Three-level NPC converter control method and system

By employing a midpoint voltage balancing strategy based on sector partitioning and small vector control, combined with narrow pulse suppression and dead-time compensation, the problems of midpoint voltage balancing, narrow pulse suppression, and dead-time compensation in three-level NPC converters are solved, simplifying the algorithm and improving system stability and output waveform stability.

CN117318583BActive Publication Date: 2026-11-17CRRC QINGDAO SIFANG ROLLING STOCK RESEARCH INSTITUTE CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202311270813.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-27
Publication Date
2026-11-17
Estimated Expiration
2043-09-27

AI Technical Summary

Technical Problem

Existing methods for midpoint voltage balance, narrow pulse suppression, and dead-time compensation in three-level NPC converters are computationally complex and isolated, affecting system stability and equipment performance.

Method used

A midpoint voltage balancing strategy combining sector partitioning and small vector control with hysteresis control is adopted. The voltage vector switching time is compensated by the midpoint voltage balancing modulation, and narrow pulse suppression and dead zone compensation are performed to simplify the algorithm and improve system stability.

Benefits of technology

It reduces the complexity of software algorithms, improves system stability and robustness, ensures the stability of output waveform and motor starting torque, and reduces low-frequency oscillations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117318583B_ABST
    Figure CN117318583B_ABST
Patent Text Reader

Abstract

This invention belongs to the field of converter control technology, and relates to a control method and system for a three-level NPC converter. From a system perspective, it comprehensively considers three issues: midpoint voltage balance, narrow pulse suppression, and dead-time compensation. It rationally addresses these three issues by employing a midpoint voltage balance control strategy combining small vector control and hysteresis control to balance the midpoint voltage. Narrow pulse suppression is applied to the voltage vector switching time obtained after compensation using midpoint voltage balance modulation. Dead-time compensation is applied to the voltage vector switching time after narrow pulse suppression. Based on the dead-time compensated voltage vector switching time, the duty cycle of the six PWM channels is assigned to control the on / off state of the power devices. This invention can balance the voltages of the upper and lower support capacitors, ensure stable output waveforms, improve motor starting torque, reduce low-frequency oscillations, and solve the problems of midpoint voltage balance control, narrow pulse suppression, and dead-time compensation in three-level NPC converters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of converter control technology, and relates to three-level converter control technology. Specifically, it relates to a three-level NPC converter control method and system. Background Technology

[0002] Multilevel converters are widely used in high-voltage, high-power applications such as traction motors, wind power converters, and scraper conveyors due to their advantages of low harmonic content, good electromagnetic compatibility, and low cost. Among them, the three-level neutral-point clamped (NPC) converter is the most widely researched and applied circuit topology. In three-level control strategies, neutral-point voltage balance, narrow pulse suppression, and dead-time compensation are key issues that require attention. Neutral-point voltage imbalance will lead to large voltage deviations between the upper and lower capacitors, causing distortion of the output voltage waveform. In severe cases, this can damage devices and endanger equipment and personal safety. Narrow pulses can cause switching failures of switching devices, resulting in output voltage waveform distortion and potentially leading to heat accumulation and burnout of switching devices. Dead-time effects can cause problems such as low motor starting torque and low-frequency oscillations, affecting the low-frequency characteristics of the equipment.

[0003] To address these issues, industry and academia have conducted extensive research. For midpoint voltage balance control, methods such as virtual vector control, small vector control, and zero-sequence vector control have been proposed. For narrow pulse suppression, methods such as direct filtering, non-nearest three-vector modulation, and zero-sequence vector control have been proposed. For dead-zone compensation, methods such as current feedback, voltage feedback, and direct pulse methods have been proposed. While these methods can solve the midpoint voltage balance, narrow pulse suppression, and dead-zone compensation problems individually, most are computationally complex and isolated, failing to consider the system as a unified whole. Solving these three problems in isolation would significantly increase algorithm complexity; moreover, because each problem is addressed in isolation, mechanical integration might lead to mutual interference, compromising the overall stability of the system. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention provides a control method and system for a three-level NPC converter, which can balance the voltages of the upper and lower support capacitors, ensure stable output waveforms, improve motor starting torque, reduce low-frequency oscillations, and solve the problems of midpoint voltage balance control, narrow pulse suppression, and dead-zone compensation in three-level NPC converters.

[0005] In a first aspect, the present invention provides a control method for a three-level NPC converter, the steps of which are as follows:

[0006] S1, Midpoint Voltage Balancing Process

[0007] Divide the three-level hexagonal space vector into large and small sectors;

[0008] Determining the location of large sectors based on three-level voltage vector coordinates;

[0009] The midpoint voltage balance modulation is obtained based on the location of the large sector, the direction of the voltage difference, and the direction of the current.

[0010] The center vector coordinates are obtained based on the position of the large sector in the three-level hexagonal space vector.

[0011] The location of a small sector is determined based on the center vector coordinates and the three-level voltage vector coordinates.

[0012] The action time of adjacent vectors is determined based on the location of the small sector, and the action time of the zero vector is compensated by the midpoint voltage balance modulation to obtain the voltage vector switching time.

[0013] S2, Narrow Pulse Suppression Processing

[0014] Determine the minimum pulse width for power devices;

[0015] Narrow pulse suppression principle is set based on minimum pulse width;

[0016] Narrow pulse suppression is applied to the voltage vector switching time obtained in step S1 based on the narrow pulse suppression principle;

[0017] S3, Dead Zone Compensation Processing

[0018] Determine the dead zone compensation coefficient;

[0019] The dead-zone-compensated voltage vector switching time is obtained by performing dead-zone compensation on the voltage vector switching time after narrow pulse suppression based on the dead-zone compensation coefficient.

[0020] S4. Based on the position of the large sector, assign values ​​to the duty cycle of the six PWM channels according to the voltage vector switching time after dead-time compensation, and control the on / off state of the power devices.

[0021] In some embodiments, in step S1, the method for dividing large and small sectors is as follows: the entire three-level hexagonal space vector is divided into six adjacent two-level hexagons, defined as six large sectors from one to six; each large sector is further divided into six adjacent triangles, defined as six small sectors from one to six.

[0022] In some embodiments, the method for determining the location of the large sector in step S1 is as follows:

[0023] Based on the three-level voltage vector coordinates in a two-phase stationary coordinate system, the large sector judgment variable is calculated according to formula (1); formula (1) is expressed as:

[0024]

[0025] In the formula, Va, Vb, and Vc are the large sector judgment variables, and Ualpha3 and Ubeta3 are the three-level voltage vector coordinates in the two-phase stationary coordinate system.

[0026] Based on the signs of the variables Va, Vb, and Vc, we can obtain the large sector number and the correspondence between the large sector number and the large sector.

[0027] In some embodiments, the method for obtaining the midpoint voltage balance modulation in step S1 is as follows:

[0028] The direction of the voltage difference is obtained using the voltage difference hysteresis loop according to formula (2), which is expressed as:

[0029]

[0030] In the formula, flag_Un represents the direction of the voltage difference, Udc_error represents the voltage difference between the upper and lower supporting capacitors, and Udc_error_Imt represents the voltage difference hysteresis limit.

[0031] The direction of the current is obtained using the current hysteresis loop according to formula (3), which is expressed as:

[0032]

[0033] In the formula, sign_Ii is the current direction, Ii is the three-phase current, i = a, b, c, and I_Imt is the current hysteresis limit.

[0034] Based on the location of the large sector, the midpoint voltage balance modulation is obtained through formula (4) according to the voltage difference direction and current direction. Formula (4) is expressed as:

[0035] k = s * flag_Un * sign_Ii * ko (4)

[0036] In the formula, k is the midpoint voltage balance modulation degree; s is the compensation direction. When the large sector is 1, 3, or 5, s = 1. When the large sector is 2, 4, or 6, s = -1. ko is the modulation coefficient, with a value range of 0-1.

[0037] In some embodiments, the method for determining the position of the small sector in step S1 is as follows:

[0038] Using the center vector coordinates, the three-level voltage vector coordinates are transformed into two-level voltage vector coordinates using formula (5), which is expressed as:

[0039]

[0040] In the formula, Ualpha2 and Ubeta2 are the two-level voltage vector coordinates, and Cer_alpha and Cer_beta are the center vector coordinates;

[0041] Calculate the small sector judgment variable according to formula (6);

[0042]

[0043] In the formula, Va2, Vb2, and Vc2 are the small sector judgment variables;

[0044] The sign of the variable is used to determine the sector number and the correspondence between the sector number and the sector.

[0045] In some embodiments, in step S1, the voltage vector switching time determination step involves: calculating the action time of adjacent vectors based on the position of the small sector, compensating the vector action time using the midpoint voltage balance modulation, and obtaining the voltage vector switching time of each small sector.

[0046] In some embodiments, in step S2, the method for determining the minimum pulse width of the power device is as follows: the minimum pulse width of the power device is determined by formula (7) based on the minimum allowable turn-on time and switching cycle of the power device. Formula (7) is expressed as:

[0047]

[0048] In the formula, Tmin is the minimum pulse width of the power device, T_on is the minimum allowable turn-on time of the power device, and T_sw is the switching period.

[0049] In some embodiments, in step S2, the narrow pulse suppression principle based on the minimum pulse width Tmin is as follows: pulses with a pulse width less than 0.5*Tmin are filtered out, pulses with a pulse width between 0.5*Tmin and 1*Tmin are widened to Tmin, and pulses with a pulse width greater than 1*Tmin are not processed.

[0050] In some embodiments, in step S3, the dead zone compensation adopts unilateral symmetrical compensation, and the dead zone compensation coefficient is calculated by formula (9), which is expressed as:

[0051]

[0052] In the formula, compen is the dead zone compensation coefficient, and T_dead is the dead zone time;

[0053] Dead-time compensation is performed on the voltage vector switching time after narrow pulse suppression based on the dead-time compensation coefficient using formula (10), which is expressed as:

[0054]

[0055] In the formula, Ti′ is the voltage vector switching time after dead zone compensation, Ii is the three-phase current, i=a,b,c.

[0056] In a second aspect, the present invention provides a three-level NPC converter control system, comprising:

[0057] The sector partitioning module divides the three-level hexagonal space vector into large and small sectors;

[0058] The large sector location determination module determines the location of large sectors based on three-level voltage vector coordinates;

[0059] The midpoint voltage balance regulation determination module obtains the midpoint voltage balance regulation based on the location of the large sector, the direction of the voltage difference, and the direction of the current.

[0060] The small sector position determination module obtains the center vector coordinates based on the position of the large sector in the three-level hexagonal space vector, and determines the position of the small sector based on the center vector coordinates and the three-level voltage vector coordinates.

[0061] The time compensation module determines the action time of adjacent vectors based on the position of the small sector, and compensates for the action time of the zero vector by using the midpoint voltage balance modulation to obtain the voltage vector switching time.

[0062] The setting module sets the narrow pulse suppression principle based on the determined minimum pulse width;

[0063] The suppression module performs narrow pulse suppression on the voltage vector switching time based on the narrow pulse suppression principle;

[0064] The dead-zone compensation module performs dead-zone compensation on the voltage vector switching time after the suppression module is processed, based on a determined dead-zone compensation coefficient, to obtain the dead-zone compensated voltage vector switching time.

[0065] The control module assigns values ​​to the duty cycle of the six PWM channels based on the location of the large sector and the voltage vector switching time after dead-time compensation, thereby controlling the on / off state of the power devices.

[0066] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0067] (1) This invention comprehensively considers three issues from a system perspective: midpoint voltage balance, narrow pulse suppression, and dead-time compensation. It rationally addresses these three issues by performing narrow pulse suppression on the voltage vector switching time obtained after compensation using midpoint voltage balance modulation, then performing dead-time compensation on the voltage vector switching time after narrow pulse suppression, and finally assigning values ​​to the duty cycles of the six PWM channels based on the dead-time compensated voltage vector switching time to control the on / off state of the power devices. This significantly reduces the complexity of the software algorithm and significantly improves the stability of the system, saving substantial debugging costs.

[0068] (2) Based on the simplification of three-level vector control, this invention adopts a midpoint voltage balance control strategy that combines small vector control and hysteresis control to balance the voltage of the upper and lower support capacitors, thereby ensuring the robustness and anti-interference capability of the control system.

[0069] (3) The present invention combines narrow pulse filtering and widening strategies to screen the output pulse, ensuring that the output waveform remains stable and that the power device can be reliably turned on and off.

[0070] (4) The present invention introduces a low-frequency dead zone compensation strategy. Through symmetrical compensation, the dead zone effect is reduced on the low-frequency range of the motor, the starting torque of the motor is improved, and the low-frequency oscillation is reduced. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the main circuit topology of an existing three-level NPC converter.

[0072] Figure 2 This is a flowchart of the three-level NPC converter control method according to an embodiment of the present invention;

[0073] Figure 3 This is a flowchart of the midpoint voltage balancing process described in an embodiment of the present invention;

[0074] Figure 4 This is a schematic diagram of the three-level space vector distribution after sector division in an embodiment of the present invention;

[0075] Figure 5 This is a flowchart of the narrow pulse suppression process described in an embodiment of the present invention;

[0076] Figure 6 This is a flowchart of the dead zone compensation process described in an embodiment of the present invention;

[0077] Figure 7 This is a schematic diagram of the PWM pulse described in an embodiment of the present invention;

[0078] Figure 8 This is a structural block diagram of the three-level NPC converter control system according to an embodiment of the present invention.

[0079] In the diagram, 1 is the sector division module, 2 is the large sector position determination module, 3 is the midpoint voltage balance modulation determination module, 4 is the small sector position determination module, 5 is the time compensation module, 6 is the setting module, 7 is the suppression module, 8 is the dead zone compensation module, and 9 is the control module. Detailed Implementation

[0080] The present invention will now be described in detail through exemplary embodiments. However, it should be understood that, without further description, elements, structures, and features in one embodiment may be advantageously incorporated into other embodiments.

[0081] See Figure 1 The existing three-level NPC converter hardware main circuit topology mainly includes supporting capacitors, IGBTs, clamping diodes, and freewheeling diodes. This topology suffers from midpoint voltage imbalance, narrow pulses, and dead-time effects. Midpoint voltage imbalance easily causes output voltage waveform distortion and can even damage components; narrow pulses easily lead to switching device failures, causing output voltage waveform distortion; and dead-time effects result in low motor starting torque, low-frequency oscillations, and poor low-frequency performance of the equipment. To address these issues, this invention provides a three-level NPC control method and system that comprehensively considers midpoint voltage balance, narrow pulse suppression, and dead-time compensation. By integrating these three aspects, the complexity of the software algorithm is significantly reduced, system stability is significantly improved, and debugging costs are saved. The following detailed description of the three-level NPC control method and system, in conjunction with the accompanying drawings, further illustrates these problems.

[0082] See Figure 2 This invention provides a control method for a three-level NPC converter, the steps of which are as follows:

[0083] S1, Midpoint voltage balancing. See [link / reference] Figure 3 The specific steps are as follows:

[0084] S11. Divide the three-level hexagonal space vector into large and small sectors.

[0085] Specifically, the method for dividing large and small sectors is as follows: the entire three-level hexagonal space vector is divided into six adjacent two-level hexagons, defined as six large sectors (1 to 6); each large sector is further divided into six adjacent triangles, defined as six small sectors (1 to 6). The distribution of the three-level space vector after sector division is as follows: Figure 4 As shown.

[0086] S12. Determine the location of large sectors based on three-level voltage vector coordinates.

[0087] Specifically, the method for determining the location of large sectors is as follows:

[0088] Based on the three-level voltage vector coordinates in a two-phase stationary coordinate system, the large sector judgment variable is calculated according to formula (1); formula (1) is expressed as:

[0089]

[0090] In the formula, Va, Vb, and Vc are the large sector judgment variables, and Ualpha3 and Ubeta3 are the three-level voltage vector coordinates in the two-phase stationary coordinate system.

[0091] Based on the signs of the large sector determination variables Va, Vb, and Vc, the large sector number and the correspondence between the large sector number and the large sector are obtained, as shown in Table 1.

[0092] Table 1

[0093] Variable judgment Large sector number Large sector Variable judgment Large sector number Large sector Va>0, Vb<0, Vc<0 1 one Va<0,Vb<0,Vc>0 4 five Va<0,Vb>0,Vc<0 2 three Va>0, Vb<0, Vc>0 5 six Va>0, Vb>0, Vc<0 3 two Va<0,Vb>0,Vc>0 6 Four

[0094] S13. Obtain the midpoint voltage balance modulation based on the location of the large sector, the direction of the voltage difference, and the direction of the current.

[0095] Specifically, the method for obtaining the midpoint voltage balance modulation is as follows:

[0096] The direction of the voltage difference is obtained using the voltage difference hysteresis loop according to formula (2), which is expressed as:

[0097]

[0098] In the formula, flag_Un represents the direction of the voltage difference, Udc_error represents the voltage difference between the upper and lower supporting capacitors, and Udc_error_Imt represents the voltage difference hysteresis limit.

[0099] The direction of the current is obtained using the current hysteresis loop according to formula (3), which is expressed as:

[0100]

[0101] In the formula, sign_Ii is the current direction, Ii is the three-phase current, i = a, b, c, and I_Imt is the current hysteresis limit.

[0102] Based on the location of the large sector, the midpoint voltage balance modulation is obtained through formula (4) according to the voltage difference direction and current direction. Formula (4) is expressed as:

[0103] k = s * flag_Un * sign_Ii * ko (4)

[0104] In the formula, k is the midpoint voltage balance modulation degree; s is the compensation direction. When the large sector is 1, 3, or 5, s = 1. When the large sector is 2, 4, or 6, s = -1. ko is the modulation coefficient, with a value range of 0-1.

[0105] S14. Obtain the center vector coordinates (Cer_alpha, Cer_beta) based on the position of the large sector in the three-level hexagonal space vector.

[0106] The relationship between the location of the large sector and the midpoint voltage balance modulation and the center vector coordinates is shown in Table 2.

[0107] Table 2

[0108]

[0109] S15. Determine the position of the small sector based on the center vector coordinates and the three-level voltage vector coordinates.

[0110] The method for determining the location of a small sector is as follows:

[0111] Using the center vector coordinates, the three-level voltage vector coordinates are transformed into two-level voltage vector coordinates using formula (5), which is expressed as:

[0112]

[0113] In the formula, Ualpha2 and Ubeta2 are the two-level voltage vector coordinates, and Cer_alpha and Cer_beta are the center vector coordinates;

[0114] Calculate the small sector judgment variable according to formula (6);

[0115]

[0116] In the formula, Va2, Vb2, and Vc2 are the small sector judgment variables;

[0117] The sign of the variable is used to determine the sector number and the correspondence between the sector number and the sector, as shown in Table 3.

[0118] Table 3

[0119] Variable judgment Small sector number Small sector Variable judgment Small sector number Small sector Va>0, Vb<0, Vc<0 1 2 Va<0,Vb<0,Vc>0 4 4 Va<0,Vb>0,Vc<0 2 6 Va>0, Vb<0, Vc>0 5 3 Va>0, Vb>0, Vc<0 3 1 Va<0,Vb>0,Vc>0 6 5

[0120] S16. Determine the action time of adjacent vectors based on the position of the small sector, and use the midpoint voltage balance modulation to compensate for the action time of the zero vector to obtain the voltage vector switching time.

[0121] The steps for determining the voltage vector switching time are as follows: Based on the location of the small sector, calculate the action times t1 and t2 of adjacent vectors, and compensate for the vector action time using the midpoint voltage balance modulation to obtain the voltage vector switching times Ta, Tb, and Tc for each small sector. The relationship between the location of the small sector and the action times t1 and t2 of adjacent vectors, and the voltage vector switching times Ta, Tb, and Tc are shown in Table 4.

[0122] Table 4

[0123]

[0124] Based on the simplification of three-level vector control, the midpoint voltage balancing process adopts a midpoint voltage balancing control strategy that combines small vector control and hysteresis control to balance the voltage of the upper and lower support capacitors, thereby achieving midpoint voltage balance and ensuring the robustness and anti-interference capability of the control system.

[0125] S2, Narrow pulse suppression processing, see [link / reference] Figure 5 The specific steps are as follows:

[0126] S21. Determine the minimum pulse width of the power device.

[0127] Specifically, the method for determining the minimum pulse width of a power device is as follows: Based on the minimum allowable turn-on time and switching cycle of the power device, the minimum pulse width of the power device is determined using formula (7), which is expressed as:

[0128]

[0129] In the formula, Tmin is the minimum pulse width of the power device, T_on is the minimum allowable turn-on time of the power device, and T_sw is the switching period.

[0130] S22. Narrow pulse suppression principle is set based on minimum pulse width.

[0131] Specifically, the narrow pulse suppression principle based on the minimum pulse width Tmin is as follows: pulses with a pulse width less than 0.5*Tmin are filtered out, pulses with a pulse width between 0.5*Tmin and 1*Tmin are widened to Tmin, and pulses with a pulse width greater than 1*Tmin are not processed.

[0132] S23. Based on the narrow pulse suppression principle, narrow pulse suppression is applied to the voltage vector switching time obtained in step S1, as follows:

[0133]

[0134] In the formula, Ti can be Ta, Tb, or Tc.

[0135] Narrow pulse suppression processing combines narrow pulse filtering and broadening strategies to filter the output pulse, thereby achieving narrow pulse suppression, ensuring that the output waveform remains stable, and ensuring that power devices can be reliably turned on and off.

[0136] S3, Dead Zone Compensation Handling, see [link / reference] Figure 6 The specific steps are as follows:

[0137] S31. Determine the dead zone compensation coefficient.

[0138] Specifically, taking the increment / decrement counting mode PWM pulse as an example, such as Figure 7As shown. Further dead-time compensation is needed for the voltage switching time Ti after narrow pulse suppression. Since the PWM pulse is centrally symmetrical, the dead-time compensation adopts unilateral symmetrical compensation. The dead-time compensation coefficient is calculated using formula (9), which is expressed as:

[0139]

[0140] In the formula, compen is the dead zone compensation coefficient, and T_dead is the dead zone time.

[0141] S32. Based on the dead-zone compensation coefficient, the voltage vector switching time after narrow pulse suppression is compensated for to obtain the voltage vector switching time after dead-zone compensation.

[0142] Specifically, dead-time compensation is performed on the voltage vector switching time after narrow pulse suppression based on the dead-time compensation coefficient using formula (10), which is expressed as:

[0143]

[0144] In the formula, Ti′ is the voltage vector switching time after dead zone compensation, Ii is the three-phase current, i=a,b,c.

[0145] The dead zone compensation process introduces a low-frequency dead zone compensation strategy. Through symmetrical compensation, it reduces the impact of the dead zone effect on the low-frequency range of the motor, improves the motor's starting torque, and reduces low-frequency oscillation.

[0146] S4. Based on the position of the large sector, assign values ​​to the duty cycle of the six PWM channels according to the voltage vector switching time after dead-time compensation, and control the on / off state of the power devices.

[0147] Specifically, the duty cycles of the six PWM channels are assigned based on the voltage vector switching time after dead-time compensation, as shown in Table 5. Here, "on" indicates a duty cycle of 100%, "off" indicates a duty cycle of 0%, and Ta1, Ta2, Tb1, Tb2, Tc1, and Tc2 represent the corresponding duty cycles of the six PWM channels.

[0148] Table 5

[0149]

[0150]

[0151] The three-level NPC converter control method of this invention comprehensively considers three issues from a system perspective: midpoint voltage balance, narrow pulse suppression, and dead-time compensation. It rationally addresses these three issues by performing narrow pulse suppression on the voltage vector switching time obtained after compensation using midpoint voltage balance modulation, followed by dead-time compensation on the voltage vector switching time after narrow pulse suppression, and finally assigning values ​​to the duty cycles of the six PWM channels based on the dead-time compensated voltage vector switching time to control the on / off state of the power devices. This significantly reduces the complexity of the software algorithm and significantly improves the stability of the system, saving substantial debugging costs.

[0152] See Figure 8 The present invention also provides a three-level NPC converter control system, comprising:

[0153] Sector partitioning module 1 divides the three-level hexagonal space vector into large and small sectors;

[0154] Large sector position determination module 2 determines the position of large sectors based on three-level voltage vector coordinates;

[0155] The midpoint voltage balance regulation determination module 3 obtains the midpoint voltage balance regulation based on the location of the large sector, the direction of the voltage difference, and the direction of the current.

[0156] The small sector position determination module 4 obtains the center vector coordinates based on the position of the large sector in the three-level hexagonal space vector, and determines the position of the small sector based on the center vector coordinates and the three-level voltage vector coordinates.

[0157] Time compensation module 5 determines the action time of adjacent vectors based on the position of the small sector, and compensates for the action time of the zero vector by using the midpoint voltage balance modulation to obtain the voltage vector switching time;

[0158] Setting module 6 sets the narrow pulse suppression principle based on the determined minimum pulse width;

[0159] Suppression module 7 performs narrow pulse suppression on the voltage vector switching time based on the narrow pulse suppression principle;

[0160] Dead zone compensation module 8 performs dead zone compensation on the voltage vector switching time after the suppression module is processed based on a determined dead zone compensation coefficient to obtain the dead zone compensated voltage vector switching time.

[0161] Control module 9 assigns values ​​to the duty cycle of the six PWM channels based on the position of the large sector and the voltage vector switching time after dead-zone compensation, thereby controlling the on / off state of the power devices.

[0162] In some embodiments, the sector partitioning module 1 divides large and small sectors as follows: the entire three-level hexagonal space vector is divided into six adjacent two-level hexagons, defined as six large sectors (1 to 6); each large sector is further divided into six adjacent triangles, defined as six small sectors (1 to 6). The distribution of the three-level space vector after sector partitioning is as follows: Figure 4 As shown.

[0163] In some embodiments, the method by which the large sector position determination module 2 determines the position of the large sector is as follows: based on the three-level voltage vector coordinates in a two-phase stationary coordinate system, the large sector determination variable is calculated according to formula (1); the formula (1) is expressed as:

[0164]

[0165] In the formula, Va, Vb, and Vc are the large sector judgment variables, and Ualpha3 and Ubeta3 are the three-level voltage vector coordinates in the two-phase stationary coordinate system.

[0166] Based on the signs of the variables Va, Vb, and Vc, we can obtain the large sector number and the correspondence between the large sector number and the large sector.

[0167] In some embodiments, the method by which the midpoint voltage balance regulation determination module 3 obtains the midpoint voltage balance regulation based on the large sector location, voltage difference direction, and current direction is as follows: the voltage difference direction is obtained using the voltage difference hysteresis loop according to formula (2), which is expressed as:

[0168]

[0169] In the formula, flag_Un represents the direction of the voltage difference, Udc_error represents the voltage difference between the upper and lower supporting capacitors, and Udc_error_Imt represents the voltage difference hysteresis limit.

[0170] The direction of the current is obtained using the current hysteresis loop according to formula (3), which is expressed as:

[0171]

[0172] In the formula, sign_Ii is the current direction, Ii is the three-phase current, i = a, b, c, and I_Imt is the current hysteresis limit.

[0173] Based on the location of the large sector, the midpoint voltage balance modulation is obtained through formula (4) according to the voltage difference direction and current direction. Formula (4) is expressed as:

[0174] k = s * flag_Un * sign_Ii * ko (4)

[0175] In the formula, k is the midpoint voltage balance modulation degree; s is the compensation direction. When the large sector is 1, 3, or 5, s = 1. When the large sector is 2, 4, or 6, s = -1. ko is the modulation coefficient, with a value range of 0-1.

[0176] In some embodiments, the small sector position determination module 4 obtains the center vector coordinates (Cer_alpha, Cer_beta) based on the position of the large sector in the three-level hexagonal space vector, and determines the position of the small sector based on the center vector coordinates and the three-level voltage vector coordinates.

[0177] Specifically, the method for determining the position of the small sector by the small sector position determination module 4 is as follows: the three-level voltage vector coordinates are converted into two-level voltage vector coordinates using the center vector coordinates through formula (5), which is expressed as:

[0178]

[0179] In the formula, Ualpha2 and Ubeta2 are the two-level voltage vector coordinates, and Cer_alpha and Cer_beta are the center vector coordinates;

[0180] Calculate the small sector judgment variable according to formula (6);

[0181]

[0182] In the formula, Va2, Vb2, and Vc2 are the small sector judgment variables;

[0183] The sign of the variable is used to determine the sector number and the correspondence between the sector number and the sector.

[0184] In some embodiments, the time compensation module 5 determines the action time of adjacent vectors based on the position of the small sector, and compensates for the action time of the zero vector by using the midpoint voltage balance modulation to obtain the voltage vector switching time.

[0185] Specifically, the time compensation module 5 determines the voltage vector switching time as follows: based on the position of the small sector, calculate the adjacent vector action times t1 and t2, and use the midpoint voltage balance modulation to compensate for the vector action time, thereby obtaining the voltage vector switching times Ta, Tb, and Tc for each small sector.

[0186] In some embodiments, the narrow pulse suppression principle set by the setting module 6 based on the determined minimum pulse width is as follows: pulses with a pulse width less than 0.5*Tmin are filtered out, pulses with a pulse width between 0.5*Tmin and 1*Tmin are widened to Tmin, and pulses with a pulse width greater than 1*Tmin are not processed.

[0187] Specifically, the method for setting module 6 to determine the minimum pulse width is as follows: the minimum pulse width of the power device is determined by formula (7) based on the minimum allowable turn-on time and switching cycle of the power device. Formula (7) is expressed as:

[0188]

[0189] In the formula, Tmin is the minimum pulse width of the power device, T_on is the minimum allowable turn-on time of the power device, and T_sw is the switching period.

[0190] In some embodiments, the suppression module 7 performs narrow pulse suppression on the voltage vector switching time based on the narrow pulse suppression principle, as shown in the following equation:

[0191]

[0192] In the formula, Ti can be Ta, Tb, or Tc.

[0193] In some embodiments, the dead-time compensation module 8 performs dead-time compensation on the voltage vector switching time after suppression module processing based on a determined dead-time compensation coefficient using formula (10), which is expressed as:

[0194]

[0195] In the formula, Ti′ is the voltage vector switching time after dead zone compensation, Ii is the three-phase current, i=a,b,c.

[0196] Specifically, when the dead-time compensation module 8 determines the dead-time compensation coefficient, taking the increment / decrement counting mode PWM pulse as an example, such as... Figure 7 As shown. Dead-time compensation is performed on the voltage switching time Ti after the suppression module. Since the PWM pulse is centrally symmetrical, the dead-time compensation adopts unilateral symmetrical compensation. The dead-time compensation coefficient is calculated by formula (9), which is expressed as:

[0197]

[0198] In the formula, compen is the dead zone compensation coefficient, and T_dead is the dead zone time.

[0199] It should be noted that the sector division module 1, the large sector position judgment module 2, the midpoint voltage balance modulation determination module 3, the small sector position judgment module 4, and the time compensation module 5 jointly complete the operation of step S1 in the above three-level NPC converter control method; the setting module 6 and the suppression module 7 jointly complete the operation of step S2 in the above three-level NPC converter control method; the dead zone compensation module 8 performs the operation of step S3 in the above three-level NPC converter control method; and the control module 9 performs the operation of step S4 in the above three-level NPC converter control.

[0200] The three-level NPC converter control system described in this invention comprehensively considers three issues from a system perspective: midpoint voltage balance, narrow pulse suppression, and dead-time compensation. It rationally addresses these three issues by performing narrow pulse suppression on the voltage vector switching time obtained after compensation using midpoint voltage balance modulation, followed by dead-time compensation on the voltage vector switching time after narrow pulse suppression, and finally assigning values ​​to the duty cycles of the six PWM channels based on the dead-time compensated voltage vector switching time to control the on / off state of the power devices. This significantly reduces the complexity of the software algorithm and significantly improves the system stability, saving substantial debugging costs.

[0201] The above embodiments are used to explain the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A three-level NPC converter control method, characterized by, The steps are as follows: S1, Midpoint Voltage Balancing Process The three-level hexagonal space vector is divided into large and small sectors. The method for dividing the large and small sectors is as follows: the entire three-level hexagonal space vector is divided into six adjacent two-level hexagons, which are defined as six large sectors from one to six; each large sector is further divided into six adjacent triangles, which are defined as six small sectors from one to six. Determining the location of large sectors based on three-level voltage vector coordinates; The midpoint voltage balance modulation is obtained based on the location of the large sector, the direction of the voltage difference, and the direction of the current. The direction of the voltage difference is obtained using the voltage difference hysteresis loop according to formula (2), which is expressed as: (2) In the formula, flag _ Un In the direction of voltage difference, Udc _ erro r is the voltage difference between the upper and lower supporting capacitors. Udc _ error _Im t This is the voltage difference hysteresis limit; The direction of the current is obtained using the current hysteresis loop according to formula (3), which is expressed as: (3) In the formula, sign _ Ii In terms of current direction, Ii It is a three-phase current. i = a , b , c , I _Im t This is the current hysteresis limit; The center vector coordinates are obtained based on the position of the large sector in the three-level hexagonal space vector. The location of a small sector is determined based on the center vector coordinates and the three-level voltage vector coordinates. The action time of adjacent vectors is determined based on the location of the small sector, and the action time of the zero vector is compensated by the midpoint voltage balance modulation to obtain the voltage vector switching time. S2, Narrow Pulse Suppression Processing Determine the minimum pulse width; the method for determining the minimum pulse width of the power device is as follows: based on the minimum allowable turn-on time and switching cycle of the power device, the minimum pulse width of the power device is determined by formula (7), which is expressed as: (7) In the formula, T min is the minimum pulse width of the power device. T _ on This refers to the minimum allowable turn-on time for power devices. T _ sw For switching cycles; Narrow pulse suppression is applied to the voltage vector switching time obtained in step S1 based on the narrow pulse suppression principle; narrow pulse suppression is applied to the voltage vector switching time based on the minimum pulse width according to formula (8), which is expressed as: (8) In the formula, Ti The voltage vector switching time after narrow pulse suppression. i = a , b , c ; S3, Dead Zone Compensation Processing Determine the dead zone compensation coefficient; calculate the dead zone compensation coefficient using formula (9), which is expressed as: (9) In the formula, compen This is the dead zone compensation coefficient. T _ dead Dead time; The dead-time compensation is obtained by performing dead-time compensation on the voltage vector switching time after narrow pulse suppression based on the dead-time compensation coefficient; the dead-time compensation is performed on the voltage vector switching time after narrow pulse suppression based on the dead-time compensation coefficient using formula (10), which is expressed as: (10) In the formula, Ti′ The voltage vector switching time after dead-time compensation. Ii It is a three-phase current. i = a , b , c ; S4. Based on the position of the large sector, assign values ​​to the duty cycle of the six PWM channels according to the voltage vector switching time after dead-time compensation, and control the on / off state of the power devices.

2. The three-level NPC converter control method as described in claim 1, characterized in that, In step S1, the method for determining the location of the large sector is as follows: Based on the three-level voltage vector coordinates in a two-phase stationary coordinate system, the large sector judgment variable is calculated according to formula (1); formula (1) is expressed as: (1) In the formula, Va , Vb , Vc For large sector judgment variables, Ualpha 3. Ubeta 3 represents the three-level voltage vector coordinates in a two-phase stationary coordinate system; Variables based on large sectors Va , Vb , Vc The symbols are used to obtain the large sector number and the correspondence between the large sector number and the large sector.

3. The three-level NPC converter control method as described in claim 1, characterized in that, In step S1, the method for obtaining the midpoint voltage balance modulation is as follows: Based on the location of the large sector, the midpoint voltage balance modulation is obtained through formula (4) according to the voltage difference direction and current direction. Formula (4) is expressed as: (4) In the formula, k For midpoint voltage balance modulation; s For compensation direction, when the large sector is 1, 3, or 5, s =1, when the large sector is 2, 4, or 6. s =-1; ko The modulation coefficient has a value range of 0-1.

4. The three-level NPC converter control method as described in claim 3, characterized in that, In step S1, the method for determining the position of the small sector is as follows: Using the center vector coordinates, the three-level voltage vector coordinates are transformed into two-level voltage vector coordinates using formula (5), which is expressed as: (5) In the formula, Ualpha 2. Ubeta 2 represents the two-level voltage vector coordinates. Ualpha 3. Ubeta 3 represents the three-level voltage vector coordinates in a two-phase stationary coordinate system. Cer _ alpha , Cer _ beta Center vector coordinates; Calculate the small sector judgment variable according to formula (6); (6) In the formula, Va 2. Vb 2. Vc 2 is the variable for judging small sectors; The sign of the variable is used to determine the sector number and the correspondence between the sector number and the sector.

5. The three-level NPC converter control method as described in claim 1, characterized in that, In step S1, the voltage vector switching time determination step is as follows: based on the position of the small sector, the action time of adjacent vectors is calculated, and the vector action time is compensated by the midpoint voltage balance modulation to obtain the voltage vector switching time of each small sector.

6. A three-level NPC converter control system, used to implement the three-level NPC converter control method according to any one of claims 1 to 5, characterized in that, include: The sector partitioning module divides the three-level hexagonal space vector into large and small sectors; The large sector location determination module determines the large sector location based on three-level voltage vector coordinates; The midpoint voltage balance regulation determination module obtains the midpoint voltage balance regulation based on the location of the large sector, the direction of the voltage difference, and the direction of the current. The small sector position determination module obtains the center vector coordinates based on the position of the large sector in the three-level hexagonal space vector, and determines the position of the small sector based on the center vector coordinates and the three-level voltage vector coordinates. The time compensation module determines the action time of adjacent vectors based on the position of the small sector, and compensates for the action time of the zero vector by using the midpoint voltage balance modulation to obtain the voltage vector switching time. The setting module sets the narrow pulse suppression principle based on the determined minimum pulse width; The suppression module performs narrow pulse suppression on the voltage vector switching time based on the narrow pulse suppression principle; The dead-zone compensation module performs dead-zone compensation on the voltage vector switching time after the suppression module is processed, based on a determined dead-zone compensation coefficient, to obtain the dead-zone compensated voltage vector switching time. The control module assigns values ​​to the duty cycle of the six PWM channels based on the location of the large sector and the voltage vector switching time after dead-time compensation, thereby controlling the on / off state of the power devices.

Citation Information

Patent Citations

  • Three-level space vector modulation and neutral point balance control method based on vector decomposition

    CN116232105A