1

[0035] Among them, u ab Initialize the harmony vector of the a-th motor signal and the control signal of the b-th motor controller, that is, the combination vector of the a-th motor signal and the control signal of the b-th motor controller; u max and u min are the upper and lower limits of the solution; r1 is a random number between (0,1);

[0036] Initialize the harmony memory HM and randomly generate m harmonics in the search domain using the above formula, i.e., the combination of the motor and the motor controller's control signal, and add them to the harmony memory HM represented by the matrix:

[0037]

[0038] Among them, F[·] is the harmony measurement function, that is, the control signal combination effect measurement function;

[0039] The harmony vector u before the k-th iteration update ab (k) Update as follows:

[0040]

[0041] BW(k)=BW min +(BW max -BW min )×r7

[0042] Among them, r2, r3, r4, r5, r6 and r7 are all random numbers between (0,1); is the randomly generated harmony vector, i.e., the control signal combination; u new,ab (k) is the harmony vector after the k-th iteration update; BW(k) is the harmony bandwidth of the k-th iteration, that is, the control signal combination limit; BW max and BW min are the maximum and minimum values ​​of the harmonic bandwidth respectively; p1 is the probability of storing the harmonic memory bank, that is, the probability of storing the control signal combination; p2 is the probability of gene fine-tuning, that is, the probability of fine-tuning the control signal combination;

[0043] The updated harmony vector u new,ab and the worst harmony vector u in the initial harmony memory HM worst The harmony metric function value is compared, if u new,ab The metric function value of (k) is less than u worst The metric function value is u new,ab (k) Replace u worst ;Otherwise, the harmony in HM remains unchanged;

[0044] Check whether the algorithm termination condition is met; if so, output the global optimal solution and the algorithm ends; otherwise, return to continue updating the harmony vector;

[0045] S3: performing predictive diagnostic maintenance on the motor and controller based on the neural network;

[0046] S4: Regularly test and verify the motor controller and motor integration structure.

[0047] Compared with the prior art, the present invention provides an aircraft electric propulsion system with the following beneficial effects:

[0048] This invention provides a distributed electric power system layout for an electric aircraft. The electric power system for propulsion uses radial magnetic motors, while the electric power system for lift uses axial magnetic motors. By optimally utilizing the advantages of different motor types, the overall weight of the power system is reduced. Furthermore, the motors utilize an integrated motor controller and motor design, which reduces wiring length and further reduces the overall weight of the power system. The integrated motor controller and motor design further enhances the aircraft's integration and optimizes its internal structure, improving its endurance and reliability, and providing strong support for applications in the aircraft field. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a schematic diagram of the system architecture mentioned in Example 1 of the present invention;

[0050] Figure 2 A schematic diagram comparing cross sections of the radial flux motor and the axial flux motor mentioned in Example 1 of the present invention;

[0051] Figure 3 This is a schematic diagram showing data comparison between the radial flux motor and the axial flux motor mentioned in Example 1 of the present invention;

[0052] Figure 4 This is a structural diagram of the heat dissipation fin mentioned in Example 1 of the present invention;

[0053] Figure 5 This is a schematic structural diagram of the second heat sink fin mentioned in Example 1 of the present invention;

[0054] Figure 6 This is a schematic diagram of the third structure of the heat dissipation fin mentioned in Example 1 of the present invention;

[0055] Figure 7 This is a schematic diagram showing how the cruising range mentioned in Example 1 of the present invention changes with aircraft weight;

[0056] Figure 8 This is a schematic diagram showing how the flight time mentioned in Example 1 of the present invention changes with the weight of the aircraft.

[0057] Meaning of the marks in the figure:

[0058] 1. DC battery; 2. Power distribution module; 3. Flight control system; 4. Lift propeller; 5. Propeller propeller. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0060] The present invention provides a distributed electric power system layout for an electric aircraft. The electric power system providing propulsion uses a radial magnetic motor, while the electric power system providing lift uses an axial magnetic motor. By optimally utilizing the advantages of different motor types, the total weight of the power system is reduced. Furthermore, the motor utilizes an integrated motor controller and motor solution, which can reduce the length of the wires, thereby further reducing the total weight of the power system. The integrated motor controller and motor design further enhances the aircraft's integration and optimizes its internal structure, improving the aircraft's endurance performance and reliability, and providing strong support for applications in the aircraft field. Specifically, the following content is included.

[0061] Example 1:

[0062] See also Figure 1-2 The present invention discloses an aircraft electric propulsion system comprising a DC battery 1, a power distribution module 2, a flight control system 3, lift propellers 4 arranged axially at four vertices in a rectangular shape, and a propeller 5 radially arranged at the top. The DC battery 1 is intelligently managed by a battery management system (BMS). It also reduces its output voltage through DC / DC conversion to provide power to the flight control system 3, which controls the operation of the lift propellers 4 and propeller 5. The DC battery 1 also provides operating voltage to the four lift propellers 4 and one propeller 5 through the power distribution module 2. The DC battery 1 has a voltage range of 600V to 800V. The BMS intelligently manages and maintains each cell within the battery, preventing overcharging and overdischarging, extending the battery's service life, and monitoring its status.

[0063] The four lift propellers 4 all utilize electronically controlled integrated axial magnetic motors, while the thrust propellers 5 utilize electronically controlled integrated radial magnetic motors. The magnetic flux direction of the electronically controlled integrated axial magnetic motor is parallel to the motor's rotation axis, while the magnetic flux direction of the electronically controlled integrated radial magnetic motor is distributed radially. Both the electronically controlled integrated axial magnetic motor and the electronically controlled integrated radial magnetic motor utilize an integrated motor controller and motor design. Water cooling is preferred, but air cooling is also acceptable.

[0064] Existing electric aircraft, particularly eVTOLs, generally adopt a combination of fixed-wing and multi-rotor configurations. Most of these propulsion systems utilize a distributed electric propulsion system, one portion of which provides the thrust or pull required for fixed-wing cruise flight, while the other provides the lift required for vertical takeoff and landing (VTOL). For these aircraft, fixed-wing cruise flight features long operating times, typically ranging from 30 minutes to several hours, with relatively low power and torque. Multi-rotor VTOL flight features shorter operating times, typically around 5 minutes, with relatively high power and torque. For fixed-wing cruise flight, the long operating times place primary demands on the powertrain motors for high durability, reliability, and thermal stability. For multi-rotor VTOL flight, the shorter operating times place primary demands on the motors for high peak power, high peak torque, a high power-to-weight ratio, and efficient heat dissipation.

[0065] Reference Figure 2 In this embodiment, the motor is divided into radial flux motor and axial flux motor. The magnetic flux direction of the radial flux motor is distributed along the radial direction of the motor, and the magnetic flux direction of the axial flux motor is parallel to the motor rotation axis. The data of the radial flux motor and the axial flux motor are compared, and the comparison results can be referred to Figure 3 , compare radial flux motors and axial flux motors. Radial flux motors have a century-long application history, and their design methods and production technologies are mature. They are characterized by low cost and high reliability, and are suitable for long-term operation. Axial flux motors have shorter and more direct flux paths, which help maintain a strong magnetic field, which helps to improve the torque, power density and efficiency of the motor. The flux path of the axial flux motor is unidirectional, and crystal-oriented steel can be used to reduce iron loss and further improve efficiency. The windings of the axial flux motor can be in direct contact with the casing, and the casing can be made of materials with higher thermal conductivity, thereby improving the cooling efficiency of the motor.

[0066] When applied to electric aircraft, the electric propulsion system provided in this embodiment, particularly the distributed electric propulsion system layout for eVTOL aircraft, utilizes radial flux motors for fixed-wing cruising flight and axial flux motors for multi-rotor vertical takeoff and landing (VTOL) flight. The electronic controls for both motors are integrated within the motors. This integration of the control and motors reduces wiring length and cooling requirements, thereby reducing the weight of the electric propulsion system.

[0067] The distributed electric power system layout technology solution for electric aircraft provided in this embodiment can optimally utilize the advantages of different motor types and reduce the overall weight of the power system. Furthermore, the motors used in this embodiment are integrated with motor controllers, which can reduce wire lengths and further reduce the overall weight of the power system.

[0068] The integrated design of the motor controller and motor uses carbon fiber composite materials to make an integrated shell. The use of carbon fiber composite materials, a lightweight structural parameter, can further reduce the weight of the entire system, improve the fuel efficiency of the aircraft or increase the cruising range.

[0069] It is also provided with heat dissipation channels and heat dissipation fins made of phase change materials or nanomaterials; the structures of the heat dissipation fins include wavy heat dissipation fins, honeycomb heat dissipation fins, spiral heat dissipation fins, hollow heat dissipation fins, multi-layer three-dimensional heat dissipation fins, curved three-dimensional heat dissipation fins, composite structure heat dissipation fins, deformable heat dissipation fins, adjustable heat dissipation fins, bionic heat dissipation fins, modular heat dissipation fins; it can also be prepared in combination with 3D printing technology. The structure of some heat dissipation fins can be referred to Figure 4-6 The shapes of the heat dissipation fins can include special-shaped structures, three-dimensional structures, composite structures and adjustable structures, etc., which are mainly used to improve heat dissipation efficiency, reduce volume and weight, enhance stability and reliability, and provide better heat dissipation solutions for the integrated design of motor controller and motor.

[0070] Electromagnetic shielding technology is also used to reduce the impact of electromagnetic interference on equipment and improve the system's electromagnetic compatibility. Using electromagnetic shielding technology can reduce the impact of electromagnetic interference on other electronic equipment and improve aircraft safety and reliability.

[0071] The integrated design of the motor controller and motor also integrates sensors, power modules and energy storage systems, integrating temperature sensors, current sensors, voltage sensors, etc. into the motor controller to achieve real-time monitoring of the motor and controller. These sensors can provide accurate operating parameters to help the intelligent control system achieve more precise control and predictive maintenance. Integrating power modules (such as IGBTs, MOSFETs, etc.) with the motor controller can reduce the size and weight of the system. At the same time, the integrated power module can improve the reliability and efficiency of the system and reduce costs. The energy storage system (such as batteries, supercapacitors, etc.) is also integrated with the motor controller and motor to form a complete electric propulsion system. This can improve the energy density and cruising range of the system, while also simplifying the structure and installation of the system.

[0072] The integrated motor controller and motor design also features an automatic intelligent control method based on the motor's operating status and external environmental conditions. The automatic intelligent control method automatically adjusts control parameters based on the motor's real-time operating status and external environmental conditions. The details are as follows:

[0073] S1: Real-time collection of the operating status and external environmental conditions of the motor controller and motor, and pre-processing and storage of data based on big data technology;

[0074] S2: Control simulation of the motor controller and motor integrated structure based on neural networks, as well as optimization and adjustment of layout and wiring;

[0075] The neural network takes the preprocessed motor controller and motor operating status and external environmental conditions as input to build a BiLSTM-FCN network. The BiLSTM-FCN network then builds a parallel improved BiLSTM network and FCN network. Finally, the outputs of the improved BiLSTM network and FCN network are connected through a connection function, and the control results are output through the connection layer and output layer.

[0076] The output of the improved BiLSTM network is as follows:

[0077]

[0078] in, and are the output results of the forward layer at time t and time t-1 respectively; and are the output results of the backward layer at time t and time t-1 respectively; and ωσ1 are the weight coefficients between the input layer and the forward layer and the backward layer respectively; is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t; is the weight coefficient between the backward propagation unit at time t-1 and the backward propagation unit at time t; and are the weight coefficients between the forward layer, backward layer and output layer respectively; f(·) is the internal calculation function of the neuron; g(·) is the function for calculating the total result of the forward layer and backward layer; y t is the output result of the BiLSTM network at time t;

[0079] The attention mechanism is also integrated into each propagation unit of the improved BiLSTM network, as follows:

[0080] Atten t =αtanh(Wx t +B)

[0081]

[0082] Among them, Atten t is the value of the attention probability distribution at time t; α, W and b are model learning parameters; w t is the attention weight; τ is the dimension of the model input vector; h t is the output at time t;

[0083] The FCN network consists of three temporal convolution blocks, each with 128, 256, and 128 filters, respectively. Each convolution layer is activated by a ReLU activation function, and then connected to a global pooling layer for output. The convolution kernel of the FCN network is a sliding window on the time series to extract the short-term features of the time series. The update is as follows:

[0084]

[0085] Among them, z i and z j are the i-th and j-th feature maps respectively; σ is the activation function; ω ij and b are model learning parameters;

[0086] The BiLSTM-FCN network also learns model parameters based on the harmony search algorithm, as follows:

[0087] The control balance degree U={u1,u2,Λ,u M} is the optimization target, M is the total number of control signal combinations of the motor and the motor controller, and the relevant parameters are initialized;

[0088] u ab =u min +(u max -u min )×r1

[0089] Among them, u ab Initialize the harmony vector of the a-th motor signal and the control signal of the b-th motor controller, that is, the combination vector of the a-th motor signal and the control signal of the b-th motor controller; u max and u min are the upper and lower limits of the solution; r1 is a random number between (0,1);

[0090] Initialize the harmony memory HM and randomly generate m harmonics in the search domain using the above formula, i.e., the combination of the motor and the motor controller's control signal, and add them to the harmony memory HM represented by the matrix:

[0091]

[0092] Among them, F[·] is the harmony measurement function, that is, the measurement function of the control signal combination effect;

[0093] The harmony vector u before the k-th iteration update ab (k) Update as follows:

[0094]

[0095] BW(k)=BW min +(BWmax -BW min )×r7

[0096] Among them, r2, r3, r4, r5, r6 and r7 are all random numbers between (0, 1); ur0 is a randomly generated harmony vector, that is, a control signal combination; u new,ab (k) is the harmony vector after the k-th iteration update; BW(k) is the harmony bandwidth of the k-th iteration, that is, the control signal combination limit; BW max and BW min are the maximum and minimum values ​​of the harmonic bandwidth respectively; p1 is the probability of storing the harmonic memory bank, that is, the probability of storing the control signal combination; p2 is the probability of gene fine-tuning, that is, the probability of fine-tuning the control signal combination;

[0097] The updated harmony vector u new,ab and the worst harmony vector u in the initial harmony memory HM worst The harmony metric function value is compared, if u new,ab The metric function value of (k) is less than u worst The metric function value is u new,ab (k) Replace u worst ;Otherwise, the harmony in HM remains unchanged;

[0098] Check whether the algorithm termination condition is met. If so, output the global optimal solution and the algorithm ends; otherwise, return to continue updating the harmony vector;

[0099] S3: Predictive diagnostic maintenance of motors and controllers based on neural networks;

[0100] S4: Regularly test and verify the motor controller and motor integration structure.

[0101] The propulsion of the aircraft is controlled based on the above automatic intelligent control method. The specific test results can be referred to Figure 7-8 In the figure, curve 1 shows the change in the cruising range of the aircraft controlled by the above-mentioned automatic intelligent control method, and curve 2 shows the change in the cruising range of the aircraft not controlled by the above-mentioned automatic intelligent control method. As can be seen from the figure, as the weight of the aircraft increases, its cruising range decreases and its flight time also decreases. The strategy based on this embodiment can reduce the weight of the aircraft and effectively improve the cruising range and flight time of the aircraft. In addition, the aircraft optimized and controlled based on the above-mentioned automatic intelligent control method can further improve the cruising range and flight time of the aircraft, thereby improving the navigation efficiency of the aircraft.

[0102] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An aircraft electric propulsion system, characterized in that: include: A DC battery (1), a power distribution module (2), a flight control system (3), lift propellers (4) axially arranged at four vertices in a rectangular shape, and a propulsion propeller (5) radially arranged at the top; the DC battery (1) is intelligently managed by a BMS system and also reduces the output voltage through DC / DC conversion to provide power for the flight control system (3), and the flight control system (3) is used to control the operation of the lift propellers (4) and the propulsion propeller (5); the DC battery (1) also provides operating voltage for the four lift propellers (4) and the propulsion propeller (5) through the power distribution module (2); The four lift propellers (4) all adopt an electrically controlled integrated axial magnetic motor, and the propulsion propeller (5) adopts an electrically controlled integrated radial magnetic motor; the magnetic flux direction of the electrically controlled integrated axial magnetic motor is parallel to the motor rotation axis, and the magnetic flux direction of the electrically controlled integrated radial magnetic motor is distributed along the radial direction of the motor; The electronically controlled integrated axial magnetic motor and the electronically controlled integrated radial magnetic motor both adopt an integrated motor controller and motor design, and are cooled by water or air; The motor controller and motor integrated design solution also sets an automatic intelligent control method according to the operating state of the motor and external environmental conditions, and the automatic intelligent control method automatically adjusts the control parameters according to the real-time operating state of the motor and the real-time external environmental conditions; The automatic intelligent control method is specifically as follows: S1: Real-time collection of the operating status and external environmental conditions of the motor controller and motor, and pre-processing and storage of data based on big data technology; S2: Control simulation of the motor controller and motor integrated structure based on neural networks, as well as optimization and adjustment of layout and wiring; The neural network takes the pre-processed motor controller and motor operating status and external environmental conditions as input to build a BiLSTM-FCN network. The BiLSTM-FCN network builds a parallel improved BiLSTM network and FCN network respectively. Finally, the outputs of the improved BiLSTM network and the FCN network are connected through a connection function, and the control results are output through the connection layer and the output layer. The output of the improved BiLSTM network is as follows: in, and are the output results of the forward layer at time t and time t-1 respectively; and are the output results of the backward layer at time t and time t-1 respectively; and are the weight coefficients between the input layer and the forward layer and the backward layer respectively; is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t; is the weight coefficient between the backward propagation unit at time t-1 and the backward propagation unit at time t; and are the weight coefficients between the forward layer, backward layer and output layer respectively; f(·) is the internal calculation function of the neuron; g(·) is the function for calculating the total result of the forward layer and backward layer; y t is the output result of the BiLSTM network at time t; The improved BiLSTM network also incorporates an attention mechanism into each propagation unit, as follows: Atten t =αtanh(Wx t +B) Among them, Atten t is the value of the attention probability distribution at time t; α, W and b are model learning parameters; w t is the attention weight; τ is the dimension of the model input vector; h t is the output at time t; The FCN network contains three temporal convolution blocks, each with 128, 256, and 128 filters respectively. Each convolution layer is activated by a ReLU activation function, and then connected to a global pooling layer for output. The convolution kernel of the FCN network is a sliding window on the time series to extract the short-term features of the time series. The update is as follows: Among them, z i and z j are the i-th and j-th feature maps respectively; σ is the activation function; ω ij and b are model learning parameters; The BiLSTM-FCN network also learns model parameters based on the harmony search algorithm, as follows: Control the balance degree U={u1,u2,...,u M } is the optimization target, M is the total number of control signal combinations of the motor and the motor controller, and the relevant parameters are initialized; in ab =in min +(in max -in min )×r1 Among them, u ab Initialize the harmony vector of the a-th motor signal and the control signal of the b-th motor controller, that is, the combination vector of the a-th motor signal and the control signal of the b-th motor controller; u max and u min are the upper and lower limits of the solution; r1 is a random number between (0,1); Initialize the harmony memory HM and randomly generate m harmonics in the search domain using the above formula, i.e., the combination of the motor and the motor controller's control signal, and add them to the harmony memory HM represented by the matrix: Among them, F[·] is the harmony measurement function, that is, the control signal combination effect measurement function; The harmony vector u before the k-th iteration update ab (k) Update as follows: BW(k)=BW min +(BW max -BW min )×r7 Among them, r2, r3, r4, r5, r6 and r7 are all random numbers between (0,1); is the randomly generated harmony vector, i.e., the control signal combination; u new,ab (k) is the harmony vector after the k-th iteration update; BW(k) is the harmony bandwidth of the k-th iteration, that is, the control signal combination limit; BW max and BW min are the maximum and minimum values ​​of the harmonic bandwidth respectively; p1 is the probability of storing the harmonic memory bank, that is, the probability of storing the control signal combination; p2 is the probability of gene fine-tuning, that is, the probability of fine-tuning the control signal combination; The updated harmony vector u new,ab and the worst harmony vector u in the initial harmony memory HM worst The harmony metric function value is compared, if u new,ab The metric function value of (k) is less than u worst The metric function value is u new,ab (k) Replace u worst ;Otherwise, the harmony in HM remains unchanged; Check whether the algorithm termination condition is met; if so, output the global optimal solution and the algorithm ends; otherwise, return to continue updating the harmony vector; S3: performing predictive diagnostic maintenance on the motor and controller based on the neural network; S4: Regularly test and verify the motor controller and motor integration structure.

2. The aircraft electric propulsion system according to claim 1, characterized in that: The voltage range of the DC battery (1) is 600-800V.

3. The aircraft electric propulsion system according to claim 1, characterized in that: The motor controller and motor integrated design also integrates a wireless communication model to enable wireless communication between the system and a ground monitoring center.

4. The aircraft electric propulsion system according to claim 3, characterized in that: The integrated design of the motor controller and motor uses carbon fiber composite materials to make an integrated housing, and is also provided with heat dissipation channels and heat dissipation fins made of phase change materials or nanomaterials; it also uses electromagnetic shielding technology to reduce the impact of electromagnetic interference on the equipment.

5. The aircraft electric propulsion system according to claim 4, characterized in that: The structure of the heat dissipation fins includes wavy heat dissipation fins, honeycomb heat dissipation fins, spiral heat dissipation fins, hollow heat dissipation fins, multi-layer three-dimensional heat dissipation fins, curved three-dimensional heat dissipation fins, composite structure heat dissipation fins, deformable heat dissipation fins, adjustable heat dissipation fins, bionic heat dissipation fins, and modular heat dissipation fins; the heat dissipation fins are also prepared in combination with 3D printing technology.

6. The aircraft electric propulsion system according to claim 5, characterized in that: The motor controller and motor integrated design also integrates sensors, power modules and energy storage systems.

Images