CN112886843B - Three-phase eight-switch model prediction control method and device for weight coefficient removal - Google Patents
Three-phase eight-switch model prediction control method and device for weight coefficient removal Download PDFInfo
- Publication number
- CN112886843B CN112886843B CN202011262144.6A CN202011262144A CN112886843B CN 112886843 B CN112886843 B CN 112886843B CN 202011262144 A CN202011262144 A CN 202011262144A CN 112886843 B CN112886843 B CN 112886843B
- Authority
- CN
- China
- Prior art keywords
- voltage
- time
- cost function
- vector
- capacitor
- 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
- 238000000034 method Methods 0.000 title claims abstract description 69
- 239000013598 vector Substances 0.000 claims abstract description 129
- 239000003990 capacitor Substances 0.000 claims abstract description 49
- 230000009471 action Effects 0.000 claims abstract description 14
- 230000007935 neutral effect Effects 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 230000000694 effects Effects 0.000 claims description 5
- 238000006467 substitution reaction Methods 0.000 claims 2
- 230000006870 function Effects 0.000 description 46
- 238000010586 diagram Methods 0.000 description 24
- 230000004907 flux Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
- H02M7/42—Conversion of dc power input into ac power output without possibility of reversal
- H02M7/44—Conversion of dc power input into ac power output without possibility of reversal by static converters
- H02M7/48—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
- H02M7/483—Converters with outputs that each can have more than two voltages levels
- H02M7/487—Neutral point clamped inverters
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M1/00—Details of apparatus for conversion
- H02M1/32—Means for protecting converters other than automatic disconnection
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
- H02M7/42—Conversion of dc power input into ac power output without possibility of reversal
- H02M7/44—Conversion of dc power input into ac power output without possibility of reversal by static converters
- H02M7/48—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
- H02M7/53—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal
- H02M7/537—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters
- H02M7/5387—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters in a bridge configuration
- H02M7/53871—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters in a bridge configuration with automatic control of output voltage or current
- H02M7/53875—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters in a bridge configuration with automatic control of output voltage or current with analogue control of three-phase output
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Inverter Devices (AREA)
Abstract
One or more embodiments of the present disclosure provide a method and an apparatus for predicting and controlling a three-phase eight-switch model with a weight coefficient removed, where the method includes: determining a voltage vector, and predicting voltage offset of voltage generated by the action of the voltage vector at the moment k+1 by combining a first voltage and a second voltage formed by the load current flowing through the first capacitor and the second capacitor; predicting a current value at the moment k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter; the cost function for generating balance action on the balance neutral point potential is subjected to weight removal coefficient processing, and meanwhile, the cost function is used for compensating delay action caused by calculation time of a processor in actual implementation, so that an optimal vector is output at the time k+1. The model of the invention does not need to adjust the weight coefficient when predicting the current control.
Description
Technical Field
One or more embodiments of the present disclosure relate to the field of model predictive control, and in particular, to a method and apparatus for model predictive control of three-phase eight switches with weight coefficients.
Background
Model predictive control is attracting attention from more and more domestic and foreign electronic people along with the development of microelectronic technology, and the three-phase eight-switch inverter is widely applied due to a simpler fault-tolerant topological structure and lower cost.
Because the fault phases in the three phases are directly linked with the middle point of the bus, the middle point potential is offset, so that the output voltage and the output current are unbalanced, an effective fault-tolerant control method is needed to be adopted for ensuring the output performance and stable operation after the system is in fault, and the existing model predictive control is a fault-tolerant control method which is more commonly adopted at present, however, the method has the following problems: at least one weighting factor needs to be introduced in the prediction process, and the adjustment of the weighting factor can only be realized by a trial-and-error method, so that the adjustment is quite time-consuming.
There is currently no method or apparatus available in the industry that can solve the above problems.
Disclosure of Invention
In view of this, an object of one or more embodiments of the present disclosure is to provide a method and an apparatus for predicting and controlling a three-phase eight-switch model with a weight coefficient removed, so as to solve the problem of time-consuming adjustment in the current predicting and controlling process of the three-phase eight-switch inverter model.
In view of the above object, one or more embodiments of the present disclosure provide a three-phase eight-switch model prediction control method for a weight coefficient removal, including:
Determining a voltage vector, and predicting voltage offset of voltage generated by the action of the voltage vector at the moment k+1 by combining a first voltage and a second voltage formed by the load current flowing through the first capacitor and the second capacitor;
predicting a current value at the moment k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter;
the cost function for generating balance action on the balance neutral point potential is subjected to weight removal coefficient processing, and meanwhile, the cost function is used for compensating delay action caused by calculation time of a processor in actual implementation, so that an optimal vector is output at the time k+1.
In combination with the above description, in another possible implementation manner of the embodiment of the present invention, the determining a voltage vector includes:
Determining alternative voltage vectors through a switching function and a voltage vector formula to obtain 1 zero vector and 6 small vectors;
and selecting an optimal vector from the 1 zero vectors and the 6 small vectors.
In combination with the above description, in another possible implementation manner of the embodiment of the present invention, the predicting the voltage offset of the voltage generated by the voltage vector action at the time k+1 includes:
Predicting a voltage offset of the voltage vector contribution generated voltage at time k+1 by the following voltage offset formula:
wherein k is time, U c is capacitor voltage, That is, the variation of the capacitor voltage is represented, U c1 is the first voltage of the first capacitor, U c2 is the second voltage of the second capacitor, and T s/C is the ratio of the sampling time to the DC side capacitor;
The predicted value of the current at time k+1 is calculated by the following current prediction formula:
Wherein, The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,The voltage vector applied at the moment k is represented, and T s/L is the ratio of the sampling time to the inductance;
through the formula, the voltage tolerance is calculated And at time k+1 current valueAnd (3) withThe predicted quantity is used to select an optimal output voltage vector.
In combination with the foregoing description, in another possible implementation manner of the embodiment of the present invention, the balancing the first voltage and the second voltage, and performing a de-weighting operation on a cost function required for balancing includes:
rewriting the cost function so as to include the final cost function And further obtaining an optimized cost function, wherein the optimized cost function is as follows:
in the method, in the process of the invention, And (3) withI.e.And k is the time of day.
In combination with the foregoing description, in another possible implementation manner of the embodiment of the present invention, the selecting an optimal vector from the voltage vectors to delay-compensate the optimal voltage vector obtained at the time k includes:
For stator current component at time k+2 、Neutral point voltage deviationMaking predictions, comprising:
the optimized cost function becomes:
Substituting the voltage vector into the above formula, and selecting the voltage vector with the minimum cost function, the prediction vector at the time k+2 is taken as the output at the time k+1.
In a second aspect, exemplary embodiments of the present invention further provide a three-phase eight-switch model prediction control device for removing weight coefficients, including:
the voltage prediction module is used for determining a voltage vector, and predicting voltage offset of voltage generated by the action of the voltage vector at the moment k+1 in combination with first voltage and second voltage formed by the fact that load current flows through the first capacitor and the second capacitor;
The current prediction module is used for predicting a current value at the moment k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter;
and the delay compensation module is used for carrying out weight removal coefficient processing on the cost function for generating the balance action on the balance midpoint potential, and compensating the delay action caused by the calculation time of the processor in practical implementation, so as to output the optimal vector at the time k+1.
The above device, the voltage prediction module is further configured to:
determining alternative voltage vectors through a switching function and a voltage vector formula to obtain 1 zero vector and 6 small vectors; and selecting an optimal vector from the 1 zero vectors and the 6 small vectors.
The above device, the voltage prediction module is further configured to:
Predicting a voltage offset of the voltage vector contribution generated voltage at time k+1 by the following voltage offset formula:
Wherein k is time, U c is capacitor voltage, U c1 is first voltage of a first capacitor, U c2 is second voltage of a second capacitor, and T s/C is ratio of sampling time to direct current side capacitor;
The predicted value of the current at time k+1 is calculated by the following current prediction formula:
Wherein, The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,The voltage vector applied at the moment k is represented, and T s/L is the ratio of the sampling time to the inductance;
through the formula, the voltage tolerance is calculated And at time k+1 current valueAnd (3) with。
The above apparatus, the delay compensation module is further configured to:
rewriting the cost function so as to include the final cost function And further obtaining an optimized cost function, wherein the optimized cost function is as follows:
in the method, in the process of the invention, The real part and the imaginary part of (a) are respectivelyAnd (3) withK is the time of day.
The above apparatus, the delay compensation module is further configured to:
For stator current component at time k+2 、Neutral point voltage deviationMaking predictions, comprising:
the optimized cost function becomes:
Substituting the voltage vector into the above formula, and selecting the voltage vector with the minimum cost function, the prediction vector at the time k+2 is taken as the output at the time k+1.
From the above, it can be seen that the method and apparatus for predicting and controlling a three-phase eight-switch model with a weight coefficient removed according to one or more embodiments of the present disclosure solve the problem of time consumption in the current three-phase eight-switch inverter model prediction and control process by removing the weight coefficient based on the idea of unifying the dimension of the cost function.
Drawings
For a clearer description of one or more embodiments of the present description or of the solutions of the prior art, the drawings that are necessary for the description of the embodiments or of the prior art will be briefly described, it being apparent that the drawings in the description below are only one or more embodiments of the present description, from which other drawings can be obtained, without inventive effort, for a person skilled in the art.
FIG. 1 is a schematic topology diagram of a three-phase eight-switch inverter according to one or more embodiments of the present disclosure;
FIG. 2 is a schematic flow diagram of a method of one or more embodiments of the present disclosure;
FIG. 3 is a schematic diagram of a three-phase eight-switch inverter voltage space vector diagram in accordance with one or more embodiments of the present disclosure;
FIG. 4 is a schematic diagram of a delay compensation principle according to one or more embodiments of the present disclosure;
FIG. 5 is a schematic diagram of a compensation framework of one or more embodiments of the present disclosure;
FIG. 6 is a graph of one or more embodiments of the present disclosure without regard to midpoint potential control Schematic of (2);
FIG. 7 is a graph of one or more embodiments of the present disclosure after consideration of midpoint potential control A schematic diagram;
FIG. 8 is a schematic diagram of a given current waveform for one or more embodiments of the present disclosure;
FIG. 9 is a schematic diagram of three-phase output current waveforms without consideration of midpoint potential control in one or more embodiments of the present disclosure;
FIG. 10 is a schematic diagram of three-phase current waveforms when midpoint potential control is considered in one or more embodiments of the present disclosure;
FIG. 11 is a diagram of a Uab voltage waveform when midpoint potential control is contemplated in one or more embodiments of the present disclosure;
FIG. 12 is a schematic diagram of three-phase current waveforms after load ramp up according to one or more embodiments of the present disclosure;
FIG. 13 is a diagram of a Uab waveform after load ramp up according to one or more embodiments of the present disclosure;
FIG. 14 is a graph showing a load after a sudden change in one or more embodiments of the present disclosure A waveform schematic;
FIG. 15 (a) is a schematic diagram of pre-load-mutation harmonic analysis according to one or more embodiments of the present disclosure;
FIG. 15 (b) is a schematic diagram of harmonic analysis after load mutation according to one or more embodiments of the present disclosure;
FIG. 16 is a schematic illustration of one or more embodiments of the present disclosure after perturbation A waveform schematic;
FIG. 17 is a schematic diagram of three-phase current waveforms after one or more embodiments of the present disclosure have been perturbed;
FIG. 18 is a schematic diagram of an apparatus according to one or more embodiments of the present disclosure;
Detailed Description
For the purposes of promoting an understanding of the principles and advantages of the disclosure, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same.
It is noted that unless otherwise defined, technical or scientific terms used in one or more embodiments of the present disclosure should be taken in a general sense as understood by one of ordinary skill in the art to which the present disclosure pertains. The use of the terms "first," "second," and the like in one or more embodiments of the present description does not denote any order, quantity, or importance, but rather the terms "first," "second," and the like are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
Multilevel converters are favored for their many advantages, with three-level Neutral-Point-Clamped (NPC) inverters being the most widely used. Compared with the two-level voltage waveform, the inverter has the advantages of better harmonic characteristic, smaller dv/dt, smaller switching loss and the like, and is always a hot spot for study of domestic and foreign students.
However, the three-level NPC inverter has more power switching devices than the conventional two-level inverter, which results in a greater probability of failure and reduced reliability. The three-phase Eight-switch inverter (weight-SWITCH THREE-PHASE INVERTER, ESTPI) is a fault-tolerant reconstruction structure after the single-phase bridge arm of the three-level NPC inverter fails. The structure is shown in figure 1 (A phase fault is taken as an example in the application of the patent), the fault-tolerant topological structure is not complex, the cost is low, and the application is wide. Because the phase A (fault phase) is directly linked with the midpoint of the bus, the midpoint potential offset is caused, so that the output voltage and the output current are unbalanced, and an effective fault-tolerant control method is needed to ensure the output performance and stable operation after the system is in fault.
In recent years, different fault-tolerant control methods have been proposed, one of which is model predictive control. The traditional method is based on model predictive control, torque, flux linkage and midpoint potential are added into a cost function to control the synchronous motor, but the method needs to introduce two weight coefficients, and the current adjustment of the weight coefficients can only be realized by a trial-and-error method, so that the method is quite time-consuming. There are some methods to propose model predictive current control and midpoint potential control for the synchronous reluctance motor 23, and the same relationship between current and midpoint potential in the cost function depends on the weight coefficient. There are also methods for selecting the optimal weight coefficient using an algorithm using an idea similar to the least square method, starting from experience of manual trial, but the method is also time-consuming and the algorithm is relatively complex.
In order to reduce the time cost of manual adjustment, in recent years, different scholars put forward the use of the weight coefficient of each method, norambuena and other scholars put forward a sequential method, namely, firstly screening out two basic voltage vectors with optimal torque performance, then comparing the two selected vectors, selecting the vector with optimal flux linkage in the two basic voltage vectors as a final output vector, and then the schrian and other scholars compare the method with the traditional model prediction method to obtain a conclusion that the performance of the method is similar to that of the traditional method, but the two researches lack theoretical analysis of a two-step method; and the students such as Rojas substitute the same basic voltage vector into the cost function of a single variable simultaneously and respectively, rank the cost function values of two seven groups from small to large, and comprehensively select the basic vector with the lowest rank as the output quantity by the two ranks, but the vector selected by the 'fair' method does not need to correspond to the vector required at a specific moment, and the error is larger.
In order to eliminate the weight coefficient in the cost function, the relation calculation is utilized to convert the flux linkage and the torque in the cost function into active torque and reactive torque; others such as: according to the relation between the electromagnetic torque and the flux linkage of the stator and the rotor, calculating is carried out based on the angular relation between the stator and the rotor, so that only flux linkage variables exist in the cost function; as another example, the control variable is converted into control of the output voltage vector. The method mentioned above has the idea of realizing the non-weight coefficient by unifying the dimensions of the variables to be controlled, but since the stator flux linkage amplitude is often determined to be invariable during calculation, errors can be caused in a low-speed interval of the motor, and the control performance is affected. The invention is based on the idea of unifying the dimension of the cost function.
The three-phase eight-switch model prediction control method for removing the weight coefficient provided by the exemplary embodiment of the invention is combined with a basic flow diagram shown in fig. 2, and the steps mainly comprise:
in step 210, a voltage vector is determined, and a voltage offset of a voltage generated by the voltage vector action at the time k+1 is predicted in combination with a first voltage and a second voltage formed by the load current flowing through the first capacitor and the second capacitor;
FIG. 3 is a voltage space vector diagram of a three-phase eight-switch inverter, with phase A directly connected to the midpoint of the capacitor, defining a switching function As shown in formula (1):
(1)
The voltage vector can be expressed as
(2)
Wherein,，AndRespectively stator phase voltages. According to (1) and (2), a three-phase eight-switch inverter space voltage vector diagram can be drawn, as shown in fig. 2, wherein the number of voltage vectors is reduced compared with a three-level inverter, including 1 zero vector (OOO), 6 small vectors (ONN, OON, OPO, OPP, OOP, ONO) and 2 medium vectors (ONP, OPN). And (3) comparing the different small vector schemes to obtain an optimal scheme using six small vectors. The control method adopted by the invention adopts 6 small vectors and 1 zero vector.
In step 220, predicting a current value at time k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter;
phase A is directly connected with midpoint O of the direct-current side capacitor, and load current flows through the capacitors C1 and C2 to cause midpoint voltage offset to cause 。
Midpoint current according to kirchhoff's law of currentCan be expressed as
(3)
Wherein,。
Thus, the capacitance current、And capacitor voltage、The relationship of (2) can be expressed as
(4)
Discretizing (4) to obtainAndIs that
(5)
Therefore, the voltage offset of the voltage vector action generation voltage at the time k+1 can be predicted by sorting (5).
(6)
The load of the three-phase eight-switch inverter is a resistive load, and for the predicted value of the current at the moment k+1, the following formula (7) can be used for calculating [1]:
(7)
Wherein, And (3) withI.e.Is used for the real and imaginary parts of (a),And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,Representing the voltage vector applied at time k. Then, the current value at time k+1 can be calculatedAnd (3) withAnd voltage-capacitance differenceThe calculation can be performed by the formula (6).
According to the traditional model predictive control method, the invention needs to control the current and balance the capacitance voltage difference, so the cost function should contain the variables and the cost functionThe following formula is shown:
(8)
Wherein, Is a weight coefficient. Currently, the value of the weight coefficient is still obtained through a trial-and-error method, which is one of limitations of the current model prediction control.
The formula (6) is rewritten as shown in the following formula (9), and the formula indicates that the difference in the amount of charge passing through the capacitor at time k+1 (the left side of the formula) is equal to the difference in the amounts of charge passing through both capacitors at time k minus the amount of charge flowing through the midpoint (the right side of the formula).
(9)
Substituting formula (9) rewrites the original cost function as shown in the following formula:
(10)
the first two terms of the molecule can be understood as the difference between the ideal amount of charge flowing and the amount of charge flowing at time k+1, while the third term, although containing However, in discrete systems this factor is an invariant, i.e. the third term of the molecule is understood to be simply proportional to the absolute value of the difference between the amounts of charge passing through the two capacitances at time k+1. Also equivalent to the fact that the three terms of the molecule have the same dimension, the denominator is an invariant (equivalent to only one coefficient), then the weight coefficient can be obtained at this timeEliminating to obtain an optimized cost function, wherein the cost function is shown as the following formula:
(11)
In step 230, the cost function for balancing the neutral potential is subjected to a de-weighting coefficient process, and is used to compensate for the delay caused by the calculation time of the processor in practical implementation, so as to output the optimal vector at time k+1.
In one implementation of the exemplary embodiment of the present invention, the best vector that minimizes the weight coefficient is selected among the seven candidate voltage vectors, and then delay compensation is performed to ensure that the best vector is output in actual execution.
In a discrete digital system, the calculation time cannot be ignored in a period, so that the optimal voltage vector calculated at the time k cannot act at the time k, but starts to act at the time k+1, as shown in fig. 4. Therefore, there is a one-beat delay in such a system, which would lead to deterioration of system performance, an increase in control error of the stator current and the midpoint voltage, and eventually even control failure if it is not time-lapse compensated.
To compensate for the adverse effects of one beat delay, it is necessary to predict the stator current component at time k+2、Neutral point voltage deviationBased on the acquired k-time correlation signal and the estimated current value and midpoint voltage value at k+1 time, the original cost function (12) can be expressed as
(12)
Then the voltage vector is substituted into the cost function formula (12), and the voltage vector minimizing the cost function is selected, and then the prediction vector at the time k+2 is taken as the output at the time k+1.
In the above formulas, k is time, U c is capacitor voltage, U c1 is first voltage of the first capacitor, U c2 is second voltage of the second capacitor, and T s/C is ratio of sampling time to direct current side capacitor; for stator current component at time k+2、Neutral point voltage deviation；The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,Representing the applied voltage vector at time k, T s/L is the ratio of sample time to inductance.
In the implementation manner of the exemplary embodiment of the present invention, the method further includes a verification step performed on the above method:
the main flow of the implementation of the model prediction control block diagram is shown in the following diagram in combination with fig. 5:
1. collecting current and midpoint potential signals;
2. substituting the known parameters and the acquired signals into a formula (6) and a formula (7) to predict the current and the midpoint voltage at the moment k+1, and then predicting the current value and the midpoint voltage value at the moment k+2 to realize delay compensation;
3. Substituting the expected value, the predicted value and the 7 voltage vectors into the weighting-coefficient-free cost function in the formula (12) respectively for calculation, and selecting an optimal vector for minimizing the cost function.
And (3) performing simulation experiments by utilizing Matlab/Simulink, and constructing a three-phase eight-switch inverter control system based on a model predictive control algorithm. The Simulink simulation parameters are as follows: sampling time; Resistance inductance load、; DC side voltage; DC side capacitor; The peak value and the frequency of a given sinusoidal current are respectively、。
The simulation results are shown in the following graph, and if the midpoint potential is not considered to be controlled, the midpoint potential is considered to be controlled as shown in FIG. 6As shown in FIG. 7, the voltage across the DC bus capacitor is unbalanced at this time, as can be seen from the simulation resultsGradually increasing from 0 to 450V, the shift of the midpoint potential causes the fluctuation and asymmetry of the three-phase current, and the output current situation is deteriorated, as can be seen from comparison of fig. 8 and 9, the a-phase current is gradually deteriorated along with the serious shift of the midpoint potential, and finally the control failure is caused. After fault-tolerant control of the model predictive current control method without weight coefficient provided by the invention is adopted, as shown in fig. 7, midpoint potential is introduced as a constraint condition of control,The periodic fluctuation within the amplitude of 20V indicates that the DC bus capacitance is effectively controlled and is in dynamic balance. Meanwhile, as shown in fig. 10, the three-phase output current is effectively improved, the a-phase current waveform is in a sine wave shape, the amplitude is close to the given current, the three-phase current is symmetrical, as shown in fig. 11, because the midpoint potential is in a dynamic balance,The amplitude of the voltage waveform also fluctuates within a small period range, and accords with the theoretical result.
When the system load suddenly changes and the three-phase load resistances are all increased to 15 omega by 10 omega in 0.8 seconds, as shown in fig. 12, the current briefly fluctuates at the moment, and the system quickly returns to a new steady state after the current fluctuates. Although the resistance value in the current prediction is a set constant value in the system as shown in formula (7), in fact, this is why the model prediction control method has a certain dependency on parameters, as can be seen from simulation results, after the parameters change, the current value deviates from the originally set resistance value, as shown in fig. 13, the waveform of the phase voltage Uab is normal, as shown in fig. 14, although the dynamic amplitude of the midpoint potential is increased but in the allowable range, and the midpoint potential is still in dynamic balance, as shown in fig. 15 (a), the current harmonic analysis before and after the load mutation is performed, as the system directly outputs the selected voltage vector, the current harmonic is reduced without adopting means such as pulse width modulation, the harmonic content is large, the difference between the harmonic content before and after the mutation is relatively small, and the system can still output stable near-sine wave current, so that the system can stably run. This illustrates that the system has good robustness.
In order to verify the stability of the system, when the system is disturbed instantaneously at the midpoint, namely, the midpoint potential suddenly changes, the voltage of one capacitor suddenly increases, and the midpoint potential waveform and the three-phase current waveform are as shown in fig. 15 (b) and fig. 16, it can be seen that when the system is disturbed at 0.3 seconds, the three-phase voltage suddenly changes, so that small fluctuation occurs in the three-phase current, but after the disturbance disappears, the current quickly returns to the normal state before the disturbance, and the waveform remains stable. The system has good stability, small fluctuation of output current under certain disturbance, and can be quickly regulated to restore to a stable state after the disturbance disappears, thereby having anti-interference capability.
Fig. 17 is a waveform of three-phase current after being disturbed.
Based on analysis of the topological structure of the three-phase eight-switch inverter, a model prediction current control method without a weight coefficient is provided for current and midpoint potential control, and the validity of the model prediction current control method is verified by utilizing a Matlab/Simulink building model, so that the conclusion is as follows:
Aiming at the resistive load, the proposed control method has good simulation result, and compared with the control of the midpoint potential, the system performance is stable, the midpoint potential balance fluctuates in a small range before the control of the midpoint potential is added, and the output voltage and current waveforms are stable and accord with the theoretical result.
Under the condition of abrupt load change, the system can enter a new steady state, the output voltage and current waveforms are stable, and the system robustness is good
The fluctuation of the output current is smaller when the neutral point potential of the system is disturbed, and the system can return to the original steady state after being disturbed, so that the system stability is good, and the control method provided by the invention is effective and good in performance.
Based on the same inventive concept, one or more embodiments of the present disclosure also provide a three-phase eight-switch model prediction control device for removing weight coefficients, corresponding to the method of any embodiment.
Referring to fig. 18, the three-phase eight-switch model predictive control device for removing weight coefficients includes:
The voltage prediction module 1810 is configured to determine a voltage vector, and predict a voltage offset of a voltage generated by applying the voltage vector at time k+1 in combination with a first voltage and a second voltage formed by flowing a load current through the first capacitor and the second capacitor;
The current prediction module 1820 is configured to combine the resistance and the inductance of the three-phase eight-switch inverter to predict a current value at time k+1;
The delay compensation module 1830 performs a de-weighting coefficient process on the cost function for balancing the neutral potential, and is used for compensating the delay caused by the calculation time of the processor in practical implementation, so as to output the optimal vector at time k+1.
In an exemplary embodiment of the present invention, the voltage prediction module 1810 is further configured to:
Determining a voltage vector through a switching function and a voltage vector formula to obtain 1 zero vector, 6 small vectors and 2 medium vectors; and selecting 1 zero vector and 6 small vectors as the voltage vectors.
In an exemplary embodiment of the present invention, the voltage prediction module 1810 is further configured to:
Predicting a voltage offset of the voltage vector contribution generated voltage at time k+1 by the following voltage offset formula:
Wherein k is time, U c is capacitor voltage, U c1 is first voltage of a first capacitor, U c2 is second voltage of a second capacitor, and T s/C is ratio of sampling time to direct current side capacitor;
The predicted value of the current at time k+1 is calculated by the following current prediction formula:
Wherein, The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,The voltage vector applied at the moment k is represented, and T s/L is the ratio of the sampling time to the inductance;
the voltage tolerance is calculated by the formula And at time k+1 current valueAnd (3) with。
In an exemplary embodiment of the present invention, the delay compensation module 1830 is further configured to:
rewriting the cost function so as to include the final cost function And further obtaining an optimized cost function, wherein the optimized cost function is as follows:
in the method, in the process of the invention, The real part and the imaginary part of (a) are respectivelyAnd (3) withK is the time of day.
In an exemplary embodiment of the present invention, the delay compensation module 1830 is further configured to:
For stator current component at time k+2 、Neutral point voltage deviationMaking predictions, comprising:
the optimized cost function becomes:
Substituting the voltage vector into the above formula, and selecting the voltage vector with the minimum cost function, the prediction vector at the time k+2 is taken as the output at the time k+1.
The device of the above embodiment is used for implementing the three-phase eight-switch model prediction control method of the corresponding weight removal coefficient in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which is not described herein.
Additionally, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown within the provided figures, in order to simplify the illustration and discussion, and so as not to obscure one or more embodiments of the present description. Furthermore, the apparatus may be shown in block diagram form in order to avoid obscuring the one or more embodiments of the present description, and also in view of the fact that specifics with respect to implementation of such block diagram apparatus are highly dependent upon the platform within which the one or more embodiments of the present description are to be implemented (i.e., such specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that one or more embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative in nature and not as restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of those embodiments will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.
The present disclosure is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Any omissions, modifications, equivalents, improvements, and the like, which are within the spirit and principles of the one or more embodiments of the disclosure, are therefore intended to be included within the scope of the disclosure.
Claims (8)
1. A three-phase eight-switch model prediction control method for a weight coefficient removal is characterized by comprising the following steps:
Determining a voltage vector, and predicting voltage offset of voltage generated by the action of the voltage vector at the moment k+1 by combining a first voltage and a second voltage formed by the load current flowing through the first capacitor and the second capacitor;
predicting a current value at the moment k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter;
the cost function for generating the balance effect on the balance neutral point potential is subjected to weight coefficient removal processing, and is used for compensating the delay effect caused by the calculation time of a processor in actual implementation, so that an optimal vector is output at the moment k+1;
further comprising balancing the first voltage and the second voltage, and performing a de-weighting operation on a cost function required for balancing, comprising:
rewriting the cost function so as to include the final cost function And further obtaining an optimized cost function, wherein the optimized cost function is as follows:
In the method, in the process of the invention, And (3) withI.e.K is the time of day;
rewriting the cost function so as to include the final cost function The method for further obtaining the optimized cost function is thatThe following formula is shown:
， Is a weight coefficient;
Substitution formula And then the original cost function is rewritten as shown in the following formula:
The first two terms of the molecule are understood to be the difference between the ideal amount of charge flowing and the amount of charge flowing at time k+1, while the third term contains However, in discrete systems, the factor is an invariant, i.e., the third term of the molecule is understood to be simply proportional to the absolute value of the difference between the amounts of charge passing through the two capacitances at time k+1; also equivalent to the fact that the three terms of the molecule have the same dimension and the denominator is an invariant, then the weight coefficient is calculatedAnd eliminating to obtain an optimized cost function.
2. The method of claim 1, wherein the determining a voltage vector comprises:
determining alternative voltage vectors comprising 1 zero vector and 6 small vectors through a switching function and a voltage vector formula;
and selecting an optimal vector from the 1 zero vector and the 6 small vectors.
3. The method of claim 1, wherein predicting a voltage offset of a voltage generated by a voltage vector contribution at time k+1 comprises:
Predicting a voltage offset of the voltage vector contribution generated voltage at time k+1 by the following voltage offset formula:
wherein k is time, U c is capacitor voltage, That is, the variation of the capacitor voltage is represented, U c1 is the first voltage of the first capacitor, U c2 is the second voltage of the second capacitor, and T s/C is the ratio of the sampling time to the DC side capacitor;
The predicted value of the current at time k+1 is calculated by the following current prediction formula:
Wherein, The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,The voltage vector applied at the moment k is represented, and T s/L is the ratio of the sampling time to the inductance;
through the formula, the voltage tolerance is calculated And at time k+1 current valueAnd (3) withThe predicted quantity is used to select an optimal output voltage vector.
4. The method of claim 1, further comprising selecting an optimal one of the voltage vectors such that the optimal voltage vector obtained at time k is delay compensated, comprising:
For stator current component at time k+2 、Neutral point voltage deviationMaking predictions, comprising:
the optimized cost function becomes:
Substituting the voltage vector into the above formula, and selecting the voltage vector with the minimum cost function, the prediction vector at the time k+2 is taken as the output at the time k+1.
5. A three-phase eight-switch model prediction control device for removing weight coefficients is characterized by comprising:
the voltage prediction module is used for determining a voltage vector, and predicting voltage offset of voltage generated by the action of the voltage vector at the moment k+1 in combination with first voltage and second voltage formed by the fact that load current flows through the first capacitor and the second capacitor;
The current prediction module is used for predicting a current value at the moment k+1 by combining the resistance and the inductance of the three-phase eight-switch inverter;
The delay compensation module is used for carrying out weight coefficient removal processing on a cost function for balancing the neutral point potential and generating a balancing effect, and is also used for compensating the delay effect caused by the calculation time of the processor in actual implementation, so that an optimal vector is output at the time k+1;
The delay compensation module is further configured to:
rewriting the cost function so as to include the final cost function And further obtaining an optimized cost function, wherein the optimized cost function is as follows:
in the method, in the process of the invention, The real part and the imaginary part of (a) are respectivelyAnd (3) withK is the time;
rewriting the cost function so as to include the final cost function The method for further obtaining the optimized cost function is thatThe following formula is shown:
， Is a weight coefficient;
Substitution formula And then the original cost function is rewritten as shown in the following formula:
The first two terms of the molecule are understood to be the difference between the ideal amount of charge flowing and the amount of charge flowing at time k+1, while the third term contains However, in discrete systems, the factor is an invariant, i.e., the third term of the molecule is understood to be simply proportional to the absolute value of the difference between the amounts of charge passing through the two capacitances at time k+1; also equivalent to the fact that the three terms of the molecule have the same dimension and the denominator is an invariant, then the weight coefficient is calculatedAnd eliminating to obtain an optimized cost function.
6. The apparatus of claim 5, wherein the voltage prediction module is further configured to:
and determining a voltage vector through a switching function and a voltage vector formula, wherein 1 zero vector and 6 small vectors are used as voltage alternative vectors, and selecting an optimal vector from the seven basic vectors.
7. The apparatus of claim 5, wherein the voltage prediction module is further configured to:
Predicting a voltage offset of the voltage vector contribution generated voltage at time k+1 by the following voltage offset formula:
Wherein k is time, U c is capacitor voltage, U c1 is first voltage of a first capacitor, U c2 is second voltage of a second capacitor, and T s/C is ratio of sampling time to direct current side capacitor;
The predicted value of the current at time k+1 is calculated by the following current prediction formula:
Wherein, The real part and the imaginary part of (a) are respectivelyAnd (3) with，And (3) withRespectively representing the resistance and inductance of a three-phase eight-switch inverter load,The voltage vector applied at the moment k is represented, and T s/L is the ratio of the sampling time to the inductance;
the module calculates the voltage tolerance through the formula And at time k+1 current valueAnd (3) withThe predicted quantity is used to select an optimal output voltage vector.
8. The apparatus of claim 5, wherein the delay compensation module is further configured to:
For stator current component at time k+2 、Neutral point voltage deviationMaking predictions, comprising:
the optimized cost function becomes:
Substituting the voltage vector into the above formula, and selecting the voltage vector with the minimum cost function, the prediction vector at the time k+2 is taken as the output at the time k+1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011262144.6A CN112886843B (en) | 2020-11-12 | 2020-11-12 | Three-phase eight-switch model prediction control method and device for weight coefficient removal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011262144.6A CN112886843B (en) | 2020-11-12 | 2020-11-12 | Three-phase eight-switch model prediction control method and device for weight coefficient removal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112886843A CN112886843A (en) | 2021-06-01 |
| CN112886843B true CN112886843B (en) | 2024-07-12 |
Family
ID=76043062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011262144.6A Active CN112886843B (en) | 2020-11-12 | 2020-11-12 | Three-phase eight-switch model prediction control method and device for weight coefficient removal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112886843B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113271028B (en) * | 2021-06-03 | 2022-05-17 | 山东大学 | Prediction control method for reconstructing neutral point balance of topology after three-level inverter fault |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110971162A (en) * | 2019-12-11 | 2020-04-07 | 浙江大学 | Improved model prediction torque control method of NPC three-level converter-PMSM system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103036460B (en) * | 2012-11-26 | 2014-10-22 | 天津大学 | Model prediction control method for three-level voltage-source-type converter |
| AT513776B1 (en) * | 2014-04-08 | 2015-09-15 | Avl List Gmbh | Method and controller for model-predictive control of a multiphase DC / DC converter |
| CN105375804B (en) * | 2015-12-17 | 2018-01-12 | 福州大学 | A kind of model prediction current control method based on NPC topology combining inverters under asymmetrical voltage |
| CN107070269B (en) * | 2017-02-14 | 2019-11-26 | 中南大学 | A kind of control method and system of three-phase four-line dc-to-ac converter |
| CN106899019B (en) * | 2017-04-01 | 2019-08-30 | 合肥工业大学 | Limited three level active filter control method of domination set model prediction single goal |
| CN107017810B (en) * | 2017-04-24 | 2023-04-07 | 东南大学盐城新能源汽车研究院 | Permanent magnet synchronous motor weight-free model prediction torque control system and method |
| CN111654219B (en) * | 2020-06-17 | 2023-09-19 | 中南大学 | Fault-tolerant control method and device for asynchronous motor |
-
2020
- 2020-11-12 CN CN202011262144.6A patent/CN112886843B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110971162A (en) * | 2019-12-11 | 2020-04-07 | 浙江大学 | Improved model prediction torque control method of NPC three-level converter-PMSM system |
Non-Patent Citations (1)
| Title |
|---|
| Finite-Set Model Predictive Control for Eight-Switch Three-Phase NPC Converter with DC-Link Voltages Fluctuation Suppression;Zeyu Wang等;2018 International Conference on Computer Modeling, Simulation and Algorithm;20181231;第151卷;85-88 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112886843A (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abdeslam et al. | A unified artificial neural network architecture for active power filters | |
| CN108028599B (en) | Controlling a three-phase power converter | |
| Long et al. | Gradient descent optimization based parameter identification for FCS-MPC control of LCL-type grid connected converter | |
| CN106533227A (en) | Modularization multi-level converter redundancy fault-tolerant control method | |
| CN112886843B (en) | Three-phase eight-switch model prediction control method and device for weight coefficient removal | |
| CN111541411A (en) | Method for controlling open winding motor model of double three-level inverter | |
| KR102275748B1 (en) | The control system and method of mmc converter | |
| CN116436273A (en) | Control method and system for three-phase three-level inverter | |
| Shah et al. | Kalman filtering technique for rooftop-PV system under abnormal grid conditions | |
| CN114938151A (en) | Control method of grid-connected inverter | |
| Zhao et al. | Model-free predictive current control of three-level grid-connected inverters with lcl filters based on kalman filter | |
| Touheed Khan et al. | Least mean sixth control approach for three-phase three-wire grid-integrated PV system | |
| CN117155075A (en) | PWM rectifier control method and related device under unbalanced power grid voltage | |
| Zainuri et al. | Simplified adaptive linear neuron harmonics extraction algorithm for dynamic performance of shunt active power filter | |
| CN113437894B (en) | Common-mode voltage suppression method for three-phase eight-switch inverter | |
| Babaie et al. | ANN based model-free sliding mode control for grid-connected compact multilevel converters: An experimental validation | |
| CN111654219B (en) | Fault-tolerant control method and device for asynchronous motor | |
| Li et al. | Adaptive model predictive control for PUC grid-connected inverter system with inaccurate parameters | |
| CN115425642A (en) | Model-free prediction control method and system for grid-connected converter | |
| CN115021598A (en) | Multi-level inverter prediction control method and system without weight coefficient | |
| CN114421794A (en) | NPC three-level virtual synchronous machine bridge arm fault tolerance model prediction control method | |
| CN112510966B (en) | Modular medium-voltage waveform generator loss balance control method and system | |
| Gutierrez et al. | Model predictive control method with preselected control options for reduced computational complexity in modular multilevel converters (MMCs) | |
| CN104393777A (en) | Half-bridge MMC (Modular Multilevel Converter) sub-module voltage control method | |
| Srivastava et al. | A Whale Optimization Algorithm Based Shunt Active Power Filter for Power Quality Improvement |
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 |