Fault-tolerant control method for electro-hydraulic actuator

Abstract

The application discloses a kind of fault-tolerant control methods of electro-hydraulic actuator, it is related to electro-hydraulic servo technical field.The method includes: constructing the mathematical model of electro-hydraulic actuator in different working conditions, the different working conditions include the normal operating state and winding open-circuit fault state of five-phase permanent magnet synchronous motor;Based on proportional-integral controller, position controller is constructed, and motor reference speed is output;Model predictive speed fault-tolerant control method is established, including predictive speed control module, cascaded extended state observer module and fault-tolerant reference current calculation module;Establish current fault-tolerant controller, including quasi-proportional-resonant controller and fault-tolerant space vector pulse width modulation method.The application can ensure that electro-hydraulic actuator fault-tolerant control operation is more stable, and has high dynamic response and high robustness.

Description

Technical Field

[0001] This invention relates to the field of electro-hydraulic servo technology, and in particular to a fault-tolerant control method for an electro-hydraulic actuator. Background Technology

[0002] Electro-hydraulic actuators (EHAs) highly integrate components such as motors, hydraulic pumps, and hydraulic cylinders, and are developing towards smaller size, higher power density, higher efficiency, and higher reliability. The adoption of five-phase permanent magnet synchronous motors (F-PMSMs) with higher power density and reliability is one of the main future development trends of electro-hydraulic actuators.

[0003] The F-PMSM has redundant windings, which can ensure the desired torque output performance and improve reliability in the event of an open-circuit fault through fault-tolerant control algorithms. Existing fault-tolerant control methods are all focused on the F-PMSM itself, and many research methods achieve low torque ripple by injecting harmonic currents to improve the steady-state torque performance of the F-PMSM during fault-tolerant control. However, current research does not consider the application requirements and characteristics of the system, making it difficult to guarantee the fault-tolerant control performance of the EHA system. EHAs often operate under reciprocating motion and alternating load conditions, inevitably encountering multi-source complex disturbances such as parameter mismatch, internal faults, and external load changes. Therefore, the decrease in control degrees of freedom and parameter mismatch will make it difficult to completely suppress the torque ripple of the F-PMSM through current injection, resulting in obvious periodic speed disturbances. Therefore, the fault-tolerant control method for EHA must have the ability to suppress periodic disturbances. In addition, the fault-tolerant control method for EHA needs to have high dynamic response and high robustness to cope with reciprocating motion conditions and rapidly changing parameter and load disturbances. To improve the fault-tolerant control capability of EHA, it is necessary to design an improved model predictive speed control with high robustness and dynamics (IMPSC-HRD) method to ensure that the fault-tolerant control of EHA operates more smoothly and has high dynamic response and high robustness. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the prior art and propose a fault-tolerant control method for electro-hydraulic actuators, which can ensure that the fault-tolerant control of electro-hydraulic actuators is more stable and has high dynamic response and high robustness.

[0005] To achieve the above objectives, this invention proposes a fault-tolerant control method for an electro-hydraulic actuator, comprising the following steps: Step 1: Construct mathematical models of the electro-hydraulic actuator under different working states, including the normal operation state and the winding open circuit fault state of the five-phase permanent magnet synchronous motor. Step 2: Construct a position controller based on a proportional-integral controller to output the motor reference speed; Step 3: Establish a model-predicted speed fault-tolerant control method, which includes a predictive speed control module, a cascaded extended state observer module, and a fault-tolerant reference current calculation module; Step 4: Establish a current fault-tolerant controller, which includes a quasi-proportional resonant controller and a fault-tolerant space vector pulse width modulation method.

[0006] Preferably, the mathematical models for the electro-hydraulic actuator under different operating states in step one include: The total electromagnetic torque model of a five-phase permanent magnet synchronous motor during normal operation is as follows: ,in, It is the electromagnetic torque under healthy operating conditions. It is an extreme logarithm. It is a permanent magnet flux chain; and They are shaft current and shaft current, and They are Shaft inductance and Shaft inductor, subscript Represents the fundamental frequency space. Represents the third harmonic space; The total electromagnetic torque during a winding open-circuit fault is: ,in, , It is the fundamental spatial electromagnetic torque during an open-circuit fault. It is the third harmonic current. It is an electrical angle; , It is the third harmonic spatial electromagnetic torque during an open-circuit fault. It is the generalized zero-sequence current; And the kinematic equations of mechanics: ,in, It is the mechanical angular velocity of a five-phase permanent magnet synchronous motor. It refers to the displacement of the plunger pump. It is load pressure. B is the moment of inertia of the five-phase permanent magnet synchronous motor, and B is the damping coefficient of the five-phase permanent magnet synchronous motor. It is electromagnetic torque; Flow continuity equation for hydraulic systems: ,in, It is the rotational speed of the plunger pump, and ; and These refer to the effective working areas of the rod-side chamber and the rodless chamber of the hydraulic cylinder, respectively. and These are the displacements of the plunger pump's oil ports A and B, respectively. It refers to the displacement of the hydraulic cylinder piston rod. It is the speed of the piston rod; This is the initial position of the hydraulic cylinder piston; It is the elastic modulus of the oil; and These are the leakage coefficients of the piston pump and the hydraulic cylinder, respectively. and These are the parameters for the hydraulic cylinder's operating mode; The mechanical equations of the hydraulic cylinder: ,in, It is the acceleration of the piston rod of the hydraulic cylinder. It is the total mass of the cylinder piston rod. It is the external load force on the piston rod of the hydraulic cylinder. It is the frictional resistance of the hydraulic cylinder.

[0007] Preferably, the predicted speed control module in step three is discretized using a hybrid Euler discrete model: Step 31: In The mechanical kinematic equations are discretized using the backward Euler method over the time interval, resulting in... ,in, and They represent the first and One control cycle; , ; It is the duration of a control cycle; Step 32: In The mechanical kinematic equations are discretized using the forward Euler method within the time interval, resulting in... ,in, They represent the first One control cycle; finally, the speed prediction model of the incremental model is obtained as follows: ,in, It is the first Incremental electromagnetic torque per control cycle No. Incremental pressure value for each control cycle; Step 33: Obtain the data from the observer and Substituting into step 32, we finally obtain Speed ​​prediction value per control cycle as follows: ; Step 34: Based on and , This can be obtained by solving the following optimization problem: ,in, This is the maximum allowable torque limit; Step 35: Solve The optimal torque increment value under unconstrained conditions can be obtained. ,as follows ; Step 36: Clamp the unconstrained solution to obtain the global optimal solution: Optimal torque for each control cycle for: ,in Represents a saturation function; Step 37: By overlay The estimated perturbation output by the cascaded extended state observer module The target torque value is obtained. .

[0008] Preferably, the cascaded extended state observer module in step three includes: The first-level extended state observer estimates a portion of the total disturbance and outputs the estimate as a known disturbance. Its discretized form is as follows: , in, , and It is a state variable. ; It is a partial total disturbance. It is the differential value of the total disturbance. , and yes , and The estimated value; , and It is the error feedback gain coefficient; and They represent the first and One control cycle, It is the duration of a control cycle; It refers to the displacement of the plunger pump. It is the load pressure; These are the nominal model parameters. ,and The nominal moment of inertia; It is the final reference electromagnetic torque; The second-stage extended state observer, containing a quasi-generalized integrator, is used to extract current harmonics and estimate the remainder of the total disturbance. Its discretized form is as follows: , in, and It is a state variable. It is the remaining total disturbance. It is the differential value of the remaining total disturbance. , and yes , and The estimated value; , and It is the error feedback gain coefficient; For the extracted current harmonics; These are the nominal model parameters; and These are the internal state variables of the i-th quasi-generalized integrator. It is the frequency of the i-th harmonic current; The first-stage extended state observer and the second-stage extended state observer use the pole placement method to adjust parameters.

[0009] Preferably, the quasi-generalized integrator is used to extract the 0th, 1st, 2nd and 4th harmonic current frequencies.

[0010] Preferably, the fault-tolerant reference current calculation module in step three ignores the reluctance torque and sets the following settings during normal operation of the five-phase permanent magnet synchronous motor. At the same time, the third harmonic reference current Set to 0 to calculate the initial reference current. and for: ,in It is a permanent magnet flux linkage in the fundamental space; by limiting the q1-axis current to satisfy... (in, The purpose of limiting phase current is to use the maximum effective value of the phase current. The target torque value; When a single-phase open-circuit fault occurs in a five-phase permanent magnet synchronous motor, by setting... , ,and At the same time, set and The following relationship must be satisfied: Let the q1-axis current satisfy... The reference electromagnetic torque considering phase current constraints is calculated. That is, the output reference electromagnetic torque : ,in It is a permanent magnet flux linkage in the third harmonic space.

[0011] Preferably, the quasi-proportional resonant current-tolerant controller in step four includes a proportional gain. Damped resonant terms ,in, For resonant gain, The damping coefficient is... It is the resonant frequency.

[0012] Preferably, the fault-tolerant space vector pulse width modulation method in step four specifically includes: Step 41: Determine the motor's operating status based on the collected actual five-phase current values, which can be divided into normal operating status and winding open circuit fault. Step 42: Establish the spatial voltage vector distribution in the corresponding static coordinate system under different working conditions, and divide the sectors according to the principle of vector average distribution; Step 43: Perform coordinate transformation on the compensated target voltage value to obtain the target voltage value in the stationary coordinate system; Step 44: Select the sector where the target voltage vector is located based on the target voltage value in the stationary coordinate system; Step 45: Select the space voltage vector used to synthesize the target voltage vector, calculate the duration of each space voltage vector, and solve for the control duty cycle of each phase of the inverter.

[0013] The beneficial effects of this invention are: 1. This invention proposes a fault-tolerant control method for EHA under single-phase open-circuit faults, which improves the reliability of EHA; 2. This invention proposes a speed fault-tolerant control method with high dynamic response and high robustness, which improves the dynamic response and robustness of EHA fault-tolerant control; 3. This invention proposes a cascaded extended state observer based on disturbance differential compensation and a quasi-generalized integrator, which enhances the estimation capability for rapidly changing load disturbances and periodic disturbances; 4. This invention proposes a reference current calculation method to maximize the torque output capability of EHA while improving speed control accuracy; 5. This invention establishes a fault-tolerant space vector pulse width modulation method to improve the synthesis accuracy of the target voltage vector, thereby improving the fault-tolerant control accuracy.

[0014] The features and advantages of the present invention will be described in detail through embodiments and in conjunction with the accompanying drawings. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the EHA principle of the present invention; Figure 2 This is the overall flowchart of the EHA fault-tolerant control of the present invention; Figure 3 This is a control block diagram of the speed fault-tolerant control method of the present invention; Figure 4 This is a block diagram of the extended state observer of the present invention; Figure 5 This is the Bode diagram of the extended state observer of the present invention; Figure 6 This is a flowchart of the reference current value calculation method of the present invention; Figure 7 This invention relates to a current-tolerant controller; Figure 8 This is a diagram showing the spatial voltage vector distribution and sector division during normal operation of the present invention. Figure 9 This is a diagram showing the spatial voltage vector distribution and sector division during a single-phase open-circuit fault according to the present invention. Figure 10 This is a block diagram of the controller structure of the present invention. Detailed Implementation

[0016] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0017] Depend on Figure 1 The EHA consists of a motor pump, 3-accumulator, 4.1-A chamber check valve, 4.2-B chamber check valve, 5.1-A chamber relief valve, 5.2-B chamber relief valve, 6-hydraulic cylinder, 7-displacement sensor, 8-rotary transformer, and 9-current sensor. The motor pump, as a key component, is driven by a 1-five-phase permanent magnet synchronous motor. During operation, the rotation of the five-phase permanent magnet synchronous motor converts electrical energy into mechanical energy, which in turn drives the hydraulic pump to generate fluid power. This fluid power drives the piston movement by changing the pressure difference across the asymmetric cylinder, thereby achieving position control.

[0018] Depend on Figure 2 One fault-tolerant control method for EHA is as follows: Construct mathematical models for EHA under different operating states; A position controller is constructed based on proportional-integral (PI). A robust and dynamically responsive model predictive speed fault-tolerant control method (IMPSC-HRD) is established, including an improved model predictive speed control (IMPSC) module, a cascaded extended state observer based on the disturbance differential compensation and quasi-generalized integrator (DDC-QGI-CESO) module, and a fault-tolerant reference current calculation module. Establish a current-tolerant controller.

[0019] The mathematical models for different operating states of the EHA specifically include: Using F-PMSM as the motor of the EHA, a mathematical model of F-PMSM is established, including mathematical models for normal operation mode and winding open circuit fault.

[0020] The F-PMSM generates electromagnetic torque through the interaction between the five-phase winding current and the permanent magnet flux linkage. During normal operation, the fundamental space current interacts with the fundamental flux linkage to generate fundamental electromagnetic torque, while the third harmonic space current interacts with the third harmonic flux linkage to generate third harmonic electromagnetic torque. The overall electromagnetic torque model of the F-PMSM can be established as follows: ; in, It is the electromagnetic torque under healthy operating conditions. It is an extreme logarithm. It is a permanent magnet flux linkage. and They are shaft current and shaft current, and They are Shaft inductance and Shaft inductor, subscript Represents the fundamental frequency space. It represents the third harmonic space.

[0021] When an open-circuit fault occurs in an F-PMSM, the symmetry of the original five-phase winding currents is disrupted, and cross-coupling occurs between the fundamental and third harmonic spaces, leading to a change in electromagnetic torque. Therefore, it is necessary to re-establish the electromagnetic torque models for the fundamental and third harmonic spaces. Since the five-phase windings of the F-PMSM are symmetrically distributed, the impact of an open-circuit fault in any phase winding is the same. Therefore, this invention selects an open-circuit fault in phase A winding as a representative for modeling. Unless otherwise specified, the open-circuit fault referred to below refers to an open-circuit fault in phase A winding.

[0022] When an open-circuit fault occurs in the A-phase winding of an F-PMSM, the fundamental space current, in addition to interacting with the fundamental flux linkage to generate fundamental electromagnetic torque, also interacts with the third harmonic flux linkage to generate electromagnetic torque pulsations. The electromagnetic torque in the fundamental space is calculated as follows: ; in, It is the fundamental spatial electromagnetic torque during an open-circuit fault. It is the third harmonic current. It is an electrical angle.

[0023] Similarly, in addition to interacting with the third harmonic flux linkage to generate third harmonic electromagnetic torque, the third harmonic space current also interacts with the fundamental flux linkage to generate electromagnetic torque pulsations. The third harmonic space electromagnetic torque is calculated as follows: ; in, It is the third harmonic spatial electromagnetic torque during an open-circuit fault. It is the generalized zero-sequence current.

[0024] In summary, the total electromagnetic torque when an open-circuit fault occurs in the F-PMSM It can be calculated as: ; Furthermore, the mechanical kinematic equations of the F-PMSM are established as follows: ; in, It is the mechanical angular velocity of the F-PMSM. It refers to the displacement of the plunger pump. It is load pressure. B is the moment of inertia of the F-PMSM, and B is the damping coefficient of the F-PMSM. It is electromagnetic torque. When the F-PMSM is operating normally, for When an open-circuit fault occurs in the F-PMSM, for .

[0025] The hydraulic components in an EHA mainly consist of piston pumps and cylinders. The flow continuity equation for the hydraulic system can be established as follows: ; in, It is the rotational speed of the plunger pump, and . and These refer to the effective working areas of the rod-side chamber and the rodless chamber of the hydraulic cylinder, respectively. and These are the displacements of the plunger pump ports A and B, respectively. It refers to the displacement of the hydraulic cylinder piston rod. It is the speed of the piston rod. This is the initial position of the hydraulic cylinder piston. It is the elastic modulus of the oil. and These are the leakage coefficients of the plunger pump and the hydraulic cylinder, respectively. and These are the parameters for the hydraulic cylinder's operating mode, and their settings are shown in Table 1.

[0026] Table 1 Working Mode Parameters

[0027] Furthermore, the mechanical equations of the hydraulic cylinder can be established as follows: ; in, It is the acceleration of the piston rod of the hydraulic cylinder. It is the total mass of the cylinder piston rod. It is the external load force on the piston rod of the hydraulic cylinder. It is the frictional resistance of the hydraulic cylinder.

[0028] The PI-based displacement controller specifically includes: Collect displacement sensing information from hydraulic cylinders ; Based on the hydraulic cylinder force balance equation and the mathematical model of the hydraulic system, a hydraulic cylinder position controller is designed based on a PI controller, with the reference displacement set as... The actual displacement is The controller output variable is set to the motor reference speed. .

[0029] Depend on Figure 3The robust and dynamically responsive model predictive speed fault-tolerant control method (IMPSC-HRD) consists of three parts: a predictive speed control module (IMPSC), a cascaded extended state observer module (DDC-QGI-CESO) based on disturbance differential compensation and a quasi-generalized integrator (QGI), and a fault-tolerant reference current calculation module. The specific modules include: The IMPSC module is used for high-precision and high-dynamic tracking of the reference speed. In actual fault-tolerant control operation, a hybrid Euler discrete model is used for discretization. Firstly... The mechanical kinematic equations are discretized using the backward Euler method over the time interval, resulting in: ; in, and They represent the first and One control cycle. , . It is the time of a control cycle.

[0030] exist Discretizing the mechanical kinematic equations using the forward Euler method over the time interval yields: ; in, They represent the first One control cycle.

[0031] Then, based on the discretized equations, the final incremental model for predicting rotational speed is: ; in, It is the first Incremental electromagnetic torque per control cycle. No. Incremental pressure value for each control cycle.

[0032] The observer obtained and Substituting into the above equation, we finally obtain Speed ​​prediction value per control cycle as follows: .

[0033] based on and , This can be obtained by solving the following optimization problem: .

[0034] in, It is the maximum allowable torque limit.

[0035] By solving The optimal torque increment value under unconstrained conditions can be obtained. ,as follows: ; Because this univariate quadratic cost function has box constraints, the global optimum can be directly obtained by clamping (limiting) the unconstrained solution. Therefore, the... Optimal torque for each control cycle for: ; in This represents the saturation function. Ultimately, it is achieved through superposition. Estimated perturbation output from the DDC-QGI-CESO module The target torque value is obtained. .

[0036] Depend on Figure 4 The DDC-QGI-CESO module consists of two parts: DDC-ESO-1 and DDC-QGI-ESO-2. DDC-ESO-1 estimates a portion of the total disturbance and then converts the estimated value... The known disturbance is output to DDC-QGI-ESO-2. DDC-QGI-ESO-2 further estimates the remaining part of the total disturbance and outputs the estimated value. Finally, the total perturbation estimated by the DDC-QGI-CESO output. .

[0037] For the velocity control loop, the established state-space equations are as follows: ; in, It is the total disturbance. These are the nominal model parameters. ,and This is the nominal moment of inertia. ,in These are the actual model parameters. It is the final reference electromagnetic torque output by the IMPSC-HRD method. This is the actual electromagnetic torque output by the F-PMSM.

[0038] The discretized representation of DDC-ESO-1 is as follows: ; in, , and It is a state variable. . It is a partial total disturbance. It is the differential value of the total disturbance. , and yes , and The estimated value. , and It is the error feedback gain coefficient.

[0039] The discrete representation of DDC-QGI-ESO-2 is designed as follows: ; in, and It is a state variable. It is the remaining total disturbance. It is the differential value of the remaining total disturbance. , and yes , and The estimated value. , and It is the error feedback gain coefficient. The extracted current harmonics are represented as: ; in, and These are the internal state variables of the i-th quasi-generalized integrator (QGI). It is the generalized integral gain. Let be the frequency of the i-th harmonic current, and i = 0, 1, 2, 4. Specifically, , , , , It is the F-PMSM electric angular velocity, and .

[0040] To achieve optimal dynamic observation performance and ensure the absolute stability of the discretized system, this invention employs a pole placement method to adjust the parameters of the DDC-ESO-1. Based on the forward Euler discretization model, the desired continuous domain poles are placed... Mapping to the Z-plane, the desired discrete closed-loop poles are defined as follows: (in , (This represents the bandwidth of DDC-QGI-CESO). This is achieved by comparing the coefficients of the actual characteristic polynomial of the system with the desired characteristic equation. Since the coefficients are equal, the analytical gain parameter can be directly derived as follows: , and Similarly, DDC-QGI-ESO-2's... , and It can be calculated as , and .

[0041] Depend on Figure 5 Figure 1 shows the Bode plot of DDC-QGI-CESO, and Figure 2 shows the Bode plot of the traditional Cascaded Extended State Observer (CESO). Compared with the traditional CESO, the proposed DDC-QGI-CESO has a smaller amplitude in the low-frequency region, indicating that its disturbance estimation error is smaller. Since aperiodic disturbances are mainly concentrated in the low-frequency region, DDC-QGI-CESO has higher estimation accuracy for aperiodic disturbances than CESO. At the frequency of periodic disturbances, the amplitude of DDC-QGI-CESO decreases significantly compared to CESO, indicating that DDC-QGI-CESO has higher estimation accuracy for periodic disturbances than CESO.

[0042] The fault-tolerant reference current calculation module includes reference torque calculation and reference current calculation. The specific process for reference torque calculation is as follows: During the speed tolerance control of the F-PMSM, it is necessary to suppress electromagnetic torque pulsation, and the corresponding reference current calculation method differs from that used during normal F-PMSM operation. Therefore, it is necessary to recalculate the tolerance reference current under current constraint conditions.

[0043] During normal operation of the F-PMSM, due to the low salient polarity of the F-PMSM studied in this invention, the reluctance torque can be ignored, and the setting is... Meanwhile, due to the relatively small harmonic flux linkage, in order to reduce copper losses, the third harmonic reference current is usually... Set it to 0. Thus, the initial reference current is calculated. and ,as follows: .

[0044] In actual EHA operation, it is usually necessary to limit the phase current to prevent damage to the hardware system. During normal operation of the F-PMSM, the five-phase current distribution is uniform, and this can be satisfied by limiting the q1 axis current. ( The maximum effective value of the phase current is used to limit the phase current.

[0045] When an open-circuit fault occurs, based on the torque model, by setting... , ,and At the same time, set and When the following relationship is satisfied, not only can the electromagnetic torque ripple of the F-PMSM be suppressed to zero, but copper losses can also be reduced to the maximum extent: .

[0046] Calculate the initial reference current and ,as follows: ; Phase current , , and With q-axis current The relationship between them is: ; in, , , and These are the currents for phases B, C, D, and E, respectively.

[0047] From the above equation, the phase current amplitudes of phases B and E are the largest. Under these conditions, it is 1.5159 times the amplitude of the phase current during normal operation (motor parameters in this invention: To ensure that the phase currents of phases B and E do not exceed hardware limitations, the q1 axis current needs to meet certain conditions. Therefore, in the speed tolerance control process of F-PMSM, the q1 axis reference current under phase current constraint conditions... The calculation is as follows: ; in, ; Furthermore, the phase current constraint conditions can be calculated. , and ,as follows: .

[0048] Then, the reference electromagnetic torque considering phase current constraints can be calculated. That is, the output reference electromagnetic torque ,as follows: .

[0049] Depend on Figure 6 The specific process of the reference current calculation method is as follows: First, Clarke and Park inverse transforms are performed on the reference currents set in the fundamental and third harmonic rotating coordinate systems, respectively, to obtain the corresponding reference current values ​​in the natural coordinate system. Then, in the natural coordinate system, the reference currents of the same phase are added together to obtain the reference current for each phase. , , and Finally, Clarke and Park transformations are performed on the reference currents of each phase to obtain the reference currents in the rotating coordinate system. , and .

[0050] The current fault-tolerant controller, such as Figure 7 As shown, it includes a quasi-proportional resonant controller and a fault-tolerant space vector pulse width modulation (SVPWM) method.

[0051] The specific process of the quasi-proportional resonant controller is as follows: Based on the collected five-phase actual current values, the current in the rotating coordinate system is obtained through coordinate transformation. The reference current command obtained from the above reference current calculation module is subtracted from the actual calculated current in the rotating coordinate system to obtain the error signal, which is then input to the quasi-proportional resonant current fault-tolerant controller. The quasi-proportional resonant current fault-tolerant controller not only includes proportional gain... It also introduces a damped resonant term. ,in, For resonant gain, The damping coefficient is... It is the resonant frequency; it can broaden the system's resonant bandwidth while achieving zero steady-state error tracking of the fundamental or specific harmonic AC signals.

[0052] The fault-tolerant space vector pulse width modulation process specifically includes: The operating status of the motor is determined based on the actual current values ​​of the five phases collected, and is divided into normal operating status and winding open circuit fault. Under different operating conditions, establish the spatial voltage vector distribution in the corresponding static coordinate system, and divide the sector according to the principle of vector average distribution, such as... Figure 8 , 9 As shown; The compensated target voltage value is then transformed to obtain the target voltage value in the stationary coordinate system. and ; Based on the target voltage value in the stationary coordinate system and Select the sector where the target voltage vector is located in the stationary coordinate system.

[0053] Select the space voltage vector used to synthesize the target voltage vector, perform voltage vector synthesis, calculate the action time of the space voltage vector participating in the synthesis of the target voltage vector, and further solve the control duty cycle of each phase of the inverter.

[0054] Depend on Figure 10 The controller structure used to implement this invention includes an STM32H723-based microcontroller responsible for coordinating all peripheral devices, receiving sensor feedback, executing the core fault-tolerant control algorithm, and calculating the action command for the next cycle. An encoder decoding module decodes the acquired resolver signals. A pulse width modulation (PWM) drive signal generation module converts digital control quantities into high-frequency pulse width modulation (PWM) signals. A temperature acquisition module monitors the temperature of the driver and EHA. A memory stores system parameters. The bus interface circuit is responsible for communication between the control board and the host computer (such as a Programmable Logic Controller (PLC) or industrial computer) or other nodes. Common protocols include CAN (Controller Area Network), RS485 (Recommended Standard 485), or Ethernet, used to receive operating commands or upload operating status. The emergency stop program, upon receiving a signal from an external hardware emergency stop button, immediately interrupts the microcontroller's normal operations, cuts off all power output, and forces the system into a safe shutdown state. Ground fault detection monitors in real time whether a phase-to-ground fault or leakage has occurred. Once leakage current exceeds the safety threshold, it immediately reports an error to the microcontroller or directly blocks the PWM output at the hardware level to protect equipment and personnel safety. The bus voltage sampling module monitors the DC bus voltage in the power circuit in real time. The signal conditioning module receives current, pressure, and position signals from external sources and converts the raw physical quantities into microcontroller ADCs (Analog to Digital) through hardware circuits such as filtering (removing noise), amplification (enhancing the signal), isolation, and level conversion. A converter (analog-to-digital converter) can safely and accurately read low-voltage analog or digital signals.

[0055] The above embodiments are illustrative of the present invention and are not intended to limit the present invention. Any simple modifications to the present invention are within the scope of protection of the present invention.

Claims

1. A fault-tolerant control method for an electro-hydraulic actuator, characterized in that, Includes the following steps: Step 1: Construct mathematical models of the electro-hydraulic actuator under different working states, including the normal operation state and the winding open circuit fault state of the five-phase permanent magnet synchronous motor. Step 2: Construct a position controller based on a proportional-integral controller to output the motor reference speed; Step 3: Establish a model-predicted speed fault-tolerant control method, which includes a predictive speed control module, a cascaded extended state observer module, and a fault-tolerant reference current calculation module; Step 4: Establish a current fault-tolerant controller, which includes a quasi-proportional resonant controller and a fault-tolerant space vector pulse width modulation method.

2. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The mathematical models for the electro-hydraulic actuator under different operating states in step one include: The total electromagnetic torque model of a five-phase permanent magnet synchronous motor during normal operation is as follows: ,in, It is the electromagnetic torque under healthy operating conditions. It is an extreme logarithm. It is a permanent magnet flux chain; and They are shaft current and shaft current, and They are Shaft inductance and Shaft inductor, subscript Represents the fundamental frequency space. Represents the third harmonic space; The total electromagnetic torque during a winding open-circuit fault is: ,in, , It is the fundamental spatial electromagnetic torque during an open-circuit fault. It is the third harmonic current. It is an electrical angle; , It is the third harmonic spatial electromagnetic torque during an open-circuit fault. It is the generalized zero-sequence current; And the kinematic equations of mechanics: ,in, It is the mechanical angular velocity of a five-phase permanent magnet synchronous motor. It refers to the displacement of the plunger pump. It is load pressure. B is the moment of inertia of the five-phase permanent magnet synchronous motor, and B is the damping coefficient of the five-phase permanent magnet synchronous motor. It is electromagnetic torque; Flow continuity equation for hydraulic systems: ,in, It is the rotational speed of the plunger pump, and ; and These refer to the effective working areas of the rod-side chamber and the rodless chamber of the hydraulic cylinder, respectively. and These are the displacements of the plunger pump's oil ports A and B, respectively. It refers to the displacement of the hydraulic cylinder piston rod. It is the speed of the piston rod; This is the initial position of the hydraulic cylinder piston; It is the elastic modulus of the oil; and These are the leakage coefficients of the piston pump and the hydraulic cylinder, respectively. and These are the parameters for the hydraulic cylinder's operating mode; The mechanical equations of the hydraulic cylinder: ,in, It is the acceleration of the piston rod of the hydraulic cylinder. It is the total mass of the cylinder piston rod. It is the external load force on the piston rod of the hydraulic cylinder. It is the frictional resistance of the hydraulic cylinder.

3. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The predicted speed control module in step three uses a hybrid Euler discrete model for discretization: Step 31: In The mechanical kinematic equations are discretized using the backward Euler method over the time interval, resulting in... ,in, and They represent the first and One control cycle; , ; It is the duration of a control cycle; Step 32: In The mechanical kinematic equations are discretized using the forward Euler method within the time interval, resulting in... ,in, They represent the first One control cycle; finally, the speed prediction model of the incremental model is obtained as follows: ,in, It is the first Incremental electromagnetic torque per control cycle No. Incremental pressure value for each control cycle; Step 33: Obtain the data from the observer and Substituting into step 32, we finally obtain Speed ​​prediction value per control cycle as follows: ; Step 34: Based on and , This can be obtained by solving the following optimization problem: ,in, This is the maximum allowable torque limit; Step 35: Solve The optimal torque increment value under unconstrained conditions can be obtained. ,as follows ; Step 36: Clamp the unconstrained solution to obtain the global optimal solution: Optimal torque for each control cycle for: ,in Represents a saturation function; Step 37: By overlay The estimated perturbation output by the cascaded extended state observer module The target torque value is obtained. .

4. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The cascaded extended state observer module in step three includes: The first-level extended state observer estimates a portion of the total disturbance and outputs the estimate as a known disturbance. Its discretized form is as follows: ,in, , and It is a state variable. ; It is a partial total disturbance. It is the differential value of the total disturbance. , and yes , and The estimated value; , and It is the error feedback gain coefficient; and They represent the first and One control cycle, It is the duration of a control cycle; It refers to the displacement of the plunger pump. It is the load pressure; These are the nominal model parameters. ,and The nominal moment of inertia; It is the final reference electromagnetic torque; The second-stage extended state observer, containing a quasi-generalized integrator, is used to extract current harmonics and estimate the remainder of the total disturbance. Its discretized form is as follows: ,in, and It is a state variable. It is the remaining total disturbance. It is the differential value of the remaining total disturbance. , and yes , and The estimated value; , and It is the error feedback gain coefficient; For the extracted current harmonics; These are the nominal model parameters; and These are the internal state variables of the i-th quasi-generalized integrator. It is the frequency of the i-th harmonic current; The first-stage extended state observer and the second-stage extended state observer use the pole placement method to adjust parameters.

5. The fault-tolerant control method for an electro-hydraulic actuator according to claim 4, characterized in that, The quasi-generalized integrator is used to extract the 0th, 1st, 2nd and 4th harmonic current frequencies.

6. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The fault-tolerant reference current calculation module in step three ignores the reluctance torque and sets the following settings during normal operation of the five-phase permanent magnet synchronous motor. At the same time, the third harmonic reference current Set to 0 to calculate the initial reference current. and for: ,in It is a permanent magnet flux linkage in the fundamental space; by limiting the q1-axis current to satisfy... (in, The purpose of limiting phase current is to use the maximum effective value of the phase current. The target torque value; When a single-phase open-circuit fault occurs in a five-phase permanent magnet synchronous motor, by setting... , ,and At the same time, set and The following relationship must be satisfied: Let the q1-axis current satisfy... The reference electromagnetic torque considering phase current constraints is calculated. That is, the output reference electromagnetic torque : ,in It is a permanent magnet flux linkage in the third harmonic space.

7. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The quasi-proportional resonant current-tolerant controller in step four includes a proportional gain. Damped resonant terms ,in, For resonant gain, The damping coefficient is... It is the resonant frequency.

8. The fault-tolerant control method for an electro-hydraulic actuator according to claim 1, characterized in that, The fault-tolerant space vector pulse width modulation method in step four specifically includes: Step 41: Determine the motor's operating status based on the collected actual five-phase current values, which can be divided into normal operating status and winding open circuit fault. Step 42: Establish the spatial voltage vector distribution in the corresponding static coordinate system under different working conditions, and divide the sectors according to the principle of vector average distribution; Step 43: Perform coordinate transformation on the compensated target voltage value to obtain the target voltage value in the stationary coordinate system; Step 44: Select the sector where the target voltage vector is located based on the target voltage value in the stationary coordinate system; Step 45: Select the space voltage vector used to synthesize the target voltage vector, calculate the duration of each space voltage vector, and solve for the control duty cycle of each phase of the inverter.

Images