CN110971162A - Improved model prediction torque control method of NPC three-level converter-PMSM system - Google Patents
Improved model prediction torque control method of NPC three-level converter-PMSM system Download PDFInfo
- Publication number
- CN110971162A CN110971162A CN201911268970.9A CN201911268970A CN110971162A CN 110971162 A CN110971162 A CN 110971162A CN 201911268970 A CN201911268970 A CN 201911268970A CN 110971162 A CN110971162 A CN 110971162A
- Authority
- CN
- China
- Prior art keywords
- vector
- voltage
- converter
- power switch
- npc
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
- H02P21/05—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation specially adapted for damping motor oscillations, e.g. for reducing hunting
-
- 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
- H02P21/30—Direct torque control [DTC] or field acceleration method [FAM]
-
- 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
Abstract
The invention discloses an improved model prediction torque control method of an NPC three-level converter-PMSM system. Calculating an offset threshold, and dividing the midpoint voltage offset of two direct current support capacitors in the converter into two types of regions according to the offset threshold: aiming at the larger midpoint voltage offset, calculating the actual amplitude and phase of a medium and small vector of the converter according to the current midpoint voltage offset in each sampling period, and dynamically dividing the sectors; according to control targets of different areas, constructing a value function and a limited control set, obtaining an optimal voltage vector, and outputting the optimal voltage vector; and adding dead time to the control time corresponding to the optimal voltage vector to output a PWM signal to the permanent magnet synchronous motor for control. The invention can improve the control performance of the torque and the flux linkage, reduce the calculated amount, and solve the problems of the reduction of the control performance of the system torque, the flux linkage and the current when the weight coefficient setting process is complex and the midpoint voltage offset amplitude is large in the existing control.
Description
Technical Field
The invention belongs to the field of motor control, and particularly relates to a method for controlling torque of a permanent magnet synchronous motor improved model supplied by a Neutral Point Clamped (NPC) three-level converter, which is used for solving the problems of large torque and current fluctuation when neutral point voltage deviation is large.
Background
Permanent Magnet Synchronous Machines (PMSM) have the advantages of high power density, simple structure, low operating noise, high efficiency, etc. The three-level converter has the advantages of small voltage stress borne by a single switch device, high output voltage quality, small switching loss under the same switching frequency and the like when the same direct-current power supply supplies power. Therefore, the NPC three-level converter-permanent magnet synchronous motor system is widely applied to medium-high voltage high-power motor driving occasions, such as traction transmission and ultra-deep well lifting.
And the model prediction torque control directly considers the on-off state of the discrete power switch tube of the converter, substitutes the voltage vector generated by the switching action of the converter into a prediction value function, and selects the on-off state of the power switch tube which can minimize the value function as the input of the next moment. The control method can realize multi-objective multivariable optimization control, is convenient for processing system constraint and easy to realize, and is widely concerned by industry and academia. When the model prediction torque control algorithm is applied to the NPC three-level converter-motor speed regulation system, a midpoint voltage item is added on the basis of an electromagnetic torque item and a stator flux linkage item in a two-level converter model prediction torque control algorithm value function, so that the torque, flux linkage and midpoint voltage multi-target optimal control is achieved.
Because the torque, flux linkage and midpoint voltage items in the value function of the traditional model prediction torque control algorithm are different in dimension, various weight coefficients need to be set, and the setting process is complex. In order to solve the problem, under the two-level converter, an electromagnetic torque item and a stator flux linkage item in a cost function can be unified into a torque item or a flux linkage item, or the torque item and the flux linkage item are unified into a stator voltage item by utilizing a dead beat principle, or the torque item and the flux linkage item are unified into vector action time. However, when the method is applied to an NPC three-level converter-motor speed regulation system, the midpoint voltage cannot be unified with the variable and does not belong to the stator voltage part, so that a midpoint voltage item still exists in the algorithm value function, and the link of weight coefficient setting still exists. When the weight coefficient is not selected reasonably or the midpoint voltage deviation amplitude is large, the motor torque and flux linkage control performance is reduced, and the midpoint voltage deviation is difficult to be effectively inhibited.
When the voltage deviation amplitude of the midpoint of the converter is large (when a capacitor network formed by all direct current support capacitors on the direct current side fails to work, the capacitance values of all the direct current support capacitors are unequal or the power supply of a direct current bus is unbalanced), the amplitude and the phase angle of small vectors and medium vectors in a space vector plane can change, and if a limited control set is constructed in a traditional space vector sector division mode, the performance of a system such as flux linkage, torque and the like can be reduced. Aiming at the neutral point voltage offset suppression of the NPC three-level converter, the traditional space vector modulation switching sequence is modified by Lewis A, Krzeminski Z and the like, and the neutral point voltage offset is balanced by using the on-off state of a small vector redundancy power switching tube (IEEE Transactions on Industrial Electronics, vol.58, No.11, pp.5076-5086,2011, 11 months). Lopez I and Ceballos et al use a non-recent three-vector composite switching sequence that would replace the vector that increases the midpoint voltage offset with a vector that has no effect on the midpoint voltage (IEEETransactions on Power Electronics, vol.31, No.2, pp.928-941,2016, 2 months). Choi U, Lee J and the like adopt virtual space vector modulation, a virtual middle vector is constructed by using the on-off state of a small-vector redundant Power switch tube, and neutral point voltage deviation is eliminated by judging the degree of unbalance and compensating voltage unbalance of a direct current side (IEEEtransactions on Power Electronics, vol.29, No.1, pp.91-100,2014, 3 months). However, the methods are all modulation methods, influence of midpoint voltage offset on system torque and flux linkage performance is not considered, and the methods cannot be directly applied to a model prediction torque control algorithm.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides an improved model prediction torque control algorithm of an NPC three-level converter-PMSM system, which reduces flux linkage, torque and current fluctuation and improves the control performance of the system while eliminating the process of setting the value function weight coefficient in the control process.
As shown in fig. 6, the technical solution adopted by the present invention is:
aiming at an NPC three-level converter-PMSM system, the control method comprises the following steps:
1) calculating an offset threshold value according to parameters such as capacitance values of the direct current support capacitors of the converter, switching action times of the converter in unit time and the like, and dividing the midpoint voltage offset of the two direct current support capacitors in the converter into two types of regions according to the offset threshold value:
if the midpoint voltage deviation is less than or equal to the deviation threshold value, the midpoint voltage deviation belongs to the I area;
if the midpoint voltage offset is greater than the offset threshold, the midpoint voltage offset is attributed to region II;
the voltage offset of the middle point of the two direct current supporting capacitors in the converter is the voltage offset of the neutral point O between the two direct current supporting capacitors.
2) Aiming at the midpoint voltage offset of the II area, calculating the actual amplitude and phase of a medium and small vector of the converter according to the current midpoint voltage offset in each sampling period, and dynamically dividing the sectors;
3) according to control targets of different areas in the step 1), a value function of a control variable in model prediction torque control and a limited control set consisting of all voltage vectors to be screened of the converter are constructed, the limited control set is substituted into the value function to be calculated to obtain a voltage vector corresponding to the minimum value function as an optimal voltage vector, and the optimal voltage vector is used as the output of the NPC three-level converter;
4) according to the incidence relation between the states of each power switching tube and each phase of output state in the NPC three-level converter, dead time is added to control time corresponding to the optimal voltage vector to prevent the power switching tubes from being directly connected, PWM signals of each power switching tube are further output to the permanent magnet synchronous motor, output of the PWM signals is achieved, and PMSM system model prediction torque control for restraining midpoint voltage deviation is achieved.
The method in the step 1) unifies the electromagnetic torque term and the stator flux linkage term dimension in the cost function into the stator voltage term, and the importance degree of the electromagnetic torque term and the stator flux linkage term in the cost function is not required to be balanced at the same time.
The invention has the innovation points that a midpoint voltage deviation partition control mode is established through the steps 1) and 2), dynamic sector division of space vectors is carried out according to midpoint voltage deviation, space vector plane output PWM signals are updated in real time to control the permanent magnet synchronous motor, model prediction torque control is improved, flux linkage, torque and current fluctuation are reduced, and torque and flux linkage control performance is improved.
The midpoint voltage deviation is divided into an I area and an II area by the midpoint voltage deviation partition control mode, so that a midpoint voltage item in the cost function is eliminated, and the complex processing step of setting the weight coefficient of each variable of the cost function in the existing control process is eliminated. The electromagnetic torque term and the stator flux linkage term dimension in the cost function are unified into a stator voltage term, and the importance degree of the electromagnetic torque term and the stator flux linkage term in the cost function does not need to be balanced at the same time.
In the step 1), the threshold is calculated according to parameters such as a capacitance value of a direct current support capacitor in the converter, the turn-on and turn-off times of a power switch tube of the converter in unit time, and the like, and specifically comprises the following steps:
in the formula, vOThe midpoint voltage between two direct current support capacitors in the converter is represented by the size of the midpoint voltage deviation; c represents the capacitance value of a direct current support capacitor in the converter; t issIs a sampling period; i.e. ixRepresenting any two-phase current value, x ∈ { a, B, C } that is equal under actual load torque.
In the step 2), for the midpoint voltage deviation of the area II, the actual amplitude and phase angle of the medium vector and the small vector in the voltage vector of the converter are calculated according to the midpoint voltage deviation value in the current sampling period, wherein the amplitude of the medium vector is
The magnitude of the small vector is Vdc/3，VdcThe method comprises the steps of representing rectified voltage output by a rectifier bridge in an NPC three-level converter-PMSM system, then dynamically dividing sectors, specifically drawing a space vector plane according to each voltage vector of the converter, dividing the space vector plane into 12 sectors with different intervals by taking a middle vector in the space vector plane and a large vector of the space vector plane as a boundary line of the sectors, and updating each voltage vector in real time under an α - β coordinate system by adopting the following formula in the current sampling period:
in the formula, vαAnd vβRespectively representing the amplitude value and the actual phase angle value of the voltage vector in the unit sampling period under the α - β coordinate system vc1And vc2Respectively representing the terminal voltages of an upper DC support capacitor and a lower DC support capacitor in the converter, the upper DC support capacitor being connected to a rectified voltage VdcA positive DC supporting capacitor connected to the rectified voltage VdcDC support capacitance of the cathode, vc1-vc2Is the midpoint voltage shift; sk1，Sk2，Sk3，Sk4The on and off states of four power switching tubes of the k-th phase of the converter are represented, for example, k belongs to { a, B, C } shown in fig. 1, a specific implementation value of "1" represents that the corresponding power switching tube is on, and a value of "0" represents that the corresponding power switching tube is off.
The α - β coordinate system is obtained by inverse park change of a dq coordinate system of the permanent magnet synchronous motor.
In the step 3), in the I-th region, if the control target is to reduce the torque and flux linkage fluctuation and consider reducing the midpoint voltage offset, a limited control set is formed by all voltage vectors; in the II area, the control target is to reduce the midpoint voltage offset, and the torque and flux linkage fluctuation are considered to be reduced, so that a limited control set is formed by only small vectors and medium vectors.
The NPC three-level converter-PMSM system comprises a voltage source, a rectifier bridge, an NPC three-level converter and a permanent magnet synchronous motor; the rectifier bridge is formed by connecting three groups of diode groups in parallel, each group of diode group is formed by connecting two diodes in series, and two diodes of the three groups of diode groups are connected to a voltage source; three groups of diode groups are connected in parallel to output rectified voltage Vdc；
The NPC three-level converter comprises three groups of power switch tube groups and two direct-current supporting capacitors C which respectively represent three-phase control, the two direct-current supporting capacitors C are connected in parallel with the three groups of power switch tube groups after being connected in series, each group of power switch tube groups is composed of four power switch tubes and two clamping diodes, the four power switch tubes are sequentially connected in series, the two clamping diodes are connected in series between a leading-out end between the first power switch tube and the second power switch tube and between a leading-out end between the third power switch tube and the fourth power switch tube, and leading-out ends i are arranged between the second power switch tubes and between the third power switch tubes of the three groups of power switch tube groupsA、iB、iCAnd the three-phase control ends are respectively connected to the three-phase control ends of the permanent magnet synchronous motor, and the leading-out ends between two clamping diodes of the three groups of power switch tube groups are respectively connected to a neutral point O between the two direct current support capacitors C.
The invention firstly obtains a reference voltage vector while eliminating the torque and flux linkage dimensions. At the moment, the midpoint voltage offset amplitude is divided into two areas, and when the offset amplitude is larger, a sector dynamic division mechanism is adopted to recalculate the medium and small vector amplitude and the phase angle, so that on one hand, a weight coefficient setting link is eliminated, on the other hand, a limited control set is reconstructed according to a main control target of each area, and the alternative vector is substituted into an improved cost function and calculated to obtain the optimal voltage vector of the next sampling period.
Compared with the prior art, the technical scheme of the invention has the following beneficial effects:
(1) the torque and flux linkage dimensions are unified by utilizing a dead beat principle, a cost function is simplified to reduce the calculated amount during prediction, a complex process of weight coefficient setting is eliminated, and the technical problems that the weight coefficient setting process is complex, the system torque and flux linkage and current control performance are reduced when the midpoint voltage offset amplitude is large and the like in the cost function of the existing model prediction torque control process are solved;
(2) when the midpoint voltage offset amplitude is larger, a sector dynamic division mechanism is introduced, so that the midpoint voltage offset inhibition time is shortened, the control performance of torque, flux linkage and current when the midpoint voltage offset amplitude is larger is ensured, and the control performance of the torque and flux linkage is improved;
(3) and the partition control of the midpoint voltage offset amplitude is provided, the number of alternative vectors in a limited control set is reduced according to each partition control target, the processing of the method is simplified, and the calculated amount is reduced.
Drawings
FIG. 1 is a diagram of an NPC three-level converter-PMSM system;
FIG. 2 is a NPC three-level converter space vector diagram;
FIG. 3 is a schematic diagram of a deadbeat voltage calculation;
FIG. 4 is a drawing showing1＝0.5、r2When the vector is 1.5, the amplitude and the phase angle of the medium and small vectors are schematically changed;
FIG. 5 is a drawing showing1＝0.5、r2When the current is 1.5, the NPC three-level converter space vector diagram;
FIG. 6 is a block diagram of an improved model predictive torque control algorithm;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention are further described in detail below with reference to the accompanying drawings:
the embodiment of the invention and the implementation working process thereof are as follows:
the structure of the NPC three-level converter-PMSM system is shown in figure 1, and the NPC three-level converter-PMSM system comprises a voltage source, a rectifier bridge, an NPC three-level converter and a permanent magnet synchronous motor; the rectifier bridge is formed by connecting three groups of diode groups in parallel, each group of diode group is formed by connecting two diodes in series, and two diodes of the three groups of diode groups are connected to a voltage source; three groups of diode groups parallel output rectificationVoltage Vdc；
The NPC three-level converter comprises three groups of power switch tube groups and two direct-current supporting capacitors C which respectively represent three-phase control, wherein the two direct-current supporting capacitors C are connected in series, then are connected in parallel with the three groups of power switch tube groups and are connected to a rectified voltage VdcThe positive DC support capacitor is used as the upper DC support capacitor and is connected to the rectified voltage VdcThe direct current support capacitor of the cathode is used as a lower direct current support capacitor; each group of power switch tube group consists of four power switch tubes and two clamping diodes, after the four power switch tubes are sequentially connected in series, the two clamping diodes are connected in series between a leading-out end between the first power switch tube and the second power switch tube and a leading-out end between the third power switch tube and the fourth power switch tube, and leading-out ends i between the second power switch tube and the third power switch tube of the three groups of power switch tube groupsA、iB、iCAnd the three-phase control ends are respectively connected to the three-phase control ends of the permanent magnet synchronous motor, and the leading-out ends between two clamping diodes of the three groups of power switch tube groups are respectively connected to a neutral point O between the two direct current support capacitors C.
Each power switch tube group is one phase, A, B, C three phases are shown in figure 1, and each phase outputs V by taking a neutral point O between two direct current support capacitors as a reference pointdcA combination of/2, 0 and-VdcThree levels, each indicated by P, O, N, for a total of 3 for three phases3The 27 power switch tube on-off state combinations correspond to 19 voltage vectors in a space vector plane and are divided into four types according to vector magnitude: big vector (2V)dc/3), medium vector
Small vector (V)dc/2) and zero vector (0).
A space vector plane is drawn for a voltage vector of the NPC three-level converter, and the space vector plane is divided into 12 sectors at intervals of pi/6 by taking a large vector and a medium vector as boundaries, as shown in fig. 2.
Specifically, after a forward Euler approximation method is used, a current prediction model under a d-q axis synchronous rotation coordinate system of the permanent magnet synchronous motor is obtained as follows:
in the formula, TsFor the sampling period, superscripts k and k +1 denote kTsAnd (k +1) TsSampling time; rsIs a stator resistor; i.e. id、iqRespectively representing d-axis and q-axis currents of the stator; l isd、LqRespectively representing d-axis inductance and q-axis inductance of the stator; v. ofd、vqRespectively representing d-axis and q-axis voltages of the stator; psifIs a rotor flux linkage; omegarIs the electrical angular velocity of the rotor.
The (k +1) th T can be obtained by the same methodsStator flux d, q axis component at sampling time
And electromagnetic torque
Because the neutral point of the NPC three-level converter direct-current support capacitor is directly connected with the load, the neutral point voltage v is caused by the charging and discharging of the load and the direct-current support capacitoroIs shifted, voThe relation among the capacitance value of the DC support capacitor, the on-off state of the power switch tube of the converter and the three-phase current is
In the formula, C represents the capacitance value of the direct current support capacitor; i.e. ioRepresents the midpoint current, iA、iB、iCRepresenting three-phase current; snRepresenting the on-off state of the corresponding phase power switch tube Sn∈{1，0，-1}，n∈{A，B，C}。
I in the formula (2)oSubstitution of expression into voAnd discretizing to obtain the midpoint voltage voAt (k +1) TsSampling time prediction value
Due to the one-step control delay of the actual digital control system, at kTsThe voltage vector selected at the sampling time is at the (k +1) th TsThe sampling time acts on the converter, and a method of predicting one step in advance is needed to compensate for the delay influence.
Establishing a cost function by taking stator flux linkage, electromagnetic torque and midpoint voltage as control targets, i.e.
In the formula, | ψs|*And
respectively setting flux linkage and electromagnetic torque; lambda [ alpha ]ψ、λTAnd λvWeighting coefficients of stator flux linkage, electromagnetic torque and midpoint voltage terms respectively; phis|k+2、
And
the stator flux linkage amplitude, the electromagnetic torque and the converter midpoint voltage are respectively (k +2) TsThe predicted value of the time.
After the traditional model predictive torque control algorithm is established, the following problems can be found to exist:
1) as can be seen from equation (3), although the midpoint voltage term in the cost function can suppress the midpoint voltage offset of the converter, when the midpoint voltage offset is large in magnitude (for example: when the direct current support capacitor network of the converter fails, the capacitance values of upper and lower capacitors are not equal or the power supply of a direct current bus is unbalanced), the amplitude and phase angle of small and medium vectors in a space vector plane can be changed, and the control performance of torque and flux linkage is reduced;
2) weighting coefficient lambda of stator flux linkage term, electromagnetic torque term and midpoint voltage termψ、λTAnd λvThe setting process is more complex;
3) the number of alternative voltage vectors in the control set is large, which results in a large calculation amount of the algorithm.
In order to solve the problems, the invention provides an improved model prediction torque control algorithm, which realizes the rapid inhibition of midpoint voltage deviation, the improvement of torque and flux control performance, the elimination of weight coefficients in a value function and the simplification of a limited control set by unifying torque flux linkage dimensions, introducing a sector dynamic division mechanism and midpoint voltage deviation amplitude value partition control.
First, neglecting the stator resistance effect, the expression of the rate of change of the electromagnetic torque after the delay compensation can be derived as follows:
note the book
After finishing, the following can be obtained:
wherein:
the expression of the flux linkage obtained by discretizing the formula (1) and the formula (2) is
According to the dead beat principle, given values of flux linkage and torque are taken as (k +2) TsThe predicted value of the time is simplified by connecting the vertical type (7) and the formula (8) in parallel. When Δ T is within the d-q coordinate systemeAt a certain time, equation (10) may be expressed as a straight line, and equation (8) may be expressed as a circle, as shown in fig. 3.
Considering the limitations of the converter output voltage and current, the corresponding N points with smaller amplitude in FIG. 3 are selected and marked as dead-beat voltage vectors, i.e. reference vectors
The d-q axis voltage corresponding to the N point can be obtained after the simultaneous reaction
And
as follows:
wherein:
to obtain
The d-q axis voltage needs to be converted into an α - β coordinate system through inverse park transformation to obtain a α axis component
And β axis component
At this time, can obtain
Amplitude and position angle.
From the above analysis, it can be known that the stator flux linkage term and the electromagnetic torque term in the cost function can be unified into the stator voltage term by using the dead beat principle, and at this time, the cost function becomes the stator voltage term
When the converter direct current support capacitor network fails, the amplitude and the phase of a medium vector and a small vector in a space vector plane are changed due to large deviation of the midpoint voltage of the converter, and the control performance of the torque, the flux linkage, the current and the midpoint voltage of a system is further influenced. To quantify the degree of midpoint voltage shift, a shift coefficient r may be defined1And r2The method comprises the following steps:
in the specific embodiment, the sectors (1) and (2) shown in fig. 5 are taken as examples, and as shown in fig. 4, when the current transformer generates midpoint voltage offset (offset coefficient r)1＝0.5、r21.5), the small vector phase angle corresponding to the on-off state POO and ONN of the power switch tube is unchanged, the amplitude is changed, and the value is V13Becomes V'13And V ″)13(ii) a The middle vector amplitude and the phase angle corresponding to the PON in the on-off state of the power switch tube can be changed from V2Becomes V'2. If the sector is still divided by pi/6 as the interval, the optimal voltage vector selection will be deviated.
With reference vectors
For example, if the power switch tube is at the position shown in fig. 4, the PON is in an on-off state andONN can reduce the midpoint voltage shift, V in the conventional model predictive torque control algorithm2And
distance d of4Less than V13And
distance d of3Therefore V is2(ONN) is an optimal voltage vector; in fact, when the converter generates midpoint voltage deviation, the small vector phase angle corresponding to the on-off state POO and ONN of the power switch tube is unchanged, and the amplitude is changed from V13Becomes V'13And V ″)13(ii) a The middle vector amplitude and the phase angle corresponding to the PON in the on-off state of the power switch tube can be changed from V2Becomes V'2. At this time, the sector should be dynamically divided according to the actual phase of the medium vector, as shown in fig. 4, the gray area is the (1) th sector, and the white area is the (2) th sector. In sector (1), V ″)13And
distance d of1Is less than V'2And
distance d of2Therefore V13(ONN) is the actual optimum voltage vector.
From the above analysis, although the midpoint voltage term in the cost function of the traditional model prediction torque control algorithm can play a role in restraining midpoint voltage offset, when the midpoint voltage offset amplitude is large, the traditional algorithm causes the selection of the optimal voltage vector to have deviation, and further influences the torque and flux linkage control effect. The invention dynamically divides the sector according to the actual phase of the vector in each sampling period, when the offset coefficient r is1＝0.5、r2When the space vector is 1.5, the sector division of the whole space vector plane is as shown in fig. 5.
Comparing fig. 2 and fig. 5, it can be seen that the large vector in each sector does not change, the small vector changes into two small vectors with the same phase and different amplitudes, and the amplitude and phase of the medium vector change. Therefore, the medium and small vector magnitude phases need to be recalculated.
The α - β axis component of the middle vector is obtained according to the invention, so that the dynamic sector division angle can be obtained.
From the above analysis, it can be known that the stator flux linkage term and the electromagnetic torque term in the cost function can be unified into the stator voltage term by using the dead beat principle. However, at this time, the cost function still has a midpoint voltage term, and the setting process of the weight coefficient is still complicated. Meanwhile, when the midpoint voltage offset amplitude is large, if the basic voltage vector which is closer to the dead-beat voltage vector is still selected according to the traditional sector division for control, the torque and flux linkage fluctuation will be large.
The method omits a midpoint voltage item in the cost function, and divides the midpoint voltage offset amplitude into I, II two regions for control by reasonably setting a threshold value. When the offset amplitude is small (I area), directly selecting an optimal voltage vector from a limited control set by using the improved cost function; when the offset amplitude is large (area II), the medium and small vector amplitudes and phase angles are corrected and the sector is dynamically divided. On the basis, a limited control set is reconstructed, and only basic vectors capable of reducing midpoint voltage offset are reserved. And finally, selecting an optimal voltage vector through the improved cost function.
(1) I region control mode
The midpoint voltage term in equation (12) is rounded off, and the improved cost function is
At this time, the sectors are still divided according to the traditional algorithm to construct a limited control set, and the distance dead beat voltage vector is selected from the limited control set through the formula (14)
The nearest base vector serves as the optimum voltage vector. If the optimal voltage vector is a minivectorAnd selecting the on-off state of the power switch tube capable of reducing the midpoint voltage offset from the on-off states of the two power switch tubes corresponding to the small vector.
(2) II zone control mode
TABLE 1. alternative Voltage vector for midpoint Voltage II region
The cost function in equation (14) is still used, but the actual phase angle of the medium vector is determined in the current sampling period, and the sector is dynamically divided according to the actual phase angle and the small vector magnitude is calculated, as shown in fig. 5. Since the midpoint voltage offset amplitude is large at this time, it is necessary to suppress the midpoint voltage offset as a main control target.
At this time, a finite control set is reconstructed, large vectors and zero vectors which cannot influence the midpoint voltage are omitted, and only medium vectors and small vectors are reserved. The reconstructed finite control set is shown in table 1. At this time, the range dead beat voltage vector is selected from the reconstructed finite control set by equation (14)
The nearest base vector serves as the optimum voltage vector.
In conclusion, the improved model prediction torque control algorithm provided by the invention realizes accurate control of torque and flux linkage while rapidly inhibiting midpoint voltage offset through unified torque flux linkage dimension, sector dynamic partition and limited control set reconstruction and midpoint voltage offset amplitude partition control. The algorithm control block diagram is shown in fig. 6.
The present invention is not limited to the above-described embodiments. The above description of the embodiments is intended to illustrate the technical solutions of the present invention, and the above embodiments are merely illustrative and not restrictive. Those skilled in the art can make many changes and modifications to the invention without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (5)
1. An improved model prediction torque control method of an NPC three-level converter-PMSM system is characterized by comprising the following steps: aiming at an NPC three-level converter-PMSM system, the control method comprises the following steps:
1) calculating an offset threshold value according to parameters such as capacitance values of the direct current support capacitors of the converter, switching action times of the converter in unit time and the like, and dividing the midpoint voltage offset of the two direct current support capacitors in the converter into two types of regions according to the offset threshold value:
if the midpoint voltage deviation is less than or equal to the deviation threshold value, the midpoint voltage deviation belongs to the I area;
if the midpoint voltage offset is greater than the offset threshold, the midpoint voltage offset is attributed to region II;
2) aiming at the midpoint voltage offset of the II area, calculating the actual amplitude and phase of a medium and small vector of the converter according to the current midpoint voltage offset in each sampling period, and dynamically dividing the sectors;
3) according to control targets of different areas in the step 1), a value function of a control variable in model prediction torque control and a limited control set consisting of all voltage vectors to be screened of the converter are constructed, the limited control set is substituted into the value function to be calculated to obtain a voltage vector corresponding to the minimum value function as an optimal voltage vector, and the optimal voltage vector is used as the output of the NPC three-level converter;
4) according to the incidence relation between the states of each power switching tube and each phase output state in the NPC three-level converter, dead time is added to the control time corresponding to the optimal voltage vector, and then PWM signals of each power switching tube are output to the permanent magnet synchronous motor, so that the PMSM system model prediction torque control for restraining the midpoint voltage deviation is realized.
2. The NPC three-level converter-PMSM system improved model predictive torque control method as claimed in claim 1, wherein: in the step 1), the threshold is calculated according to parameters such as a capacitance value of a direct current support capacitor in the converter, the turn-on and turn-off times of a power switch tube of the converter in unit time, and the like, and specifically comprises the following steps:
in the formula, vOThe middle point voltage between two direct current support capacitors in the converter is obtained; c represents the capacitance value of a direct current support capacitor in the converter; t issIs a sampling period; i.e. ixRepresenting any two-phase current values that are equal at the actual load torque.
3. The NPC three-level converter-PMSM system improved model predictive torque control method as claimed in claim 1, wherein: in the step 2), for the midpoint voltage deviation of the area II, the actual amplitude and phase angle of the medium vector and the small vector in the voltage vector of the converter are calculated according to the midpoint voltage deviation value in the current sampling period, wherein the amplitude of the medium vector is
The magnitude of the small vector is Vdc/3，VdcThe method comprises the steps of representing rectified voltage output by a rectifier bridge in an NPC three-level converter-PMSM system, then dynamically dividing sectors, specifically drawing a space vector plane according to each voltage vector of the converter, dividing the space vector plane into 12 sectors with different intervals by taking a middle vector in the space vector plane and a large vector of the space vector plane as a boundary line of the sectors, and updating each voltage vector in real time under an α - β coordinate system by adopting the following formula in the current sampling period:
in the formula, vαAnd vβRespectively representing the amplitude value and the actual phase angle value of the voltage vector in the unit sampling period under the α - β coordinate system vc1And vc2Respectively representing the terminal voltages of the upper direct current support capacitor and the lower direct current support capacitor in the converter,the upper DC supporting capacitor being connected to the rectified voltage VdcA positive DC supporting capacitor connected to the rectified voltage VdcDC support capacitance of the cathode, vc1-vc2Is the midpoint voltage shift; sk1，Sk2，Sk3，Sk4And the on and off states of four power switching tubes of the k phase of the converter are shown.
4. The NPC three-level converter-PMSM system improved model predictive torque control method as claimed in claim 1, wherein: in the step 3), in the I-th region, if the control target is to reduce the torque and flux linkage fluctuation and consider reducing the midpoint voltage offset, a limited control set is formed by all voltage vectors; in the II area, the control target is to reduce the midpoint voltage offset, and the torque and flux linkage fluctuation are considered to be reduced, so that a limited control set is formed by only small vectors and medium vectors.
5. The NPC three-level converter-PMSM system improved model predictive torque control method as claimed in claim 1, wherein: the NPC three-level converter-PMSM system comprises a voltage source, a rectifier bridge, an NPC three-level converter and a permanent magnet synchronous motor; the rectifier bridge is formed by connecting three groups of diode groups in parallel, each group of diode group is formed by connecting two diodes in series, and two diodes of the three groups of diode groups are connected to a voltage source; three groups of diode groups are connected in parallel to output rectified voltage Vdc(ii) a The NPC three-level converter comprises three groups of power switch tube groups and two direct-current supporting capacitors C which respectively represent three-phase control, the two direct-current supporting capacitors C are connected in parallel with the three groups of power switch tube groups after being connected in series, each group of power switch tube groups is composed of four power switch tubes and two clamping diodes, the four power switch tubes are sequentially connected in series, the two clamping diodes are connected in series between a leading-out end between the first power switch tube and the second power switch tube and between a leading-out end between the third power switch tube and the fourth power switch tube, and leading-out ends i are arranged between the second power switch tubes and between the third power switch tubes of the three groups of power switch tube groupsA、iB、iCAnd the three-phase control ends are respectively connected to the three-phase control ends of the permanent magnet synchronous motor, and the leading-out ends between two clamping diodes of the three groups of power switch tube groups are respectively connected to a neutral point O between the two direct current support capacitors C.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911268970.9A CN110971162B (en) | 2019-12-11 | 2019-12-11 | Model prediction torque control method of NPC three-level converter-PMSM system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911268970.9A CN110971162B (en) | 2019-12-11 | 2019-12-11 | Model prediction torque control method of NPC three-level converter-PMSM system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110971162A true CN110971162A (en) | 2020-04-07 |
| CN110971162B CN110971162B (en) | 2021-04-06 |
Family
ID=70033815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911268970.9A Active CN110971162B (en) | 2019-12-11 | 2019-12-11 | Model prediction torque control method of NPC three-level converter-PMSM system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110971162B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112564567A (en) * | 2020-12-09 | 2021-03-26 | 天津工业大学 | Three-level inverter driving permanent magnet synchronous motor system finite set prediction control method |
| CN112737452A (en) * | 2020-12-25 | 2021-04-30 | 珠海格力电器股份有限公司 | Control method and device of three-level permanent magnet synchronous motor |
| CN112821816A (en) * | 2021-02-23 | 2021-05-18 | 大连海事大学 | PMSM three-vector model prediction current control method based on NPC type three-level inverter |
| CN114142758A (en) * | 2021-12-07 | 2022-03-04 | 浙江大学先进电气装备创新中心 | Novel modulation method suitable for line voltage cascade type triple converter |
| CN116915096A (en) * | 2023-09-14 | 2023-10-20 | 武汉商学院 | Dead-beat prediction voltage control method of PMSM three-level NPC inverter |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102013824A (en) * | 2010-09-21 | 2011-04-13 | 深圳市大族激光科技股份有限公司 | Controller for inhibiting neutral point potential drift in three-level inverter |
| CN102035463A (en) * | 2010-12-13 | 2011-04-27 | 天津电气传动设计研究所 | 6 kV medium voltage frequency converter based on neutral-point-clamped three-level technology |
| CN102710163A (en) * | 2012-05-22 | 2012-10-03 | 华中科技大学 | Neutral-point voltage control method of NPC (neutral-point converter) type three-level inverter based on interval selection |
| EP2654190A2 (en) * | 2012-04-18 | 2013-10-23 | GE Energy Power Conversion GmbH | Method for operating an electric circuit |
| CN105680712A (en) * | 2016-03-24 | 2016-06-15 | 山东大学 | SHEPWM (selective harmonic elimination pulse width modulation) control circuit, double-T-type three-level SHEPWM inverter parallel system and method therefor |
| CN105680711A (en) * | 2016-03-22 | 2016-06-15 | 山东大学 | SHEPWM adopted T type three-level inverter neutral-point voltage balance control method |
| CN205453533U (en) * | 2016-03-24 | 2016-08-10 | 山东大学 | SHEPWM control circuit , two three level SHEPWM dc -to -ac converter parallel system on T type |
| TW201729482A (en) * | 2016-02-05 | 2017-08-16 | 國立清華大學 | Voltage balancing control system and method thereof |
| CN107196536A (en) * | 2017-05-03 | 2017-09-22 | 浙江大学 | A kind of three level SVPWM methods with neutral balance and common-mode voltage rejection ability |
| CN107306516A (en) * | 2014-12-24 | 2017-10-31 | 东芝三菱电机产业系统株式会社 | Power conversion device |
| CN107623457A (en) * | 2017-09-28 | 2018-01-23 | 湘潭大学 | NPC types three-level inverter suppresses DC side midpoint low-frequency oscillation modulator approach |
| CN107769595A (en) * | 2017-11-21 | 2018-03-06 | 中国矿业大学 | A kind of three-level PWM rectifier model predictive control method |
| CN108649855A (en) * | 2018-06-14 | 2018-10-12 | 天津工业大学 | A kind of model prediction method for controlling torque based on duty ratio |
| CN108988664A (en) * | 2018-07-19 | 2018-12-11 | 山东大学 | Three level VIENNA rectifier Model forecast systems and method under the conditions of unbalanced power supply |
| CN109302119A (en) * | 2018-10-10 | 2019-02-01 | 山东大学 | Control method, controller and the system of complete period low common-mode voltage operation |
| CN110460281A (en) * | 2019-03-28 | 2019-11-15 | 南通大学 | The double vector models of three level permanent magnet synchronous motor of one kind predict flux linkage control method |
| CN110971168A (en) * | 2019-12-18 | 2020-04-07 | 福州大学 | Improved direct torque control method for predicting induction machine model |
-
2019
- 2019-12-11 CN CN201911268970.9A patent/CN110971162B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102013824A (en) * | 2010-09-21 | 2011-04-13 | 深圳市大族激光科技股份有限公司 | Controller for inhibiting neutral point potential drift in three-level inverter |
| CN102035463A (en) * | 2010-12-13 | 2011-04-27 | 天津电气传动设计研究所 | 6 kV medium voltage frequency converter based on neutral-point-clamped three-level technology |
| EP2654190A2 (en) * | 2012-04-18 | 2013-10-23 | GE Energy Power Conversion GmbH | Method for operating an electric circuit |
| CN102710163A (en) * | 2012-05-22 | 2012-10-03 | 华中科技大学 | Neutral-point voltage control method of NPC (neutral-point converter) type three-level inverter based on interval selection |
| CN107306516A (en) * | 2014-12-24 | 2017-10-31 | 东芝三菱电机产业系统株式会社 | Power conversion device |
| TW201729482A (en) * | 2016-02-05 | 2017-08-16 | 國立清華大學 | Voltage balancing control system and method thereof |
| CN105680711A (en) * | 2016-03-22 | 2016-06-15 | 山东大学 | SHEPWM adopted T type three-level inverter neutral-point voltage balance control method |
| CN105680712A (en) * | 2016-03-24 | 2016-06-15 | 山东大学 | SHEPWM (selective harmonic elimination pulse width modulation) control circuit, double-T-type three-level SHEPWM inverter parallel system and method therefor |
| CN205453533U (en) * | 2016-03-24 | 2016-08-10 | 山东大学 | SHEPWM control circuit , two three level SHEPWM dc -to -ac converter parallel system on T type |
| CN107196536A (en) * | 2017-05-03 | 2017-09-22 | 浙江大学 | A kind of three level SVPWM methods with neutral balance and common-mode voltage rejection ability |
| CN107623457A (en) * | 2017-09-28 | 2018-01-23 | 湘潭大学 | NPC types three-level inverter suppresses DC side midpoint low-frequency oscillation modulator approach |
| CN107769595A (en) * | 2017-11-21 | 2018-03-06 | 中国矿业大学 | A kind of three-level PWM rectifier model predictive control method |
| CN108649855A (en) * | 2018-06-14 | 2018-10-12 | 天津工业大学 | A kind of model prediction method for controlling torque based on duty ratio |
| CN108988664A (en) * | 2018-07-19 | 2018-12-11 | 山东大学 | Three level VIENNA rectifier Model forecast systems and method under the conditions of unbalanced power supply |
| CN109302119A (en) * | 2018-10-10 | 2019-02-01 | 山东大学 | Control method, controller and the system of complete period low common-mode voltage operation |
| CN110460281A (en) * | 2019-03-28 | 2019-11-15 | 南通大学 | The double vector models of three level permanent magnet synchronous motor of one kind predict flux linkage control method |
| CN110971168A (en) * | 2019-12-18 | 2020-04-07 | 福州大学 | Improved direct torque control method for predicting induction machine model |
Non-Patent Citations (2)
| Title |
|---|
| CHANGLIANG XIA.ETC: "《Discontinuous Space Vector PWM Strategy of Neutral一Point一Clamped Three一Level Inverters for Output Current Ripple Reduction》", 《IEEE TRANSACTIONS ON PO认ER ELECTRONICS》 * |
| 夏长亮等: "《结合开关表的三电平逆变器永磁同步电机模型预测转矩控制》", 《电工技术学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112564567A (en) * | 2020-12-09 | 2021-03-26 | 天津工业大学 | Three-level inverter driving permanent magnet synchronous motor system finite set prediction control method |
| CN112737452A (en) * | 2020-12-25 | 2021-04-30 | 珠海格力电器股份有限公司 | Control method and device of three-level permanent magnet synchronous motor |
| CN112821816A (en) * | 2021-02-23 | 2021-05-18 | 大连海事大学 | PMSM three-vector model prediction current control method based on NPC type three-level inverter |
| CN112821816B (en) * | 2021-02-23 | 2022-07-22 | 大连海事大学 | PMSM model prediction current control method based on NPC type three-level inverter |
| CN114142758A (en) * | 2021-12-07 | 2022-03-04 | 浙江大学先进电气装备创新中心 | Novel modulation method suitable for line voltage cascade type triple converter |
| CN114142758B (en) * | 2021-12-07 | 2023-05-19 | 浙江大学先进电气装备创新中心 | Novel modulation method suitable for line voltage cascading type triple converter |
| CN116915096A (en) * | 2023-09-14 | 2023-10-20 | 武汉商学院 | Dead-beat prediction voltage control method of PMSM three-level NPC inverter |
| CN116915096B (en) * | 2023-09-14 | 2023-12-15 | 武汉商学院 | Dead-beat prediction voltage control method of PMSM three-level NPC inverter |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110971162B (en) | 2021-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110971162B (en) | Model prediction torque control method of NPC three-level converter-PMSM system | |
| CN109274303B (en) | Asynchronous motor model prediction control method based on extended control set | |
| CN106936356B (en) | Vector screening and duty ratio combined motor model prediction control system and method | |
| US6819078B2 (en) | Space vector PWM modulator for permanent magnet motor drive | |
| CN110995076A (en) | Permanent magnet synchronous motor model prediction current control method | |
| CN108736778B (en) | Dual-vector prediction flux linkage control method for permanent magnet synchronous motor | |
| CN110557074A (en) | prediction current control method for permanent magnet motor system of cascaded H-bridge inverter | |
| CN112737444B (en) | Double three-phase permanent magnet synchronous motor control method for alternatively executing sampling and control programs | |
| CN111541411B (en) | Double-three-level inverter open winding motor model control method | |
| CN109600085B (en) | Permanent magnet synchronous motor direct prediction duty ratio control method based on variable control set | |
| CN109038673B (en) | Model prediction optimization control method of photovoltaic power generation system | |
| CN111525859B (en) | Finite switch sequence model prediction current control method of OW-PMSM system | |
| CN112217437A (en) | Permanent magnet synchronous motor three-vector model prediction current control circuit and method | |
| CN107994792A (en) | Double-permanent-magnet synchronous motor control inverter and compensation control method | |
| CN110620539B (en) | Direct torque control optimization method for open-winding motor under variable bus voltage working condition | |
| Liu et al. | An optimal model predictive control method for five-level active NPC inverter | |
| CN113783456A (en) | Low common mode vector modulation method and system of three-level SNPC inverter | |
| CN112290858B (en) | Five-phase permanent magnet synchronous motor interphase short circuit fault-tolerant control method based on multi-vector model prediction | |
| CN113746108B (en) | T-type three-level SAPF open circuit fault sequence model prediction fault tolerance control method | |
| CN113437894B (en) | Common-mode voltage suppression method for three-phase eight-switch inverter | |
| CN114362626B (en) | Prediction control method and system for supercoiled sliding mode observer of permanent magnet synchronous motor | |
| CN115459553B (en) | Midpoint potential spwm control method for t-type three-level pwm rectifier | |
| Chekireb et al. | Generalised algorithm of novel space vector modulation: for N‐level three‐phase voltage source inverter | |
| Dong et al. | A Modified Modulated Model Predictive Control for PMSM in Over-Modulation Region | |
| CN115622465B (en) | Multi-level converter weightless factor model prediction control method |
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 |