CN117277888A - Method and device for predicting voltage vector of stator of permanent magnet synchronous motor - Google Patents

Method and device for predicting voltage vector of stator of permanent magnet synchronous motor Download PDF

Info

Publication number
CN117277888A
CN117277888A CN202311570761.6A CN202311570761A CN117277888A CN 117277888 A CN117277888 A CN 117277888A CN 202311570761 A CN202311570761 A CN 202311570761A CN 117277888 A CN117277888 A CN 117277888A
Authority
CN
China
Prior art keywords
vector
stator
axis
voltage
current
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.)
Granted
Application number
CN202311570761.6A
Other languages
Chinese (zh)
Other versions
CN117277888B (en
Inventor
逯超
孟庆辉
王汉瑞
任彬
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.)
Weichai Power Co Ltd
Original Assignee
Weichai Power 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 Weichai Power Co Ltd filed Critical Weichai Power Co Ltd
Priority to CN202311570761.6A priority Critical patent/CN117277888B/en
Publication of CN117277888A publication Critical patent/CN117277888A/en
Application granted granted Critical
Publication of CN117277888B publication Critical patent/CN117277888B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H02P21/0014—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using neural networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/14—Estimation or adaptation of machine parameters, e.g. flux, current or voltage
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/22—Current control, e.g. using a current control loop
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P25/00—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details
    • H02P25/02—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details characterised by the kind of motor
    • H02P25/022—Synchronous motors
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P2207/00—Indexing scheme relating to controlling arrangements characterised by the type of motor
    • H02P2207/05—Synchronous machines, e.g. with permanent magnets or DC excitation

Abstract

The method comprises the steps of fusing a priori estimation method and a neural network topology prediction method to obtain an optimized predicted current vector, reconstructing a vector expression, solving the minimum value of the vector expression, determining the voltage of the d axis of the stator corresponding to the minimum value and the voltage of the q axis of the stator as an optimal predicted voltage vector, and obtaining the optimal predicted voltage vector.

Description

Method and device for predicting voltage vector of stator of permanent magnet synchronous motor
Technical Field
The present disclosure relates to the field of permanent magnet synchronous motors, and in particular, to a method and an apparatus for predicting a voltage vector of a stator of a permanent magnet synchronous motor, a method for controlling a permanent magnet synchronous motor, a computer readable storage medium, and an electronic device.
Background
The existing scheme often adopts a corresponding model to predict the voltage changes of the d axis and the q axis of the stator of the permanent magnet synchronous motor in a period of time in the future, but a model used for the traditional permanent magnet synchronous motor model prediction control is a mechanism model, and the mechanism model cannot truly reflect the change of the motor along with the environment in the real world, and cannot simulate the uncertainty and the unknown of the external environment.
Therefore, a method for predicting the voltage vector of the stator of the permanent magnet synchronous motor is needed to solve the problem of lower motor model accuracy in the existing scheme of permanent magnet synchronous motor model prediction control.
Disclosure of Invention
The main objective of the present application is to provide a method and an apparatus for predicting a voltage vector of a stator of a permanent magnet synchronous motor, a method for controlling a permanent magnet synchronous motor, a computer readable storage medium and an electronic device, so as to at least solve the problem of low accuracy of a motor model in the model prediction control of the permanent magnet synchronous motor in the existing scheme.
In order to achieve the above object, according to one aspect of the present application, there is provided a method of predicting a voltage vector of a stator of a permanent magnet synchronous motor, the method comprising:
acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment; obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator; weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current; and constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving a minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as the optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage.
Optionally, a priori estimation method is adopted, and a first predicted current vector is obtained according to the first stator voltage, the second stator voltage and the current rotor angular speed, including: and constructing a first predicted current vector expression by adopting the first stator voltage, the second stator voltage and the current rotor angular speed, wherein the first predicted current vector expression comprises a proportional relation between an inductance value of a q-axis of the stator and an inductance value of a d-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of the q-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of the d-axis of the stator, a proportional relation between a sampling period and an inductance value of the q-axis of the stator, and a proportional relation between the sampling period and an inductance value of the d-axis of the stator, and determining the first predicted current vector according to the first predicted current vector expression.
Optionally, the neural network in the neural network topology model includes a plurality of input layers, a plurality of hidden layers and a plurality of output layers, and each of the input layers, each of the hidden layers and each of the output layers are connected in a fully connected manner.
Optionally, in the process of processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, the method further includes:
Constructing a d-axis prediction performance expression and a q-axis prediction performance expression, wherein the d-axis prediction performance expression comprises a difference value between a current of a t-th predicted d-axis obtained by adopting the neural network topology model and a current of a t+1th predicted d-axis obtained by adopting the neural network topology model, and the q-axis prediction performance expression comprises a difference value between a current of a t-th predicted q-axis obtained by adopting the neural network topology model and a current of a t+1th predicted q-axis obtained by adopting the neural network topology model;
a predictive performance vector is determined, the predictive performance vector being a vector comprising d-axis predictive performance and q-axis predictive performance.
Optionally, after determining the predictive performance vector, the method further comprises:
according to，
Determining a connection weight between neurons of the input layer and the hidden layer and a connection weight between neurons of the hidden layer and the output layer when the neural network topology model is adopted for t+1th prediction;
wherein,for the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t+1th prediction, the weight is +.>To use the neural network topology model to predict the t+1st time Connection weights between neurons of said hidden layer and said output layer at the time,/->For the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t-th prediction, the +_>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted for the t-th prediction, the +_>For a preset coefficient, loss is the predictive performance vector, i is used for representing the ith neuron of the input layer, j is used for representing the jth neuron of the hidden layer, and k is used for representing the kth neuron of the output layer.
Optionally, constructing a vector expression by using the optimized predicted current vector and the stator voltage vector, calculating a minimum value of the vector expression, and determining a voltage of a d axis of the stator and a voltage of a q axis of the stator corresponding to the minimum value as an optimal predicted voltage vector, including:
constructing a vector expression by adopting the optimized predicted current vector and the stator voltage vector, solving the minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as the optimized predicted voltage vector, wherein the method comprises the following steps:
According to，
Determining the optimal predicted voltage vector, wherein,for the optimal predicted voltage vector, +.>For predicting the future Nth stepIs +.>Is->Transposed matrix of>For predicting the difference between said optimized predicted current vector of future step a and the actual value of the corresponding current vector,/v>Is->Transposed matrix of>For the stator voltage vector,/a>Is->The transposed matrix P, R, Q of (a) is a preset weight matrix, and the actual value of the current vector is the current of the d axis of the stator and the current of the q axis of the stator fed back by the permanent magnet synchronous motor in real time.
According to another aspect of the present application, there is provided a predicting apparatus for a voltage vector of a stator of a permanent magnet synchronous motor, the apparatus comprising:
the device comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment;
The first processing unit is configured to obtain a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular velocity by using an a priori estimation method, and process the first stator voltage, the second stator voltage and the current rotor angular velocity by using a neural network topology model to obtain a second predicted current vector, where the first predicted current vector is a vector including a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector including a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using multiple sets of training data, and each set of training data in the multiple sets of training data includes a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
the second processing unit is used for carrying out weighting processing on the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current;
And the third processing unit is used for constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving the minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as the optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage.
According to another aspect of the present application, there is provided a control method of a permanent magnet synchronous motor, the method including: acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment; obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator; weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current; constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving a minimum value of the vector expression, and determining a voltage of a d axis of the stator and a voltage of a q axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage; and carrying out coordinate inverse transformation on the optimal predicted voltage vector to obtain a target three-phase voltage, and controlling the permanent magnet synchronous motor based on the target three-phase voltage.
According to another aspect of the present application, there is provided a computer readable storage medium, where the computer readable storage medium includes a stored program, and when the program runs, the apparatus where the computer readable storage medium is controlled to execute any one of the methods for predicting a voltage vector of a stator of a permanent magnet synchronous motor.
According to another aspect of the present application, there is provided an electronic device comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising a prediction method for performing a voltage vector of a stator of any one of the permanent magnet synchronous motors.
By applying the technical scheme, the prior estimation method and the neural network topology prediction method are fused to obtain the optimized prediction current vector, and the vector expression is reconstructed to obtain the optimal prediction voltage vector, so that the accuracy of prediction is improved due to the combination of the prior estimation method and the neural network topology prediction method, and the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
fig. 1 shows a flow chart of a method for predicting a voltage vector of a stator of a permanent magnet synchronous motor according to an embodiment of the present application;
FIG. 2 shows a topology schematic of a neural network topology model;
FIG. 3 shows a schematic diagram of a process for deriving an optimized predicted current vector from a first stator voltage, a second stator voltage, and a present rotor angular velocity;
fig. 4 shows a flow diagram of a method for controlling a permanent magnet synchronous motor;
fig. 5 shows a block diagram of a device for predicting a voltage vector of a stator of a permanent magnet synchronous motor according to an embodiment of the present application;
fig. 6 shows a flow chart of another control method of the permanent magnet synchronous motor.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
As introduced in the background art, the existing scheme often adopts a corresponding model to predict the voltage changes of the d axis and the q axis of the stator of the permanent magnet synchronous motor in a period of time in the future, but the model used in the traditional permanent magnet synchronous motor model prediction control is a mechanism model, and the mechanism model cannot truly reflect the change of the motor along with the environment in the real world, and cannot simulate the uncertainty and the unknowness of the external environment, so as to solve the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
In the present embodiment, a method of predicting a voltage vector of a stator of a permanent magnet synchronous motor is provided, it is to be noted that the steps shown in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from that herein.
Fig. 1 is a flow chart of a method for predicting a voltage vector of a stator of a permanent magnet synchronous motor according to an embodiment of the present application. As shown in fig. 1, the method comprises the steps of:
step S101, obtaining a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of the d axis of the stator of the permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of the q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of the rotor of the permanent magnet synchronous motor at the current moment;
specifically, the voltages of the q axis and the d axis of the stator of the permanent magnet synchronous motor at the current moment are beneficial to predicting the subsequent voltage, so that the voltages of the q axis and the d axis of the stator at the current moment need to be obtained, and the angular speed of the rotor of the permanent magnet synchronous motor at the current moment is similar, so that the prediction of the subsequent voltage is also beneficial to;
，
of the formulaFor a matrix comprising the derivative of the d-axis current and the derivative of the q-axis current of the stator, x is the d-axis comprising the statorA matrix of current and q-axis current, u is a matrix including d-axis voltage and q-axis voltage of the stator, a represents a system matrix, B represents a control matrix, and G is a constant term; / >And->D-axis and q-axis voltages of the stator, respectively; />And->D-axis and q-axis currents of the stator, respectively; />And->The resistance value of the stator and the angular velocity of the rotor, respectively; />The preset constant is the inherent characteristic of the permanent magnet synchronous motor, and if a certain motor is determined, the preset constant of the permanent magnet synchronous motor is determined;
for unknown disturbance current parameters related to state, +.>Unknown disturbance current parameter for q-axis, +.>The current parameter is perturbed for the d-axis position.
Step S102, a priori estimation method is adopted, a first predicted current vector is obtained according to the first stator voltage, the second stator voltage and the current rotor angular velocity, a neural network topology model is adopted, the first stator voltage, the second stator voltage and the current rotor angular velocity are processed, and a second predicted current vector is obtained, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
Specifically, the prior estimation method is as follows:
and constructing a first predicted current vector expression by using the first stator voltage, the second stator voltage and the current rotor angular velocity, wherein the first predicted current vector expression comprises a proportional relation between an inductance value of a q-axis of the stator and an inductance value of a d-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of a q-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of a d-axis of the stator, a proportional relation between a sampling period and an inductance value of a q-axis of the stator, and a proportional relation between the sampling period and an inductance value of a d-axis of the stator, and determining the first predicted current vector according to the first predicted current vector expression.
One specific example (without regard to unknown disturbances) is:
according to，
Determining the first predicted current vector, wherein,in order to obtain k+1 moment by adopting the prior estimation methodPredicted current of d-axis of the stator, < >>For the sampling period +.>For the resistance value of the stator, +.>For the d-axis inductance of the stator, < >>For the predicted current of the d-axis of the stator at the k-time using the prior estimation method,/- >For the inductance value of the q-axis of the stator, < >>For the angular velocity of the rotor at time k, < >>For the voltage of the d-axis of the stator at time k,/>To obtain the q-axis voltage of the stator at the k time by the prior estimation method,is a preset constant->For the predicted current of the q-axis of the stator at time k obtained by the prior estimation method,a predicted current of the q-axis of the stator at time k+1;
the current predicted by the current round is calculated by adopting the current predicted by the previous round, so that the accuracy of calculating the current predicted by the current round is improved.
In addition, regarding a neural network topology model, firstly, based on a permanent magnet synchronous motor with a determined model, a motor open-loop control test is carried out on a motor pair dragging platform to acquire training data. Under the condition of determining the direct current bus voltage, the first stator voltage, the second stator voltage and the current rotor angular speed of the control input quantity are discretized, wherein the discretized interval of the first stator voltage and the second stator voltage is 5V, and the voltage ranges of all the stator d-axis and q-axis are required to be obtained. The current rotor angular speed of the load motor is 20rpm at discrete intervals, and the rotating speed range of the motor to be tested is taken. A script language (Python or MATLAB) is compiled, at the upper computer end, different first stator voltages, second stator voltages and current rotor angular speeds are respectively input in an automatic mode, the script automatically outputs a d-axis predicted current and a q-axis predicted current, and the different inputs and outputs form the whole data set.
Data preprocessing: and collecting the obtained data, and carrying out data abnormal points and elimination and normalization processing on the data. The data normalization method adopts a minimum value and maximum value method, and the expression is as follows:
；
wherein,for normalizing the processed data, +.>Raw data in the data sequence (i.e. matrix comprising d-axis current and q-axis current of stator),>minimum in data sequence, +.>Maximum value in the data sequence.
As shown in fig. 2, the neural network in the neural network topology model includes a plurality of input layers, a plurality of hidden layers, and a plurality of output layers, wherein each of the input layers, each of the hidden layers, and each of the output layers are connected in a fully connected manner, three values, which are respectively a first stator voltage, a second stator voltage, and a current rotor angular velocity, need to be input in the input layers, and two values, which are respectively a second q-axis predicted current and a second d-axis predicted current, need to be output in the output layers.
Wherein, in the process of adopting a neural network topology model to process the first stator voltage, the second stator voltage and the current rotor angular velocity to obtain a second predicted current vector, the method further comprises:
Constructing a d-axis predictive performance expression and a q-axis predictive performance expression, wherein the d-axis predictive performance expression comprises a difference value between a current of a t-th predicted d-axis obtained by using the neural network topology model and a current of a t+1th predicted d-axis obtained by using the neural network topology model, and the q-axis predictive performance expression comprises a difference value between a current of a t-th predicted q-axis obtained by using the neural network topology model and a current of a t+1th predicted q-axis obtained by using the neural network topology model;
a predictive performance vector is determined, which is a vector including d-axis predictive performance and q-axis predictive performance.
The d-axis prediction performance and the q-axis prediction performance are ensured to be respectively obtained, so that the accuracy of obtaining the corresponding weight by utilizing the two prediction performances is improved.
In one particular embodiment of this invention,
according toDetermining a d-axis prediction performance, wherein the d-axis prediction performance is a performance of predicting a d-axis current of the stator by the neural network topology model, and lossd is the d-axis prediction performance,for the current of d-axis of the t-th prediction obtained by using the neural network topology model,/I>The current of d axis of the t+1st prediction obtained by adopting the neural network topology model;
According toDetermining a q-axis prediction performance, wherein the q-axis prediction performance is a performance of predicting a q-axis current of the stator by the neural network topology model, and lossq is the d-axis prediction performance,for the current of the q-axis of the t-th prediction obtained by using the above neural network topology model,/I>The current of the q-axis is predicted for the t+1st time obtained by adopting the neural network topology model;
and determining a prediction performance vector, wherein the prediction performance vector comprises d-axis prediction performance and q-axis prediction performance, and improving the accuracy of the prediction of the neural network topology model.
In addition, after determining the predicted performance vector, the method further includes:
according to，
Determining a connection weight between neurons of the input layer and the hidden layer and a connection weight between neurons of the hidden layer and the output layer when the t+1th prediction is performed by using the neural network topology model;
wherein, first, the number of hidden layers and the node number of each hidden layer, the hidden layer neuron threshold value and the output layer neuron threshold value are determined. Training parameters including learning rate, maximum iteration number and minimum tolerance error are set. Then, loss functions lossd and lossq are constructed to evaluate the prediction performance of the neural network, wherein a smaller loss indicates a better prediction of the neural network, and the activation functions of neurons in the hidden layer and the output layer of the neural network are both sigmoid (i.e. S-type functions commonly found in biology) type activation functions.
For the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is used for the t+1th prediction, < + >>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted to carry out the t+1th prediction, < + >>For the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t-th prediction, +.>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted to perform the t-th prediction, +.>For a predetermined coefficient, loss is the predictive performance vector, i is used to represent the ith neuron of the input layer, j is used to represent the jth neuron of the hidden layer, and k is used to represent the kth neuron of the output layer.
Therefore, the connection weight between the neurons of the hidden layer and the neurons of the output layer and the connection weight between the neurons of the input layer and the hidden layer are obtained, so that the parameters input by the input layer can be converted into the parameters output by the output layer according to the weight, and the accuracy of the prediction of the neural network topology model is improved.
Step S103, weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector comprises an optimized q-axis predicted current and an optimized d-axis predicted current;
specifically, the optimized predicted current vector is an optimized predicted current vector obtained by weighting the first predicted current vector and the second predicted current vector, and the objective of fusing a priori estimation method and a neural network topology prediction method is achieved by weighting the first predicted current vector and the second predicted current vector, so that the accuracy of the predicted current vector is improved, and as shown in fig. 3, fig. 3 shows a process of obtaining the optimized predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular velocity.
One of the specific examples is that,
1) Firstly, writing a mathematical model of a permanent magnet synchronous motor into a form of a state space expression, and predicting a value a priori(i.e., corresponding to the first predicted current vector of step a):
；
a first predicted current vector for step a-1, >A voltage of a d axis of the stator and a voltage of a q axis of the stator corresponding to a minimum value of the vector expression of the a-1 st step;
2) Calculating covariance matrix of a priori estimation error of step a：
；
A covariance matrix of a posterior estimation error in the step a-1, wherein W1 is a constant term;
A T a transposed matrix of A;
3) Calculating the system gain K of the step a a ：
；
H is a measurement matrix, i.e. a preset matrix, H T For the transposed matrix of H, W2 is also a constant term;
4) Calculating the optimal estimated value of the step aOptimized predicted current vector of step a:
；
z a a second predicted current vector which is an observed value, namely the step a;
5) Obtaining covariance matrix of a posterior estimation error of the step a：
；
I is an identity matrix.
Step S104, constructing a vector expression by using the optimized predicted current vector and the stator voltage vector, obtaining a minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage.
Determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value of the vector expression as optimal predicted voltage vectors;
In the method, the prior estimation method and the neural network topology prediction method are fused to obtain the optimized prediction current vector, and the vector expression is built to obtain the optimal prediction voltage vector, so that the accuracy of prediction is improved due to the combination of the prior estimation method and the neural network topology prediction method, and the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control is solved.
In particular according to，
The optimal predicted voltage vector is determined, wherein,for the optimal predicted voltage vector, +.>For predicting the above optimized predicted current vector of the future N-th step, a +.>Is->Transposed matrix of>To predict the difference between the above-mentioned optimized predicted current vector of step a in the future and the actual value of the corresponding current vector, and (2)>Is->Transposed matrix of>For the stator voltage vector, < >>Is->The transposed matrix P, R, Q of (a) is a preset weight matrix, and the actual value of the current vector is the current of the d axis of the stator and the current of the q axis of the stator fed back by the permanent magnet synchronous motor in real time.
The accuracy of the optimal predicted voltage vector is improved by considering the predicted optimal predicted current vector of the future Nth step, the difference value between the predicted optimal predicted current vector of the future a-th step and the actual value of the corresponding current vector, and the stator voltage vector.
The two models (namely the model corresponding to the prior estimation method and the neural network topology model) are fused to obtain the model, and then the fused model is used for carrying out model predictive control on the current loop of the permanent magnet synchronous motor, so that better predictive voltage vectors (d-axis and q-axis voltages) can be obtained, and the control precision is improved.
In order to enable those skilled in the art to more clearly understand the technical solutions of the present application, the implementation procedure of the method for predicting the voltage vector of the stator of the permanent magnet synchronous motor of the present application will be described in detail below with reference to specific embodiments.
The embodiment relates to a specific control method of a permanent magnet synchronous motor, as shown in fig. 4, comprising the following steps:
step S1: acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment;
step S2: obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, and processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
Step S3: weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector comprises an optimized q-axis predicted current and an optimized d-axis predicted current;
step S4: constructing a vector expression by using the optimized predicted current vector and a stator voltage vector, obtaining a minimum value of the vector expression, and determining a voltage of a d-axis of the stator and a voltage of a q-axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector including the first stator voltage and the second stator voltage;
step S5: and controlling the permanent magnet synchronous motor based on the optimal predicted voltage vector.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
The embodiment of the application also provides a device for predicting the voltage vector of the stator of the permanent magnet synchronous motor, and it is to be noted that the device for predicting the voltage vector of the stator of the permanent magnet synchronous motor in the embodiment of the application can be used for executing the method for predicting the voltage vector of the stator of the permanent magnet synchronous motor provided by the embodiment of the application. The device is used for realizing the above embodiments and preferred embodiments, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
The following describes a device for predicting a voltage vector of a stator of a permanent magnet synchronous motor provided in an embodiment of the present application.
Fig. 5 is a block diagram of a device for predicting a voltage vector of a stator of a permanent magnet synchronous motor according to an embodiment of the present application. As shown in fig. 5, the apparatus includes:
an obtaining unit 51, configured to obtain a first stator voltage, a second stator voltage, and a current rotor angular velocity, where the first stator voltage is a voltage of a d-axis of a stator of the permanent magnet synchronous motor at a current time, the second stator voltage is a voltage of a q-axis of the stator of the permanent magnet synchronous motor at the current time, and the current rotor angular velocity is an angular velocity of a rotor of the permanent magnet synchronous motor at the current time;
a first processing unit 52, configured to obtain a first predicted current vector according to the first stator voltage, the second stator voltage, and the current rotor angular velocity by using an a priori estimation method, and process the first stator voltage, the second stator voltage, and the current rotor angular velocity by using a neural network topology model to obtain a second predicted current vector, where the first predicted current vector is a vector including a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector including a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained using a plurality of sets of training data, and each set of training data in the plurality of sets of training data includes a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
A second processing unit 53, configured to perform weighting processing on the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, where the optimized predicted current vector is a vector including an optimized q-axis predicted current and an optimized d-axis predicted current;
and a third processing unit 54 configured to construct a vector expression using the optimized predicted current vector and a stator voltage vector, determine a minimum value of the vector expression, and determine a voltage of a d-axis of the stator and a voltage of a q-axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector including the first stator voltage and the second stator voltage.
In the device, the prior estimation method and the neural network topology prediction method are fused to obtain the optimized prediction current vector, and the vector expression is reconstructed to obtain the optimal prediction voltage vector, so that the accuracy of prediction is improved due to the combination of the prior estimation method and the neural network topology prediction method, and the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control is solved.
In one embodiment of the present application, the first processing unit includes a first processing module and a second processing module, where the first processing module is configured to construct a first predicted current vector expression using the first stator voltage, the second stator voltage, and the current rotor angular velocity, where the first predicted current vector expression includes a proportional relationship between an inductance value of a q-axis of the stator and an inductance value of a d-axis of the stator, a proportional relationship between a resistance value of the stator and an inductance value of a q-axis of the stator, a proportional relationship between a resistance value of the stator and an inductance value of a d-axis of the stator, a proportional relationship between a sampling period and an inductance value of a q-axis of the stator, and a proportional relationship between the sampling period and an inductance value of a d-axis of the stator; the second processing module is used for determining the first predicted current vector according to the first predicted current vector expression.
In an embodiment of the present application, the neural network in the neural network topology model includes a plurality of input layers, a plurality of hidden layers, and a plurality of output layers, and each of the input layers, each of the hidden layers, and each of the output layers are connected in a fully connected manner.
In one embodiment of the present application, the first processing unit includes a third processing module and a fourth processing module, and in the process of processing the first stator voltage, the second stator voltage, and the current rotor angular velocity by using a neural network topology model, to obtain a second predicted current vector,
the third processing module is configured to construct a d-axis prediction performance expression and a q-axis prediction performance expression, where the d-axis prediction performance expression includes a difference between a current of a d-axis predicted at a t-th time obtained by using the neural network topology model and a current of a d-axis predicted at a t+1th time obtained by using the neural network topology model, and the q-axis prediction performance expression includes a difference between a current of a q-axis predicted at a t-th time obtained by using the neural network topology model and a current of a q-axis predicted at a t+1th time obtained by using the neural network topology model;
the fourth processing module is configured to determine a predicted performance vector that is a vector that includes a d-axis predicted performance and a q-axis predicted performance.
In one embodiment of the present application, the first processing unit includes a first determination module that, after determining the predictive performance vector,
the first determining module is used for according to，
Determining a connection weight between neurons of the input layer and the hidden layer and a connection weight between neurons of the hidden layer and the output layer when the t+1th prediction is performed by using the neural network topology model;
for the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is used for the t+1th prediction, < + >>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted to carry out the t+1th prediction, < + >>For the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t-th prediction, +.>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted to perform the t-th prediction, +.>For a predetermined coefficient, loss is the predictive performance vector, i is used to represent the ith neuron of the input layer, j is used to represent the jth neuron of the hidden layer, and k is used to represent the kth neuron of the output layer.
In one embodiment of the present application, the third processing module includes a second determining module;
the second determining module is used for according to，
The optimal predicted voltage vector is determined, wherein,for the optimal predicted voltage vector, +.>For predicting the above optimized predicted current vector of the future N-th step, a +.>Is->Transposed matrix of>To predict the difference between the above-mentioned optimized predicted current vector of step a in the future and the actual value of the corresponding current vector, and (2)>Is->Transposed matrix of>For the stator voltage vector, < >>Is->The transposed matrix P, R, Q of (a) is a preset weight matrix, and the actual value of the current vector is the current of the d axis of the stator and the current of the q axis of the stator fed back by the permanent magnet synchronous motor in real time.
The device for predicting the voltage vector of the stator of the permanent magnet synchronous motor comprises a processor and a memory, wherein the acquisition unit, the first processing unit, the second processing unit, the third processing unit and the like are all stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions. The modules are all located in the same processor; alternatively, the above modules may be located in different processors in any combination.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one kernel, and the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control is solved by adjusting kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
The embodiment of the invention provides a computer readable storage medium, which comprises a stored program, wherein the program is used for controlling equipment where the computer readable storage medium is positioned to execute a method for predicting a voltage vector of a stator of a permanent magnet synchronous motor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the program runs to execute the method for predicting the voltage vector of the stator of the permanent magnet synchronous motor.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program stored in the memory and capable of running on the processor, wherein the processor realizes at least the following steps when executing the program: acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment; obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, and processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator; weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector comprises an optimized q-axis predicted current and an optimized d-axis predicted current; and constructing a vector expression by using the optimized predicted current vector and the stator voltage vector, obtaining a minimum value of the vector expression, and determining a voltage of the d axis of the stator and a voltage of the q axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector including the first stator voltage and the second stator voltage. The device herein may be a server, PC, PAD, cell phone, etc.
The present application also provides a computer program product adapted to perform a program initialized with at least the following method steps when executed on a data processing device: acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment; obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, and processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator; weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector comprises an optimized q-axis predicted current and an optimized d-axis predicted current; and constructing a vector expression by using the optimized predicted current vector and the stator voltage vector, obtaining a minimum value of the vector expression, and determining a voltage of the d axis of the stator and a voltage of the q axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector including the first stator voltage and the second stator voltage.
The application also provides a control method of the permanent magnet synchronous motor, as shown in fig. 6, the method comprises the following steps:
step S601, obtaining a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of the d axis of the stator of the permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of the q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of the rotor of the permanent magnet synchronous motor at the current moment;
step S602, a priori estimation method is adopted, a first predicted current vector is obtained according to the first stator voltage, the second stator voltage and the current rotor angular velocity, a neural network topology model is adopted, the first stator voltage, the second stator voltage and the current rotor angular velocity are processed, and a second predicted current vector is obtained, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
Step S603, weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current;
step S604 of constructing a vector expression using the optimized predicted current vector and a stator voltage vector, obtaining a minimum value of the vector expression, and determining a voltage of a d-axis of the stator and a voltage of a q-axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector including the first stator voltage and the second stator voltage;
and step S605, performing coordinate inverse transformation processing on the optimal predicted voltage vector to obtain a target three-phase voltage, and controlling the permanent magnet synchronous motor based on the target three-phase voltage.
In the method, the prior estimation method and the neural network topology prediction method are fused to obtain the optimized prediction current vector, and the vector expression is built to obtain the optimal prediction voltage vector, so that the accuracy of prediction is improved due to the combination of the prior estimation method and the neural network topology prediction method, and the problem of lower motor model accuracy in the existing scheme permanent magnet synchronous motor model prediction control is solved.
The application also provides an electronic device comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs comprise a prediction method for executing the voltage vector of the stator of any one of the permanent magnet synchronous motors. The prior estimation method and the neural network topology prediction method are fused to obtain an optimized prediction current vector, and a vector expression is built to obtain an optimal prediction voltage vector, so that the accuracy of prediction is improved due to the combination of the prior estimation method and the neural network topology prediction method, and the problem of lower motor model accuracy in the existing scheme of permanent magnet synchronous motor model prediction control is solved.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may be implemented in program code executable by computing devices, so that they may be stored in a storage device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:
1) According to the method for predicting the voltage vector of the stator of the permanent magnet synchronous motor, the prior estimation method and the neural network topology prediction method are fused to obtain the optimal predicted current vector, and the vector expression is built to obtain the optimal predicted voltage vector.
2) According to the prediction device for the voltage vector of the stator of the permanent magnet synchronous motor, the prior estimation method and the neural network topology prediction method are fused to obtain the optimal prediction current vector, and the vector expression is built to obtain the optimal prediction voltage vector.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims (10)

1. A method of predicting a voltage vector of a stator of a permanent magnet synchronous motor, comprising:
acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment;
obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
Weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current;
and constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving a minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as the optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage.
2. The method of claim 1, wherein deriving a first predicted current vector from the first stator voltage, the second stator voltage, and the present rotor angular velocity comprises:
constructing a first predicted current vector expression by adopting the first stator voltage, the second stator voltage and the current rotor angular speed, wherein the first predicted current vector expression comprises a proportional relation between an inductance value of a q-axis of the stator and an inductance value of a d-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of the q-axis of the stator, a proportional relation between a resistance value of the stator and an inductance value of the d-axis of the stator, a proportional relation between a sampling period and an inductance value of the q-axis of the stator, and a proportional relation between the sampling period and an inductance value of the d-axis of the stator;
And determining the first predicted current vector according to the first predicted current vector expression.
3. The method of claim 1, wherein the neural network in the neural network topology model comprises a plurality of input layers, a plurality of hidden layers, and a plurality of output layers, each of the input layers, each of the hidden layers, and each of the output layers being connected in a fully connected manner.
4. A method according to claim 3, wherein in processing the first stator voltage, the second stator voltage and the current rotor angular velocity using a neural network topology model, the method further comprises:
constructing a d-axis prediction performance expression and a q-axis prediction performance expression, wherein the d-axis prediction performance expression comprises a difference value between a current of a t-th predicted d-axis obtained by adopting the neural network topology model and a current of a t+1th predicted d-axis obtained by adopting the neural network topology model, and the q-axis prediction performance expression comprises a difference value between a current of a t-th predicted q-axis obtained by adopting the neural network topology model and a current of a t+1th predicted q-axis obtained by adopting the neural network topology model;
A predictive performance vector is determined, the predictive performance vector being a vector comprising d-axis predictive performance and q-axis predictive performance.
5. The method of claim 4, wherein after determining the predictive performance vector, the method further comprises:
according to，
Determining a connection weight between neurons of the input layer and the hidden layer and a connection weight between neurons of the hidden layer and the output layer when the neural network topology model is adopted for t+1th prediction;
wherein,for the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t+1th prediction, the weight is +.>To adopt the neural network topology modelConnection weights between neurons of the hidden layer and the output layer when making the t+1st prediction,/for>For the connection weight between the neurons of the input layer and the hidden layer when the neural network topology model is adopted for the t-th prediction, the +_>For the connection weight between the neurons of the hidden layer and the output layer when the neural network topology model is adopted for the t-th prediction, the +_>For a preset coefficient, loss is the predictive performance vector, i is used for representing the ith neuron of the input layer, j is used for representing the jth neuron of the hidden layer, and k is used for representing the kth neuron of the output layer.
6. The method of claim 1, wherein constructing a vector expression using the optimized predicted current vector and a stator voltage vector, finding a minimum value of the vector expression, and determining a voltage of a d-axis of the stator and a voltage of a q-axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, comprises:
according to，
Determining the optimal predicted voltage vector, wherein,for the optimal predicted voltage vector, +.>For predicting said optimized predicted current vector of the future N-th step of the prediction,/th step>Is->Transposed matrix of>For predicting the difference between said optimized predicted current vector of future step a and the actual value of the corresponding current vector,/v>Is->Transposed matrix of>For the stator voltage vector,/a>Is->The transposed matrix P, R, Q of (a) is a preset weight matrix, and the actual value of the current vector is the current of the d axis of the stator and the current of the q axis of the stator fed back by the permanent magnet synchronous motor in real time.
7. A prediction apparatus for a voltage vector of a stator of a permanent magnet synchronous motor, comprising:
the device comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment;
The first processing unit is configured to obtain a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular velocity by using an a priori estimation method, and process the first stator voltage, the second stator voltage and the current rotor angular velocity by using a neural network topology model to obtain a second predicted current vector, where the first predicted current vector is a vector including a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector including a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using multiple sets of training data, and each set of training data in the multiple sets of training data includes a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
the second processing unit is used for carrying out weighting processing on the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current;
And the third processing unit is used for constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving the minimum value of the vector expression, and determining the voltage of the d axis of the stator and the voltage of the q axis of the stator corresponding to the minimum value as the optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage.
8. A control method of a permanent magnet synchronous motor, characterized by comprising:
acquiring a first stator voltage, a second stator voltage and a current rotor angular speed, wherein the first stator voltage is the voltage of a d axis of a stator of a permanent magnet synchronous motor at the current moment, the second stator voltage is the voltage of a q axis of the stator of the permanent magnet synchronous motor at the current moment, and the current rotor angular speed is the angular speed of a rotor of the permanent magnet synchronous motor at the current moment;
obtaining a first predicted current vector according to the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a priori estimation method, processing the first stator voltage, the second stator voltage and the current rotor angular speed by adopting a neural network topology model to obtain a second predicted current vector, wherein the first predicted current vector is a vector comprising a first q-axis predicted current and a first d-axis predicted current, the second predicted current vector is a vector comprising a second q-axis predicted current and a second d-axis predicted current, the neural network topology model is trained by using a plurality of sets of training data, and each set of training data in the plurality of sets of training data comprises a vector obtained in a historical time period: input data including a voltage of a d-axis of the stator, a voltage of a q-axis of the stator, and an angular velocity of the rotor, and output data corresponding to the input data, the output data including a predicted current of the q-axis of the stator and a predicted current of the d-axis of the stator;
Weighting the first predicted current vector and the second predicted current vector to obtain an optimized predicted current vector, wherein the optimized predicted current vector is a vector comprising an optimized q-axis predicted current and an optimized d-axis predicted current;
constructing a vector expression by adopting the optimized predicted current vector and a stator voltage vector, solving a minimum value of the vector expression, and determining a voltage of a d axis of the stator and a voltage of a q axis of the stator corresponding to the minimum value as an optimized predicted voltage vector, wherein the stator voltage vector is a vector comprising the first stator voltage and the second stator voltage;
and carrying out coordinate inverse transformation on the optimal predicted voltage vector to obtain a target three-phase voltage, and controlling the permanent magnet synchronous motor based on the target three-phase voltage.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform a method of predicting a voltage vector of a stator of a permanent magnet synchronous motor according to any one of claims 1 to 6.
10. An electronic device, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising a method for performing the prediction of the voltage vector of the stator of the permanent magnet synchronous motor of any of claims 1 to 6.
CN202311570761.6A 2023-11-23 2023-11-23 Method and device for predicting voltage vector of stator of permanent magnet synchronous motor Active CN117277888B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311570761.6A CN117277888B (en) 2023-11-23 2023-11-23 Method and device for predicting voltage vector of stator of permanent magnet synchronous motor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311570761.6A CN117277888B (en) 2023-11-23 2023-11-23 Method and device for predicting voltage vector of stator of permanent magnet synchronous motor

Publications (2)

Publication Number Publication Date
CN117277888A true CN117277888A (en) 2023-12-22
CN117277888B CN117277888B (en) 2024-03-19

Family

ID=89203129

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311570761.6A Active CN117277888B (en) 2023-11-23 2023-11-23 Method and device for predicting voltage vector of stator of permanent magnet synchronous motor

Country Status (1)

Country Link
CN (1) CN117277888B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010166714A (en) * 2009-01-16 2010-07-29 Denso Corp Controller and control system for rotating machine
JP2013038946A (en) * 2011-08-09 2013-02-21 Denso Corp Control unit for rotary machine
US20150039545A1 (en) * 2013-08-05 2015-02-05 City University of London Systems, methods and devices for vector control of permanent magnet synchronous machines using artificial neural networks
CN108880394A (en) * 2018-06-04 2018-11-23 江苏大学 A kind of wavelet neural network position-less sensor of switched reluctance motor forecast Control Algorithm
CN110535396A (en) * 2019-08-30 2019-12-03 长安大学 Surface permanent magnetic Synchronous Machine Models forecast Control Algorithm based on BP neural network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010166714A (en) * 2009-01-16 2010-07-29 Denso Corp Controller and control system for rotating machine
JP2013038946A (en) * 2011-08-09 2013-02-21 Denso Corp Control unit for rotary machine
US20150039545A1 (en) * 2013-08-05 2015-02-05 City University of London Systems, methods and devices for vector control of permanent magnet synchronous machines using artificial neural networks
CN108880394A (en) * 2018-06-04 2018-11-23 江苏大学 A kind of wavelet neural network position-less sensor of switched reluctance motor forecast Control Algorithm
CN110535396A (en) * 2019-08-30 2019-12-03 长安大学 Surface permanent magnetic Synchronous Machine Models forecast Control Algorithm based on BP neural network

Also Published As

Publication number Publication date
CN117277888B (en) 2024-03-19

Similar Documents

Publication Publication Date Title
KR102170105B1 (en) Method and apparatus for generating neural network structure, electronic device, storage medium
CN111638958B (en) Cloud host load processing method and device, control equipment and storage medium
CN109284956B (en) Task duration determining method and device and electronic equipment
CN109405843B (en) Path planning method and device and mobile device
Witczak et al. Unknown input observers and filters
CN112230146A (en) Method, system and equipment for predicting battery charging remaining time
CN117277888B (en) Method and device for predicting voltage vector of stator of permanent magnet synchronous motor
Yin et al. A variable‐structure gradient RBF network with its application to predictive ship motion control
Van Gorp et al. Learning neural networks with noisy inputs using the errors-in-variables approach
CN114358274A (en) Method and apparatus for training neural network for image recognition
CN112308337A (en) Prediction method, prediction device and processor for probabilistic short-term load of power system
CN117074951A (en) Lithium battery state of charge estimation method, device, equipment and readable storage medium
CN112884160B (en) Meta learning method and related device
CN111738407B (en) Clock error prediction method, device, medium and terminal based on deep learning
CN109436980B (en) Method and system for detecting state of elevator component
Kukurowski et al. Fault-tolerant tracking control for Takagi–Sugeno fuzzy systems under actuator and sensor faults
CN113657844B (en) Task processing flow determining method and device
CN113740064B (en) Rolling bearing fault type diagnosis method, device, equipment and readable storage medium
CN112861951B (en) Image neural network parameter determining method and electronic equipment
EP4124887A1 (en) Methods and systems for predicting trajectory data of an object and methods and systems for training a machine learning method for predicting trajectory data of an object
Zhang et al. LPV system common state basis estimation from independent local LTI models
CN117370728A (en) Method and device for determining water level of underground water in mining area and electronic equipment
Alidoust AGI brain II: the upgraded version with increased versatility index
CN116418260A (en) Torque estimation method, device, equipment and storage medium of permanent magnet synchronous motor
CN117674656A (en) Permanent magnet synchronous motor model prediction control parameter optimizing method based on neural network

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