Model-free control method for driving motor of deep-sea robot

Abstract

The invention discloses a model-free control method for a driving motor of a deep-sea robot. The model-free control method comprises the following steps: acquiring the position, rotating speed and bus voltage of the motor and phase currents of different stator modules; acquiring a rotating speed sampling value to predict the rotating speed and rotating speed ring disturbance at the next moment; acquiring a rotating speed instruction, calculating an electromagnetic torque instruction by combining the rotating speed and a rotating speed loop disturbance prediction value, converting the electromagnetic torque instruction into a current instruction, and averagely distributing the current instruction to each stator module; injecting current into the d axis, and acquiring a current sampling value to predict current and current loop disturbance at the next moment; obtaining a current instruction and calculating a voltage instruction by combining the predicted values of the current and the current loop disturbance; obtaining a gradient function for an adaptive adjustment system and adaptively adjusting a coefficient based on a finite time gradient descent method; and S3, based on the voltage instruction in the step S3, generating a PWM pulse signal through coordinate transformation and space vector modulation, and driving each stator module inverter to control the operation of the motor.

Description

Technical Field

[0001] The present invention relates to the field of deep-sea motor control technology, and more specifically, to a model-free control method for a deep-sea robot driving motor. Background Art

[0002] Deep-sea robot drive motors are widely used in underwater equipment such as deep-sea pipeline inspection robots, submarine cable inspection robots, deep-sea cable laying robots and deep-sea cleaning robots. In order to ensure normal operation in the deep sea, deep-sea robot drive motors need to be filled with oil to ensure internal and external pressure balance.

[0003] like Figure 1 As shown, it is a deep-sea robot driving motor, including a housing 100, in which a rotating shaft 200 is axially arranged, and a plurality of modular stators 300 are axially spaced and arranged on the inner wall of the housing, and are arranged one by one with the rotor modules 400. Each modular stator 300 can cooperate with the corresponding rotor module 400 to make the rotating shaft 200 output torque, so as to reduce the radial size of the modular stator 300 on the basis of outputting substantially the same torque, and at the same time, the modular stators 300 are independent of each other, and when one of the modular stators 300 is damaged, the remaining modular stators 300 can still operate normally; as shown in FIG. Figure 2 As shown, the modular stator 300 includes a plurality of separately arranged stator modules 330. By splitting the modular stator 300 into a plurality of mutually independent stator modules 330, when one of the stator modules 330 is damaged, the remaining stator modules 330 can still operate normally.

[0004] For a single modular stator, four stator modules are independently wound into a three-phase motor. Therefore, when in use, three types of equivalent three-phase motors, equivalent three three-phase motors, and equivalent twelve three-phase motors are involved. According to the traditional driving method, a single three-phase controller, three three-phase controllers, or twelve three-phase controllers can be used for driving. However, as the temperature changes during operation, the resistance, inductance, and magnetic flux parameters of the deep-sea robot drive motor will change. At the same time, the change in the viscosity coefficient of the oil inside the motor will cause the rotor friction coefficient to change. The changes in the above parameters will affect the performance of the traditional controller, including increased motor current pulsation and increased losses; changes in the overshoot and adjustment time of the speed control, increased transient current, increased transient losses, and reduced speed regulation performance.

[0005] Therefore, it is necessary to design a drive control method for a deep-sea robot drive motor with multiple modular stators. Summary of the invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a model-free control method for a deep-sea robot drive motor.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A model-free control method for a deep-sea robot driving motor comprises the following steps:

[0009] Step S1, obtaining the position, speed, bus voltage and phase current of different stator modules of the motor;

[0010] Step S2, obtaining the speed sampling value to predict the speed and speed ring disturbance at the next moment; obtaining the speed command and calculating the electromagnetic torque command in combination with the predicted values ​​of the speed and speed ring disturbance, and converting it into a current command and evenly distributing it to each stator module;

[0011] Step S3, inject current into the d-axis, obtain the current sampling value to predict the current and current loop disturbance at the next moment; obtain the current command and calculate the voltage command in combination with the predicted value of the current and current loop disturbance; obtain the gradient function for the adaptive regulation system and adaptively adjust the coefficient based on the finite time gradient descent method;

[0012] Step S4, based on the voltage instruction of step S3, a PWM pulse signal is generated through coordinate transformation and space vector modulation to drive the inverters of each stator module to control the operation of the motor.

[0013] Furthermore, step S2 includes the following steps:

[0014] Step S201, designing the speed and disturbance prediction equation in the discrete domain based on the speed sampling value:

[0015] (1)

[0016] In formula (1), Represents the predicted value of mechanical angular velocity, Represents the mechanical angular velocity sampling value, represents the speed control cycle, represents electromagnetic torque, g represents speed loop disturbance, and represents the speed loop prediction coefficient, represents the speed loop controller coefficient;

[0017] Step S202, obtaining an electromagnetic torque instruction according to a given value and the predicted value of step S201:

[0018] (2)

[0019] In formula (2), represents the electromagnetic torque, represents the speed loop controller coefficient, represents the speed control cycle, Represents the mechanical angular velocity given value, represents the mechanical angular velocity sampling value, and g represents the speed loop disturbance.

[0020] Furthermore, step S3 includes the following steps:

[0021] Step S301, injecting current into the d-axis, and the injected current is in the following form:

[0022] (3)

[0023] In formula (3), represents the injected current, represent Divide by The remainder of represents the period of the injected current, represents the current control period, represents the amplitude of the injected current, Represents time;

[0024] Step S302, designing the current and disturbance prediction equation in the discrete domain based on the current sampling value:

[0025] (4)

[0026] In formula (4), represent Shaft current, represent Shaft voltage, represents the predicted current value, and represents the current loop prediction coefficient, represent Axis disturbance, represents the electrical angular velocity, represents the current loop controller coefficient, represents the current control period, represents the error;

[0027] Step S303, based on the given value and the predicted value of step S302, obtain Voltage command under the axis:

[0028] (5)

[0029] In formula (5), represent Shaft voltage, represents the current loop controller coefficient, represents the current control period, represent Axis current command, represents the electrical angular velocity, represents the current control period, represents the predicted current value, represent Axis disturbance, j is an imaginary unit;

[0030] Step S304, calculating the adaptive adjustment coefficient The gradient function is:

[0031] (6)

[0032] In formula (6), represents the gradient, represents the optimal value of the coefficient, represents the amplitude of the injected current, represents the current loop controller coefficient, represents the period of the injected current, represent The sampling value of shaft current;

[0033] Step S305, based on the finite time gradient descent method, the coefficients in the current model-free controller are adaptively adjusted to adapt to real-time parameter changes:

[0034] (7)

[0035] In formula (7), represents the current loop controller coefficient, represents the current control period, and is the adaptive adjustment coefficient.

[0036] The beneficial effects of the present invention are:

[0037] 1. In the present invention, the speed loop is controlled without model, so that it is independent of the rotational inertia parameters of the deep-sea robot driving motor, and has stronger adaptability to the change of friction coefficient caused by the change of viscosity coefficient, which can ensure that the speed control has no overshoot and rapid adjustment capability.

[0038] 2. In the present invention, the current loop is improved on the basis of traditional zero-beat control in combination with the model-free theory, and a coefficient adaptive method is designed based on the zero-beat characteristic. As the resistance, inductance and magnetic flux parameters of the deep-sea robot drive motor change, the current can also achieve zero-beat control, with fast dynamic performance, small current pulsation, the current is always in the optimal control state, and the loss is small. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1A schematic diagram of the structure of a deep-sea robot driving motor in this embodiment;

[0040] Figure 2 is a structural schematic diagram of a modular stator in this embodiment;

[0041] Figure 3 A block diagram of the model-free control method for the deep-sea robot driving motor in this embodiment;

[0042] Figure 4 This is a flow chart of the model-free control method for the deep-sea robot driving motor in this embodiment.

[0043] Reference numerals: housing 100 , rotating shaft 200 , modular stator 300 , stator module 330 , rotor module 400 . DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Embodiment: A model-free control method for a deep-sea robot driving motor, such as Figure 3-Figure 4 As shown, the following steps are included:

[0046] Step S1, obtaining the position, speed, bus voltage and phase current of different stator modules of the motor;

[0047] The position of the motor is obtained through the encoder and the orthogonal decoding function of the chip. The chip can be TMS320F28346. The speed is obtained through the differential operation of the position, and the bus voltage and phase current can be obtained through the sensor and the analog-to-digital conversion module.

[0048] Step S2, speed model-free control: obtain the speed sampling value to predict the speed and speed ring disturbance at the next moment; obtain the speed command and calculate the electromagnetic torque command in combination with the predicted values ​​of the speed and speed ring disturbance, and convert it into a current command and evenly distribute it to each stator module;

[0049] Specifically, step S2 includes the following steps:

[0050] Step S201, without relying on the rotor motion model, directly designs the following speed and disturbance prediction equation in the discrete domain based on the speed sampling value according to the first-order differential equation:

[0051] (1)

[0052] In formula (1), Represents the predicted value of mechanical angular velocity, Represents the mechanical angular velocity sampling value, represents the speed control cycle, represents the electromagnetic torque and can be expressed as The shaft current is equivalent, the equivalent coefficient is the torque coefficient, g represents the speed loop disturbance, and represents the speed loop prediction coefficient, represents the speed loop controller coefficient, is a natural number;

[0053] Step S201 can be implemented by TMS320F28346 chip software programming. According to formula (1), the actual operation sequence is: calculate the mechanical angular velocity sampling value at time k And the mechanical angular velocity prediction value The error is calculated, and the speed loop disturbance g in the next calculation cycle is predicted at time k+1, and the mechanical angular velocity prediction value for the next calculation cycle is calculated. ;

[0054] Step S202, obtaining an electromagnetic torque instruction according to a given value and the above-mentioned predicted value:

[0055] (2)

[0056] In formula (2), represents the electromagnetic torque, represents the speed loop controller coefficient, represents the speed control cycle, Represents the mechanical angular velocity given value, represents the mechanical angular velocity sampling value, and g represents the speed loop disturbance.

[0057] The above speed prediction value is used to reduce the influence of control delay and improve the speed dynamic performance. The disturbance prediction value is used to eliminate the disturbance caused by the moment of inertia parameter error and underwater random load.

[0058] In step S2, the speed loop is controlled using model-free control to calculate the electromagnetic torque set value required by the current inner loop, so that it is independent of the rotational inertia parameters of the deep-sea robot drive motor, and has stronger adaptability to changes in the friction coefficient caused by changes in the viscosity coefficient, ensuring that the speed control has no overshoot and rapid adjustment capabilities.

[0059] Step S3, current model-free control: inject current into the d-axis, obtain current sampling values ​​to predict the current and current loop disturbance at the next moment; obtain current instructions and calculate voltage instructions in combination with the predicted values ​​of current and current loop disturbance; obtain the gradient function for the adaptive regulation system and adaptively adjust the coefficient based on the finite time gradient descent method;

[0060] Specifically, step S3 includes the following steps:

[0061] Step S301, injecting current into the d-axis, and the injected current is in the following form:

[0062] (3)

[0063] In formula (3), represents the injected current, represent Divide by The remainder of represents the period of the injected current, represents the current control period, represents the amplitude of the injected current, Represents time;

[0064] The process of injecting current in the d-axis is achieved by superimposing the signal expressed by equation (3) on the given value of the d-axis current and by the unbiased tracking performance of the current closed loop. d (ref)=0 control mode, therefore, the implementation method of injecting current in the d-axis can be equivalent to setting the d-axis current given as i d (ref)=i ds (t), where i d (ref) is the given value of d-axis current.

[0065] Step S302, without relying on the electromagnetic model, directly based on the current sampling value according to the first-order differential equation, the following current and disturbance prediction equations in the discrete domain are designed:

[0066] (4)

[0067] In formula (4), represent Shaft current, represent Shaft voltage, represents the predicted current value, and represents the current loop prediction coefficient, represent Axis disturbance, represents the electrical angular velocity, represents the current loop controller coefficient, represents the current control period, represents the error;

[0068] Step S302 can be implemented by TMS320F28346 chip software programming. According to formula (4), the actual operation sequence is: calculate the dq axis current sampling value at time k and dq axis current prediction value The error is calculated at time k+1, which is the prediction of the current loop disturbance in the next calculation cycle. , calculate the predicted dq axis current for the next calculation cycle .

[0069] Step S303, based on the given value and the predicted value of step S302, obtain Voltage command under the axis:

[0070] (5)

[0071] In formula (5), represent Shaft voltage, represents the current loop controller coefficient, represents the current control period, represent Axis current command, represents the electrical angular velocity, represents the current control period, represents the predicted current value, represent Axis disturbance, j is an imaginary unit;

[0072] The q-axis current command is generated in step S2 , and the d-axis current command is generated in step S301 .

[0073] The above-mentioned current prediction value is used to reduce the influence of control delay and realize zero-beat control, and the disturbance prediction value is used to eliminate the disturbance caused by parameter error and improve the control accuracy.

[0074] Step S304, calculating the adaptive adjustment coefficient The gradient function is:

[0075] (6)

[0076] In formula (6), represents the gradient, represents the optimal value of the coefficient, represents the amplitude of the injected current, represents the current loop controller coefficient, represents the period of the injected current, represent The sampling value of shaft current;

[0077] Specifically, the gradient is obtained according to the response after the d-axis current is injected in step S301. For the d-axis current injected at time k, it is expected that the sampling current equal to the given value at time k can be obtained at time k+2, that is, after two control cycles. If they are not equal, the difference between the two and the adaptive adjustment coefficient The gradient described in formula (6) is That is, the adjustment coefficient generated according to the above description , the gradient In the TMS320F28346 chip implementation, it can be directly calculated by formula (6), and this step is executed after the voltage instruction is generated in step S303.

[0078] Step S305, based on the finite time gradient descent method, the coefficients in the current model-free controller are adaptively adjusted to adapt to real-time parameter changes:

[0079] (7)

[0080] In formula (7), represents the current loop controller coefficient, represents the current control period, and is the adaptive adjustment coefficient, is the adaptive adjustment step size, used to adjust the coefficient The convergence speed.

[0081] Step S305 is implemented by TMS320F28346 chip software programming. According to formula (7), the actual operation sequence is: Calculate the adaptive adjustment step length according to the gradient calculated in step S304 , and then adaptively adjust the coefficient Finally, the calculation result is It will be used in the next current control cycle, namely step S302 and step S303.

[0082] In step S3, the current loop is improved on the basis of traditional zero-beat control in combination with the model-free theory, and a coefficient adaptive method is designed based on the zero-beat characteristic. As the resistance, inductance and magnetic flux parameters of the deep-sea robot drive motor change, the current can also achieve zero-beat control, with fast dynamic performance, small current pulsation, and the current is always in the optimal control state with low loss.

[0083] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A model-free control method for a deep-sea robot driving motor, characterized in that: The steps include: Step S1, obtaining the position, speed, bus voltage and phase current of different stator modules of the motor; Step S2, obtaining the speed sampling value to predict the speed and speed ring disturbance at the next moment; obtaining the speed command and calculating the electromagnetic torque command in combination with the predicted values ​​of the speed and speed ring disturbance, and converting it into a current command and evenly distributing it to each stator module; Step S3, injecting current into the d-axis, obtaining current sampling values ​​to predict the current and current loop disturbance at the next moment; Obtain current command and calculate voltage command in combination with predicted values ​​of current and current loop disturbance; obtain gradient function for adaptive regulation system and adaptively adjust coefficient based on finite time gradient descent method; Step S4, based on the voltage instruction of step S3, a PWM pulse signal is generated through coordinate transformation and space vector modulation to drive the inverters of each stator module to control the operation of the motor.

2. According to claim 1, a model-free control method for a deep-sea robot driving motor is characterized in that: The step S2 comprises the following steps: Step S201, designing the speed and disturbance prediction equation in the discrete domain based on the speed sampling value: (1) In formula (1), Represents the predicted value of mechanical angular velocity, Represents the mechanical angular velocity sampling value, represents the speed control cycle, represents electromagnetic torque, g represents speed loop disturbance, and represents the speed loop prediction coefficient, represents the speed loop controller coefficient; Step S202, obtaining an electromagnetic torque instruction according to a given value and the predicted value of step S201: (2) In formula (2), represents the electromagnetic torque, represents the speed loop controller coefficient, represents the speed control cycle, Represents the mechanical angular velocity given value, represents the mechanical angular velocity sampling value, and g represents the speed loop disturbance.

3. A model-free control method for a deep-sea robot driving motor according to claim 1, characterized in that: The step S3 comprises the following steps: Step S301, injecting current into the d-axis, and the injected current is in the following form: (3) In formula (3), represents the injected current, represent Divide by The remainder of represents the period of the injected current, represents the current control period, represents the amplitude of the injected current, Represents time; Step S302, designing the current and disturbance prediction equation in the discrete domain based on the current sampling value: (4) In formula (4), represent Shaft current, represent Shaft voltage, represents the predicted current value, and represents the current loop prediction coefficient, represent Axis disturbance, represents the electrical angular velocity, represents the current loop controller coefficient, represents the current control period, represents the error; Step S303, based on the given value and the predicted value of step S302, obtain Voltage command under the axis: (5) In formula (5), represent Shaft voltage, represents the current loop controller coefficient, represents the current control period, represent Axis current command, represents the electrical angular velocity, represents the current control period, represents the predicted current value, represent axis disturbance, j is an imaginary unit; Step S304, calculating the adaptive adjustment coefficient The gradient function is: (6) In formula (6), represents the gradient, represents the optimal value of the coefficient, represents the amplitude of the injected current, represents the current loop controller coefficient, represents the period of the injected current, represent The sampling value of shaft current; Step S305, based on the finite time gradient descent method, the coefficients in the current model-free controller are adaptively adjusted to adapt to real-time parameter changes: (7) In formula (7), represents the current loop controller coefficient, represents the current control period, and is the adaptive adjustment coefficient.

Images