240results about How to "Suppress disturbance" patented technology

A 3D memory device includes an array of semiconductor body pillars and bit line pillars, dielectric charge trapping structures, and a plurality of levels of word line structures arranged orthogonally to the array of semiconductor body pillars and bit line pillars. The semiconductor body pillars have corresponding bit line pillars on opposing first and second sides, providing source and drain terminals. The semiconductor body pillars have first and second channel surfaces on opposing third and fourth sides. Dielectric charge trapping structures overlie the first and second channel surfaces, providing data storage sites on two sides of each semiconductor body pillar in each level of the 3D array. The device can be operated as a 3D AND-decoded flash memory.

The invention relates to a piezoelectric switch type valve positioner control method and a piezoelectric switch type valve positioner control system. The system consists of a 4-20mA current source, a current-to-voltage (I / V) conversion circuit, a voltage conversion module, a control signal amplifying and filtering circuit, a feedback signal amplifying and filtering circuit, 2 external analog-to-digital converter (ADC), an ultra-low power consumption singlechip, a liquid crystal keyboard module, an electrically erasable programmable read-only memory (EEPROM), a passive crystal, and a piezoelectric valve driven amplifying circuit. The control method comprises the steps of: determining the optimum pulse width modulation (PWM) duty cycle during the self-tuning of parameters; and adopting a five-step switching method with reverse PWM during the control. The piezoelectric switch type valve positioner control method and the piezoelectric switch type valve positioner control system can effectively avoid the overshoot phenomenon during the control of a valve positioner, solve the non-linear problem of an actuating mechanism, and have stronger universal applicability.

The invention belongs to the field of power systems, and in particular to a method which uses an additional damping controller to modulate active and reactive output of a doubly-fed wind generator to increase low-frequency oscillation damping of a grid connection power system of a large-scale wind farm. According to the invention, the method uses auto-disturbance rejection control technology, introduces the theory of data driving control, engineers an additional damping controller of the wind generator, and makes feature signal of the connecting points of a wind farm as input of the additional damping controller which outputs a signal to be added to active and reactive modulation quantity of a machine side current converter. According to the invention, simulation of the method proves that negative damping low frequency oscillation and forced power oscillation are better controlled, and the method has stronger robustness and engineering practicality. The invention provides a novel method for adjusting parameters of a state observer of a second-order non-linear auto-disturbance controller based on broadband concept.

The invention discloses a parallel-connection active power filter suitable for a three-phase four-wire power grid system, which mainly comprises a three-phase full-bridge inverter circuit, a half-bridge chopper circuit and two capacitors; the three-phase full-bridge inverter circuit is electrically connected with the three-phase four-wire power grid system and connected in parallel with the half-bridge chopper circuit; and the other side of the half-bridge chopper circuit is electrically connected with a midpoint of the two capacitors at a direct-current side through an inductance, and the half-bridge chopper circuit is used for controlling the direct-circuit side voltage of the parallel-connection active power filter. A double close-loop control mode is adopted in the invention, wherein the inner ring is a voltage ring which is used for providing energy feedforward for controlling the direct-current end voltage stabilization and restraining zero-sequence current perturbance; and the outer ring is a current ring which is used for controlling the output current of the parallel-connection active power filter. Compared with the prior art, the parallel-connection active power filter with the double close-loop control structure can be used for preferably compensating the power grid harmonic current and restraining three-phase imbalance.

The invention relates to an automatic solar tracking system which tracks the sun accurately in real time and is used for efficiently collecting and using solar energy. The automatic solar tracking device comprises a solar energy photovoltaic array, an adjusting device connected with the solar energy photovoltaic array and used for adjusting the position of the solar energy photovoltaic array, a stand column for supporting the adjusting device and a control system. The adjusting device is composed of a pitch angle adjusting assembly and a position angle adjusting assembly, wherein the pitch angle adjusting assembly is connected with the position angle adjusting assembly through a vertical shaft, the position angle adjusting assembly is arranged on the stand column, and the control system is used for performing closed loop control on the adjusting device. The accuracy in solar tracking of the automatic solar tracking device is smaller than 0.1, the energy consumption only occupies for 0.5% of the electric energy production of the driving photovoltaic array, and the automatic solar tracking system has the advantages of being long in service life, intact in protection function and the like.

The invention provides a flexible satellite neural network backstepping sliding mode attitude control method, relates to a flexible spacecraft attitude control method, and aims at solving a problem of perturbation generated by plate flexible vibration and antenna rotation and a problem of enhancement of steady state precision and stability of existing attitude control methods. According to the method, a flexible satellite attitude kinetic model is established firstly according to a spacecraft, and then a model formula is processed; a sliding mode attitude controller (which is expressed in the specification) based on a backstepping method is designed; then an RBF neural network approach (eta+htau)sgn(sigma) is adopted; the controller (which is expressed in the specification) is designed; and finally a complete attitude controller (which is expressed in the specification) is obtained. A three-axis attitude controller is designed respectively according to the aforementioned process. The flexible satellite neural network backstepping sliding mode attitude control method is suitable for the field of flexible spacecraft attitude control.

The invention relates to an autopilot flight control system. The autopilot flight control system comprises a flight control panel, a radio station, a servo mechanism, a receiver and a remote controller, wherein the radio station is communicated with the flight control panel in a two-way mode on remote control and telemetering, the servo mechanism receives servo control instructions of the flight control panel, the receiver transmits manual control instructions to the flight control panel, and the remote controller transmits remote manual control instructions to the receiver. The flight control panel further comprises a central processing module, a communication module which communicates with the central processing module in a two-way mode, a sensor module which communicates with the central processing module in the two-way mode, and a servo output module which receives instructions of the central processing module, wherein the communication module of the flight control panel receives instructions of the receiver and communicates with the radio station in the two-way mode. The radio station communicates in the two-way mode with a ground control system outside the autopilot flight control system. The servo output module transmits servo information to the servo mechanism. The autopilot flight control system is small in structure, simple and clear in system design, high in computation efficiency and reliability and very good in expansibility and portability.

The invention discloses a range hood centrifugal fan with a double air inlet structure. The range hood centrifugal fan comprises a volute and an impeller arranged in the volute, the volute comprises a front cover plate, a rear cover plate, an annular middle shell and volute tongues, the annular middle shell is connected between the front cover plate and the rear cover plate, a motor is arranged at the center of the impeller, each volute tongue is provided with a volute tongue surface projecting along a direction far away from the joint of the volute tongue and the shell, each volute tongue surface comprises a front inclined curved surface and a rear inclined curved surface which are asymmetric and are in smooth transition connection, and the intersecting line of the two inclined curved surfaces is the outermost projection of the volute tongue surface. By the aid of the bidirectionally inclined volute tongues, distribution of the inclined curved surfaces conforms to airflow velocity distribution, a new noise source generated at the volute tongue more effectively offsets part of original noise, and noise reduction capacity is improved. The intersecting line of the two inclined curved surfaces of each bidirectionally inclined volute tongue is aligned to a central disk of the impeller as far as possible, the inclined directions of the volute tongues corresponding to air inlets in two sides of the double air inlet impeller are consistent, and better noise reduction effects are achieved.

The invention relates to a control method for switching power supply, whose specific practices are as following: a voltage checking electric circuit (VCC) examines the output voltage V<0> of a converter then delivered to an error amplifier (VA), the error amplifier (VA) carries on the comparison with an output reference voltage V and the output voltage V<0> to get an error voltage value delta V at the start time of each switching period; an error interval judgment (VC) compares the error voltage value delta V with error interval delta n, n=1,2,... N (N=2-5) of the output voltage, control pulse selection signals are outputted to cause a multistage pulse generator (MPG) to generate multistage control pulse with different dutyfactors according to the comparison result, so that the switch tube (TD)of the converter is carried on multistage control. When using the control method, the output voltage ripple of the converter is small, the dynamic response is good, anti-interference capability is strong, and the method is suitable for converters of various topologies.

The invention discloses a centrifugal fan impeller, which comprises a front ring and a back ring, wherein a plurality of blades are fixedly arranged between the front ring and the back ring. The centrifugal fan impeller is characterized in that the outlet ends of the blades form a wave-shaped sawtooth type by a plurality of sawteeth. The outlet ends (tail edges) of the blades are provided with structures in the wave-shaped sawtooth type, and the wake vortex shedding positions of the tail edges of the blades are changed, so the distance between the vortex centers is increased, the disturbance of the shedding vortex on the wake flowing is inhabited, and the pneumatic noise caused by the wake vortex and the unsteady pressure fluctuation of the blade surfaces is further reduced. The height and the width of the sawteeth are limited, the noise is further reduced, and the influence of the sawteeth on the pressure and air quantity reduction of a fan can also be reduced.

The invention relates to a method for the multi-pulse sequence control of a switch power supply working in a pseudo continuous mode, which comprises the following steps that: a voltage detection circuit (VCC) detects the output voltage V0 of a convertor and transmits the output voltage to an error amplifier (VA); at the beginning time of each switch cycle, the error amplifier (VA) compares an output reference voltage Vref with the output voltage V0 to generate an error voltage value delta V; and an error burst judger (VC) compares the error voltage value delta V with a preset error burst value deltan ( n is an integer from 1 to N) of N(N is more than or equal to 1) output voltages, and outputs a control pulse selection signal according to the result of the comparison to make a multi-pulse generator (MPG) generate control pulses Pn1 and Pn2 with different duty cycles to control the switching tubes S1 and S2 of the a converter TD respectively, wherein the n is an integer from 1 to N+1. The method can realize the multi-pulse sequence control of the switch power supply working in a pseudo continuous mode. The converter adopting the control method has current critical condition limitation-free output power, small output voltage ripples, excellent dynamic response and high disturbance resistance, and is applicable to convertors of various topological structures.

The invention discloses a small-sized unmanned rotary-wing aircraft high-precision control method based on an adaptive neural network, and relates to the design of a composite controller with the combination of the construction and optimization of the feedback control and non-sample training adaptive neural network of a small-sized unmanned rotary-wing aircraft. Firstly, as for a small-sized unmanned rotary-wing aircraft dynamical model, a feedback control coefficient matrix is constructed through a pole assignment method to ensure the preliminary stability of a system. Secondly, the adaptive neural network with independent updating weight features is designed, an adaptive network weight updating matrix is constructed based on error messages to update a weight matrix of the neural network in an online mode, and estimation and restraint of disturbance are achieved. An adaptive threshold value optimizing strategy is designed, online updating is carried out on a control residual error upper limit threshold value of the adaptive neural network on the basis of the mean square error between the actual position and the expectation position in a time window, and the small-sized unmanned rotary-wing aircraft high-precision attitude control under the complex environment is achieved. The small-sized unmanned rotary-wing aircraft high-precision control method has the advantages of being good in real-time performance, fast in dynamic parameter response, strong in multi-source interference adaptability and the like, can be used for the high-precision control over the small-sized unmanned aircraft under the complex multi-source interference environment and the like.

The invention discloses an INS (inertial navigation system)-assisted wireless indoor mobile robot positioning method, belonging to the technical field of wireless positioning of a robot. The positioning method comprises a training stage and a pre-estimating stage. The training stage comprises the following steps of: integrating an INS and a WSN (web shell navigator) in a local relative coordinate system; and carrying out data fusion on obtained synchronized navigation data by expanding a Kalman filtering wave to obtain continuous and stable navigation information. The pre-estimating stage comprises the following steps of: inputting the position and speed information measured by the INS into the training stage, and carrying out error compensation by a neural-network-trained INS error model to obtain optimal navigation information. According to the method provided by the invention, the INS positioning precision can be improved, and the positioning range of the indoor robot can be expanded on the basis that the WSN network scale is reduced.

The invention provides a bi-axis inertially-stabilized platform high-precision control method based on a self-adaptive backstepping sliding mode, and relates to design of a composite controller for backstepping sliding mode control based on an auxiliary integral sliding mode surface and adaptive neural network construction and optimization. The method is characterized by, to begin with, designing a backstepping sliding mode control method based on the auxiliary integral sliding mode surface according to a bi-axis inertially-stabilized platform dynamical model, and generating a control command according to state error information to realize suppression of indeterminacy and interference of dynamical model parameters; and then, constructing an adaptive neural network, constructing an adaptive neural network weight updating matrix based on the error information to update a weight matrix of the neural network online, estimating upper bound of interference error in real time, and realizing bi-axis inertially-stabilized platform high-precision control under complex environment. The method has the advantages of good real-time performance, fast dynamic parameter response and high multisource interference adaptability and the like, and can used for high-precision control and the like under the complex multisource interference environment.

The invention relates to a distributed flight control system fault diagnosis design method based on the multi-agent technology. The method is characterized by including the steps that firstly, a multi-agent system connected graph with a leader is established and represented through an undirected graph and a Laplacian matrix L is acquired; secondly, a state equation and an output equation of each node of a flight control system are established; thirdly, according to each node, a distributed error equation and a global error equation based on the undirected graph are established; fourthly, according to the established undirected graph, the collected state equations of the flight control system, the collected output equations of the flight control system, and the established global error equations, a distributed fault diagnosis observer gain matrix of the flight control system can be acquired. According to the method, a multi-target distributed fault diagnosis observer is designed, online diagnosis can be conducted when any node in the system breaks down or multiple nodes break down simultaneously, and online fault diagnosis and real-time fault estimation can be conducted on the flight control system.

The invention relates to a parallel micro-positioning platform based on the driving of piezoelectric ceramic drivers and a system, which comprises a micropositioner, a baseplate and a control system, wherein the micropositioner comprises a movable platform and four flexible hinges, a single piezoelectric ceramic driver and a pair of side-by-side piezoelectric ceramic drivers are respectively arranged at the two adjacent sides of the movable platform, one end of each piezoelectric ceramic driver and the movable platform are connected through a universal hinge, the other end of each piezoelectric ceramic driver is connected with a pre-tightening spring through the other universal hinge, and the pre-tightening spring is pre-tightened through a bolt; a capacitance sensor is respectively arranged on the upper surface of the movable platform along the axis direction of each piezoelectric ceramic driver; and one capacitance sensor is positioned on a sensor bracket fixed at the outer side of the micropositioner, the other capacitance sensor is positioned on the other sensor bracket fixed on the upper surface of the movable platform, and computer closed-loop control is adopted. Through theparallel micro-positioning platform and system, the displacement straightness accuracy and the positioning precision are effectively improved, coupling is eliminated, driver separation is avoided, and piezoelectric drivers are protected. By adopting an LQG (Linear Quadratic Gaussian) algorithm of an augmented noise model, narrowband foundation disturbance is effectively inhibited, and the positioning precision is improved.

The invention mainly provides a novel soft phase-locked loop, which is an improved soft phase-locked loop based on a wave trap and a low-pass filter. The novel soft phase-locked loop can accurately and quickly detect the amplitude and the phase of a positive-sequence component of power network voltage when a power network has an unsymmetrical fault, so the novel soft phase-locked loop not only can effectively inhibit phase 100 Hz disturbances, but also has a quick response speed, has high stability and high disturbance resisting capacity, is helpful for improving the dynamic control performance of a network-side converter when the power network has the unsymmetrical fault, and provides a guarantee for a low-voltage traversing technique of a wind power generation system.

The invention discloses an electric field electrorheological jet polishing device. The electrorheological jet polishing device comprises a work chamber, a rotary table, a machine tool, a jetting device, a filter, a stirring tank, a cooling device, a boosting device, flow guide pipes and an electrorheological polishing liquid. The machine tool does three-dimensional linear movement; the jetting device is fixed on a horizontal moving shaft driving device and faces the object carrying tabletop of the rotary table; the jetting device and the object carrying table of the rotary table are sealed by the work chamber; the filter is arranged between the work chamber and the stirring tank; the cooling device is arranged on the outer wall of the stirring tank; a boosting system is arranged between the jetting device and the stirring tank; and all parts of a system are connected through the flow guide pipes to form a liquid loop. When the electrorheological jet polishing device is in work, the electrorheological polishing liquid is boosted by the boosting device, input into the electrified jetting device, shaped by the jetting device and subjected to an electrorheological effect to form stable standard straight jets for polishing a workpiece; and waste liquid can be recycled for use after treatment. The electric field electrorheological jet polishing device combines an electrorheological polishing technology and a jet polishing technology and belongs to the domain of optical fine machining.

The invention discloses a multi-fighter cooperative fault diagnosis method under switching topology. The method includes the steps that firstly, a multi-fighter system connecting diagram of each topology structure is constructed and represented with a digraph to obtain the Laplacian matrix and self-cycle matrix of each topology structure; secondly, a state equation and an output equation of each node fighter are obtained, and a system model with an executor fault is set up for each node; then, a global augmented system state equation based on the digraphs is constructed for each node; finally, a multi-fighter cooperative fault diagnosis observer is set up, collected input data and output data of the all the nodes are sent into the observer to obtain the fault estimated value of each node, and therefore the executor faults of the flight control system are estimated. By means of the method, the fault of any node and the faults simultaneously appearing in multiple nodes can be diagnosed on line in the cooperative flight control system, and online fault diagnosis and real-time fault estimation of the flight control system can be carried out.

The invention discloses a boundary control method of restraining vibration of a flexible industrial conveyor belt, a flexible industrial conveyor belt system is analyzed and a mathematic model is established, the mathematic model is combined to design the adaptive feature of boundary disturbance of the flexible industrial conveyor belt system and a boundary controller, then stability of the flexible industrial conveyor belt system is subjected to simulated analysis, lastly, parameters of the flexible industrial conveyor belt system are adjusted according to a simulation result, and finally, the design requirements are achieved. According to the boundary control method, the vibration of the flexible industrial conveyor belt in a working process can be effectively restrained; meanwhile, the boundary control method has a certain adaptive feature on the boundary disturbance.

The invention belongs to the field of controller design, and is used for allowing tracking error of an observer to be converged to an original point within finite time. According to auto-disturbance-rejection controller design, a classic third-order finite extended-state observer serves as an example, and when the classic third-order finite extended-state observer meets the property of finite-time convergence, the parameters of the observer are solved. It should be understood that for other-order finite-time extended-state observers, parameters can be similarly solved with the method. According to the technical scheme, a method of an auto-disturbance-rejection controller based on the finite-time extended-state observer includes the specific steps that 1, the third-order finite-time extended-state observer is built; 2, for the extended-state observer, the parameter of the extended-state observer is solved; 3, disturbance feedback u is designed, wherein u=(u<0>-z<3>) / b, and z<3>=f (y,w,t), and the system is approximated as a second-order integrator series type. The auto-disturbance-rejection controller design is mainly applied to the controller design occasion.

The invention discloses a force / position switching control method of joints of a single leg of hydraulic four-leg robot. The control method comprises fuzzy control and multi-model control. The method comprises steps of carrying out segment control on the movement of the robot; carrying out joint force control when a foot end of the robot touches the ground so as to reduce impact and disturbance generated by touch between the foot end and the ground; and carrying out joint position control when the foot end of the robot gets away from the ground so as to achieve precise position of the robot. However, transition from no touch to touch or from touch to no touch between the robot and the ground is finished via force / position switching control, and harmful effects such as jumping of system parameters and shaking of the system when a robot system carries out the force / position switching will be produced. According to the invention, switching of control between force and position is finished via the fuzzy multi-model switching algorithm, so stable transition is ensured; and shaking generated during the force and position switching process can be effectively reduced and control precision is quite high.

The invention relates to a low-frequency air spring vibration isolator with a motion converter and belongs to the field of precision measurement and precision mechanical / optical processing. The low-frequency air spring vibration isolator is characterized in that the motion converter is assembled inside a traditional air spring passive vibration isolator, is a combined part and mainly comprises a flywheel mass, a screw rod and the like; the load vertical vibration is converted into the rotating motion of a flywheel in the load vibration process. The vibration isolator has the benefits that a certain mechanical damping is provided, so that a vibration signal can be can quickly attenuated; the inherent frequency of the vibration isolator can be further reduced on the basis of not changing the air spring stiffness, so that the stability of a vibration isolation system under the large load is greatly improved.

The present invention relates to a three-phase grid-connected inverter grid-connected current control method, and relates to a three-phase LCL-type grid-connected inverter grid-connected current control method based on an improved repetitive controller, especially to a grid-connected inverter current control method based on an improved repetitive controller. The voltage of a direct current side ofthe inverter is controlled by a voltage regulation outer ring, the output of a direct current IP regulator is taken as a given electric current as shown in the description of a d axis in a two-phaserotating coordinate system; currents as shown in the description are subjected to inverse park transformation to obtain current set values i[Alpha]ref and i[Beta]ref in an [Alpha][Beta] coordinate system; an inner film structure with repetitive control is improved to shorten the delay time to half of the original delay time; the improvement of the repetitive control is that a proportional controller and a repetition controller are connected in parallel to improve the dynamic response capacity of the repetitive control; and a voltage of a power grid is subjected to clark transformation and thenis output after processing through a feed-forward function as output of a feed-forward link of the voltage of the power grid. The grid-connected inverter current control method based on the improvedrepetitive controller shortens the inherent delay time of the repetitive control to achieve rapid and accurate tracking of the set reference current by employing the inverter grid-connected current soas to have better steady accuracy, response speed and robustness.

The invention discloses a reverse modulation free space optical communication system based on a pseudo phase conjugate reflector. The system is composed of an optical transceiver end and a reverse modulator end. The optical transceiver end comprises an optical transmitting module and a first optical detector module. The reverse modulator end comprises a second optical detector module and a reverse modulator module. The reverse modulator module is composed of a pseudo phase conjugate reflector, a spatial light modulator and a signal input unit, can modulate and reversely reflect an optical signal, and can eliminate the influence of atmospheric turbulence of any strength under certain conditions. The reverse modulator module can be integrated with a cat's eye lens or other optical systems with similar functions, and the field angle of the reverse modulator module after integration can be expanded. The reverse modulator end of the system has low power consumption and a compact structure, and is suitable for working in places with limited resources. Therefore, the reverse modulation free space optical communication system has many important applications, such as micro satellite / unmanned aerial vehicle communication, target identification, and environmental monitoring.

The invention discloses a working space solving method of a flexible supporting industrial robot. The working space solving method comprises the steps that a reachable working space is subjected to primary refining at first by utilizing industrial robot kinematics Jacobi local condition number indexes proposed based on a fixing substrate, so that a flexible working space is obtained, that is, theindustrial robot can be away from singularities in the working space when working in the working space, and accordingly kinematics precision is ensured; then, force system (force and moment) ellipsoidindexes are proposed for the flexible supporting industrial robot so that the flexible working space of the industrial robot can be further refined, and the flexible working space is named as a combined vibration suppression working space; and when the industrial robot works in the working space, a counterforce system borne by a flexible supporting platform from the industrial robot can be effectively controlled, so that the terminal precision of the industrial robot is not damaged and is the same with the kinematics terminal precision of a fixing substrate industrial robot.

The invention discloses a path maintaining and wheel sideslip control method for four-wheel steering / driving vehicles. The path maintaining and wheel sideslip control method for the four-wheel steering / driving vehicles comprises (a), establishing a vehicle body dynamic equation; (b), establishing a wheel dynamic equation; (c), establishing a relational expression between the frictional force and the moment of force; (d), establishing a computational formula for the longitudinal slip ratio and the lateral slip ratio; (e), establishing a path maintaining equation; (f), establishing a control model by controlling the wheel slip ratio; (g), designing a state feedback controller, namely, (g-1), selecting a performance index; (g-2), working out a matrix sum constant and (g-3), designing the controller; (h), obtaining wheel moment of force and a steering angle for a controlled vehicle. The path maintaining and wheel sideslip control method for the four-wheel steering / driving vehicles has the advantages of obtaining the appropriate wheel torque and the steering angle due to control of the longitudinal and lateral slip ratios of wheels and establishment of the control model, limiting the vehicle slip ratio to be within a certain range, restraining vehicle disturbance, enabling the vehicle to drive according to a preset path and guaranteeing lateral deviation not exceeding standard.

The invention provides an auto-disturbance-rejection control method used for nuclear reactor power. Aiming at a problem that a current nuclear reactor object has no controller which is simple and can achieve a good control effect, the invention provides a method of using an auto-disturbance-rejection controller in nuclear reactor power control. Through derivation, an original model is converted into a non-linear model suitable for designing the auto-disturbance-rejection controller. Model information is fully used and an ESO burden is mitigated. A characteristic that system output y is measurable is used so that the auto-disturbance-rejection controller based on RESO is designed. Finally, through simply adjusting a bandwidth, a controller parameter can be easily obtained. Compared to advanced control algorithms of prediction control, fuzzy control and the like, the control system has a simple structure and a good control effect.

The invention discloses a photoelectric system visual axis stabilization method based on carrier attitude angle information feedforward. The visual axis stabilization problem of a horizontal photoelectric system installed on a moving carrier when the carrier has three-axis attitude angle vibration is solved. For a traditional moving carrier photoelectric system, the method for stabilizing a visualaxis is to install angular rate gyros on frame pitch and azimuth axes, and the orientation of the pitch and azimuth axes of the frame is stabilized through a closed loop. The vibration of the movingcarrier is usually distributed in the three axial directions of pitch, azimuth and rolling, and the azimuth axis gyro can not measure the angular vibration of a rolling axis due to an orthogonal relationship so that a disturbance to a visual axis can not be restrained. In the invention, the azimuth and pitch axis angular velocity closed loops of an original photoelectric system are canceled, encoders are installed on the azimuth and pitch axes to form an angular position closed loop; and an inertial attitude measuring unit is installed on the substrate of the photoelectric system, and the disturbance quantity of the carrier to the visual axis is fed forward to the angular position closed-loop controllers of the azimuth and pitch axes so as to achieve the stability of the visual axis.