A multi-rate asynchronous sampling control method for deep-sea robot drive motors

Abstract

The present invention discloses a multi-rate asynchronous sampling control method for a deep-sea robot's drive motor. The method includes the following steps: configuring the basic electrical and mechanical parameters of a motor system with multiple independent partitions; constructing a parameter matrix and achieving control signal decoupling through structural decomposition; setting a base sampling period and configuring differentiated sampling periods that are integer multiples of the base sampling period for each partition to discretize the control signal; constructing a selection matrix to determine the control signal sequence under multi-rate sampling; discretizing the parameter matrix and solving the feedback control gain matrix; constructing an autonomous system to describe the speed tracking signal and solving the feedforward control gain matrix; and finally, calculating discrete control voltages based on the real-time detected armature current according to the multi-rate period to drive the motor. This multi-rate asynchronous sampling strategy effectively matches the dynamic characteristics of different partitions with the control requirements, significantly improving motor control efficiency and responsiveness in deep-sea environments.

Description

Technical Field

[0001] The present invention relates to the field of deep-sea motor control technology, and more specifically, to a multi-rate asynchronous sampling control method for a deep-sea robot drive motor. Background Art

[0002] As the core power source, the deep-sea robot drive motor is responsible for driving the robot's movement and performing delicate operations to meet various mission requirements in complex environments. It plays a vital role in deep-sea exploration and operations. To address the problems of high power density and heat dissipation difficulties, the segmented deep-sea robot drive motor divides the cavity into multiple independent working areas equipped with stators, armatures, and impellers. Each partition drives the rotating shaft in coordination, thereby improving heat dissipation efficiency. If asynchronous sampling control is implemented for each armature, the sampling frequency is allocated and adjusted according to the characteristics and needs of each partition, which can provide a more optimized control solution for the deep-sea robot drive motor. Compared with traditional sampling control, the asynchronous sampling control method can effectively reduce electromagnetic coupling and cross-interference between armatures, reduce computational burden and processing delay, and improve the flexibility and applicability of the motor system.

[0003] Currently, many methods have adopted multi-rate sampling and asynchronous sampling control methods to improve system performance. For example, Chinese Patent Publication No. CN109773799A discloses a multi-rate sampling coordinated control method for a rigid humanoid manipulator, Chinese Patent Publication No. CN117376066A discloses a digital signal modulation mode identification method for asynchronous sampling, and Chinese Patent Publication No. CN117277047A discloses a miniaturized asynchronous sampling dual-optical frequency comb system. However, a multi-rate asynchronous sampling control method for segmented deep-sea robot drive motors has not yet been developed. Furthermore, applying asynchronous sampling control to segmented motors still faces several challenges: Segmented motors may contain multiple independent armatures with significantly different dynamic characteristics. How to accurately assign an appropriate sampling frequency to each armature and model it? How to overcome electromagnetic coupling and cross-interference between different armatures and design an asynchronous sampling controller to ensure smooth motor operation? And how to coordinately control them to achieve a specified speed output when asynchronous sampling is performed on each armature at different times?

[0004] Therefore, it is necessary to develop a new control method for the multi-rate asynchronous sampling requirements of the segmented deep-sea robot drive motor to improve the management and control efficiency of the motor and improve the overall performance of the deep-sea robot. Summary of the Invention

[0005] The purpose of the present invention is to overcome the above-mentioned deficiencies in the prior art and to provide a multi-rate asynchronous sampling control method for a deep-sea robot drive motor.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A multi-rate asynchronous sampling control method for a deep-sea robot drive motor comprises the following steps:

[0008] Step S1, considering a segmented deep-sea robot drive motor with n independent partitions, setting basic electrical parameters and mechanical dynamic parameters of each partition to describe the input and output characteristics of the motor;

[0009] Step S2, constructing a motor parameter matrix that characterizes the dynamic characteristics of the motor system, taking the control voltage signals of each independent partition as a reference, and realizing equivalent block upper triangularization of the motor parameter matrix through matrix structure decomposition;

[0010] Step S3: setting a basic sampling period for the motor system, and configuring a differentiated sampling period for each independent partition based on the period, with the value being a positive integer multiple of the basic sampling period; discretizing the control voltage signal according to the sampling period of each partition, and determining an update rule for the discrete control voltage signal of each partition;

[0011] Step S4, calculating the least common multiple of the sampling periods of each partition and using it as the minimum common period of the discrete control voltage signals, constructing a selection matrix according to the update rule of the discrete control voltage signals of each partition, and determining a multi-rate update rule for a complete discrete control voltage signal sequence composed of the discrete control voltage signals of all partitions within the minimum common period based on the selection matrix;

[0012] Step S5, discretizing the motor parameter matrix after structural decomposition in step S2 according to the configured independent partition sampling periods, and calculating the feedback control gain matrix of the control system;

[0013] Step S6, constructing an autonomous system for generating a motor speed tracking signal, and calculating a feedforward control gain matrix based on a motor parameter matrix of the motor system, a constructed selection matrix, a discretized motor parameter matrix, and the obtained feedback control gain matrix;

[0014] Step S7: construct a control system based on the obtained feedback control gain matrix and feedforward control gain matrix, detect the specific value of the armature current of each partition in real time according to the sampling period of each partition for calculating the discrete control voltage signal, and act on the motor.

[0015] Furthermore, the step S1 includes the following steps:

[0016] Step S101, set the basic electrical variables and parameters of the motor, the speed of the armature is the variable , the real-time control voltage signal of the jth partition is variable , the real-time armature current is variable , armature resistance is a parameter , armature inductance is the parameter , the electromotive force constant is a parameter , where the continuous-time independent variable Emphasize the dynamic characteristics of variables changing over time. The subscript j is an identifier used to distinguish the parameters of the jth partition, j = 1, 2, ..., n;

[0017] Step S102, setting the mechanical dynamic parameters of the motor, the moment of inertia of the armature is the parameter , the friction coefficient is a parameter , the torque constant of the jth armature is parameter .

[0018] Furthermore, step S2 includes the following steps:

[0019] Step S201, constructing a motor parameter matrix for describing the dynamic characteristics of the motor system, based on the basic parameters configured in step S1 dimensional parameter matrix ,and dimensional parameter matrix ,as follows:

[0020] , (1)

[0021] In formula (1), the superscript “T” represents the transpose operation of the matrix; Represents the j-th column element of the parameter matrix B and corresponds to the control voltage signal ; The resistance values ​​of armature 1, armature 2 to armature n are in sequence; These are the inductance values ​​of armature 1, armature 2 to armature n, respectively; These are the electromotive force constants of armature 1, armature 2 to armature n, respectively; The torque constants of armature 1, armature 2 to armature n are in sequence; and are the moment of inertia and friction coefficient of the motor armature respectively;

[0022] Step S202: the parameter matrix in equation (1) , and the parameter matrix The first column in Perform structural decomposition, for the parameter matrix The first column Constructing a Matrix , , from the matrix Select the maximum linearly independent group and add the linearly independent vector to construct a full-rank n-dimensional square matrix , using square matrix For parameter matrix and parameter matrix Perform linear changes to obtain the corresponding equivalent block matrix:

[0023] , (2)

[0024] In formula (2), 、 、 、 is a sub-matrix block; It is a sub-matrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored; the superscript “ " represents the matrix inversion operation;

[0025] Step S203, repeat the structural decomposition in step S202, and transform the parameter matrix and parameter matrix Transformed into the corresponding equivalent block upper triangular matrix:

[0026] , (3)

[0027] In formula (3), and is a diagonal submatrix block, It is a sub-matrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored.

[0028] Furthermore, step S3 includes the following steps:

[0029] Step S301, set the basic sampling period of the motor system to , the control voltage signal for each independent partition Set the differentiated sampling period, recorded as parameter , set the control voltage signal Update cycle Basic sampling period Positive integer multiples of :

[0030] (4)

[0031] In formula (4), is a positive integer, The least common multiple of ;

[0032] Step S302: Determine the expression of the discretized control voltage signal. The update cycle is , the control voltage signal between cycles Keep unchanged, remember for to Discrete control voltage signal within a time period:

[0033] (5)

[0034] In formula (5), the discrete time independent variable Indicates signal The discrete time sequence number, k=1,2,….

[0035] Furthermore, step S4 includes the following steps:

[0036] Step S401, determine is the discrete control voltage signal sequence of the update period, with is the initial moment, the period All control voltage signals in the stack are stacked vertically to form a dimensional discrete control voltage signal sequence:

[0037] (6)

[0038] In formula (6), , , ,Right now represents the row vector composed of all control voltage signals at time k, represents the row vector composed of all control voltage signals at time (k+1), Represents the row vector composed of all control voltage signals at time (k+N-1), It is an ordered sequence composed of all discrete control voltage signals of armature 1 to armature n from time k to (k+N-1) arranged in a specific order. The arrangement rule is: at each sampling moment, the armature voltage signals are arranged in sequence according to the armature numbers 1 to n, and then the armature voltage signal groups at each moment are arranged in sequence from k to (k+N-1) in chronological order;

[0039] Step S402: determine the updating rule of the discrete control voltage signal sequence in formula (6): each discrete control voltage signal The update cycle is set to , every Basic sampling period Update once, not update the moment, Keep the value of the previous moment unchanged;

[0040] Step S403: construct a selection matrix , used to extract or copy the signal value from formula (6) to implement the update rule described in step S402.

[0041] Furthermore, the discrete control voltage signal exist The specific update rules of the moment are as follows: ,but Update to the current value, otherwise, Keep the old value. ;in, Represents the modulo operation.

[0042] Furthermore, the selection matrix The value rules are as follows:

[0043] a. Select Matrix for dimensional square matrix;

[0044] b. Selection matrix Each row has only one 1 element, and the rest are 0;

[0045] c. Selection matrix The The row corresponds to the discrete control voltage signal , where the non-negative integer time-scale variable The value range is ;

[0046] d. Selection matrix The submatrix composed of the first n rows and first n columns of elements is the unit matrix;

[0047] e. Selection matrix Starting from row n+1, if ,but Need to be updated, set the selection matrix No. OK The column element is 1, otherwise, No update is required, so the selection matrix No. Row element copy row elements to keep the two rows of elements consistent.

[0048] Furthermore, step S5 includes the following steps:

[0049] Step S501: Based on the basic sampling period , the parameter matrix is ​​converted to Discretize into matrix pairs :

[0050] , (7)

[0051] In formula (7), and is the parameter matrix of the motor; is a natural constant; is the integral variable in the continuous time domain; yes The infinitesimal element of

[0052] Step S502: Based on the sampling period , the equivalent block upper triangular matrix submatrix pair is converted into Discretize into matrix pairs :

[0053] , (8)

[0054] In formula (8), is the parameter matrix The equivalent block upper triangular matrix submatrix block of , is the parameter matrix The equivalent block upper triangular matrix submatrix block of ;

[0055] Step S503, based on the matrix obtained in step S502 , solve the following discrete Riccati equations:

[0056] (9)

[0057] In formula (9), it contains the matrix Identity matrices of the same dimensions and positive definite matrices , among which is the solution matrix of the discrete Riccati equations;

[0058] Step S504: Based on the solution matrix obtained in step S503 Construct the feedback control gain matrix of the control system:

[0059] (10)

[0060] In formula (10), is the constructed feedback control gain matrix.

[0061] Furthermore, step S6 includes the following steps:

[0062] Step S601: construct an autonomous system model without external input, relying only on the initial state and internal dynamic generation signals, which is equivalent to generating a tracking signal of the motor speed. The specific form is:

[0063] (11)

[0064] In formula (11), is the state vector of the autonomous system; It is the motor speed tracking signal output by the autonomous system; yes dimensional state matrix; yes dimensional output matrix;

[0065] Step S602: construct a parameter matrix for calculating the feedforward control gain matrix of the control system:

[0066] , ,

[0067] , ,

[0068] ,

[0069]

[0070] in, 、 、 and is the parameter matrix constructed based on the matrix pair, state matrix, and feedback control gain matrix, for dimensional matrix, where the first n elements are 0 and the last element is 1;

[0071] Step S603: Solve the linear matrix equations to calculate the feedforward control gain matrix of the control system:

[0072] (12)

[0073] In formula (12), the matrix is the unknown solution matrix of the linear matrix equations; the matrix is the output matrix defined in equation (11); is the constructed selection matrix; , The unknown matrix in For the feedforward control gain matrix that needs to be solved for specific parameters, for dimensional unknown matrix.

[0074] Furthermore, step S7 includes the following steps:

[0075] Step S701: Detect the armature current value of each partition in real time, construct a control system based on the obtained feedback control gain matrix and feedforward control gain matrix, and calculate the discrete control voltage signal of each partition according to the following formula: :

[0076] (13)

[0077] In formula (13), To obtain the feedback control gain matrix; To obtain the feedforward control gain matrix No. OK; for The measured time Armature current value of each partition; is the state vector of the autonomous system;

[0078] Step S702, Update the discrete control voltage signal according to formula (13) at all times The value of The first A partition for controlling the motor.

[0079] The beneficial effects of this invention are: by dynamically assigning differentiated sampling frequencies to each partition, it effectively matches the dynamic characteristics and control requirements of different regions of a segmented motor, ensuring high-precision control in critical areas while avoiding unnecessary waste of computing resources. This invention is not only applicable to deep-sea robot drive systems, but its multi-rate asynchronous control concept can also be extended to other motor systems requiring partitioned control, thus possessing broad engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A schematic diagram of a control structure of a multi-rate asynchronous sampling control method for a deep-sea robot driving motor in this embodiment;

[0081] Figure 2 This is a control effect diagram of the multi-rate asynchronous sampling control method for the deep-sea robot drive motor in this embodiment;

[0082] Figure 3 This is a control voltage signal diagram of the multi-rate asynchronous sampling control method for the deep-sea robot drive motor in this embodiment. DETAILED DESCRIPTION

[0083] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0084] Example: A multi-rate asynchronous sampling control system for a deep-sea robot drive motor, such as Figure 1 As shown in Figure 1, it includes three parts: motor system, control system and autonomous system. The parameters of the control system and autonomous system models need to be calculated step by step and implemented in the computer system. The motor system is segmented and contains n independent partitions. Each partition contains an armature, a stator and an impeller. Each partition is asynchronously sampled at multiple rates to obtain a control voltage signal. The control system uses a multi-rate asynchronous sampling strategy to control the voltage signal of each partition. Set different update cycles , remaining constant within the cycle. The control system includes dedicated feedback gain and feedforward gain modules for each partition. The feedback control variable is calculated by real-time detection of the armature current signal. A feedforward compensation variable is generated based on the autonomous system state variable. The two are superimposed to form the control voltage signal for each partition.

[0085] A multi-rate asynchronous sampling control method for a deep-sea robot's drive motor is proposed. This method addresses the multi-zone characteristics of a segmented deep-sea robot's drive motor and achieves efficient and flexible control by assigning differentiated sampling control frequencies to each zone. The method includes the following steps:

[0086] Step S1, basic parameter configuration and characteristic description of the motor system: Consider a segmented deep-sea robot drive motor with n independent partitions, each partition contains an armature, a stator and an impeller, and the partitions jointly drive the rotating shaft. The basic electrical parameters and mechanical dynamic parameters of each partition are set separately to describe the input and output characteristics of the motor.

[0087] Specifically, step S1 includes the following steps:

[0088] Step S101, set the basic electrical variables and parameters of the motor, the speed of the armature is the variable , the real-time control voltage signal of the jth partition is variable , the real-time armature current is variable , armature resistance is a parameter , armature inductance is the parameter , the electromotive force constant is a parameter , where the continuous-time independent variable Emphasize the dynamic characteristics of variables changing over time. The subscript j is an identifier used to distinguish the parameters of the jth partition, j = 1, 2, ..., n;

[0089] Step S102, setting the mechanical dynamic parameters of the motor, the moment of inertia of the armature is the parameter , the friction coefficient is a parameter , the torque constant of the jth armature is parameter .

[0090] Step S2, motor parameter matrix structure decomposition and control signal decoupling: Based on the basic electrical parameters and mechanical dynamic parameters of the motor determined in step S1, a motor parameter matrix that characterizes the dynamic characteristics of the motor system is constructed. Taking the control voltage signals of each independent partition as the benchmark, the equivalent block upper triangularization of the motor parameter matrix is ​​achieved through matrix structure decomposition, thereby completing the decoupling of the control voltage signal and providing conditions for configuring differentiated sampling periods for each armature.

[0091] Specifically, step S2 includes the following steps:

[0092] Step S201, constructing a motor parameter matrix for describing the dynamic characteristics of the motor system, based on the basic parameters configured in step S1 dimensional parameter matrix ,and dimensional parameter matrix ,as follows:

[0093] , (1)

[0094] In formula (1), the superscript “T” represents the transpose operation of the matrix; Represents the j-th column element of the parameter matrix B and corresponds to the control voltage signal ; The resistance values ​​of armature 1, armature 2 to armature n are in sequence; These are the inductance values ​​of armature 1, armature 2 to armature n, respectively; These are the electromotive force constants of armature 1, armature 2 to armature n, respectively; The torque constants of armature 1, armature 2 to armature n are in sequence; and are the moment of inertia and friction coefficient of the motor armature respectively;

[0095] Step S202: the parameter matrix in equation (1) , and the parameter matrix The first column in Perform structural decomposition, for the parameter matrix The first column Constructing a Matrix , , from the matrix Select the maximum linearly independent group and add the linearly independent vector to construct a full-rank n-dimensional square matrix , using square matrix For parameter matrix and parameter matrix Perform linear changes to obtain the corresponding equivalent block matrix:

[0096] , (2)

[0097] In formula (2), 、 、 、 is a sub-matrix block; It is a sub-matrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored; the superscript “ " represents the matrix inversion operation;

[0098] Step S203, repeat the structural decomposition in step S202, and finally transform the parameter matrix and parameter matrix Convert it into an equivalent block upper triangular matrix, that is, use the structural decomposition method in step S202 to decompose the submatrix in formula (2) and submatrix The first column of the structure decomposition is performed to calculate the equivalent block matrix. Continuing this process will eventually give the parameter matrix and parameter matrix The equivalent block upper triangular matrix of :

[0099] , (3)

[0100] In formula (3), and is a diagonal submatrix block, It is a sub-matrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored.

[0101] Step S3, configuration and discretization of the sampling period of the control voltage signal of each partition: set the basic sampling period of the motor system, and based on this period, configure each independent partition with a differentiated sampling period that is a positive integer multiple of the basic sampling period, discretize the control voltage signal according to the sampling period of each partition, and determine the update rules of the discrete control voltage signal of each partition.

[0102] Specifically, step S3 includes the following steps:

[0103] Step S301, set the basic sampling period of the motor system to , the control voltage signal for each independent partition Set the differentiated sampling period, recorded as parameter , set the control voltage signal Update cycle Basic sampling period A positive integer multiple of , that is:

[0104] (4)

[0105] In formula (4), is a positive integer, The least common multiple of ;

[0106] Step S302: Determine the expression of the discretized control voltage signal. The update cycle is , the control voltage signal between cycles Keep unchanged, remember for to The discrete control voltage signal within a time period is:

[0107] (5)

[0108] In formula (5), the discrete time independent variable Indicates signal The discrete time sequence number, k=1,2,….

[0109] Step S4, constructing a multi-rate control voltage signal update rule based on the selection matrix: calculating the least common multiple of the sampling periods of each partition and using it as the minimum common period of the discrete control voltage signal, constructing a selection matrix according to the update rule of the discrete control voltage signal of each partition, and determining the multi-rate update rule of the complete discrete control voltage signal sequence composed of the discrete control voltage signals of all partitions within the minimum common period based on the selection matrix.

[0110] Specifically, step S4 includes the following steps:

[0111] Step S401, determine is the discrete control voltage signal sequence of the update period, with is the initial moment, the period All control voltage signals in the stack are stacked vertically to form a dimensional discrete control voltage signal sequence:

[0112] (6)

[0113] In formula (6), , , ,Right now represents the row vector composed of all control voltage signals at time k, represents the row vector composed of all control voltage signals at time (k+1), Represents the row vector composed of all control voltage signals at time (k+N-1), It is an ordered sequence composed of all discrete control voltage signals of armature 1 to armature n from time k to (k+N-1) arranged in a specific order. The arrangement rule is: at each sampling moment, the armature voltage signals are arranged in sequence according to the armature numbers 1 to n, and then the armature voltage signal groups at each moment are arranged in sequence from k to (k+N-1) in chronological order;

[0114] Step S402: determine the update rule of the discrete control voltage signal sequence in formula (6). The update rule is to convert each discrete control voltage signal into The update cycle is set to , that is, every Basic sampling period Update once, not update the moment, Keep the value of the previous moment unchanged;

[0115] Discrete control voltage signal exist The specific update rules of the moment are as follows: ,but Update to the current value, otherwise, Keep the old value, that is ;in, Represents the modulo operation, which is to find the remainder after dividing two integers;

[0116] Step S403: To implement the update rule described in step S402, a selection matrix needs to be constructed. , used to extract or copy the signal value from formula (6), select the matrix The value rules are as follows:

[0117] a. Select Matrix for dimensional square matrix;

[0118] b. Selection matrix Each row has only one 1 element, and the rest are 0;

[0119] c. Selection matrix The The row corresponds to the discrete control voltage signal , where the non-negative integer time-scale variable The value range is ;

[0120] d. Selection matrix The submatrix composed of the first n rows and first n columns of elements is the unit matrix;

[0121] e. Selection matrix Starting from row n+1, if ,but Need to be updated, set the selection matrix No. OK The column element is 1, otherwise, No update is required, so the selection matrix No. Row element copy row elements to keep the two rows of elements consistent.

[0122] According to the above rules, the selection matrix is ​​successfully constructed ,but That is, considering the discrete control voltage signal sequence after multi-rate sampling, the matrix is ​​selected The specific parameters of the control gain matrix will be calculated later to build a control system.

[0123] Step S5, discretization of the motor parameter matrix and solution of the feedback control gain matrix: discretize the motor parameter matrix after structural decomposition in step S2 according to the sampling periods of each independent partition configured in step S3, and calculate the feedback control gain matrix of the control system based on the discretized motor parameter matrix.

[0124] Specifically, step S5 includes the following steps:

[0125] Step S501: Based on the basic sampling period , the parameter matrix in formula (1) is converted to Discretize into matrix pairs ,Right now:

[0126] , (7)

[0127] In formula (7), and is the parameter matrix of the motor; is a natural constant; is the integral variable in the continuous time domain; yes The infinitesimal element of

[0128] Step S502: Based on the sampling period , the sub-matrix in equation (3) of step S203 is converted into Discretized into , that is, the submatrix pair Discretize into matrix pairs , the value range of subscript j is j=1,2,…,n;

[0129] , (8)

[0130] In formula (8), is the parameter matrix The equivalent block upper triangular matrix submatrix block of , is the parameter matrix The equivalent block upper triangular matrix submatrix block of ;

[0131] Step S503, based on the matrix obtained in step S502 , solve the following discrete Riccati equations:

[0132] (9)

[0133] In formula (9), it contains the matrix Identity matrices of the same dimensions and positive definite matrices , among which is the solution matrix of the discrete Riccati equations;

[0134] Step S504: Based on the solution matrix obtained in step S503 Construct the feedback control gain matrix of the control system:

[0135] (10)

[0136] In formula (10), is the constructed feedback control gain matrix.

[0137] Step S6, autonomous system construction and feedforward control gain matrix solution: Construct an autonomous system for generating a motor speed tracking signal, and calculate the feedforward control gain matrix based on the motor parameter matrix of the motor system in step S2, the selection matrix constructed in step S4, the discretized motor parameter matrix in step S5, and the feedback control gain matrix obtained in step S5.

[0138] Specifically, step S6 includes the following steps:

[0139] Step S601: construct an autonomous system model without external input, relying only on the initial state and internal dynamic generation signals, which is equivalent to generating a tracking signal of the motor speed. The specific form is:

[0140] (11)

[0141] In formula (11), is the state vector of the autonomous system; It is the motor speed tracking signal output by the autonomous system, that is, the reference signal of the motor speed; yes dimensional state matrix; yes dimensional output matrix; and The specific parameters are determined according to the specific form of the motor speed tracking signal;

[0142] Step S602: construct a parameter matrix for calculating the feedforward control gain matrix of the control system:

[0143] , , ,

[0144] , ,

[0145] in, and is the matrix pair defined by equation (7) in step S501, is the state matrix of the autonomous system formula (11) in step S601, is the feedback control gain matrix constructed according to formula (10) in step S504, 、 、 and is the parameter matrix constructed based on the above matrix, for dimensional matrix, where the first n elements are 0 and the last element is 1, a positive integer is defined in step S301 the lowest common multiple of ;

[0146] Step S603: Solve the linear matrix equations to calculate the feedforward control gain matrix of the control system:

[0147] (12)

[0148] In formula (12), the matrix is the unknown solution matrix of the linear matrix equation system of formula (12); is the output matrix defined in equation (11); is the selection matrix constructed in step S403; , The unknown matrix in That is the feedforward control gain matrix that needs to solve the specific parameters, for The specific value of the unknown matrix is ​​calculated by solving the linear matrix equation group (12).

[0149] Step S7, calculate the discrete control voltage signal in real time and act on the motor: construct a control system based on the feedback control gain matrix obtained in step S5 and the feedforward control gain matrix obtained in step S6, and detect the specific value of the armature current of each partition used to calculate the discrete control voltage signal in real time according to the sampling period of each partition, and act on the motor.

[0150] Specifically, step S7 includes the following steps:

[0151] Step S701: detect the armature current value of each partition in real time, construct a control system based on the feedback control gain matrix obtained by formula (10) and the feedforward control gain matrix obtained by formula (12), and calculate the discrete control voltage signal of each partition according to the following formula: :

[0152] (13)

[0153] In formula (13), is the feedback control gain matrix obtained in equation (10); is the feedforward control gain matrix obtained in equation (12) No. OK; for The measured time Armature current value of each partition; is the state vector of the autonomous system;

[0154] Step S702, Update the discrete control voltage signal according to formula (13) at all times The value of The first A partition for controlling the motor.

[0155] In this embodiment, the controlled object is a segmented motor system with 4 partitions, that is, the number of partitions is , installation step S1, set the basic electrical parameters of the motor:

[0156] , , , ,

[0157] , , , ,

[0158] , , , ;

[0159] Set the mechanical dynamic parameters of the motor:

[0160] , , , , , .

[0161] According to step S2, the parameter matrix in equation (1) is constructed:

[0162] ,

[0163] Perform structural decomposition on the above parameter matrix to obtain the parameter matrix and parameter matrix The equivalent block upper triangular matrix of the form (3) is as follows:

[0164] , , , , .

[0165] According to step S3, set the basic sampling period of the motor system , and then set the update cycle of each partition control voltage signal , , .

[0166] According to step S4, construct the selection matrix The specific values ​​are as follows:

[0167] ,

[0168] in, , ,

[0169] , ,

[0170] .

[0171] According to step S5, the discrete Riccati equations in equation (9) are solved, and the feedback control gain matrix is ​​constructed according to equation (10):

[0172] , , .

[0173] According to step S6, setting the motor speed requires tracking the following cosine signal:

[0174]

[0175] To generate the above cosine signal, it is necessary to construct an autonomous system as shown in Equation (11), where the parameter matrix is:

[0176] , ,

[0177] Solving the linear matrix equations in equation (12), we get the feedforward control gain matrix as follows:

[0178] .

[0179] According to step S7, according to the update cycle , , The control voltage signal of each partition is calculated and applied to the armature of each partition.

[0180] Figure 2 This image demonstrates the effectiveness of multi-rate asynchronous sampling control for the deep-sea robot's drive motors. The blue solid line represents the actual motor speed, and the green dashed line represents the target tracking speed. The results demonstrate that the system achieves excellent speed tracking performance.

[0181] Figure 3 The control voltage signal waveform of each partition is shown. From the figure, we can see that the control voltage signal of different partitions , , , A differentiated update cycle is adopted.

[0182] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A multi-rate asynchronous sampling control method for a deep-sea robot drive motor, characterized in that: The steps include: Step S1, considering a segmented deep-sea robot drive motor with n independent partitions, setting basic electrical parameters and mechanical dynamic parameters of each partition to describe the input and output characteristics of the motor; Step S2, constructing a motor parameter matrix that characterizes the dynamic characteristics of the motor system, taking the control voltage signals of each independent partition as a reference, and realizing equivalent block upper triangularization of the motor parameter matrix through matrix structure decomposition; Step S3: setting a basic sampling period for the motor system, and configuring a differentiated sampling period for each independent partition based on the period, with the value being a positive integer multiple of the basic sampling period; discretizing the control voltage signal according to the sampling period of each partition, and determining an update rule for the discrete control voltage signal of each partition; Step S4, calculating the least common multiple of the sampling periods of each partition and using it as the minimum common period of the discrete control voltage signals, constructing a selection matrix according to the update rule of the discrete control voltage signals of each partition, and determining a multi-rate update rule for a complete discrete control voltage signal sequence composed of the discrete control voltage signals of all partitions within the minimum common period based on the selection matrix; Step S5, discretizing the motor parameter matrix after structural decomposition according to the configured independent partition sampling period, and calculating the feedback control gain matrix of the control system; Step S6, constructing an autonomous system for generating a motor speed tracking signal, and calculating a feedforward control gain matrix based on a motor parameter matrix of the motor system, a constructed selection matrix, a discretized motor parameter matrix, and the obtained feedback control gain matrix; Step S7: construct a control system based on the obtained feedback control gain matrix and feedforward control gain matrix, detect the specific value of the armature current of each partition in real time according to the sampling period of each partition for calculating the discrete control voltage signal, and act on the motor.

2. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 1, characterized in that: The step S1 comprises the following steps: Step S101, set the basic electrical variables and parameters of the motor, the speed of the armature is the variable , the real-time control voltage signal of the jth partition is variable , the real-time armature current is variable , armature resistance is a parameter , armature inductance is the parameter , the electromotive force constant is a parameter , where the continuous-time independent variable Emphasize the dynamic characteristics of variables changing over time. The subscript j is an identifier used to distinguish the parameters of the jth partition, j = 1, 2, ..., n; Step S102, setting the mechanical dynamic parameters of the motor, the moment of inertia of the armature is the parameter , the friction coefficient is a parameter , the torque constant of the jth armature is parameter .

3. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 1, characterized in that: The step S2 comprises the following steps: Step S201, constructing a motor parameter matrix for describing the dynamic characteristics of the motor system, based on the basic parameters configured in step S1 dimensional parameter matrix ,and dimensional parameter matrix ,as follows: ， (1) In formula (1), the superscript "T" represents the transpose operation of the matrix; Represents the j-th column element of the parameter matrix B and corresponds to the control voltage signal ; The resistance values ​​of armature 1, armature 2 to armature n are in sequence; These are the inductance values ​​of armature 1, armature 2 to armature n, respectively; These are the electromotive force constants of armature 1, armature 2 to armature n, respectively; The torque constants of armature 1, armature 2 to armature n are in sequence; and are the moment of inertia and friction coefficient of the motor armature respectively; Step S202: the parameter matrix in equation (1) , and the parameter matrix The first column in Perform structural decomposition, for the parameter matrix The first column Constructing a Matrix , , from the matrix Select the maximum linearly independent group and add the linearly independent vector to construct a full-rank n-dimensional square matrix , using square matrix For parameter matrix and parameter matrix Perform linear changes to obtain the corresponding equivalent block matrix: ， (2) In formula (2), 、 、 、 is a sub-matrix block; is a submatrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored; the superscript " " represents the matrix inversion operation; Step S203, repeat the structural decomposition in step S202, and transform the parameter matrix and parameter matrix Transformed into the corresponding equivalent block upper triangular matrix: ， (3) In formula (3), and is a diagonal submatrix block, It is a sub-matrix block that is irrelevant to subsequent calculations and whose specific values ​​can be ignored.

4. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 1, characterized in that: The step S3 comprises the following steps: Step S301, set the basic sampling period of the motor system to , the control voltage signal for each independent partition Set the differentiated sampling period, recorded as parameter , set the control voltage signal Update cycle Basic sampling period Positive integer multiples of : (4) In formula (4), is a positive integer, The least common multiple of ; Step S302: Determine the expression of the discretized control voltage signal. The update cycle is , the control voltage signal between cycles Keep unchanged, remember for to Discrete control voltage signal within a time period: (5) In formula (5), the discrete time independent variable Indicates signal The discrete time sequence number, k=1,2,….

5. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 4, characterized in that: The step S4 comprises the following steps: Step S401, determine is the discrete control voltage signal sequence of the update period, with is the initial moment, the period All control voltage signals in the stack are stacked vertically to form a dimensional discrete control voltage signal sequence: (6) In formula (6), , , ,Right now represents the row vector composed of all control voltage signals at time k, represents the row vector composed of all control voltage signals at time (k+1), Represents the row vector composed of all control voltage signals at time (k+N-1), It is an ordered sequence composed of all discrete control voltage signals of armature 1 to armature n from time k to (k+N-1) arranged in a specific order. The arrangement rule is: at each sampling moment, the armature voltage signals are arranged in sequence according to the armature numbers 1 to n, and then the armature voltage signal groups at each moment are arranged in sequence from k to (k+N-1) in chronological order; Step S402: determine the updating rule of the discrete control voltage signal sequence in formula (6): each discrete control voltage signal The update cycle is set to , every Basic sampling period Update once, not update the moment, Keep the value of the previous moment unchanged; Step S403: construct a selection matrix , used to extract or copy the signal value from formula (6) to implement the update rule described in step S402.

6. A multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 5, characterized in that: Discrete control voltage signal exist The specific update rules of the moment are as follows: ,but Update to the current value, otherwise, Keep the old value. ;in, Represents the modulo operation.

7. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 5, characterized in that: The selection matrix The value rules are as follows: a. Select Matrix for dimensional square matrix; b. Selection matrix Each row has only one 1 element, and the rest are 0; c. Selection matrix The The row corresponds to the discrete control voltage signal , where the non-negative integer time-scale variable The value range is ; d. Selection matrix The submatrix composed of the first n rows and first n columns of elements is the unit matrix; e. Selection matrix Starting from row n+1, if ,but Need to be updated, set the selection matrix No. OK The column element is 1, otherwise, No update is required, so the selection matrix No. Row element copy row elements to keep the two rows of elements consistent.

8. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 3, characterized in that: The step S5 comprises the following steps: Step S501: Based on the basic sampling period , the parameter matrix is ​​converted to Discretize into matrix pairs : ， (7) In formula (7), and is the parameter matrix of the motor; is a natural constant; is the integral variable in the continuous time domain; yes The infinitesimal element of Step S502: Based on the differentiated sampling period , the equivalent block upper triangular matrix submatrix pair is converted into Discretize into matrix pairs : ， (8) In formula (8), is the parameter matrix The equivalent block upper triangular matrix submatrix block of , is the parameter matrix The equivalent block upper triangular matrix submatrix block of ; Step S503, based on the matrix obtained in step S502 , solve the following discrete Riccati equations: (9) In formula (9), it contains the matrix Identity matrices of the same dimensions and positive definite matrices , among which is the solution matrix of the discrete Riccati equations; Step S504: Based on the solution matrix obtained in step S503 Construct the feedback control gain matrix of the control system: (10) In formula (10), is the constructed feedback control gain matrix.

9. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 8, characterized in that: The step S6 comprises the following steps: Step S601: construct an autonomous system model without external input, relying only on the initial state and internal dynamic generation signals, which is equivalent to generating a tracking signal of the motor speed. The specific form is: (11) In formula (11), is the state vector of the autonomous system; It is the motor speed tracking signal output by the autonomous system; yes dimensional state matrix; yes dimensional output matrix; Step S602: construct a parameter matrix for calculating the feedforward control gain matrix of the control system: ， ， ， ， ， in, 、 、 and is the parameter matrix constructed based on the matrix pair, state matrix, and feedback control gain matrix, for dimensional matrix, where the first n elements are 0 and the last element is 1; Step S603: Solve the linear matrix equations to calculate the feedforward control gain matrix of the control system: (12) In formula (12), the matrix is the unknown solution matrix of the linear matrix equations; the matrix is the output matrix defined in equation (11); is the constructed selection matrix; , The unknown matrix in For the feedforward control gain matrix that needs to be solved for specific parameters, for dimensional unknown matrix.

10. The multi-rate asynchronous sampling control method for a deep-sea robot drive motor according to claim 9, characterized in that: The step S7 includes the following steps: Step S701: Detect the armature current value of each partition in real time, construct a control system based on the obtained feedback control gain matrix and feedforward control gain matrix, and calculate the discrete control voltage signal of each partition according to the following formula: : (13) In formula (13), To obtain the feedback control gain matrix; To obtain the feedforward control gain matrix No. OK; for The measured time Armature current value of each partition; is the state vector of the autonomous system; Step S702, Update the discrete control voltage signal according to formula (13) at all times The value of The first A partition for controlling the motor; , where Is a positive integer.

Images