Intelligent switching cooperative output method and system of aircraft gasoline-electric hybrid power system

Abstract

The invention discloses an intelligent switching cooperative output method and system for an aircraft gas-electric hybrid power system. The method comprises the following steps: acquiring operation data and flight attitude information of an engine, a generator, a battery and an electric driving system, and constructing a state vector; identifying a flight stage based on the flight speed, the height change rate and the target track parameter, and generating a corresponding power demand and change rate sequence; establishing a complementary relationship between the fuel power and the electric drive power according to the electrical state, and forming a linear complementary constraint model; an oil-electricity power matching initial value meeting boundary conditions is obtained through a Lemke algorithm, and a final power distribution result is obtained through a Fischer-Burmeister function structure and semi-smooth Newton iteration; calculating modulation ratio control quantities of an engine throttle valve, generator excitation, a battery DC / DC converter and an inverter according to a power distribution result to form a control instruction; multi-source power collaborative output is executed, feedback data are collected in real time, the state vector is updated, and closed-loop intelligent control over the oil-electricity hybrid power system of the aircraft is achieved.

Description

Technical Field

[0001] This invention relates to the field of aviation power system control technology, and in particular to a method and system for intelligent switching and coordinated output of an aircraft hybrid power system. Background Technology

[0002] With the development of new energy technologies and intelligent control algorithms, aircraft propulsion systems are gradually evolving towards hybrid electric power systems to improve energy efficiency and flight range. Currently, hybrid electric power systems mostly employ fixed power distribution strategies or simple threshold switching logic, using preset operating condition tables or SOC (State of Charge) thresholds to switch energy and control output between fuel and electric drive.

[0003] Existing hybrid power management methods have significant limitations in multi-condition and dynamic flight environments. On the one hand, traditional power switching logic lacks multi-dimensional perception of flight attitude, acceleration changes, and real-time power demands, leading to delayed switching response or uneven energy distribution during complex flight phases (such as takeoff, landing, climb, cruise, and descent), which can easily cause system oscillations or uncoordinated use of fuel and electrical energy. On the other hand, power allocation strategies based on empirical mapping or simple linear control cannot fully consider multi-source constraints such as bus voltage, current extremes, and SOC boundaries, making it difficult to achieve complementary coordination between fuel power and electric drive power, and easily resulting in energy redundancy, boundary exceedances, or control instability.

[0004] Therefore, how to provide a method and system for intelligent switching and coordinated output of aircraft hybrid power systems is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] One objective of this invention is to propose an intelligent switching and coordinated output method and system for an aircraft's hybrid power system. This invention comprehensively utilizes flight condition identification, linear complementary problem modeling, Lemke's algorithm, and semi-smooth Newton iteration method to construct a complementary constraint relationship between fuel power and electric drive power, thereby realizing dynamic allocation and closed-loop control of multi-source energy. It has the advantages of smooth switching, high power coordination accuracy, and fast system response.

[0006] The intelligent switching and cooperative output method and system for an aircraft hybrid power system according to an embodiment of the present invention includes the following steps: The status acquisition module collects engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, drive motor electromagnetic parameters, and flight attitude and aerodynamic state information, and generates a state vector. The operating condition identification module receives the state vector and outputs the flight operating condition label and the required power sequence. The complementary modeling module receives flight condition labels, constructs a linear complementary problem model, and outputs a constraint mapping structure. The initial value generation module receives the constraint mapping structure, uses the Lemke algorithm to perform basis vector pivot iteration on the linear complementarity problem model, and extracts the initial value of the oil-electric power ratio, the active constraint index, and the updated constraint mapping structure under the current working condition. The smooth iteration module receives the initial value of the oil-electric power ratio, the active constraint index, and the updated constraint mapping structure. It uses the semi-smooth Newton iteration method to perform local convergence solution on the initial value of the oil-electric power ratio to obtain the final power allocation solution that conforms to the current flight conditions. The control command generation module receives the final power allocation solution and the active constraint index, and outputs the control command vector. The execution feedback module receives control command vectors, drives the engine set, generator set and power battery system to jointly output power, collects execution response parameters and updates the state vector.

[0007] Optionally, modules can be integrated using the following methods: Collect engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, drive motor operating parameters, and flight attitude and aerodynamic state data to construct a state vector; Based on state vectors, the upper and lower boundaries of power demand and power change rate under the flight phase are identified and calculated to generate flight condition labels and demand power sequences. Based on the flight condition labels, power demand sequence and electrical states in the state vector, construct the complementary relationship between fuel power output and electric drive power output, and establish a constraint mapping structure in the form of a linear complementary problem. The Lemke algorithm is used to perform basis vector pivot iteration on the constraint mapping structure to determine the feasible solution set that satisfies the complementary relationship and boundary conditions, and to extract the initial value of the oil-electricity power ratio, the active constraint index and the updated constraint mapping structure. The updated constraint mapping structure is transformed into a continuously differentiable structure of the Fischer–Burmeister form. The semi-smooth Newton method is used to perform iterative convergence on the initial value of the oil-electric power ratio to obtain the final power allocation solution. Based on the final power allocation solution and active constraint index, calculate the engine throttle opening, generator excitation voltage, battery-side bidirectional DC / DC converter duty cycle and electric drive inverter modulation ratio control quantities, and apply dwell time and rate of change limits to form control commands; Control commands are applied to the engine, generator and battery systems to drive multi-source coordinated power output, and the response results are collected to update the state vector, thus constructing a closed-loop control path for the hybrid electric system.

[0008] Optionally, generating flight condition labels and power demand sequences includes: Flight speed data and altitude change rate data are extracted from the state vector. Based on the condition that the flight speed data is lower than the first speed threshold and the altitude change rate data is positive, it is identified as the vertical takeoff phase. Based on the condition that the flight speed data is between the first speed threshold and the second speed threshold and the altitude change rate data is positive, it is identified as the climb phase; The cruise phase is identified based on the condition that the absolute value of the altitude change rate data is less than the first change rate threshold. Based on the condition that the flight speed data is higher than the second speed threshold and the altitude change rate data is negative, it is identified as the descent phase; Based on the condition that the flight speed data is below the third speed threshold and the altitude change rate data is negative, it is identified as the vertical descent phase; Based on the flight phases identified above, flight condition tags are assigned according to fixed identifier numbers, and the flight condition tags are arranged and stored in the time sequence of flight phase identification to generate time index identifier information. Acceleration boundary data is extracted from the state vector; target flight speed data and target flight altitude data are extracted from the target trajectory parameters; based on the difference between the current flight speed data and the target flight speed data, the speed error value is determined; based on the difference between the current flight altitude data and the target flight altitude data, the altitude error value is determined. Based on the flight stage corresponding to the flight condition label, the power demand estimation logic associated with that flight stage is invoked, and the power demand value under the current flight stage is calculated using the speed error value and altitude error value as input parameters. Based on the maximum and minimum acceleration boundary values ​​recorded in the acceleration boundary data, positive and negative rate of change constraints are applied to the power demand value to form the upper and lower boundary values ​​of the power change rate corresponding to the current flight phase. The flight condition label, power demand value, upper boundary value of power change rate, and lower boundary value of power change rate are combined according to the time sequence of flight phase identification to form the flight demand power sequence.

[0009] Optionally, the constraint mapping structure for establishing the form of the linear complementarity problem includes: Determine the current flight phase type based on the flight condition label, retrieve the power demand value corresponding to the flight phase type from the flight demand power sequence, and set it as the current total power demand target value. Extract the current maximum allowable output power value of the fuel generator, the current maximum allowable output power value of the battery system, and the total power demand range of the drive motor from the electrical states in the state vector. Limit the fuel power output variable to zero and its maximum allowable output power value, and limit the electric drive power output variable to zero and its maximum allowable output power value. The complementarity determination of fuel power output variables and electric drive power output variables is performed according to the following steps: If the fuel power output variable is greater than zero, then the electric drive power output variable will be set to zero. If the electric drive power output variable is greater than zero, then the fuel power output variable will be set to zero. If both variables are zero, then the target value for total power demand is set to zero; Extract the current bus voltage, current and power battery state of charge values ​​from the electrical state in the state vector, and obtain the upper and lower limits of the bus voltage, the current extreme value and the upper and lower boundary values ​​of the state of charge from the system settings; The following judgments are made for the fuel power output variable and the electric drive power output variable respectively: Substitute the sum of the two variables into the current voltage model to calculate the corresponding bus voltage value and determine whether it is between the upper and lower limits of the bus voltage. The instantaneous output changes of the two variables are converted into current changes, and it is determined whether they are within the current extreme range. The electric drive power output variable is converted into the battery energy consumption rate, and combined with the current state of charge to determine whether it is below the state of charge limit; The combinations of fuel power output variables and electric drive power output variables that satisfy all the above conditions are respectively taken as feasible output solution pairs, and their corresponding complementary judgment results and state boundary judgment results are recorded. Each feasible output solution pair is assigned a set of input vectors, and its corresponding complementarity results and state boundary constraint results are recorded respectively. These are listed in a matrix data table in a linear order, and the matrix data table is used as the constraint input table for the linear complementarity problem form.

[0010] Optionally, the extraction of the initial value of the oil-electric power ratio, the active constraint index, and the updated constraint mapping structure includes: Extract fuel power output variables, electric drive power output variables, battery state of charge, voltage upper limit, current extreme value, and current power demand information of the aircraft from the constraint mapping structure, and construct a complementary variable vector. ,in: Indicates the first One control variable; Each pair of complementary variables Satisfying complementary relationship ; Constructing the standard linear complementarity problem: ,in A column vector of complementary variables; It is a complementary coefficient matrix; A constant offset vector; Introducing artificial variables to satisfy all Items converted This leads to an initial, feasible artificial solution. Initialize the set of basic variables, set artificial variables and relaxation variables as initial basic variables, and complementary variables as non-basic variables. Select the variable with the largest absolute value of negative offset as the input variable and execute the first pivot iteration. Execute the following iterative process until the termination condition is met: Construct based on current basic variables Solving linear systems Calculate the next state; Calculate the ratios of all non-basic variables to their corresponding values. ,in It is a unit vector; Select the variable with the smallest positive ratio to enter the basis, and its complementary variable to leave the basis. Update the variable state and the basis set. If artificial variables Leaving the base set or all complementary variables satisfy If so, then the process terminates; Extract the fuel power and electric drive power variables corresponding to the final state. This value serves as the initial value for the oil-electric power ratio. Extract the corresponding electrical constraint number from the currently active variables to form an active constraint index; The initial value of the oil-electric power ratio, the active constraint index, and the variable-constraint mapping structure at the termination time are combined to form the updated constraint mapping structure.

[0011] Optionally, generating the final power allocation solution includes: Extract the initial value vector of oil-electric power ratio from the output of the Lemke algorithm. ,in Indicates the first The power distribution values ​​of the complementary variables in the initial state This represents the number of complementary variables, with each component in the vector corresponding to the fuel power output component and the electric drive power output component. Extracting the complementary coefficient matrix from the constraint mapping structure With offset vector ,in Indicates the first The power allocation variable affects the first The linear response coefficients of complementary conditions, Indicates the first The offset of each complementary condition under the current operating condition; Constructing the Fischer–Burmeister function Non-smooth complementary conditions Convert to a continuously differentiable form; Initialize state vector Set convergence accuracy threshold Maximum number of iterations And set the iteration counter. ; In satisfying and Under the given conditions, perform the following iterative steps: Calculate the current complementary mapping vector and construct the Jacobian approximation matrix. Solve the linear system to obtain the correction vector. Update the state variables and the iteration counter; When satisfied or The iteration process terminates when the time is right; The final convergence vector The fuel power output component and the electric drive power output component are used as the final power allocation solution for the current flight phase.

[0012] Optionally, the generated control commands include: Extract the corresponding fuel power output and electric drive power output from the final power allocation solution, and denot them as the fuel power target value and the electric drive power target value, respectively. Extract the identifiers related to the boundary constraints of the fuel system, electrical system, and power distribution from the active constraint index to determine the dynamic boundary range that the current output control quantity must satisfy; Based on the target fuel power value and the current engine speed, consult the engine performance mapping table to determine the corresponding target throttle opening value for that fuel power, and mark it as... ; Based on the target fuel power value and the aircraft's real-time total power demand, the generator-side power output value is calculated, and based on the generator excitation curve table, the required excitation voltage target value is determined and marked as follows. ; Based on the target electric drive power value and the current voltage, current direction, and state of charge of the power battery, the current operating mode of the bidirectional DC / DC converter is determined. Combining the internal open-loop characteristics and lookup table data, the required duty cycle target value is calculated and marked as follows: ; Based on the target electric drive power and the current bus voltage, the required motor voltage amplitude for the inverter output is calculated. According to the preset modulation strategy, the corresponding inverter modulation ratio target value is determined and marked as follows. ; Apply dwell time limits to all the above target values ​​and determine whether the dwell time of each target value since the previous period is lower than the minimum allowable dwell time. If it is lower, retain the control value of the previous period unchanged. Apply a rate of change limit to all the above target values, and determine whether the change of each target value relative to the control value of the previous cycle exceeds the preset rate of change threshold. If it does, correct the current target value with the rate of change threshold to avoid control jumps. The revised , , and The parameters are assembled into a structured control parameter set, which forms a control command package and is output to the aircraft control system execution module.

[0013] Optionally, constructing a closed-loop control system for a hybrid electric vehicle includes: The throttle opening control quantity in the control command The output is sent to the engine unit execution module to adjust the fuel injection system and control the actual fuel supply rate. Excitation voltage control quantity The output is sent to the generator set execution module to adjust the excitation winding current, change the generator magnetic flux intensity, and affect the generator voltage amplitude and power output; Duty cycle control of bidirectional DC / DC converter The output is sent to the DC / DC controller in the battery management system, which selects either boost or buck logic based on the current mode to control the battery charging and discharging current. Modulation ratio control quantity of electric drive inverter The output is sent to the inverter control module in the drive motor controller to adjust the duty cycle of the pulse width modulation signal, control the amplitude and phase of the output voltage, and drive the motor to output the target power; After all control commands are applied, the system response acquisition process is initiated to collect real-time data including: actual engine speed, generator output voltage and current, battery terminal voltage and current, SOC status, motor output torque, bus voltage and current, and feedback data on flight attitude and aerodynamic status. The collected feedback data is filtered and time-series calibrated to construct a new state vector, which includes the current physical and power states of the engine set, generator set, battery system, and electric drive system. The constructed state vector is input into the flight phase identification and power demand assessment module of the next control cycle, forming a closed-loop control path of control-feedback-control, and continuously performing dynamic updates and iterative cycles.

[0014] The beneficial effects of this invention are: This invention introduces a flight condition identification and state vector construction mechanism, and uses multi-source parameters from the engine, generator, battery and electric drive system for fusion analysis to achieve accurate determination of flight phase and power demand. It overcomes the shortcomings of traditional oil-electric switching that relies on a single threshold control and has a lagging response to operating conditions, enabling the system to dynamically adapt to changes in energy demand under complex flight environments.

[0015] This invention establishes a linear complementary problem model of fuel power output and electric drive power output, and uses the Lemke algorithm to solve the basis vector pivot iteratively, thereby determining the feasible solution set that satisfies the extreme values ​​of bus voltage and current and the SOC boundary constraints. This effectively avoids the problems of unbalanced power distribution, excessive constraints, and control oscillations in traditional energy management methods.

[0016] This invention introduces the Fischer–Burmeister function and a semi-smooth Newton iteration method to transform non-smooth complementary conditions into a continuously differentiable form, thereby achieving fast convergence and global smooth solution of power allocation and significantly improving the performance of the control algorithm in terms of real-time performance and convergence stability. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 The flowchart shows the intelligent switching and cooperative output method and system for the aircraft's hybrid electric power system proposed in this invention. Detailed Implementation

[0018] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0019] refer to Figure 1 A method and system for intelligent switching and coordinated output of an aircraft's hybrid electric power system, comprising the following steps: The status acquisition module collects engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, drive motor electromagnetic parameters, and flight attitude and aerodynamic state information, and generates a state vector. The operational condition identification module receives the state vector, calculates the current flight stage and the corresponding power requirements and rate of change limits based on the flight speed, altitude change rate, acceleration boundary and target trajectory instructions, and outputs the flight operational condition label and the required power sequence. The complementary modeling module receives flight condition labels, power demand sequence and electrical parameters in the state vector, establishes complementary constraint relationships between fuel power output and electric drive power output, and introduces bus voltage, current extreme values ​​and SOC boundary constraints to construct a linear complementary problem model and output constraint mapping structure. The initial value generation module receives the constraint mapping structure, uses the Lemke algorithm to perform basis vector pivot iteration on the linear complementarity problem model, determines the feasible solution set that satisfies the complementarity condition and all boundary constraints, and extracts the initial value of the oil-electricity power ratio, the active constraint index and the updated constraint mapping structure under the current working condition. The smooth iteration module receives the initial value of the oil-electric power ratio, the active constraint index, and the updated constraint mapping structure output by the initial value generation module. It converts the constraint mapping structure into a continuously differentiable expression in the form of a Fischer–Burmeister function, and uses the semi-smooth Newton iteration method to locally converge the initial value of the oil-electric power ratio to obtain the final power allocation solution that conforms to the current flight conditions. The control command generation module receives the final power allocation solution and active constraint index, calculates the engine throttle opening, generator excitation voltage, battery bidirectional DC / DC converter duty cycle and electric drive inverter modulation ratio control, applies dwell time and rate of change limits, and outputs control command vector. The execution feedback module receives control command vectors, drives the engine set, generator set and power battery system to jointly output power, collects execution response parameters and updates the state vector, forming a closed-loop control cycle for the hybrid electric system.

[0020] In this embodiment, the modules are connected through the following method: Collect engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, drive motor operating parameters, and flight attitude and aerodynamic state data to construct a state vector; Based on the flight speed, altitude change rate, acceleration boundary and target trajectory parameters in the state vector, the flight phase is identified, and the power demand value and upper and lower boundaries of the power change rate under the flight phase are calculated to generate flight condition labels and power demand sequences. Based on the flight condition labels, power demand sequence and electrical states in the state vector, a complementary relationship between fuel power output and electric drive power output is constructed. By superimposing bus voltage, current extreme values ​​and charge state boundaries, a constraint mapping structure in the form of a linear complementary problem is established. The Lemke algorithm is used to perform basis vector pivot iteration on the constraint mapping structure to determine the feasible solution set that satisfies the complementary relationship and boundary conditions, and to extract the initial value of the oil-electricity power ratio, the active constraint index and the updated constraint mapping structure. The updated constraint mapping structure is transformed into a continuously differentiable structure of the Fischer–Burmeister form. The semi-smooth Newton method is used to perform iterative convergence on the initial value of the oil-electric power ratio to obtain the final power allocation solution. Based on the final power allocation solution and active constraint index, calculate the engine throttle opening, generator excitation voltage, battery-side bidirectional DC / DC converter duty cycle and electric drive inverter modulation ratio control quantities, and apply dwell time and rate of change limits to form control commands; Control commands are applied to the engine, generator and battery systems to drive multi-source coordinated power output, and the response results are collected to update the state vector, thus constructing a closed-loop control path for the hybrid electric system.

[0021] In this embodiment, generating flight condition labels and power demand sequences includes: Flight speed and altitude change rate data are extracted from the state vector. The vertical takeoff phase is identified when the flight speed is below a first speed threshold and the altitude change rate is positive; the climb phase is identified when the flight speed is between a first and a second speed threshold and the altitude change rate is positive; the cruise phase is identified when the absolute value of the altitude change rate is less than the first change rate threshold; the descent phase is identified when the flight speed is above the second speed threshold and the altitude change rate is negative; and the vertical landing phase is identified when the flight speed is below a third speed threshold and the altitude change rate is negative. Based on the flight phases identified above, flight condition tags are assigned according to fixed identifier numbers, and the flight condition tags are arranged and stored in the time sequence of flight phase identification to generate time index identifier information. Acceleration boundary data is extracted from the state vector; target flight speed data and target flight altitude data are extracted from the target trajectory parameters; based on the difference between the current flight speed data and the target flight speed data, the speed error value is determined; based on the difference between the current flight altitude data and the target flight altitude data, the altitude error value is determined. Based on the flight stage corresponding to the flight condition label, the power demand estimation logic associated with that flight stage is invoked, and the power demand value under the current flight stage is calculated using the speed error value and altitude error value as input parameters. Based on the maximum and minimum acceleration boundary values ​​recorded in the acceleration boundary data, positive and negative rate of change constraints are applied to the power demand value to form the upper and lower boundary values ​​of the power change rate corresponding to the current flight phase. The flight condition labels, power demand values, upper boundary values ​​of power change rate, and lower boundary values ​​of power change rate are combined according to the time sequence of flight phase identification to form a flight demand power sequence. The flight demand power sequence is then output to the constraint mapping structure construction step in the form of a linear complementary problem.

[0022] In this embodiment, establishing the constraint mapping structure for the linear complementarity problem includes: Determine the current flight phase type based on the flight condition label, retrieve the power demand value corresponding to the flight phase type from the flight demand power sequence, and set it as the current total power demand target value. Extract the current maximum allowable output power value of the fuel generator, the current maximum allowable output power value of the battery system, and the total power demand range of the drive motor from the electrical states in the state vector. Limit the fuel power output variable to zero and its maximum allowable output power value, and limit the electric drive power output variable to zero and its maximum allowable output power value. The complementarity determination of fuel power output variables and electric drive power output variables is performed according to the following steps: If the fuel power output variable is greater than zero, then the electric drive power output variable will be set to zero. If the electric drive power output variable is greater than zero, then the fuel power output variable will be set to zero. If both variables are zero, then the target value for total power demand is set to zero; Extract the current bus voltage, current and power battery state of charge values ​​from the electrical state in the state vector, and obtain the upper and lower limits of the bus voltage, the current extreme value and the upper and lower boundary values ​​of the state of charge from the system settings; The following judgments are made for the fuel power output variable and the electric drive power output variable respectively: Substitute the sum of the two variables into the current voltage model to calculate the corresponding bus voltage value and determine whether it is between the upper and lower limits of the bus voltage. The instantaneous output changes of the two variables are converted into current changes, and it is determined whether they are within the current extreme range. The electric drive power output variable is converted into the battery energy consumption rate, and combined with the current state of charge to determine whether it is below the state of charge limit; The combinations of fuel power output variables and electric drive power output variables that satisfy all the above conditions are respectively taken as feasible output solution pairs, and their corresponding complementary judgment results and state boundary judgment results are recorded. Each feasible output solution pair is assigned a set of input vectors, and its corresponding complementarity results and state boundary constraint results are recorded respectively. These are listed in a matrix data table in a linear order, and the matrix data table is used as a constraint input table for the linear complementarity problem form, which is then provided for subsequent solution steps.

[0023] In this embodiment, the Lemke algorithm includes: Extract fuel power output variables, electric drive power output variables, battery state of charge, voltage upper limit, current extreme value, and current power demand information of the aircraft from the constraint mapping structure, and construct a complementary variable vector. ,in Indicates the first One control variable, such as fuel power or electric power; Each pair of complementary variables Satisfying complementary relationship ; The standard linear complementarity problem is expressed as follows: ; in, : Complementary variable column vector, containing ratio control variables such as fuel power and electric power; The coefficient matrix is ​​calculated from the coupling relationship between the variables in the constraint mapping structure, including energy supply and demand coupling, bus voltage limiting, drive current distribution, etc. : Constant offset vector, composed of the current flight status, including power demand, current bus voltage, current margin, and battery SOC upper and lower limit offsets; Introducing artificial variables to satisfy all Items converted To form an artificially feasible preliminary solution; Initialize the set of basic variables, set artificial variables and relaxation variables as initial basic variables, and complementary variables as non-basic variables. Select the variable with the largest absolute value of negative offset as the input variable and execute the first pivot iteration. Execute the following iterative process until the termination condition is met: Construct based on current basic variables Solving linear systems Calculate the next state; Calculate the ratios of all non-basic variables to their corresponding values. ,in It is a unit vector; Select the variable with the smallest positive ratio to enter the basis, and its complementary variable to leave the basis. Update the variable state and the basis set. If artificial variables Leaving the base set or all complementary variables satisfy If so, then the process terminates; Extract the fuel power and electric drive power variables corresponding to the final state. This value serves as the initial value for the oil-electric power ratio. Extract the corresponding electrical constraint number from the currently active variables to form an active constraint index; The initial value of the oil-electric power ratio, the active constraint index, and the variable-constraint mapping structure at the termination time are combined to form an updated constraint mapping structure, which is used for subsequent smooth solution steps.

[0024] In this embodiment, generating the final power allocation solution includes: Extract the initial value vector of oil-electric power ratio from the output of the Lemke algorithm. ,in Indicates the first The power distribution values ​​of the complementary variables in the initial state This represents the number of complementary variables, with each component in the vector corresponding to the fuel power output component and the electric drive power output component. Extracting the complementary coefficient matrix from the constraint mapping structure With offset vector ,in Indicates the first The power allocation variable affects the first The linear response coefficients of complementary conditions, Indicates the first The offset terms of each complementary condition under the current operating condition reflect the combined effect of flight power demand, bus voltage, current boundary and state of charge boundary. Constructing the Fischer–Burmeister function Used to apply non-smooth complementary conditions Converted to continuously differentiable form, the specific function expression is as follows: ; in, For the first Complementary variables, The corresponding complementary mapping function value is calculated as follows: ; Initialize state vector Set convergence accuracy threshold Maximum number of iterations And set the iteration counter. ; In satisfying and Under the given conditions, perform the following iterative steps: Calculate the current complementary mapping vector: ; Construct the Jacobian approximation matrix Its elements are:

[0025] in, For the first Current values ​​of complementary functions These are the elements in the input coefficient matrix; Solving linear systems: The corrected vector is obtained. ; Update state variables: Update iterative counter ; When satisfied or The iteration process terminates when the time is right; The final convergence vector The fuel power output component and the electric drive power output component are used as the final power allocation solution for the current flight phase.

[0026] In this embodiment, the generation of control commands includes: Extract the corresponding fuel power output and electric drive power output from the final power allocation solution, and denot them as the fuel power target value and the electric drive power target value, respectively. Extract the identifiers related to the boundary constraints of the fuel system, electrical system, and power distribution from the active constraint index to determine the dynamic boundary range that the current output control quantity must satisfy; Based on the target fuel power value and the current engine speed, consult the engine performance mapping table to determine the corresponding target throttle opening value for that fuel power, and mark it as... ; Based on the target fuel power value and the aircraft's real-time total power demand, the generator-side power output value is calculated, and based on the generator excitation curve table, the required excitation voltage target value is determined and marked as follows. ; Based on the target electric drive power value and the current voltage, current direction, and state of charge of the power battery, the current operating mode of the bidirectional DC / DC converter is determined. Combining the internal open-loop characteristics and lookup table data, the required duty cycle target value is calculated and marked as follows: ; Based on the target electric drive power and the current bus voltage, the required motor voltage amplitude for the inverter output is calculated. According to the preset modulation strategy, the corresponding inverter modulation ratio target value is determined and marked as follows. ; Apply dwell time limits to all the above target values ​​and determine whether the dwell time of each target value since the previous period is lower than the minimum allowable dwell time. If it is lower, retain the control value of the previous period unchanged. Apply a rate of change limit to all the above target values, and determine whether the change of each target value relative to the control value of the previous cycle exceeds the preset rate of change threshold. If it does, correct the current target value with the rate of change threshold to avoid control jumps. The revised , , and The parameters are assembled into a structured control parameter set, which forms a control command package and is output to the aircraft control system execution module to drive the engine set, generator set, battery system and drive inverter to work together.

[0027] In this embodiment, the closed-loop control path for constructing the hybrid electric vehicle system includes: The throttle opening control quantity in the control command The output is sent to the engine unit execution module to adjust the fuel injection system and control the actual fuel supply rate. Excitation voltage control quantity The output is sent to the generator set execution module to adjust the excitation winding current, change the generator magnetic flux intensity, and affect the generator voltage amplitude and power output; Duty cycle control of bidirectional DC / DC converter The output is sent to the DC / DC controller in the battery management system, which selects either boost or buck logic based on the current mode to control the battery charging and discharging current. Modulation ratio control quantity of electric drive inverter The output is sent to the inverter control module in the drive motor controller to adjust the duty cycle of the pulse width modulation signal, control the amplitude and phase of the output voltage, and drive the motor to output the target power; After all control commands are applied, the system response acquisition process is initiated to collect real-time data including: actual engine speed, generator output voltage and current, battery terminal voltage and current, SOC status, motor output torque, bus voltage and current, and feedback data on flight attitude and aerodynamic status. The collected feedback data is filtered and time-series calibrated to construct a new state vector, which includes the current physical and power states of the engine set, generator set, battery system, and electric drive system. The constructed state vector is input into the flight phase identification and power demand assessment module of the next control cycle, forming a closed-loop control path of control-feedback-control, and continuously performing dynamic updates and iterative cycles.

[0028] Example: To verify the feasibility and technical advantages of the intelligent switching and coordinated output method and system for the aircraft's hybrid power system proposed in this invention, it was applied to a hybrid power system experimental platform of a medium-sized fixed-wing vertical takeoff and landing unmanned aerial vehicle (UAV). This UAV uses a 45kW range-extended internal combustion engine, a 22kW dual-motor electric drive system, and a 12kWh lithium-ion battery pack to form a hybrid power system. The system adopts a three-phase bus structure with a voltage level of 400V, and is equipped with a bidirectional DC / DC converter and inverter drive module, as well as a real-time status acquisition and control unit (FCS).

[0029] The experiment was conducted at a flight test center from May 10th to May 18th, 2025, under clear skies, with wind speeds not exceeding 5 m / s and an average temperature of 22°C. To ensure the representativeness and repeatability of the experiment, the test flight consisted of five typical phases: vertical takeoff, climb, cruise, descent, and vertical landing. Data on fuel power, electric drive power, battery state of charge (SOC), bus voltage, current, and attitude changes were collected during each phase.

[0030] Under traditional control strategies, the system employs fixed power ratio control, where the allocation ratio of fuel power to electric drive power is set according to empirical values ​​and switched between different flight phases using simple thresholds. When flight power demand changes abruptly, the system often experiences switching lag, voltage disturbances, and excessive current transients, leading to reduced energy utilization and decreased attitude stability. To address these issues, this invention introduces flight condition identification, complementary constraint modeling, and a complementary solution mechanism based on the Lemke algorithm, combined with a semi-smooth Newton iterative solution strategy in the form of a Fischer–Burmeister function, to achieve continuous complementary adjustment of fuel and electric power.

[0031] In the experimental scenario, the flight control computer first acquires real-time data such as engine speed, generator terminal current, battery voltage and SOC, electromagnetic parameters of the electric drive motor, and flight attitude angular velocity through the state acquisition module, constructing a state vector. Based on flight speed, rate of altitude change, and target trajectory, the system automatically identifies the flight phase and outputs flight condition labels and power demand sequences. Subsequently, the complementary modeling module defines fuel output power and electric drive output power as complementary variables and introduces bus voltage, current extrema, and SOC boundary conditions to construct a linear complementary problem model. The Lemke algorithm is used to perform a pivotal iteration of the basis vectors to obtain an initial feasible power ratio that satisfies all boundary constraints.

[0032] After obtaining the initial values, the system further transforms the complementary constraint structure into a continuous form of the Fischer–Burmeister function and uses a semi-smooth Newton iteration method to achieve local convergence of the power allocation solution, thereby obtaining the optimal power allocation solution. This allocation result is then converted by the control command generation module into control signals for the fuel throttle opening, generator excitation voltage, battery DC / DC converter duty cycle, and inverter modulation ratio. Upon receiving the commands, the execution module drives the fuel engine, generator, and electric drive motor to work together to generate power, and the feedback module monitors the response signals in real time, forming a closed-loop control circuit.

[0033] Through this dynamic constraint complementary solution mechanism, the system can optimize the fuel-electric distribution ratio in real time according to different flight phases, achieving coordinated output of multi-source energy. For example, during vertical takeoff, the system identifies that the aircraft is in a high acceleration positive change range, automatically increases the electric drive output to meet short-term peak power demand, and controls the engine to smoothly accelerate, avoiding voltage fluctuations caused by sudden power changes. During cruise, the system automatically adjusts the fuel-electric ratio to maintain the SOC within the target range by monitoring the bus voltage and SOC in real time, thus extending the overall endurance.

[0034] Throughout the flight test, the system updated the state vector with a control cycle of 10ms and recorded all electrical and attitude parameters via FCS. Finally, after each mission, the performance of the proposed method was compared with that of traditional control strategies, with a focus on power switching response time, bus voltage fluctuation amplitude, system energy utilization, SOC fluctuation range, and flight attitude stability.

[0035] The table below shows the objective data results obtained based on the average statistics of 12 test flight missions: Table 1 Comparative experimental results of traditional fixed power control and the intelligent switching cooperative output method of this invention ; The data results show that the method of this invention significantly improves the overall system performance under dynamic operating conditions. The power switching response time is reduced from 228ms in the traditional strategy to 86ms, an improvement of approximately 62%. Due to the introduction of the complementary constraint model and the application of the smooth solution mechanism, the bus voltage fluctuation amplitude is reduced from ±4.8% to ±1.7%, effectively avoiding the voltage surge problem in the traditional switching mode, and significantly enhancing flight control stability. Energy utilization is improved by approximately 9%, the overall system efficiency is increased from 81.7% to 89.2%, and fuel consumption is reduced by approximately 10%.

[0036] Regarding flight endurance, the method of this invention, by adjusting the fuel-electric ratio in real time, makes the coupling of fuel and electrical energy more reasonable, increasing the overall flight endurance from 66.5 minutes to 72.1 minutes, an extension of approximately 8.4%. The battery SOC fluctuation range is significantly narrowed, indicating that the system can maintain good energy balance and energy storage stability at different power demand stages.

[0037] Further analysis revealed that traditional systems are prone to power surges during climb and descent, resulting in bus voltage deviations exceeding ±15V. In contrast, the system of this invention, through dynamic complementary constraint adjustments during the same phase, keeps voltage fluctuations within ±6V, improving stability by nearly 60%. Simultaneously, the attitude disturbance angle is reduced to 1.1°, indicating that the flight attitude remains more stable during power switching.

[0038] From an engineering application perspective, the core technical feature of this invention is: (1) By constructing flight state vectors, multi-source information fusion and real-time perception of engine, generator, battery and electric drive system are realized; (2) By establishing a linear complementary constraint relationship between fuel and electric drive power, and introducing bus voltage, current extreme values ​​and SOC boundary conditions, the constraint coordination of multi-source power is achieved; (3) The Lemke algorithm is used to solve the linear complementarity problem to obtain the initial power allocation solution that satisfies all constraints; (4) By introducing the Fischer–Burmeister function and the semi-smooth Newton iteration algorithm, the non-smooth complementary conditions are transformed into a continuously differentiable problem, thus achieving a fast convergence solution for power allocation. (5) Combined with the control command generation module, the precise scheduling of multi-source energy is achieved through the joint control of throttle, excitation voltage, converter duty cycle and modulation ratio; (6) Form a closed-loop feedback control path to realize a cyclic optimization mechanism of control-feedback-control.

[0039] In this embodiment, these technical features have been fully verified in actual flight missions. Taking a typical mission as an example, when the UAV transitions from the vertical takeoff phase to the cruise phase, and the total power demand increases from 30kW to 42kW, the system completes the redistribution of fuel and electric power within 260ms. The fuel power smoothly increases to 32.5kW, the electric drive power smoothly decreases to 9.5kW, and the bus voltage remains within the range of 397.2V±1.9V without significant transient oscillations. Compared to traditional systems where the bus voltage fluctuates by more than ±4% under the same operating conditions, this invention achieves significant optimization.

[0040] Furthermore, the method of this invention exhibits high robustness under various external disturbance conditions. In a 5 m / s gust environment, the aircraft's attitude change rate remains within 0.03 rad / s, and the system automatically suppresses attitude deviations caused by aerodynamic disturbances through a combined oil and electric power compensation.

[0041] In summary, this embodiment verifies the engineering practicality and superiority of the intelligent switching and cooperative output method for the aircraft's hybrid power system. This method achieves intelligent allocation and closed-loop control of hybrid power through multi-source data fusion, linear complementary modeling, and smooth iterative solution. This not only improves the system's dynamic response capability and control accuracy but also significantly enhances flight energy efficiency and stability, providing a reliable technical approach and application value for hybrid power energy management in future UAVs and general aviation fields.

[0042] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent switching and cooperative output method and system for an aircraft oil-electric hybrid power system, characterized in that, Comprise: State acquisition module, collect engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, driving motor electromagnetic parameters and flight attitude and aerodynamic state information, generate state vector; Working condition recognition module, receive state vector, output flight working condition label and demand power sequence; Complementary modeling module, receive flight working condition label, construct linear complementary problem model, output constraint mapping structure; Initial value generation module, receive constraint mapping structure, use Lemke algorithm to perform basis vector pivot iteration on linear complementary problem model, extract oil-electric power ratio initial value, active constraint index and updated constraint mapping structure under current working condition; Smooth iteration module, receive oil-electric power ratio initial value, active constraint index and updated constraint mapping structure, use semi-smooth Newton iteration method to locally converge and solve oil-electric power ratio initial value, obtain final power distribution solution conforming to current flight working condition; Control instruction generation module, receive final power distribution solution and active constraint index, output control instruction vector; Execution feedback module, receive control instruction vector, drive engine group, generator set and power battery system to jointly output power, collect execution response parameters and update state vector.

2. The intelligent switching and collaborative output method for an aircraft oil-electric hybrid power system, characterized in that, The modules are realized by the following methods: Collect engine speed, generator terminal voltage and current, power battery voltage, current and state of charge, driving motor operating parameters, and flight attitude and aerodynamic state data, and construct a state vector; Based on the state vector, identify the power demand value and power change rate upper and lower boundary in the flight phase, generate the flight working condition label and demand power sequence; According to the flight working condition label, the demand power sequence and the electrical state in the state vector, the complementary relationship between the fuel power output and the electric drive power output is constructed, and the constraint mapping structure in the form of linear complementary problem is established; Determine the feasible solution set that meets the complementary relationship and boundary conditions by performing basis vector pivot iteration on the constraint mapping structure through Lemke algorithm, extract the oil-electric power ratio initial value, the active constraint index and the updated constraint mapping structure; Transform the updated constraint mapping structure into a continuous and differentiable structure in the Fischer-Burmeister form, and use the semi-smooth Newton method to perform iterative convergence on the oil-electric power ratio initial value to obtain the final power distribution solution; According to the final power distribution solution and the active constraint index, calculate the engine throttle opening, generator excitation voltage, battery side bidirectional DC / DC converter duty ratio and electric drive inverter modulation ratio control quantity, and apply the residence time and change rate limit to form the control instruction; Apply the control instruction to the engine group, generator set and power battery system to drive multi-source collaborative output, and collect the response results to update the state vector, and construct the closed-loop regulation and control path of the oil-electric hybrid power system.

3. The intelligent switching and cooperative output method of the hybrid power system of the aircraft according to claim 2, characterized in that, Generating flight working condition label and demand power sequence includes: Extract flight speed data and height change rate data from state vector, and according to the condition that flight speed data is lower than first speed threshold and height change rate data is positive, identify vertical take-off phase; According to the condition that the flight speed data is between the first speed threshold value and the second speed threshold value and the height rate of change data is positive, it is identified as a climbing stage; According to the condition that the absolute value of the height rate of change data is less than the first change rate threshold value, it is identified as a cruising stage; According to the condition that the flight speed data is higher than the second speed threshold value and the height rate of change data is negative, it is identified as a descending stage; According to the condition that the flight speed data is lower than the third speed threshold value and the height rate of change data is negative, it is identified as a vertical landing stage; Based on the above-identified flight stage, the flight condition label is assigned according to the fixed identification number, and the flight condition label is arranged and stored in the time sequence of flight stage identification, and the time index identification information is generated; From the state vector, acceleration boundary data is extracted, target flight speed data and target flight height data are extracted from the target flight path parameters, based on the difference between the current flight speed data and the target flight speed data, the speed error value is determined; based on the difference between the current flight height data and the target flight height data, the height error value is determined; According to the flight stage corresponding to the flight condition label, the power demand estimation logic associated with the flight stage is called, and the speed error value and the height error value are used as input parameters to calculate the power demand value in the current flight stage; Based on the maximum acceleration boundary value and the minimum acceleration boundary value recorded in the acceleration boundary data, the positive change rate constraint and the negative change rate constraint are applied to the power demand value respectively, and the power change rate upper boundary value and the power change rate lower boundary value corresponding to the current flight stage are formed; The flight condition label, the power demand value, the power change rate upper boundary value and the power change rate lower boundary value are combined in the time sequence of flight stage identification to form the flight demand power sequence.

4. The intelligent switching and collaborative output method of the aircraft oil-electric hybrid power system according to claim 2 characterized in that The constraint mapping structure in the form of linear complementarity problem is established, including: According to the flight stage type determined by the flight condition label, the power demand value corresponding to the flight stage type in the flight demand power sequence is called as the current total power demand target value; From the electrical state in the state vector, the current maximum allowed output power value of the fuel generator, the current maximum allowed output power value of the battery system and the total power demand range of the drive motor are extracted, and the fuel power output variable is limited between zero and its maximum allowed output power value, and the electric drive power output variable is limited between zero and its maximum allowed output power value; The complementarity of the fuel power output variable and the electric drive power output variable is determined, and the following steps are taken: If the fuel power output variable is greater than zero, the electric drive power output variable is assigned to zero; If the electric drive power output variable is greater than zero, the fuel power output variable is assigned to zero; If both variables are zero, the total power demand target value is assigned to zero; From the electrical state in the state vector, the current bus voltage value, the current value and the power battery state of charge value are extracted, and the bus voltage upper limit value and the lower limit value, the current extreme value and the state of charge upper and lower boundary values are obtained from the system setting; The following judgments are made on the fuel power output variable and the electric drive power output variable respectively: The sum of the two variables is substituted into the current voltage model to calculate the corresponding bus voltage value, and it is judged whether it is between the upper and lower limits of the bus voltage; The instantaneous output change of the two variables is converted into a current change, and it is judged whether it is within the current limit range; The electric drive power output variable is converted into a battery energy consumption rate, and combined with the current state of charge to judge whether it is lower than the lower limit of the state of charge; The combination of the fuel power output variable and the electric drive power output variable that meets all the above conditions is respectively taken as a feasible output solution pair, and its corresponding complementary judgment result and state boundary judgment result are recorded; Each set of feasible output solution pairs is assigned as a set of input vectors, and their corresponding complementary results and state boundary limit results are recorded respectively, and they are sequentially listed in the matrix data table according to the linear relationship, and the matrix data table is taken as the constraint input table in the form of linear complementarity problem.

5. The method of claim 2, wherein, The initial value of the oil-electric power ratio, the active constraint index and the updated constraint mapping structure include: The fuel power output variable, the electric drive power output variable, the battery state of charge, the voltage upper limit, the current extreme value and the current power demand information of the aircraft are extracted from the constraint mapping structure to construct a complementary variable vector wherein: represents the th control variable each pair of complementary variables satisfy the complementary relationship ; Construct the standard linear complementarity problem: where is a complementarity variable column vector; is a complementarity coefficient matrix; is a constant offset vector; Introducing artificial variables Convert all terms that do not satisfy to to form an artificial feasible initial solution; Initialize the base variable set, set the artificial variable and the slack variable as the initial base variable, the complementary variable as the non-base variable, select the variable corresponding to the largest negative offset in absolute value as the entering base variable, and perform the first pivot iteration; The following iteration process is executed until the termination condition is met: constructing according to the current base variable , solving linear system computing next state; Compute all non-base variables with their corresponding ratios where is a unit vector; Select the variable corresponding to the minimum positive ratio into the base, and its complementary variable out of the base, update the variable state and the base set; If artificial variable Leave the base set or all complementary variables meet Then terminate; Extract the fuel power and electric drive power variables corresponding to the final state. This value serves as the initial value for the oil-electric power ratio. Extract the corresponding electrical constraint number from the variables currently in the active state to form the active constraint index; The initial value of the oil-electric power ratio, the active constraint index and the variable-constraint mapping structure at the termination time are combined to form the updated constraint mapping structure.

6. The method of claim 2, wherein, Generating the final power distribution solution includes: extracting an initial value vector of oil-electric power ratio from the output result of the Lemke algorithm wherein denotes the power distribution value in the initial state of the complementary variable, denotes the number of complementary variables, and each component in the vector corresponds to a fuel power output component and an electric drive power output component; Extracting complementary coefficient matrices from constraint mapping structures with offset vectors wherein denotes the linear response coefficient of the th power allocation variable to the th complementary condition, denotes the offset term of the th complementary condition at the current operating condition; Constructing Fischer-Burmeister function non-smooth complementarity conditions into a continuously differentiable form; initializing a state vector , setting a convergence precision threshold , maximum number of iteration rounds , and setting an iteration counter ; under the condition that and the following iteration steps are performed: computing a current complementary mapping vector, constructing a jacobian approximation matrix solving a linear system to obtain a correction vector updating the state variable and updating the iteration counter the iteration process is terminated when or the iteration process is terminated when the final converged vector the fuel power output component and the electric drive power output component in the final converged vector as the final power distribution solution for the current flight phase.

7. The method of claim 2, wherein, Generating control instructions includes: Extract the corresponding fuel power output and electric drive power output from the final power distribution solution, and record them as fuel power target value and electric drive power target value respectively; Extract the identification content about the fuel system, the electrical system and the power distribution boundary constraint from the active constraint index to determine the dynamic boundary range that the current output control quantity needs to meet; According to the fuel power target value and the current engine speed, the engine performance mapping table is inquired to determine the throttle opening target value corresponding to the fuel power, and marked as ; According to the fuel power target value and the real-time total power demand of the aircraft, the generator-side power output value is calculated, and based on the generator excitation curve table, the required excitation voltage target value is determined, marked as ; According to the electric drive power target value and the current voltage, current direction and state of charge of the power battery, the current operation mode of the bidirectional DC / DC converter is determined, and the required duty cycle target value is calculated in combination with the internal open loop characteristics and the lookup table data, which is marked as ; According to the electric drive power target value and the current bus voltage, the motor voltage amplitude required by the inverter output is calculated reversely, and according to the preset modulation strategy, the corresponding inverter modulation ratio target value is determined, which is marked as ; Apply the residence time limit to all target values respectively, and judge whether the holding time of each target value since the previous cycle is lower than the minimum allowed residence time, if it is lower, keep the control value of the previous cycle unchanged; Apply the change rate limit to all target values, and judge whether the change amplitude of each target value relative to the control value of the previous cycle exceeds the preset change rate threshold, if it exceeds, correct the current target value with the change rate threshold to avoid control jumping; The modified , , and are assembled into a structured control parameter group, forming a control instruction package, and output to the aircraft control system execution module.

8. The method of claim 2, wherein, The closed-loop regulation and control of the oil-electric hybrid power system includes: The throttle opening control amount in the control command is calculated is output to an engine group execution module, which adjusts a fuel injection system to control the actual fuel supply rate Controlling the excitation voltage Output to the generator set execution module, adjust the excitation winding current, change the generator magnetic flux intensity, affect the generator voltage amplitude and power output; Duty cycle control for bidirectional dc / dc converter The DC / DC controller outputs to the battery management system, according to the mode selection of boost or buck logic, control the battery charge and discharge current; Controlling the modulation ratio of an electric drive inverter The output to the inverter control module in the drive motor controller, adjusting the pulse width modulation signal duty cycle, control output voltage amplitude and phase, drive motor output target power; After all the control instructions are applied, start the system response collection process, and collect real-time feedback data including: engine actual speed, generator output voltage and current, battery terminal voltage and current, SOC state, electric motor output torque, bus voltage, current, and flight attitude and aerodynamic state; Filter and time sequence calibrate the above collected feedback data to build a new state vector, which contains the current physical state and power state of the engine group, generator group, battery system and electric drive system; The constructed state vector is input to a flight phase identification and power demand assessment module of the next control cycle to form a control-feedback-control closed loop control path for continuous dynamic updating and cyclic iteration.

Images