121results about How to "Implement fault-tolerant control" patented technology

The invention discloses an aircraft posture robust inversion fault tolerant control method based on a neural network observer. Firstly, a posture dynamic equation of a near space aircraft X-33 is given, and a control surface fault model is built; the adaptive neural network observer is designed according to the control surface fault model of a posture angular rate loop, and a posture angle loop and a designed observer dynamic equation are combined; by the adoption of an instruction filtering inversion method, an angle ring controller and an angular speed ring controller are designed. According to the aircraft posture robust inversion fault tolerant control method based on the neural network observer, precise fault information and interference information are not needed and are respectively hidden in the designed adaptive neural network observer, the hidden information is fed back to the controllers in real time, and robust fault tolerant control is achieved; the problem that robustness and sensibility are needed by FDI is solved, and robust fault tolerant control is achieved; finally, the method is applied to stability control and tracking control over the near space aircraft under control surface fault situations, robust fault tolerant control over the flying posture is achieved, and good control performance and a good control effect are achieved.

The invention relates to a dynamic control reallocation based unmanned aerial vehicle attitude fault-tolerance control method. According to the method, dynamic inversion control is adopted, an attitude angle rate loop dynamic equation corresponding to a near-space vehicle is given firstly, further a control surface stuck and damage model is established, then an adaptive sliding mode observer is designed according to the control surface stuck and damage model of an attitude angle rate loop, the attitude angle rate loop and a designed observer dynamic equation are simultaneous, an angle loop controller and an angle rate loop controller are designed, a designed fault-tolerance control system does not need accurate damage information, interference information and the like which are all implied in the designed adaptive sliding mode observer, the implicit information is fed back to a pseudocontrol control law and an online reallocation algorithm LMI in real time so as to realize robust fault-tolerance control, the designed method is applied to attitude stabilization control and tracking control of the near-space vehicle under the condition of control surface faults, and flight attitude robust fault-tolerance control is realized.

The invention discloses an aircraft attitude fault-tolerant control method based on dynamic control redistribution. The method includes the steps that firstly, an attitude angle rate loop dynamic equation of a near-space aircraft X-33 is given, and a control surface jamming and damage fault model is established; then according to the control surface damage fault model of an attitude angle rate loop, a self-adaptive sliding-mode observer is designed; the attitude angle loop and the designed observer dynamic equation are combined, and an angle loop controller and an angular speed loop controller are designed. Accurate damage fault information, interference information and the like are not needed, the information is implied in the designed self-adaptive sliding-mode observer, the implicit information is fed back to a pseudo control law and online distribution algorithm LMI in real time, and then robust fault-tolerant control is achieved. Finally, the designed method is applied to attitude stability control and tracking control of the near-space aircraft under the control surface fault condition, flight attitude robust fault-tolerant control is achieved, and good control performance and effects are achieved.

The invention provides a fault diagnosis method for a fuel cell system. The fault diagnosis method comprises the steps of executing offline processing and executing on-line processing; the off-line processing is used for establishing a fault database according to operation and fault data of a fuel cell, collected by a plurality of sensors and calculating a fault threshold value; and the on-line processing is used for identifying the type of a fault according to the fault database and the fault threshold value, judging the type of the fault according to the fault database, if the fault is reversible and a fault component is a non-core component, reconstructing a sensor signal corresponding to the fault component to enable the fuel cell system to continue running, and if the fault is irreversible or the fault component is a core component, shutting down and maintaining the fuel cell system. The method has the advantages that signal authenticity verification, fault diagnosis, fault positioning and fault-tolerant control are realized, and signal reconstruction is realized through an algorithm to maintain power output in a fault state, so that the safety and reliability of the fuel cellsystem for a vehicle are ensured; and the method can be suitable for different fuel cell systems.

The invention relates to an open winding motor system and a drive circuit and belongs to the technical field of opening winding motor control. Redundant bridge arms and redundant inductors are set inthe drive circuit. When an open circuit fault occurs in certain phase of winding or certain bridge arm, the fault winding or bridge arm stops working, the redundant inductors and the redundant bridgearms are put into operation, and fault-tolerant control is realized. Even under fault condition, a connection method of motor windings is not limited, and a switching drive signal originally sent to the fault bridge arm is forwarded to the redundant bridge arms, so the control is simple, secure and reliable. Even if the fault occurs in certain phase, the circuit still can be operated normally, normal drive is realized, and reliability of an inverter circuit is improved.

The present invention relates to a fault-tolerant control method for the position sensor fault of a brushless direct current motor rotor. The invention is structurally characterized in that the hall control mode and the non-hall control mode are integrated to realize the automatic switching function. According to the technical scheme of the method, the abnormal condition of the position signal of the motor rotor is detected, and then the system continues to control the motor when being switched from the position sensor mode to the non-position sensor mode. Meanwhile, the fault-tolerant control is realized at the same time. In the position sensor mode, a hall signal is adopted as a motor rotor phase-change basis. In the non-position sensor mode, the three-phase counter electromotive force signal of a motor is adopted as a motor rotor phase-change basis. In this way, the fault diagnosis method can quickly locate a fault and to select a corresponding mode. As a result, the fault-tolerant control of the motor is realized.

The invention discloses a fault tolerance control method based on an asymmetric double three-phase permanent magnet synchronous motor uniform model. The objects of the invention are to establish an asymmetric double three-phase permanent magnet synchronous motor uniform model, to realize the decoupling control of the asymmetric double three-phase permanent magnet synchronous motor based on the uniform model and to improve the redundancy of the system. Aiming for asymmetry of the double three-phase permanent magnet synchronous motor under various faulted conditions and the condition that traditional coordinate transformation cannot realize decoupling control, the fault tolerance control method comprises steps of establishing a double three-phase permanent magnet synchronous motor uniform model under various faulted conditions, analyzing a coupling relation of a voltage equation in a d-q subplane, putting forward two times of rotation transformation to realize the decoupling of the d-q subplane, and putting forward the fault tolerance control method having minimum stator copper consumption or minimum stator current. The invention can effectively reduce the electromagnetic torque pulsation when the double-three-phase permanent magnet synchronous motor is in phase-deficient operation, and can realize the magnetic field directive decoupling control in the phase-deficient operation.

The invention discloses a current space vector error projection correction-based fault-tolerant method for a current sensor of a position sensorless driving system, and relates to the field of motor control, and aims to solve the problem that in an existing vector control technology of a permanent magnet synchronous motor, a sensor is required to collect a signal to guarantee reliable driving of the permanent magnet synchronous motor, but once the sensor breaks down, the permanent magnet synchronous motor cannot normally run, so that the reliability is poor. An estimated speed and estimated position information of the permanent magnet synchronous motor are obtained; when a certain phase current sensor breaks down, a current error reconstruction link is used for obtaining a current space vector error according to a phase current error, and projecting the error to obtain a reconstructed phase current error; according to the reconstructed phase current error, the phase current error, theestimated speed, an alpha-axis voltage and a beta-axis voltage, an estimated alpha-axis current value and an estimated beta-axis current value are obtained; and performing closed-loop control on the permanent magnet synchronous motor by using values obtained after conversion of the estimated values. The method is used for correcting a phase current to control the motor.

The invention discloses a fault-tolerance control method of a multi-mode dual-channel switched reluctance motor system. A stator, a rotor, two sets of windings and two standard three-phase inverters are included. A star connection mode is adopted by the two sets of windings, and neutral points are connected together. The two standard three-phase inverters drive the two sets of windings to form themulti-mode dual-channel switched reluctance motor system. The system has two operation modes. An operation mode 1 is a traditional three-phase switched reluctance motor operating mode driven by an H-bridge inverter, and driving current is a square wave. An operation mode 2 is a double-inductance coupled switched reluctance motor mode, and the drive current is a sine wave. When the system has a switch device fault, the operation mode 1 is selected to carry out fault-tolerance control, and when the system has a phase open-loop fault, the operation mode 2 is selected to carry out fault-tolerancecontrol. According to the fault-tolerance method, the remaining control freedom of the system after the fault is fully used, the invariability of an average output torque before and after the fault is ensured, and the torque ripple can be effectively reduced.

The invention discloses an event-driving-based fault-tolerant control method under network attack of an unmanned surface vehicle and relates to a fault-tolerant control method under the network attackof an unmanned surface vehicle, thereby solving problems of high network communication bandwidth occupancy rate and high data transmission energy consumption in the existing unmanned surface vehiclesystem. The method comprises the following steps: step one, establishing a state space equation of an unmanned surface vehicle system according to a kinematics equation of the unmanned surface vehicle; step two, introducing a random spoofing attack based on the state-space equation of the unmanned surface vehicle system and establishing a mathematical model of the random spoofing attack; step three, establishing a state space equation of the unmanned surface vehicle system with the spoofing attack introduced according to the mathematical model of the random spoofing attack; step four, designing a fault-tolerant controller gain matrix based on dynamic event driving; and step five5, designing a dynamic event driving mechanism based on the dynamic-event-driving-based fault-tolerant controllergain matrix. The method is used for fault-tolerant control under the network attack of the unmanned surface vehicle.

The invention provides a multi-motor total amount collaborative consistency fault-tolerant control method based on a virtual total axis. Based on the feedback characteristic of a virtual main axis control strategy, once a certain wheel set of a high-speed train has a slip or idling fault, to prevent the fault from continuously happening, the traction total torque is reduced by using the feedback regulation characteristic of the virtual total axis; a multi-motor system can also keep dynamically tracking the moment change so as to keep the total amount of traction power of the system consistent;the total moment is restored to be normal after the adhesion fault condition is restored to be normal, and the total amount consistency of the traction power is guaranteed while the system recovers and maintains the steady state.

The invention provides a dual three-level inverter model prediction fault-tolerant control strategy. In a value function, by using the stator current as a tracking target, the constraint of mid-pointpotential balance is introduced to adjust the DC bus voltage, and the constraint term of the switching frequency is added. Then, a zero-sequence loop is limited based on the degree of freedom of the zero-sequence voltage, and during a single-arm fault, an available zero common-mode vector is selected as an optimized target vector so as to eliminate the zero-sequence current and greatly reduce thecalculation delay. In the case of a single-transistor fault, the available zero common-mode vector cannot satisfy the control needs, and the subset of the optimized switch vector is selected discretely. Finally, the proposed scheme is verified by simulation and compared with a traditional scheme, and a result verifies the effectiveness and feasibility of the proposed method.

The invention provides an automatic fault-tolerant attitude control method for a six-rotor unmanned plane, and the method comprises the following steps: collecting currents and/voltages of a pluralityof motors on the six-rotor unmanned plane, and judging whether each motor has a fault or not according to the currents and/or voltages; turning off the motor which is adjacent to the motor with the fault and rotates in a direction opposite to the rotating direction of the motor with the fault when the motor is judged to be the motor with the fault; selecting a corresponding mapping matrix Ri according to the number of a propeller wing i driven by the motor with the fault, converting an input attitude angle into a control attitude angle according to the mapping matrix Ri, wherein i is a natural number which is greater than or equal to zero and is less than or equal to six; and inputting the control attitude angle into a flight control module of the six-rotor unmanned plane so as to controlthe six-rotor unmanned plane to fly. According to the invention, when each motor in the six-rotor unmanned plane has an inclined installation angle, the method can achieve the optimization of the power distribution according to the fault condition of the motor, and achieves the fault-tolerant control of the six-rotor unmanned plane.

The invention discloses a displacement sensor fault-tolerant control system and method for an active radial magnetic bearing. The displacement sensor fault-tolerant control system and method are matched with the active radial magnetic suspension bearing; the system comprises a displacement feedback part, a fault diagnosis circuit, a fault-tolerant control module and a feedback execution part, wherein the displacement feedback part is fixedly arranged on the periphery of a rotor of the active radial magnetic suspension bearing in an asymmetric mode; the output of the displacement feedback partis electrically connected with the input of the fault diagnosis circuit and the input of the fault-tolerant control module respectively; the output of the fault diagnosis circuit is electrically connected with the input of the fault-tolerant control module; the output of the fault-tolerant control module is electrically connected with the input of the feedback execution part; and the output of thefeedback execution part is electrically connected with an electromagnetic coil in the active radial magnetic suspension bearing. According to the displacement sensor fault-tolerant control system andmethod, the number of displacement sensors is greatly reduced, the fault types capable of being subjected to fault-tolerant control are increased, and it is ensured that the displacement feedback part can have the same sensitivity and measurement precision before and after the fault.

Disclosed in the invention is an intelligent control method of an electrically-driven motor. According to the control method, with two sets of windings of the motor, the load is driven to be operated by the motor state by using a PI controller, an integral limiter, a prediction controller and an inverter; and the speed adjustment performance is good. The prediction controller is used for replacing the traditional current PI controller, so that the defect of the hysteresis characteristic of the current PI controller is overcome and the band width and dynamic performances of the system are improved. Besides, functions of redundancy communication, fault diagnosis, position signal fault detection and calibration are added and thus the winding open-circuit and short-circuit existence at the control system or position signal faults can be diagnosed and processed respectively. According to the invention, the method has characteristics of simpleness, easy implement, and high intelligence and reliability and the like; fault-tolerant control of the system open circuit, short circuit or position signal fault can be realized well; and the method is suitable for power driven occasions with high requirements of the dynamic performance and reliability.

The invention discloses a rapid self-adaptive sliding mode fault-tolerant control method for a hypersonic flight vehicle. The method comprises the following steps that 1, an accurate feedback linearization model for the hypersonic flight vehicle is established; 2, a fault model for the hypersonic flight vehicle is established on the basis of the first step; 3, a basic controller is designed; and 4, self-adaptive fault-tolerant controller is designed by adding a self-adaptive law on the basis of the third step. The rapid self-adaptive sliding mode fault-tolerant control method for the hypersonic flight vehicle has the advantages that under the condition that the failure time and the efficiency loss coefficient are not known, the fault is processed in a self-adaptive mode, an asymptotic tracking flying output reference command is completed, and high efficiency is achieved.

The invention discloses an inverter-stage fault tolerant control method of a full bidirectional switch-type two-stage matrix converter, belonging to the technical field of fault tolerant control of an electronic power device. A one-way switching tube in the original two-stage matrix converter is substituted by a bidirectional switch unit to form the full directional switch-type two-stage matrix converter; fault tolerant control methods for inverter-stage single tube short circuit of the full bidirectional switch-type two-stage matrix converter, inverter-stage bidirectional switch unit open circuit, inverter-stage bidirectional switch unit short circuit and inverter-stage single tube open circuit are proposed; the converter has the ability of bearing reverse voltage by additionally arranging six switching tubes at an inverter stage; and the input and output performance are compatible as well as the fault tolerant strategy of the two-stage matrix converter is simplified.

The invention discloses a short-circuit compensation control method of a four-phase permanent-magnetic fault-tolerant motor. A concentrated winding is adopted to be connected with an independent H-bridge-driving inverter to realize complete fault isolation. When a winding of a certain phase is in a short-circuit fault, a big short-circuit current may be generated at the phase to cause unbalanced running of the motor. By utilizing a torque equivalence principle and adjusting amplitude values and phase positions of other non-fault-phase currents, equivalence in electromagnetic torque output by the motor before and after faults is realized to realize short-circuit fault-tolerant control.

The invention provides a PMSM (permanent magnet synchronous motor) control device and method based on a single current sensor, and the device comprises a PMSM connected with a power supply. The exterior of the power supply is sequentially provided with four branch lines from the left to the right, and the four branch lines sequentially comprise a voltage dividing branch line and three phase-splitting branch lines. The voltage dividing branch line is provided with two capacitors which are longitudinally adjacent to each other in a width direction, and each of the three phase-splitting branch lines is provided with an inverter switching tube. Each phase-splitting branch line is connected with a three-phase winding in the PMSM through a first wire, and a midpoint between the two capacitors and the midpoint of the three-phase winding are connected through a second wire. The lower ends of the phase-splitting branch lines are connected with the negative electrode of the power supply througha DC bus, and the single current sensor is disposed outside the DC bus and the second wire. The device can achieve the control of the PMSM in a normal operation mode and a fault-tolerant operation mode through one single current sensor, is simple in structure, is simple in control method, and is higher in control efficiency.