CN113726253B - Method for improving efficiency of permanent magnet synchronous motor for electric automobile - Google Patents
Method for improving efficiency of permanent magnet synchronous motor for electric automobile Download PDFInfo
- Publication number
- CN113726253B CN113726253B CN202111029633.1A CN202111029633A CN113726253B CN 113726253 B CN113726253 B CN 113726253B CN 202111029633 A CN202111029633 A CN 202111029633A CN 113726253 B CN113726253 B CN 113726253B
- Authority
- CN
- China
- Prior art keywords
- current
- permanent magnet
- magnet synchronous
- synchronous motor
- efficiency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001360 synchronised effect Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 15
- 230000033228 biological regulation Effects 0.000 claims abstract description 12
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 229910052742 iron Inorganic materials 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 9
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 8
- 229910052802 copper Inorganic materials 0.000 claims description 8
- 239000010949 copper Substances 0.000 claims description 8
- 230000004907 flux Effects 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 238000002759 z-score normalization Methods 0.000 claims description 2
- 238000013178 mathematical model Methods 0.000 abstract description 2
- 238000005457 optimization Methods 0.000 description 11
- 238000005259 measurement Methods 0.000 description 9
- 238000013135 deep learning Methods 0.000 description 6
- 238000011217 control strategy Methods 0.000 description 4
- 230000007423 decrease Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- H02P21/22—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
- 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
-
- 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
- H02P21/24—Vector control not involving the use of rotor position or rotor speed sensors
- H02P21/28—Stator flux based control
-
- 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
- H02P27/00—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage
- H02P27/04—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage
- H02P27/06—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters
- H02P27/08—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters with pulse width modulation
- H02P27/085—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters with pulse width modulation wherein the PWM mode is adapted on the running conditions of the motor, e.g. the switching frequency
-
- 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
- H02P27/00—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage
- H02P27/04—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage
- H02P27/06—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters
- H02P27/08—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters with pulse width modulation
- H02P27/12—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters with pulse width modulation pulsing by guiding the flux vector, current vector or voltage vector on a circle or a closed curve, e.g. for direct torque control
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Control Of Ac Motors In General (AREA)
Abstract
The invention discloses a method for improving the efficiency of a permanent magnet synchronous motor for an electric automobile, which improves the operation efficiency of a built-in permanent magnet synchronous motor for the electric automobile through coordination control of current, firstly, a mathematical model of the motor efficiency is established, the distribution rules of direct-axis current and quadrature-axis current when the motor efficiency is optimal in different operation areas are qualitatively analyzed, and a vector control system based on the current distribution model is designed; secondly, an AVL dynamometer rack system experimental platform is built, and a sample space is built through measured data; secondly, introducing a current regression model of the deep confidence network, and learning training network parameters in two stages by adopting 'pre-training-fine tuning parameters'; finally, embedding the regression model into a control system to realize the coordination control of the current. The method has accurate proportioning current and can improve the operation efficiency of the permanent magnet synchronous motor in the whole speed regulation range.
Description
Technical Field
The invention relates to the technical field of permanent magnet synchronous motors, in particular to a method for improving the efficiency of a permanent magnet synchronous motor for an electric automobile.
Background
Because the running condition of the electric automobile is complex, the electric automobile is started, stopped and accelerated and decelerated frequently, and the comfort of driving and riding of a driver is considered, the electric automobile has high requirements on a driving system of the electric automobile. The embedded permanent magnet synchronous motor has been widely used in the field of electric automobiles due to the advantages of high efficiency, high power density and the like. However, complicated operating conditions of electric vehicles can greatly reduce motor efficiency. Therefore, research on efficiency optimization of the driving motor of the electric automobile has important significance for saving energy and reducing environmental pollution. The existing method for improving efficiency mainly comprises the steps of designing a motor body and controlling strategies. The design of the motor body is to realize energy conservation by using new materials and improved technology. However, when determining the motor body, optimization of the control strategy is a cost-effective approach.
The existing motor efficiency optimization control strategy methods are quite many and mainly divided into two major types, namely an efficiency optimization method of a motor loss model and an input power minimum strategy. The control efficiency of the loss model control method is globally optimal and the control speed is high, but the method is too dependent on an accurate motor mathematical model and cannot adapt to variable electric automobile operation conditions. The minimum input power strategy is to adjust the dq axis current to minimize the input power of the power supply under the same working condition, the calculated amount of the algorithm is large, the optimization speed is low, and phenomena such as oscillation can occur.
The prior art related to the present invention is: an efficiency optimization algorithm based on a motor loss model; a hybrid-based fuzzy search efficiency optimization method; a minimum loss prediction current control strategy based on online calculation of iron loss; and (5) efficiency optimization control based on an online search method. The disadvantages of the prior art are: too dependent on accurate motor parameters, the influence factors are many, and the requirements of the complex operation conditions of the electric automobile can not be met; the algorithm has the advantages of large calculated amount, low optimization speed, complex control system, difficult parameter setting and easy torque jitter generation.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the method comprises the steps of providing a current coordination control to realize the efficiency optimization of the permanent magnet synchronous motor for the electric automobile under different working conditions, utilizing discontinuous discrete actual measurement data, realizing continuous accurate prediction of multiple inputs and multiple outputs under any working condition point, simplifying the structure of the system, reducing the operation quantity, providing accurate current input quantity for a PI regulator, matching the current accurately, and improving the operation efficiency of the permanent magnet synchronous motor in the whole speed regulation range.
The invention adopts the technical scheme that: the method for optimizing the efficiency of the permanent magnet synchronous motor for the electric automobile based on the deep belief network regression algorithm comprises the following steps of collecting measured data as a sample library, constructing a deep belief network regression model, embedding the model into a vector control system, and realizing the efficiency optimization of the motor under different working conditions through the coordinated control of currents:
step one: according to stator voltage equation, electromagnetic torque equation, motor input power, copper consumption power, iron consumption power and motor output power under the condition of steady-state operation of permanent magnet synchronous motor, motor efficiency and direct-axis current i are established sd And quadrature axis current i sq Is a mathematical relationship of (a);
step two: solving the direct-axis current i when the efficiency is optimal under the constant torque and constant power operation area of the permanent magnet synchronous motor according to the actual operation condition of the permanent magnet synchronous motor by utilizing the mathematical relation established in the first step sd And quadrature axis current i sq According to the solving of the direct axis current i under different operation conditions sd Current of intersecting axis i sq Deducing a current schematic diagram of the established analytical model;
step three: the vector control system is established according to the optimal current in the second step, the vector control system comprises a current distribution module, a PI regulation module, a coordinate transformation module and an SVPWM module, the rotating speed and the torque given by the permanent magnet synchronous motor are used as input values, and the current distribution module distributes current according to the input values to obtain reference direct-axis currentQuadrature axis current->Output reference current straight-axis current +.>Quadrature axis current->The control signals of the inverter are generated by PI regulation and coordinate transformation and then are input as a module for SVPWM generation, so that the corresponding control effect is achieved;
step four: selecting a current distribution analysis model as a control i sd and isq The control model of the proportion is used for measuring current discrete data and various system parameters affecting current distribution when the permanent magnet synchronous motor runs under all working conditions on an AVL experimental platform, and using the current discrete data and various system parameters as a sample library for constructing a deep confidence network current regression model;
step five: preprocessing measured data of the permanent magnet synchronous motor under the operation conditions of a constant torque area and a constant power area, wherein the preprocessing result is used as a feature set in a sample library for distributing current by a deep belief network regression algorithm; inputting the feature set, searching a mapping relation between an input feature x and an output feature y by using a RBM of a limited Boltzmann machine, learning and training network parameters in a deep confidence network regression model by adopting two stages of pre-training and fine-tuning parameters to obtain a trained deep confidence network current regression model, and finally obtaining an output result, namely a straight-axis currentAnd quadrature axis current->
Step six: and replacing the trained deep confidence network current regression model with the current distribution model in the vector control system in the third step, finally determining the control model for improving the efficiency of the permanent magnet synchronous motor, and verifying the effectiveness of the model through an AVL experimental platform.
In the first step: according to a stator voltage equation, an electromagnetic torque equation, motor input power, copper consumption power, iron consumption power and motor output power under the condition of steady-state operation of the permanent magnet synchronous motor, the mathematical relationship among motor efficiency, torque current and exciting current is established as follows:
in the third step: the current distribution vector control system calculates the reference current under different working conditions according to formulas according to different working conditions of the motorOutput reference current->The control signals of the inverter are generated by the PI adjusting module and the coordinate transformation module and then are input as a module for SVPWM generation.
The fifth specific implementation process is as follows:
(1) Performing normalization pretreatment on measured data of the permanent magnet synchronous motor under the operation conditions of a constant torque zone and a constant power zone by adopting a z-score normalization method;
wherein ,σ x the mean value and standard deviation of a certain characteristic x in the measured data are obtained; x' is a value subjected to normalization processing;
i when the efficiency of the permanent magnet synchronous motor under the constant torque and constant power operation area is optimal according to the second step sd and isq The expression of (2) selects the electromagnetic torque T in the constant torque region e The rotating speed n and the permanent magnet flux linkage psi f D-axis inductance L d Inductance L of q axis q Stator phase current I s And electromagnetic torque T in constant power region e The rotating speed n and the permanent magnet flux linkage psi f D-axis inductance L d Inductance L of q axis q Copper consumption power P cu Iron loss power P fe Input power P in Output power P out Maximum stator voltage U max Maximum stator current I max D-axis voltage u sd Voltage on q-axisu sq As an input feature set;
(2) Pre-training: training a limited Boltzmann machine RBM from bottom to top each time by taking the feature set as input, and at each layer, the parameters w ij Constructing according to the data obtained by the calculation of the previous layer, calculating the states of the units according to the formulas (1) and (2), and updating the weight by adopting a single-step contrast bifurcation algorithm;
wherein ,wij Is the connection weight between neurons i, j; a, a i 、x i The bias and state of the ith neuron in the visible layer, respectively; b j 、x j The bias and state of the j-th neuron in the hidden layer respectively; sigma (x) is a sigmiod activation function in the neural network;
(3) Fine tuning parameters: after the training is finished, the weight of the whole network is finely adjusted from top to bottom by using a wake-sleep algorithm until the set training times or errors meet the requirements, and finally an output result, namely the direct-axis current, is obtainedAnd quadrature axis current
Compared with the prior art, the invention has the advantages that:
(1) From the deep learning point of view, the vector control system of the current prediction model based on the deep learning belief network regression algorithm is modeled, so that the excitation current and the torque current of the permanent magnet synchronous motor are subjected to regression prediction, and the operation efficiency of the whole speed regulation range of the permanent magnet synchronous motor is improved through coordination control of the current.
(2) According to the method, discontinuous discrete actual measurement data modeling is utilized, continuous accurate prediction of multiple inputs and multiple outputs is realized at any working point, the structure of the system is simplified, the operand is reduced, an accurate current input amount is provided for a PI (proportion) regulator, the current is accurate, and the influence of parameter change in actual operation of a motor is effectively reduced.
Drawings
FIG. 1 is a schematic diagram of the current of a analytical model built by formula derivation;
FIG. 2 is a diagram of a vector control system based on a current distribution model;
FIG. 3 is a block diagram of a deep belief network current regression algorithm;
FIG. 4 is a graph comparing a measured current curve in a constant torque region with a current regression curve through a deep belief network regression model;
FIG. 5 is a graph comparing a current regression curve of a constant power region current actual measurement curve with a current regression curve of a deep belief network regression model;
FIG. 6 is a graph comparing motor efficiency of a vector control system of a constant torque zone conventional vector control with a embedded deep belief network current regression model;
FIG. 7 is a graph comparing motor efficiency of a vector control system of a constant power region conventional vector control and embedded deep belief network current regression model;
FIG. 8 is a map of the efficiency of the vector control system embedded in the deep belief network current regression model over the entire speed range of the motor;
fig. 9 is a flow chart of an implementation of the method of the present invention.
Detailed Description
The invention is further described with reference to the drawings and detailed description.
As shown in fig. 9, the specific implementation steps of the present invention are as follows:
step one: in the rotor magnetic field orientation vector control of the permanent magnet synchronous motor, a stator voltage equation (1) and an electromagnetic torque equation (2) of the motor are adopted in steady-state operation. The motor loss mainly comprises two parts, namely mechanical loss and electrical loss, wherein the mechanical loss is generally uncontrollable; the electrical losses include stator copper losses (3) and core lossesConsumption (4). The efficiency of the motor can be expressed by the ratio of the output power (5) of the motor to the input power of the motor, i.e., formula (6); as can be seen from the above, for a certain operating condition (a given torque T e And rotational speed omega r ) Optimal control of motor efficiency can be achieved by coordinated control of torque current and field current.
T e ＝1.5n p i q [i d (L d -L q )+ψ f ] (2)
P fe ＝C fe ω r [(L d i sd +ψ f ) 2 +(L q i sd ) 2 ] (4)
P out ＝ω r T e (5)
in the formula ：usd 、u sq Respectively dq axis components of the stator voltage; i.e sd 、i sq Respectively dq-axis components of the stator current; r is the resistance of the stator; psi phi type d 、ψ q Is the dq-axis component of the stator flux linkage; omega e Is the electrical angular velocity; l (L) d 、L q Respectively dq axis inductance components; psi phi type f Is a permanent magnet flux linkage; t (T) e For electromagnetic torque, n p Is the pole pair number; c (C) fe Is the iron loss coefficient.
Step two: and establishing a current distribution model of the motor in different operation areas. (constant torque zone) when the rotational speed is lower than the base rotational speed, the principle of selecting the maximum current ratio controls the stator current, determines the stator current vector (7), minimizes the copper consumption of the motor, and thereby improves the electricityThe machine's operating efficiency. According to the Lagrangian multiplier method, equation (2) is a constraint condition, the optimal current distribution of a constant torque operation area is solved, and the calculation result is as shown in equation (10); in actual operation, the motor usually adopts rated direct-axis current i by considering factors such as magnetic saturation and the like sdrated As a means of(constant power zone) the motor is subject to current and voltage limitations in the constant power zone, and the motor input efficiency P (i sd ,i sq ) Obtaining a minimum value; defining Lagrangian as equation (12), and applying equation (12) to i sd and isq And solving the partial derivative and making the partial derivative zero, substituting the partial derivative equation into the formula (2), and obtaining an approximate result as shown in the formula (14).
From the above solution, i under different working conditions sd 、i sq As shown in FIG. 1, when ω r <ω rbase Handle i sdrated By taking into the equation (7), the point coordinates can be obtained, and the current vector curve is o-p when the efficiency obtained by the equation (8) is optimal 1 . When omega r ≥ω rbase A-B and A '-B' are two cases when the torques are different, and the upper part of the curve is a selectable current scheme in a constraint range. If the minimum point of formula (12) is within the constraint range, namely A '-B' and o-p 1 The intersection point of the two extreme points meets the constraint of voltage and current, and the extreme point is the optimal solution; and if the minimum value point exceeds the constraint range, the abscissa of the point B on the voltage ring is nearest to the minimum value point, namely the point B is the optimal solution.
in the formula ：Is Is the stator current; i max Is the maximum stator current; u (U) max Is the maximum stator voltage; u (u) sd 、u sq Respectively dq axis components of the stator voltage; i.e sd 、i sq Respectively dq-axis components of the stator current; r is the resistance of the stator; psi phi type d 、ψ q Is the dq-axis component of the stator flux linkage; omega r Is the mechanical angular velocity; l (L) d 、L q Respectively dq axis inductance components; psi phi type f Is a permanent magnet flux linkage; t (T) e For electromagnetic torque, n p Is polar logarithmic.
wherein ,
step three: analyzing and influencing current distribution general rules according to different operation conditionsConstructing a current distribution vector control system shown in fig. 2, wherein the control effect of the control system mainly depends on a current distribution module in a controller, the input values such as rotating speed, torque and the like are set according to the motor requirement, and a current distribution model performs current distribution according to the input values to obtain a reference command currentOutput reference current command->The control signal of the inverter is generated by PI regulation and coordinate transformation and then is input as a module for SVPWM generation, so that the corresponding control effect is achieved, and the control of the high-performance motor is realized.
Step four: an experimental platform based on an AVL dynamometer rack system is built, and efficiency points under different working conditions are tested. The system shown in fig. 2 is used as a control system of the experimental platform permanent magnet synchronous motor, and actual measurement data are obtained. In the experiment, the speed regulation range is 200r/min to 3000r/min, the rotating speed is set to be a fixed value through a control console, the torque is regulated on an upper computer, the actual measurement motor efficiency is obtained through AVL, and the regulation is stopped when the power reaches the upper limit. And measuring several groups of parameters of motor efficiency, voltage, quadrature axis current, direct axis current and the like corresponding to different torques at the same rotating speed. And changing the rotating speed set value until the speed regulation full range is covered, and finally, taking the measured data as a measured data sample for constructing a current distribution model based on a deep learning confidence network regression algorithm.
Step five: the deep learning confidence network current regression model is schematically shown in FIG. 3, and the output layer outputsAnd (3) the regression prediction result shows that each limited Boltzmann machine (RBM) layer completes feature extraction, and the invention can realize a multi-input multi-output operation mode. The optimal number of hidden layers, the optimal number of neurons in each layer and the initial learning rate of each RBM layer are required to be continuously debugged according to training experience. The invention adopts the two-stage learning training of' pre-training-fine tuning parametersThe method comprises the steps of initializing network parameters, namely initializing the whole model parameters layer by utilizing a greedy unsupervised learning algorithm in a first stage; the second stage utilizes wake-sleep algorithm to supervise fine-tune the relevant parameters of the network space from top to bottom.
The selection of the input feature set is one of important factors influencing the quality of the deep confidence network prediction result. To ensure i sd 、i sq The characteristic value should be selected by considering the different current analysis relations of the motor in the constant torque area and the constant power area. In the constant torque region, (T) is selected according to the formulas (7), (8) and (10) e 、n、ψ f 、L d 、L q 、I s ) 5-dimensional eigenvalues; after multiple attempts, the DBN model is set to be 5 layers, the number of neurons in each layer is set to be 108, 80, 30 and 201, and the corresponding learning rates of each RBM layer are respectively 0.02, 0.1, 0.02, 0.01 and 0.01. In the constant power region, according to the formulas (11), (13) and (14), the value (T) is selected e 、n、ψ f 、L d 、L q 、P cu 、P fe 、P in 、P out 、U max 、U max 、u sd 、u sq ) 12-dimensional eigenvalues; after multiple adjustments, the DBN model is set to 7 layers, the number of neurons in each layer is set to 202, 100, 40 and 201, and the corresponding learning rates of each RBM layer are respectively 0.03, 0.1, 0.02, 0.1, 0.03 and 0.01. And verifying the accuracy of the regression modeling current prediction of the deep belief network by utilizing the actually measured current distribution discrete data in the constant torque area and the constant power area respectively. The invention adopts the following two evaluation standards, namely the average absolute value error shown in the formula (15) and the root mean square error shown in the formula (16), wherein the smaller the numerical value of the two evaluation standards is, the more accurate the predicted value is.
in the formula ：yp To predict the result, y t For the actual measurement result, N is the number of predicted samples.
Step six: the constructed deep confidence network current regression model is embedded into a vector control system, and the operation result pairs are shown in fig. 4, 5, 6 and 7. FIG. 4 is a graph showing the comparison of predicted and measured d-q axis currents for a constant torque region; fig. 4 (a) and fig. 4 (b) are current regression curves at different torques given a rotation speed of 600 rpm; fig. 4 (c) and fig. 4 (d) are current prediction curves at different rotational speeds for a given torque of 500Nm, illustrating the high accuracy of current prediction according to the present invention. The analysis according to the formula (15) shows that the overall average error of the measured value and the predicted value is lower than 2%, and the error is within the range of the actual engineering requirement. FIG. 5 is a graph showing the comparison between the predicted current value and the measured current value of the d-q axis current in the constant power range. FIGS. 5 (a) and 5 (b) are current regression curves at different torques at a given rotational speed of 1400 rpm; fig. 5 (c) and fig. 5 (d) are current prediction curves at different rotational speeds for a given torque of 300 Nm. According to the analysis of the formula (15), the overall average error of the measured value and the predicted value is lower than 1.5%, and the actual engineering error requirement is met. In summary, the DBN predicted current curve with deep learning capability has a better fit to the measured current curve. The errors of the predicted value and the measured value of the current distribution under different working conditions are less than 2%, so that the requirements of actual engineering errors are met.
In order to verify the effectiveness of the current control strategy provided by the invention, the motor is experimentally verified in an AVL dynamometer. Fig. 6 and fig. 7 are efficiency comparison graphs of the DBN current prediction method and the traditional vector control method when the motor operation reaches a steady state by selecting a mechanical rotation speed range of 200r/min to 3000r/min when the torque is fixed under the conditions of constant torque and constant power, which illustrate that the influence of parameter variation in the actual operation of the motor is effectively reduced by processing and analyzing actual measurement data under different operation conditions by the deep confidence network current regression model. As can be seen from fig. 6, the efficiency improvement effect is better when the motor operation speed is lower, and the efficiency is improved by about 5% when the rotation speed is 400r/min, and as can be seen from fig. 7, in the constant power region, the motor operation efficiency of the conventional VC control decreases with the increase of the rotation speed, and the motor operation phase rate using the DBN current prediction method relatively decreases slowly. At a rotational speed of 2700r/min, the efficiency is improved by about 3%. Fig. 8 is a map diagram of efficiency of a motor under full working conditions by adopting a DBN current prediction control method, which illustrates that the invention adopts discrete actual measurement data modeling, so that a deep learning confidence network regression model solves the problem of discrete actual measurement data discontinuity, realizes optimal current matching at any working point, and can improve the running efficiency of the permanent magnet synchronous motor in the whole speed regulation range, and as can be seen from fig. 8, the middle block represents an efficient area in which an electric automobile is operated when running, and the area of the efficient area is more than 80% to meet the requirements of actual engineering.
Claims (4)
1. The method for improving the efficiency of the permanent magnet synchronous motor for the electric automobile is characterized by comprising the following steps of: the method comprises the following steps:
step one: according to stator voltage equation, electromagnetic torque equation, motor input power, copper consumption power, iron consumption power and motor output power under the condition of steady-state operation of permanent magnet synchronous motor, motor efficiency and direct-axis current i are established sd And quadrature axis current i sq Is a mathematical relationship of (a);
step two: solving the direct-axis current i when the efficiency is optimal under the constant torque and constant power operation area of the permanent magnet synchronous motor according to the actual operation condition of the permanent magnet synchronous motor by utilizing the mathematical relation established in the first step sd And quadrature axis current i sq According to the solving of the direct axis current i under different operation conditions sd Current of intersecting axis i sq Deducing a current schematic diagram of the established analytical model;
step three: the optimal straight axis current i according to step two sd And quadrature axis current i sq Establishing a vector control system, wherein the vector control system comprises a current distribution module, a PI regulation module, a coordinate transformation module and an SVPWM module, the given rotating speed and torque of the permanent magnet synchronous motor are used as input values, and the current distribution module performs current distribution according to the input values to obtain reference direct-axis currentQuadrature axis current->Output reference current straight-axis current +.>Quadrature axis current->The control signals of the inverter are generated by PI regulation and coordinate transformation and then are input as a module for SVPWM generation, so that the corresponding control effect is achieved;
step four: selecting a current distribution analysis model as a control i sd and isq The control model of the proportion is used for measuring current discrete data and various system parameters affecting current distribution when the permanent magnet synchronous motor runs under all working conditions on an AVL experimental platform, and using the current discrete data and various system parameters as a sample library for constructing a deep confidence network current regression model;
step five: preprocessing measured data of the permanent magnet synchronous motor under the operation conditions of a constant torque area and a constant power area, wherein the preprocessing result is used as a feature set in a sample library for distributing current by a deep belief network regression algorithm; inputting the feature set, searching a mapping relation between an input feature x and an output feature y by using a RBM of a limited Boltzmann machine, learning and training network parameters in a deep confidence network regression model by adopting two stages of pre-training and fine-tuning parameters to obtain a trained deep confidence network current regression model, and finally obtaining an output result, namely a straight-axis currentAnd quadrature axis current->I when the efficiency of the permanent magnet synchronous motor under the constant torque and constant power operation area is optimal according to the second step sd and isq Expression of (2)Selecting electromagnetic torque T in constant torque region e The rotating speed n and the permanent magnet flux linkage psi f D-axis inductance L d Inductance L of q axis q Stator phase current i s And electromagnetic torque T in constant power region e The rotating speed n and the permanent magnet flux linkage psi f D-axis inductance L d Inductance L of q axis q Copper consumption power P cu Iron loss power P fe Input power P in Output power P out Maximum stator voltage U max Maximum stator current I max D-axis voltage u sd Q-axis voltage u sq As an input feature set;
step six: and replacing the trained deep confidence network current regression model with the current distribution model in the vector control system in the third step, finally determining the control model for improving the efficiency of the permanent magnet synchronous motor, and verifying the effectiveness of the model through an AVL experimental platform.
2. The method for improving the efficiency of the permanent magnet synchronous motor for the electric automobile according to claim 1, wherein the method comprises the following steps: in the first step: according to a stator voltage equation, an electromagnetic torque equation, motor input power, copper consumption power, iron consumption power and motor output power under the condition of steady-state operation of the permanent magnet synchronous motor, the mathematical relationship among motor efficiency, torque current and exciting current is established as follows:
in the formula ：Rs Is the resistance of the stator; omega r Is the mechanical angular velocity; c (C) fe Is the iron loss coefficient.
3. The method for improving the efficiency of the permanent magnet synchronous motor for the electric automobile according to claim 1, wherein the method comprises the following steps: in the third step: the current distribution vector control system calculates the reference current under different working conditions according to formulas according to different working conditions of the motorOutput reference current->The control signals of the inverter are generated by the PI adjusting module and the coordinate transformation module and then are input as a module for SVPWM generation.
4. The method for improving the efficiency of the permanent magnet synchronous motor for the electric automobile according to claim 1, wherein the method comprises the following steps: the fifth specific implementation process is as follows:
(1) Performing normalization pretreatment on measured data of the permanent magnet synchronous motor under the operation conditions of a constant torque zone and a constant power zone by adopting a z-score normalization method;
wherein ,σ x the mean value and standard deviation of a certain characteristic x in the measured data are obtained; x' is a value subjected to normalization processing;
(2) Pre-training: training a limited Boltzmann machine RBM from bottom to top each time by taking the feature set as input, and at each layer, the parameters w ij Constructing according to the data obtained by the calculation of the previous layer, calculating the states of the units according to the formulas (1) and (2), and updating the weight by adopting a single-step contrast bifurcation algorithm;
wherein ,wij Is the connection weight between neurons i, j; a, a j 、x i The bias and state of the ith neuron in the visible layer, respectively; b j 、x j The bias and state of the j-th neuron in the hidden layer respectively; sigma (x) is a sigmiod activation function in the neural network;
(3) Fine tuning parameters: after the training is finished, the weight of the whole network is finely adjusted from top to bottom by using a wake-sleep algorithm until the set training times or errors meet the requirements, and finally an output result, namely the direct-axis current, is obtainedAnd quadrature axis current->
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111029633.1A CN113726253B (en) | 2021-09-03 | 2021-09-03 | Method for improving efficiency of permanent magnet synchronous motor for electric automobile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111029633.1A CN113726253B (en) | 2021-09-03 | 2021-09-03 | Method for improving efficiency of permanent magnet synchronous motor for electric automobile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113726253A CN113726253A (en) | 2021-11-30 |
| CN113726253B true CN113726253B (en) | 2023-10-27 |
Family
ID=78681359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111029633.1A Active CN113726253B (en) | 2021-09-03 | 2021-09-03 | Method for improving efficiency of permanent magnet synchronous motor for electric automobile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113726253B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114611402B (en) * | 2022-03-18 | 2023-01-06 | 四川大学 | Direct-drive fan dynamic fitting method based on block interconnection model and data driving |
| CN115149859A (en) * | 2022-05-18 | 2022-10-04 | 湘潭大学 | Efficiency optimization control method and system for permanent magnet synchronous motor |
| CN115356624B (en) * | 2022-08-15 | 2023-07-18 | 小米汽车科技有限公司 | Motor iron loss determination method and device, vehicle, storage medium and chip |
| CN117040326B (en) * | 2023-08-04 | 2024-03-15 | 上海大学 | Cooperative optimization and driving control method and system for vehicular disc type permanent magnet motor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109981016A (en) * | 2019-03-25 | 2019-07-05 | 安徽大学 | A kind of optimal fast square output method of electric bus asynchronous machine based on random forest regression algorithm |
| CN112052796A (en) * | 2020-09-07 | 2020-12-08 | 电子科技大学 | Permanent magnet synchronous motor fault diagnosis method based on deep learning |
| CN112926728A (en) * | 2021-03-22 | 2021-06-08 | 南京理工大学 | Small sample turn-to-turn short circuit fault diagnosis method for permanent magnet synchronous motor |
| CN113009338A (en) * | 2021-02-26 | 2021-06-22 | 中国长江三峡集团有限公司 | Offshore wind power variable pitch motor stator turn-to-turn short circuit fault diagnosis method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109800537B (en) * | 2019-02-20 | 2022-11-18 | 大连理工大学 | Machine tool thermal error model reliability calculation method based on deep neural network and Monte Carlo method |
-
2021
- 2021-09-03 CN CN202111029633.1A patent/CN113726253B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109981016A (en) * | 2019-03-25 | 2019-07-05 | 安徽大学 | A kind of optimal fast square output method of electric bus asynchronous machine based on random forest regression algorithm |
| CN112052796A (en) * | 2020-09-07 | 2020-12-08 | 电子科技大学 | Permanent magnet synchronous motor fault diagnosis method based on deep learning |
| CN113009338A (en) * | 2021-02-26 | 2021-06-22 | 中国长江三峡集团有限公司 | Offshore wind power variable pitch motor stator turn-to-turn short circuit fault diagnosis method |
| CN112926728A (en) * | 2021-03-22 | 2021-06-08 | 南京理工大学 | Small sample turn-to-turn short circuit fault diagnosis method for permanent magnet synchronous motor |
Non-Patent Citations (2)
| Title |
|---|
| Learning features from vibration signals for induction motor fault diagnosis;Siyu Shao et al.;《2016 International Symposium on Flexible Automation (ISFA)》;第71-76页 * |
| 基于深度置信网络的双馈风机变换器开路故障诊断;夏候凯顺 等;《电力工程技术》;第40卷(第1期);第188-194页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113726253A (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113726253B (en) | Method for improving efficiency of permanent magnet synchronous motor for electric automobile | |
| CN107645257B (en) | Machine learning device, motor control device, and machine learning method | |
| CN106357184B (en) | The temperature-compensation method of automobile permanent magnet synchronous motor output torque based on neural network | |
| CN109981016B (en) | Random forest regression algorithm-based optimal speed and moment output method for asynchronous motor for electric bus | |
| CN106712627B (en) | A kind of acquisition methods, device and the electric car of permanent magnet synchronous motor key parameter | |
| CN112468035B (en) | Method for selecting full-speed-domain optimal-efficiency control magnetization state of adjustable-flux permanent magnet synchronous motor and online control method | |
| CN111211716A (en) | Efficiency-optimized PMSM current prediction control method and system | |
| CN113472262A (en) | MTPA control method for identifying d-q axis inductance parameters of permanent magnet synchronous motor by adopting fuzzy logic control | |
| CN106817058A (en) | The drive control method of electri forklift, device and electri forklift | |
| CN115632584A (en) | Loss optimization control method for embedded permanent magnet synchronous motor | |
| CN113268040A (en) | Optimization method for control parameters of servo feeding system of numerical control machine tool | |
| CN102750422B (en) | Design method of electric car driving system | |
| CN114915225A (en) | Permanent magnet synchronous motor parameter identification method based on optimized BP neural network | |
| CN110518847B (en) | Surface permanent magnet synchronous motor model prediction control method based on BP neural network | |
| Xie et al. | Optimal speed–torque control of asynchronous motor for electric cars in the field-weakening region based on voltage vector optimization | |
| CN112468034B (en) | Permanent magnet synchronous motor weak magnetic area efficiency optimal control current track searching method and online control method | |
| Xie et al. | Current distribution method of induction motor for electric vehicle in whole speed range based on Gaussian process | |
| CN110518860B (en) | Permanent magnet synchronous motor model prediction control method | |
| CN112468033B (en) | Permanent magnet synchronous motor maximum power control current track searching method and online control method | |
| CN110376884A (en) | A kind of building method of new-energy automobile driving motor Intelligent Dynamic anti-interference controller | |
| CN112468037B (en) | Permanent magnet synchronous motor MTPV control current track searching method and online control method | |
| CN112468036A (en) | Permanent magnet synchronous motor full-speed domain efficiency optimal control current track searching method and online control method | |
| CN110224649A (en) | A method of DTC PREDICTIVE CONTROL is used for based on support vector machines | |
| CN111055920B (en) | Construction method of multi-model corner controller of automobile EPS (electric power steering) system | |
| Xie et al. | Efficiency optimization control of permanent magnet synchronous motors for pure electric vehicles based on GBDT |
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 |