CN110569767B - Motion parameter identification method, system, equipment and computer readable storage medium - Google Patents

Motion parameter identification method, system, equipment and computer readable storage medium Download PDF

Info

Publication number
CN110569767B
CN110569767B CN201910804638.3A CN201910804638A CN110569767B CN 110569767 B CN110569767 B CN 110569767B CN 201910804638 A CN201910804638 A CN 201910804638A CN 110569767 B CN110569767 B CN 110569767B
Authority
CN
China
Prior art keywords
period
motion
current
motion parameter
current period
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910804638.3A
Other languages
Chinese (zh)
Other versions
CN110569767A (en
Inventor
黄金飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Inovance Technology Co Ltd
Original Assignee
Suzhou Inovance Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Suzhou Inovance Technology Co Ltd filed Critical Suzhou Inovance Technology Co Ltd
Priority to CN201910804638.3A priority Critical patent/CN110569767B/en
Publication of CN110569767A publication Critical patent/CN110569767A/en
Application granted granted Critical
Publication of CN110569767B publication Critical patent/CN110569767B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/72—Electric energy management in electromobility

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

The invention provides a motion parameter identification method, a system, equipment and a computer readable storage medium, wherein the motion parameter identification method is applied to motor control and comprises the following steps: when the motor runs, acquiring the torque and the rotating speed of the motor according to a preset period; performing enhancement processing on the torque and the rotating speed, and acquiring sample data after the enhancement processing; and acquiring the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule. According to the embodiment of the invention, through a dynamic sampling and self-adaptive gradient descent method, the motion parameters can be obtained in real time in the operation of the motor, and the problem that the machine needs to be stopped every time in an offline identification manner or the load cannot be adjusted when changing is avoided.

Description

Motion parameter identification method, system, equipment and computer readable storage medium
Technical Field
Embodiments of the present invention relate to the field of motor control, and more particularly, to a method, system, apparatus, and computer-readable storage medium for identifying motion parameters.
Background
The servo system is widely applied to industries such as numerical control machine tools, robots, automatic production lines and the like, and needs to have high speed, high precision and high response performance. Because the servo motor often needs to work under different working conditions, and the working state of the servo motor can also change in the running process, the system parameters of the servo motor can change in a large range in the running process. The motion parameters such as load inertia, friction and the like have the greatest influence on the stability and the responsiveness of the servo controller, and the motion parameters need to be set according to actual values.
Currently, the motion parameters of a servo controller are generally obtained in two ways: offline identification and online identification.
Before the servo system works, the off-line identification needs to give an operation instruction from the outside or the inside of the servo controller, so that the servo motor acts according to a preset mode, and then the load parameter is estimated. Therefore, the offline identification scheme needs to be executed one by one device before installation every time, and is long in operation time for a plurality of multi-axis occasions, inconvenient to execute in short-stroke occasions and has certain requirements on experience of users. And, when the load or the posture change is replaced, the dynamic parameters need to be re-detected and set, so that the optimal control performance can be ensured.
On-line identification is currently carried out on schemes such as model reference self-adaption, least square method, kalman filtering, neural network and the like, wherein the model reference self-adaption scheme needs to model an actual system, and the influence of random nonlinear factors is ignored, so that estimated values are easy to be not converged; the least square method scheme needs to store data, is large in calculation amount, and is low in identification consistency for multiple times; the schemes such as Kalman filtering, neural network and the like have complex operation and low practicability.
Disclosure of Invention
Aiming at the problems that the motion parameter offline identification operation of the servo controller is complicated, the load or gesture change needs to be monitored and set again, the model reference adaptive online identification needs to model an actual system, estimated values are easy to cause unconvergence, the least square method online identification calculation amount is large, the multiple identification consistency is low, and the Kalman filtering and neural network online identification operation is complex, the embodiment of the invention provides a novel motion parameter identification method, a novel motion parameter identification system, novel motion parameter identification equipment and a computer readable storage medium.
The technical scheme for solving the technical problems in the embodiment of the invention is to provide a motion parameter identification method which is applied to motor control and comprises the following steps:
when the motor runs, acquiring the torque and the rotating speed of the motor according to a preset period;
performing enhancement processing on the torque and the rotating speed, and acquiring sample data after the enhancement processing;
and acquiring the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule.
Preferably, the method comprises: judging whether the iterative operation reaches a stable state, and updating the motion parameters of the motor controller by using the motion parameters of the current period when the iterative operation reaches the stable state.
Preferably, the sample data includes an output quantity, a first input quantity, and a second input quantity, and the enhancing process is performed on the torque and the rotation speed, including:
integrating the torque, and taking the difference between the torque integral value of the current period and the torque integral value of the previous period as output quantity;
obtaining acceleration according to the rotating speed, carrying out integration processing on the acceleration, and taking the difference between an acceleration integral value of a current period and an acceleration integral value of a previous period as a first input quantity;
and integrating the rotation speed, and taking the difference between the rotation speed integral value of the current period and the speed integral value of the previous period as a second input quantity.
Preferably, the motion parameters include load inertia and viscosity coefficient, and the iteratively acquiring the motion parameters of the current period includes:
iteratively calculating the motion parameters of the current period by the following calculation steps:
θ k ＝θ k-1 -α k ×L′(θ k-1 )
where k is the current period, θ k For the motion parameter of the current period, θ k ＝[J k ,B k ] T And J k For the load inertia of the current period, B k The viscosity coefficient of the current period; θ k-1 Alpha is the motion parameter of the previous period k For the iteration step of the current cycle, L' (θ k-1 ) A first derivative of a loss function which is characterized by a motion parameter of a previous cycle, and the loss function isX is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1.
Preferably, the motion parameter includes load inertia, and the iteratively acquiring the motion parameter of the current period includes:
iteratively calculating the motion parameters of the current period by the following calculation steps:
θ k ＝θ k-1 -α k ×L′(θ k-1 )
where k is the current period, θ k For the motion parameter of the current period, θ k ＝J k And J k Load inertia for the current period; θ k-1 For the motion parameter of the previous cycle，α k For the iteration step of the current cycle, L' (θ k-1 ) A first derivative of a loss function which is characterized by a motion parameter of a previous cycle, and the loss function isX is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1.
Preferably, the iteration step of the current cycle: alpha k ＝(X T ×X) -1 ；
Alternatively, the iteration step of the current period: alpha k ＝[τ×α k-1 -1 +x k ×x k T ] -1 Where τ is the adaptation factor, α k-1 For the iteration step of the previous cycle, x k Is a matrix of the first input quantity and the second input quantity of the current period.
The embodiment of the invention also provides a motion parameter identification system which is applied to motor control, wherein the system comprises a sampling unit, a preprocessing unit and an identification unit, wherein:
the sampling unit is used for acquiring the torque and the rotating speed of the motor according to a preset period when the motor runs;
the preprocessing unit is used for carrying out enhancement processing on the torque and the rotating speed and acquiring sample data after the enhancement processing;
the identification unit is used for obtaining the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule.
Preferably, the system further comprises a setting unit for updating the motion parameters of the motor controller using the motion parameters of the current cycle when the iterative operation of the identification unit reaches a steady state.
Preferably, the sample data includes an output quantity, a first input quantity, and a second input quantity, the motion parameter includes a load inertia and a viscosity coefficient, and the preprocessing unit includes a first subunit, a second subunit, and a third subunit, wherein:
the first subunit is configured to perform integration processing on the torque, and take a difference between a torque integral value in a current period and a torque integral value in a previous period as an output quantity;
the second subunit is configured to obtain an acceleration according to the rotational speed, perform integration processing on the acceleration, and use a difference between an acceleration integral value in a current period and an acceleration integral value in a previous period as a first input quantity;
the third subunit performs integration processing on the rotation speed, and takes a difference between a rotation speed integral value of a current period and a speed integral value of a previous period as a second input quantity.
The identification unit iteratively calculates the motion parameters of the current cycle by:
θ k ＝θ k-1 -α k ×L′(θ k-1 )
where k is the current period, θ k For the motion parameter of the current period, θ k ＝[J k ,B k ] T And J k For the load inertia of the current period, B k The viscosity coefficient of the current period; θ k-1 Alpha is the motion parameter of the previous period k For the iteration step of the current cycle, L' (θ k-1 ) A first derivative of a loss function which is characterized by a motion parameter of a previous cycle, and the loss function isX is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1.
Preferably, the motion parameter includes load inertia, and the identification unit iteratively calculates the motion parameter of the current period by:
θ k ＝θ k-1 -α k ×L′(θ k-1 )
wherein k isCurrent period, θ k For the motion parameter of the current period, θ k ＝J k And J k Load inertia for the current period; θ k-1 Alpha is the motion parameter of the previous period k For the iteration step of the current cycle, L' (θ k-1 ) A first derivative of a loss function which is characterized by a motion parameter of a previous cycle, and the loss function isX is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1. The embodiment of the invention also provides a motion parameter identification device, which comprises a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and the processor realizes the steps of the motion parameter identification method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the steps of the motion parameter identification method according to any one of the above steps when being executed by a processor.
According to the method, the system, the equipment and the computer readable storage medium for identifying the motion parameters, the motion parameters can be obtained in real time in the running process of the motor through the dynamic sampling and the self-adaptive gradient descent method, and the condition that the machine is required to be stopped every time in offline identification or the load cannot be adjusted when the load changes is avoided. Compared with the common online identification method, the method has the advantages that the accuracy is little influenced by nonlinear factors, the implementation is relatively simple, the requirements on hardware resources and computing capacity are low, and the feasibility is high. The motion parameters identified by the embodiment of the invention can be directly used for a motor controller, so that the robot can stably run under various load conditions, and the robot motion control method is particularly suitable for the fast motion occasion of the robot.
Drawings
Fig. 1 is a schematic flow chart of a motion parameter identification method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a motion parameter identification system according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a motion parameter identification apparatus according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
According to the motion parameter identification method, the load inertia and the viscosity coefficient are converted according to the dynamic equation, and then the motion parameter is identified in real time in a self-adaptive gradient descent mode. The motion parameter identification method can carry out iterative computation on each sampling, so that the variable load can be detected in real time, and meanwhile, the precision is not influenced by nonlinear factors, vibration and other factors. Because of small calculation amount, the motion parameter identification method is easy to realize on a low-calculation-force motor controller.
As shown in fig. 1, a flow chart of a motion parameter identification method provided by the embodiment of the invention is shown, and the method can be applied to motor (for example, a servo motor) control, and can realize real-time identification of motion parameters in a self-adaptive gradient descending mode, so that the motor can stably operate under various load conditions, and is particularly suitable for a robot rapid motion occasion. The motion parameter identification method of the present embodiment may be integrated into a motor controller, and includes:
step S11: when the motor operates, the torque and the rotating speed of the motor are obtained according to a preset period.
After the motor starts to run, the motor can be controlled to run through a preset initial value of the motion parameter (for example, the inertia of the motor body is taken as load inertia, and the viscosity coefficient is zero). In the running process of the motor, an encoder or a speed measuring device can be utilized to collect motor position or speed signals in real time, and the real-time rotating speed of the motor is obtained through conversion; the motor torque can then be obtained by means of a sensor (current sensor) or a sensorless estimation method.
Specifically, when the motor torque is obtained, the current sensor can be used for sampling the working current of the motor, and then the working current is converted into the motor output torque through physical conversion, a lookup table, data fitting and other modes. For example, for a surface-mounted permanent magnet synchronous motor, the motor torque may be obtained using a control strategy with a d-axis current command of zero, i.e., the motor torque is obtained according to the following torque equation (1):
T e0 ＝K t ×I q (1)
wherein: i q To coordinate-transform the sampled current, the q-axis current, K t Is a moment coefficient.
Step S12: and performing enhancement processing on the torque and the rotating speed, and acquiring sample data after the enhancement processing.
According to the following servo motor dynamics equation (2):
where J is the inertia of the moving parts of the servo system,for acceleration of moving parts, B is the viscosity coefficient, T d0 Is a disturbance term (mainly comprising friction, load torque, noise, etc.).
That is, the raw data acquired in step S11 is affected by the disturbance term, so that appropriate processing is required to eliminate the influence of data distortion due to nonlinear factors such as friction, clearance, noise, vibration, etc. in actual operation, and to enhance the effective signal component. Specifically, in the step S12, the method may combine digital denoising (such as bandpass filtering, lowpass filtering, smoothing filtering, wavelet transforming, etc.), signal observation, integral operation, etc., to process the raw data, so as to convert the calculation formula (2) into the formula (3):
for a system which acts in real time under different working conditions, effective characteristic signals can be further selected to participate in operation, the robustness and stability of recognition are improved, and the following formula (4) can be obtained by methods of differential operation, threshold denoising and the like:
since the disturbance term in the formula (3) is generally constant or has small variation, the disturbance term is far smaller than the first two terms after being selected, and can be eliminated, so that the formula (4) can be converted into the following formula (5):
at this time, sampling is performed for a plurality of times, and the motion system can establish a model as follows in consideration of random errors existing in actual sampling:
F＝X×θ+E (6)
wherein:
F＝[y 1 y 2 … y n ] T a matrix (n is an integer greater than or equal to 1) representing a motion system comprising n outputs;
X＝[x 1 x 2 … x n ] T a matrix representing the motion system comprising n input quantities;
θ＝[J B] T the characteristic parameter matrix of the motion system is represented, wherein J is load inertia, and B is viscosity coefficient;
e, representing random error between predicted output value X X theta and sampling true output value F of motion system
Wherein, the output and input of the kth sample are respectively:
Δ∫ k T e as the output quantity, it may specifically be a difference value between the torque integral value of the current cycle and the torque integral value of the previous cycle;as the first input amount, it may specifically be a difference value between the speed integrated value of the current period and the speed integrated value of the previous period; delta ≡ k ω is a second input amount, which may specifically be a difference between the acceleration integrated value of the current cycle and the acceleration integrated value of the previous cycle.
Step S13: and obtaining the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule.
In this step, an optimization objective of gradient descent may be established first, specifically, L1 and L2 optimization may be adopted, and in order to reduce the calculation amount, in one embodiment of the present invention, L2 mean square error minimization may be adopted as a loss function, that is:
the smaller L (theta) is, the closer the calculated characteristic parameter theta value is to the actual characteristic parameter theta value of the motion system. The gradient descent optimization rule can be established based on the loss function, and generally, random gradient descent, batch gradient descent, small batch gradient descent, gradient descent with momentum and the like can be adopted. For low-dimensional operations, in order to improve real-time performance, in one embodiment of the present invention, a random gradient descent method may be used. According to the random gradient descent principle, the following iterative solution formula (9) can be established to solve the load inertia J and the viscosity coefficient B:
θ k ＝θ k-1 -α k ×L′(θ k-1 ) (9)
where k represents the kth sampling period (i.e., k is an integer greater than or equal to 1), i.e., the current period, θ k The characteristic parameter value calculated for the kth time is the motion parameter of the current period; θ k ＝[J k ,B k ] T And J k For the load inertia of the current period, B k The viscosity coefficient of the current period; θ k-1 Alpha is the motion parameter of the previous period k The step length of gradient descent, namely the iteration step length of the current period; l' (θ) k-1 ) The first derivative of the loss function, which is characterized by the motion parameter of the previous cycle.
In this step, θ can be set 0 =[J 0 B 0 ] T Is an initial iteration condition, in which J 0 For the inertia of the motor body, B 0 May be set to zero. Step alpha of the gradient decrease k Can be set to a constant.
In particular, when the calculation resources of the apparatus (e.g., motor controller) implementing the above-described motion parameter identification method are limited, the viscosity coefficient may be directly discarded, and the characteristic parameter value θ calculated the kth time in the calculation formula (9) may be caused to be calculated k ＝J k Thereby further conserving computing resources.
Furthermore, the step size α of the gradient decrease k Or a variable step value based on a rule such as the number of time steps. Preferably, the step size α of the gradient descent can be automatically calculated as follows k ：
Based on L' (θ) =0, L (θ) is minimized, and the solution is performed to obtain:
θ＝(X T ×X) -1 ×X T ×F (10)
and (3) developing the formula (10) according to the current sampling point to obtain:
taking equations (8) and (11) into equation (9), the step length α of the gradient decrease is calculated k ：
α k ＝(X T ×X) -1 (12)
Preferably, the step size alpha of the gradient descent can also be improved by setting an adaptation factor tau (which can be empirically set) k Data saturation is prevented, specifically as follows:
α k ＝[τ×α k-1 -1 +x k ×x k T ] -1 (13)
according to the motion parameter identification method, the motion parameters can be obtained in real time in the operation of the motor through the dynamic sampling and self-adaptive gradient descent method, so that the problem that the machine needs to be stopped every time in an offline identification mode or the load cannot be adjusted when the load changes is avoided. Compared with the common online identification method, the method has the advantages that the accuracy is little influenced by nonlinear factors, the implementation is relatively simple, the requirements on hardware resources and computing capacity are low, and the feasibility is high.
In another embodiment of the present invention, in addition to the above steps S11 to S13, the following steps may be included: judging whether the iterative operation reaches a stable state, and when the iterative operation reaches the stable state, updating the motion parameters of the motor controller by using the motion parameters of the current period, so that the motion parameters of the current period can be used for controlling the operation of the motor of the next period.
Specifically, whether the iteration reaches stability can be judged according to the combination of parameter variation, iteration time and iteration times. For example, the parameter variation amount ε may be calculated by the following equation (14), and then compared with a set threshold to determine whether the iterative operation has stabilized:
ε＝|(θ k -θ k-1 )/θ k | (14)
and when the parameter variation epsilon is less than 0.01, confirming that the calculation is stable, and outputting the identified load inertia J and viscosity coefficient B as motion parameters of the next period to perform motor operation control.
Fig. 2 is a schematic diagram of a motion parameter identification system according to an embodiment of the present invention, where the system may be applied to a motor controller and assist in motor control. The motion parameter identification system of the present embodiment includes a sampling unit 21, a preprocessing unit 22, and an identification unit 23, where the sampling unit 21, the preprocessing unit 22, and the identification unit 23 may be integrated into a motor controller, and be formed by combining hardware and software of the motor controller.
The sampling unit 21 is used for acquiring the torque and the rotation speed of the motor according to a preset period when the motor is running. The sampling unit 21 performs a sampling operation after the motor starts to operate (the motor controller may control the motor to operate by a preset initial value of the motion parameter). For example, the sampling unit 21 may collect the motor position or speed signal in real time by using an encoder or a speed measuring device, and convert the motor position or speed signal into the real-time rotation speed of the motor; the motor torque can then be obtained by means of a sensor (current sensor) or a sensorless estimation method.
The preprocessing unit 22 is used for performing enhancement processing on the torque and the rotation speed, and acquiring sample data after the enhancement processing. By the processing of the preprocessing unit 22, it is possible to eliminate the influence of data distortion due to nonlinear factors such as friction, backlash, noise, vibration, etc. at the time of actual operation, among the torque and the rotational speed acquired by the sampling unit 11, and to enhance the effective signal component.
The identifying unit 23 is configured to obtain the motion parameter of the current cycle in an iterative manner according to the sample data, the motion parameter of the previous cycle, and the gradient descent optimization rule.
Specifically, the sample data obtained by the preprocessing unit 22 may include an output quantity, a first input quantity and a second input quantity, the corresponding motion parameters include a load inertia and a viscosity coefficient, and the preprocessing unit 22 includes a first subunit, a second subunit and a third subunit, where the first subunit is configured to perform an integration process on the torque, and takes, as the output quantity, a difference between a torque integral value in a current period and a torque integral value in a previous period; the second subunit is used for obtaining acceleration according to the rotating speed, carrying out integration processing on the acceleration, and taking the difference between the acceleration integral value of the current period and the acceleration integral value of the previous period as a first input quantity; the third subunit performs an integration process on the rotational speed, and uses a difference between the rotational speed integral value of the current period and the speed integral value of the previous period as a second input quantity.
The recognition unit 23 iteratively calculates the motion parameter of the current cycle by the following calculation formula (15):
θ k ＝θ k-1 -α k ×L′(θ k-1 ) (15)
where k is the current period, θ k For the motion parameter of the current period, θ k ＝[J k ,B k ] T And J k For the load inertia of the current period, B k The viscosity coefficient of the current period; θ k-1 Is the motion parameter of the previous period; alpha k For the iterative step of the current cycleLong; l' (θ) k-1 ) The first derivative of the loss function being a characteristic variable of the motion parameter of the previous cycle, and the loss function beingX is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1.
Alternatively, when the computational power is limited, the viscosity coefficient B of the current period is discarded k For example, the viscosity coefficient is directly set to be constant zero, and at this time, θ in the formula (15) is calculated k ＝J k 。
In addition, the system further comprises a setting unit for updating the motion parameters of the motor controller by using the motion parameters of the current period when the iterative operation of the identification unit reaches a stable state, so that the motor controller can use the new motion parameters for motor control in the next period.
The motion parameter identification system in this embodiment belongs to the same concept as the motion parameter identification method in the corresponding embodiment of fig. 1, and detailed implementation process of the motion parameter identification system is shown in the corresponding method embodiment, and technical features in the method embodiment are correspondingly applicable in the device embodiment, which is not described herein.
The embodiment of the present invention further provides a motion parameter identification apparatus 3, where the apparatus 3 may be a servo controller for driving a permanent magnet synchronous motor, and as shown in fig. 3, the motion parameter identification apparatus 3 includes a memory 31 and a processor 32, a computer program executable by the processor 32 is stored in the memory 31, and the processor 32 implements the steps of the motion parameter identification method described above when executing the computer program.
The motion parameter identification device 3 in this embodiment belongs to the same concept as the motion parameter identification method in the corresponding embodiment of fig. 1, and detailed implementation process of the motion parameter identification device is shown in the corresponding method embodiment, and technical features in the method embodiment are correspondingly applicable in this device embodiment, which is not repeated herein.
The embodiment of the invention also provides a computer readable storage medium (for example, located in a servo driver), on which a computer program is stored which, when executed by a processor, implements the steps of the motion parameter identification method as described above. The computer readable storage medium in this embodiment belongs to the same concept as the motion parameter identification method in the corresponding embodiment of fig. 1, and the specific implementation process is detailed in the corresponding method embodiment, and the technical features in the method embodiment are correspondingly applicable in the device embodiment, which is not repeated herein.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present application in any way.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional allocation may be performed by different functional units and modules according to needs. The functional units and modules in the embodiment may be integrated in one processor, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed motion parameter identification method, system and apparatus may be implemented in other manners. For example, the motion parameter identification system embodiments described above are merely illustrative.
In addition, each functional unit in the embodiments of the present application may be integrated in one processor, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each method embodiment described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or interface switching device, recording medium, USB flash disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), electrical carrier wave signals, telecommunications signals, and software distribution media, among others, capable of carrying the computer program code. It should be noted that the computer readable medium may include content that is subject to appropriate increases and decreases as required by jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is not included as electrical carrier signals and telecommunication signals.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims (9)

1. A method for identifying motion parameters, applied to motor control, comprising the steps of:
when the motor runs, the torque and the rotating speed of the motor are obtained according to a preset period;
performing enhancement processing on the torque and the rotating speed, and acquiring sample data after the enhancement processing;
acquiring the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule;
the sample data includes an output quantity, a first input quantity and a second input quantity, the motion parameters include a load inertia, or the motion parameters include a load inertia and a viscosity coefficient, and when the motion parameters of the current period are obtained in an iterative manner, the motion parameters are based on the following loss functionEstablishing a gradient descent optimization rule:
wherein X is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1;
the step of carrying out enhancement treatment on the torque and the rotating speed comprises the following steps:
integrating the torque, and taking the difference between the torque integral value of the current period and the torque integral value of the previous period as output quantity;
obtaining acceleration according to the rotating speed, carrying out integration processing on the acceleration, and taking the difference between an acceleration integral value of a current period and an acceleration integral value of a previous period as a first input quantity;
and integrating the rotation speed, and taking the difference between the rotation speed integral value of the current period and the speed integral value of the previous period as a second input quantity.
2. The method of motion parameter identification of claim 1, further comprising: judging whether the iterative operation reaches a stable state, and updating the motion parameters of the motor controller by using the motion parameters of the current period when the iterative operation reaches the stable state.
3. The method of claim 1, wherein the motion parameters include a load inertia and a viscosity coefficient, and the iteratively obtaining the motion parameters for the current cycle includes:
iteratively calculating the motion parameters of the current period by the following calculation steps:
where k is the current period of time,for the motion parameter of the current cycle, +.>And->For the load inertia of the current cycle, +.>The viscosity coefficient of the current period; />For the motion parameter of the previous cycle, +.>For the iteration step of the current cycle, +.>The first derivative of the loss function, which is characterized by the motion parameter of the previous cycle.
4. The method of claim 1, wherein the motion parameters include load inertia, and the iteratively obtaining the motion parameters of the current cycle includes:
iteratively calculating the motion parameters of the current period by the following calculation steps:
where k is the current period of time,for the motion parameter of the current cycle, +.>And->Load inertia for the current period;for the motion parameter of the previous cycle, +.>For the iteration step of the current cycle, +.>The first derivative of the loss function, which is characterized by the motion parameter of the previous cycle.
5. The method according to claim 3 or 4, characterized in that the iteration step of the current cycle:；
alternatively, the iteration step of the current period:wherein->Is adaptive factor->For the iteration step of the previous cycle, +.>Is a matrix of the first input quantity and the second input quantity of the current period.
6. A motion parameter identification system applied to motor control, which is characterized by comprising a sampling unit, a preprocessing unit and an identification unit, wherein:
the sampling unit is used for acquiring the torque and the rotating speed of the motor according to a preset period when the motor operates;
the preprocessing unit is used for carrying out enhancement processing on the torque and the rotating speed and acquiring sample data after the enhancement processing;
the identification unit is used for obtaining the motion parameters of the current period in an iterative mode according to the sample data, the motion parameters of the previous period and the gradient descent optimization rule;
the sample data includes an output quantity, a first input quantity and a second input quantity, the motion parameters include a load inertia and a viscosity coefficient, and the identification unit is based on the following loss function when obtaining the motion parameters of the current period in an iterative mannerEstablishing a gradient descent optimization rule:
wherein X is an input quantity matrix comprising n first input quantities and n second input quantities, F is an output quantity matrix comprising n output quantities, and n is an integer greater than or equal to 1;
the preprocessing unit comprises a first subunit, a second subunit and a third subunit, wherein:
the first subunit is configured to perform integration processing on the torque, and take a difference between a torque integral value in a current period and a torque integral value in a previous period as an output quantity;
the second subunit is configured to obtain an acceleration according to the rotational speed, perform integration processing on the acceleration, and use a difference between an acceleration integral value in a current period and an acceleration integral value in a previous period as a first input quantity;
the third subunit performs integration processing on the rotation speed, and takes a difference between a rotation speed integral value of a current period and a speed integral value of a previous period as a second input quantity.
7. The motion parameter identification system according to claim 6, wherein the identification unit iteratively calculates the motion parameter of the current cycle by:
where k is the current period of time,for the motion parameter of the current cycle, +.>And->For the load inertia of the current cycle, +.>A viscosity coefficient of the current period or zero; />For the motion parameter of the previous cycle, +.>For the iteration step of the current cycle, +.>The first derivative of the loss function, which is characterized by the motion parameter of the previous cycle.
8. A motion parameter identification apparatus comprising a memory and a processor, the memory having stored thereon a computer program executable on the processor, the processor executing the computer program to perform the steps of the motion parameter identification method according to any one of claims 1 to 5.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the motion parameter identification method according to any of claims 1 to 5.
CN201910804638.3A 2019-08-28 2019-08-28 Motion parameter identification method, system, equipment and computer readable storage medium Active CN110569767B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910804638.3A CN110569767B (en) 2019-08-28 2019-08-28 Motion parameter identification method, system, equipment and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910804638.3A CN110569767B (en) 2019-08-28 2019-08-28 Motion parameter identification method, system, equipment and computer readable storage medium

Publications (2)

Publication Number Publication Date
CN110569767A CN110569767A (en) 2019-12-13
CN110569767B true CN110569767B (en) 2024-01-26

Family

ID=68776608

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910804638.3A Active CN110569767B (en) 2019-08-28 2019-08-28 Motion parameter identification method, system, equipment and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN110569767B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111923045B (en) * 2020-08-07 2021-10-29 珠海格力智能装备有限公司 Robot control method, device, computer readable storage medium and processor
CN112666829B (en) * 2020-12-02 2022-05-27 深圳先进技术研究院 System optimization control method and device, terminal equipment and readable storage medium
CN114177865A (en) * 2021-12-01 2022-03-15 昆明理工大学 Rigid-flexible splicing stirring device and chaotic variable-frequency stirring method thereof

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106026822A (en) * 2016-06-13 2016-10-12 上海电气集团股份有限公司 Inertia online identification method and system of servo motor drive system

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100349352C (en) * 2005-12-09 2007-11-14 天津理工大学 Ambiguity type power system stabilizer parameter self-optimization method and self-optimization device
CN102355193A (en) * 2011-09-30 2012-02-15 哈尔滨工业大学 On-line rotational inertia identification device for alternate current permanent magnet servo system and identification method
CN104391497B (en) * 2014-08-15 2017-08-29 北京邮电大学 A kind of joint servo system parameter identification and control parameter method for on-line optimization
JP6506219B2 (en) * 2016-07-21 2019-04-24 ファナック株式会社 Machine learning device, motor control device and machine learning method for learning current command of motor

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106026822A (en) * 2016-06-13 2016-10-12 上海电气集团股份有限公司 Inertia online identification method and system of servo motor drive system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
时军.高速直线伺服系统的学习前馈控制算法及谐振抑制研究.中国优秀硕士学位论文全文数据库信息科技辑(月刊).2017,(第07期),正文第29-50页. *

Also Published As

Publication number Publication date
CN110569767A (en) 2019-12-13

Similar Documents

Publication Publication Date Title
CN110569767B (en) Motion parameter identification method, system, equipment and computer readable storage medium
CN111788040B (en) Kinetic parameter identification method of robot, robot and storage device
CN105680750B (en) PMSM servo system control methods based on improved model compensation ADRC
CN109927032A (en) A kind of mechanical arm Trajectory Tracking Control method based on High-Order Sliding Mode observer
CN107612445B (en) Control method of follow-up speed regulation system with load acceleration feedback
CN104638999B (en) Dual-servo-motor system control method based on segmentation neutral net friction model
CN109586645B (en) Permanent magnet synchronous motor inertia identification method and equipment
CN106130426B (en) Based on EKF without sensor ultrahigh speed permanent magnet synchronous motor method for controlling number of revolution
CN104009696B (en) Interactive model reference adaptive speed and stator resistance identification method based on sliding-mode control
CN109491251B (en) AC servo system model identification method and equipment considering data disturbance compensation
CN111965976B (en) Robot joint sliding mode control method and system based on neural network observer
CN109639200B (en) Rotational inertia online identification method based on motor load torque detection
CN111162721A (en) Load parameter identification method, control system, device and readable storage medium
JP2018536381A (en) Method and system for controlling an electric motor
CN116423522A (en) Robot dragging teaching zero-force control method with self-adaptive friction compensation
CN106950842B (en) Decoupling control method and device, robot and storage medium
CN112910328B (en) Permanent magnet synchronous motor acceleration arrangement method based on torque observation compensation
CN115566954B (en) Embedded motor speed regulation control compensation method and system
CN114665780A (en) Control method, device and equipment of permanent magnet synchronous motor and storage medium
CN114035436B (en) Backstepping control method, storage medium and device based on saturation adaptive law
CN115313931A (en) Sensor-free vector control method of permanent magnet synchronous motor based on AEKF
CN104734591B (en) The tandem system stabilization speed regulating method of motorcar electric steering motor Field orientable control
JP2906255B2 (en) Servo control device
CN116885993B (en) Servo motor parameter identification method and system integrating cloud end and edge end
CN113067514B (en) Rapid online rotational inertia identification method and system suitable for servo system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant