Two parallel converter vector timing construction method with duty ratio allocation freedom degree

Abstract

This invention proposes a method for constructing vector timing sequences for two parallel converters with duty cycle allocation degrees of freedom. The method includes: dividing the equivalent vector plane of the parallel converters into 12 basic sectors, each sector containing four basic vectors and their redundant vectors; within each basic sector, selecting four basic vectors and their corresponding redundant vectors to construct a vector timing sequence with duty cycle allocation degrees of freedom; simultaneously, introducing duty cycle allocation constraints between two vectors with the same vector angle, and adjusting the duty cycle allocation according to the position of the reference voltage vector, which can achieve optimization of single-phase circulating current, zero-sequence circulating current, or output current ripple, as well as multi-objective comprehensive optimization; finally, calculating the modulation wave based on the duty cycle allocation coefficients, determining the corresponding carrier comparison relationship and switching operation mode, and generating the corresponding vector timing sequence through carrier modulation. This invention solves the problem of single-objective optimization or multi-objective collaborative optimization of important indicators of two parallel converters.

Description

Technical Field

[0001] This invention relates to methods for optimizing ripple, single-phase circulating current, and zero-sequence circulating current in two parallel converters, and specifically to a method for constructing vector timing sequences for two parallel converters with duty cycle allocation degrees of freedom. Background Technology

[0002] With the continuous growth of high-power demands in modern industry, parallel converters have been widely used in high-power motor drives, grid-connected wind power generation, and other scenarios. Output current ripple, switching losses, and circulating current are key indicators for the reliable operation of converters. Excessive current ripple can cause significant torque pulsation in motor drive systems, leading to mechanical vibration and noise; high switching losses reduce the energy utilization rate of the converter; circulating current causes additional energy loss, and the peak value of the circulating current will affect the design of the filter inductor. Common-mode voltage is crucial to the reliability of parallel converter systems, as it creates voltage stress on the insulating medium, increasing the risk of breakdown. Existing modulation strategies optimized for these high-frequency indicators often focus on performance optimization in one aspect, or sacrifice other key indicators when optimizing a specific indicator, making it difficult to achieve simultaneous performance improvements. Summary of the Invention

[0003] The purpose of this invention is to provide a vector timing construction method for two parallel converters with duty cycle allocation degrees of freedom, which solves the single-objective optimization or multi-objective collaborative optimization problems of important indicators of two parallel converters such as ripple, single-phase circulating current, and zero-sequence circulating current.

[0004] The technical solution to achieve the purpose of this invention is as follows:

[0005] A method for constructing vector timing sequences for two parallel converters with duty cycle allocation degrees of freedom includes:

[0006] The equivalent vector plane of the two parallel converters is evenly divided into 12 basic sectors, each of which contains four basic vectors and their corresponding redundant vectors.

[0007] In each basic sector, four basic vectors and their redundant vectors are selected to construct a vector timing sequence. The four basic vectors include a zero vector, two non-zero vectors with the same angle but different magnitudes, and a third non-zero vector with different magnitudes and angles from the other vectors.

[0008] A duty cycle allocation coefficient is introduced to allocate the duty cycle for two vectors with the same angle but different magnitudes in each vector timing sequence; and the duty cycle allocation coefficient is adjusted according to the position of the reference voltage vector in the basic sector to obtain the optimal duty cycle allocation coefficient, thereby optimizing single-phase circulating current, zero-sequence circulating current or output current ripple, or achieving comprehensive optimization of single-phase circulating current, zero-sequence circulating current and output current ripple;

[0009] Calculate the modulation wave according to the optimal duty cycle allocation coefficient, determine the corresponding carrier comparison relationship and switching action mode, and generate a carrier modulation scheme.

[0010] Furthermore, classify the four basic vectors into long vectors, medium vectors, short vectors, and zero vectors according to the magnitude of the vector modulus. The modulus of the long vector is 2 / 3 of the DC bus voltage, and the modulus of the medium vector is the bus voltage, the modulus of the short vector is 1 / 3 of the bus voltage, and the modulus of the zero vector is zero.

[0011] Furthermore, constructing the vector timing specifically includes: in each basic sector, select four basic vectors and their redundant vectors. First, arrange them in the order of "medium vector - long vector - short vector - zero vector" to form a 4-segment sequence, and use the last zero vector in this 4-segment sequence as the mirror point. Corresponding to select the redundant vector whose zero-sequence circulating current change rate is the opposite of that in the 4-segment sequence. Finally, construct the vector timing in the order of "medium vector - long vector - short vector - zero vector - short vector - long vector - medium vector".

[0012] Furthermore, perform duty cycle allocation on two vectors with the same angle but different magnitudes in each vector timing, specifically including:

[0013] For any two adjacent basic sectors, both include two basic vectors with the same angle but different magnitudes. Introduce a duty cycle allocation coefficient k between the two basic vectors. If k = 0, then in the constructed basic timing, one of the basic vectors does not participate in the action of the vector timing; if k = 1, then in the constructed basic timing, the other basic vector does not participate in the action of the vector timing; if 0 < k < 1, then in the constructed basic timing, the two common basic vectors act together.

[0014] Furthermore, the value range of the duty cycle allocation coefficient k satisfies:

[0015] When a < 0 in the basic sector, if b ≥ 0.5 - 3|a|, the change range of the duty cycle allocation coefficient k is: -1 + 6a / (2b - 1) ≤ k ≤ 1; if b < 0.5 - 3|a|, then the change range of the duty cycle allocation coefficient k is: 0 ≤ k ≤ 1;

[0016] When a > 0 in the basic sector, if b ≥ 0.5 - 3|a|, the change range of the duty cycle allocation coefficient k is: -1 - 6a / (2b - 1) ≤ k ≤ 1; if b < 0.5 - 3|a|, then the change range of the duty cycle allocation coefficient k is: 0 ≤ k ≤ 1;

[0017] Among them, the variables a and b satisfy:

[0018]

[0019] Where u* max u* mid and u* min These represent the DC bus voltage V. DC The per-unit values ​​of the maximum, intermediate, and minimum three-phase voltages are used as a reference. These are the per-unit values ​​of the three-phase reference voltage vectors, respectively.

[0020] Furthermore, the duty cycle allocation coefficient is adjusted according to the position of the reference voltage vector in the basic sector. Specifically, this includes: establishing a time-domain mathematical model of output current ripple, single-phase circulating current, and zero-sequence circulating current, and introducing the duty cycle allocation coefficient into the time-domain mathematical model; obtaining the optimal duty cycle allocation coefficient under different optimization requirements by differentiating or finding the maximum boundary of the time-domain mathematical model with the duty cycle allocation coefficient introduced, and the optimal duty cycle allocation coefficient satisfies the constraint that the action time of the four vectors is greater than or equal to zero.

[0021] Furthermore, the carrier comparison relationship and switching operation mode are as follows:

[0022] When u * mid <0 o'clock:

[0023]

[0024] When u * mid When ≥0:

[0025]

[0026] Among them, u mmax1 and u mmax2 This represents the two modulating waves of the maximum phase of the three-phase voltage; u mmin1 and u mmin2 This represents the two modulating waves of the least phase of a three-phase voltage; u* max u* mid and u* min These represent the DC bus voltage V. DC The per-unit values ​​of the maximum, intermediate, and minimum three-phase voltages are used as a reference; ActionA~ActionE represent the switching operation modes, and k is the duty cycle allocation coefficient.

[0027] A vector timing construction system for two parallel converters with duty cycle allocation degrees of freedom includes:

[0028] The sector partitioning unit evenly divides the equivalent vector plane of the two parallel converters into 12 basic sectors. Each basic sector contains four basic vectors and their corresponding redundant vectors.

[0029] The vector timing construction unit selects four basic vectors and their redundant vectors in each basic sector to construct the vector timing. The four basic vectors include a zero vector, two non-zero vectors with the same angle but different magnitudes, and a third non-zero vector with different magnitudes and angles from the other vectors.

[0030] The duty cycle allocation coefficient solving unit introduces a duty cycle allocation coefficient to allocate the duty cycle for two vectors with the same angle but different magnitudes in each vector timing sequence; and adjusts the duty cycle allocation coefficient according to the position of the reference voltage vector in the basic sector to obtain the optimal duty cycle allocation coefficient, thereby optimizing single-phase circulating current, zero-sequence circulating current or output current ripple, or achieving comprehensive optimization of single-phase circulating current, zero-sequence circulating current and output current ripple;

[0031] The carrier modulation scheme generation unit calculates the modulation wave based on the optimal duty cycle allocation coefficient, determines the corresponding carrier comparison relationship and switching operation mode, and generates the carrier modulation scheme.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] This invention innovatively proposes a vector timing construction method for two parallel converters with a duty cycle allocation degree of freedom. The designed method can achieve targeted optimization of the performance of two parallel converters, such as ripple, zero-sequence circulating current, and single-phase circulating current, as well as comprehensive optimization of multiple objectives. Starting from increasing the degree of freedom in vector timing construction, the basic timing is constructed through four basic vectors. A duty cycle allocation coefficient is introduced between vectors with different magnitudes but the same vector angle. Based on different reference voltage positions, the action time of each vector is flexibly allocated without changing the vector synthesis relationship. The optimal allocation method is finally determined according to the optimization objective, and then the final optimal modulation region is determined, thus achieving comprehensive optimization of various performance aspects. Attached Figure Description

[0034] Figure 1 This is a topology diagram of two parallel converters.

[0035] Figure 2 This is the equivalent vector plane diagram of two parallel converters and the diagram of 12 basic sectors.

[0036] Figure 3 This is a diagram showing the sub-sectors within the 12 basic sectors.

[0037] Figure 4 This is a partition map for the first basic sector, which requires different ranges of allocation coefficients.

[0038] Figure 5The optimal allocation coefficient distribution for the first and second basic sectors, optimized separately for ripple, and the corresponding numerical variation diagram.

[0039] Figure 6 The graph shows the variation of the peak values ​​of zero-sequence circulation and single-phase circulation with the distribution coefficient.

[0040] Figure 7 The diagram shows the optimal distribution of ripple and circulation for the first and second basic sectors, as well as the required action mode.

[0041] Figure 8 A flowchart for achieving comprehensive optimization of ripple and circulation.

[0042] Figure 9 The simulation results of the proposed scheme, MDPWM, and ISVM are shown in the figure when the modulation index is 0.7.

[0043] Figure 10 The graph shows the RMS ripple and THD curves for the three modulation methods across the entire modulation range.

[0044] Figure 11 The graphs show the zero-sequence circulating current and single-phase circulating current curves for the three modulation methods across the entire modulation range. Detailed Implementation

[0045] The technical methods in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] A method for constructing vector timing sequences for two parallel converters with duty cycle allocation degrees of freedom, specifically including:

[0047] (1) The topology of two parallel converters is as follows Figure 1 As shown, where L s R s As parallel inductors and parallel resistors, L1, R1 and L2, R2 are the bridge-side inductance and parasitic resistance on the first and second converters, respectively. a i b i c Indicates the three-phase parallel current, i a1 i b1 i c1 and i a2 i b2 i c2 These represent the three-phase currents on the first and second converters, respectively. a e b e cThis represents the grid-connected three-phase voltage, V. DC The voltage is the DC bus voltage, where the inductance on the bridge arm side generally satisfies L1=L2=L and R1=R2=R.

[0048] Based on Kirchhoff's laws, the expression for the equivalent output voltage of two parallel converters in a two-phase stationary coordinate system can be derived:

[0049] (1)

[0050] Among them, u α u β These are the components of the equivalent output voltage in the two-phase stationary coordinate system, S. a1 S b1 S c1 S a2 S b2 S c2 These represent the switching states of the switching transistors on the first and second converters, respectively. When the upper transistor of the corresponding phase on each converter is turned on, S... x =1, when the down transistor is on, S x =0. Therefore, for a single converter, there are a total of 8 switching combinations, and two parallel converters have 8... 2 =64 switch combinations, which can ultimately form 19 equivalent voltage output vectors. Based on this, an equivalent vector plane diagram is drawn, such as Figure 2 As shown in Table 1, the relationship between the equivalent output voltage vector of the two parallel converters and the switching combination is shown in Table 1.

[0051] Table 1. Equivalent Output Voltage Vectors of Two Parallel Converters

[0052]

[0053] All the basic voltage vectors in Table 1 are defined by their vector magnitudes in a two-dimensional plane. The 19 different basic voltage vectors include one zero vector (V0), six long vectors (V1~V6), and six medium vectors (V7~V6). 12 ) and 6 short vectors (V 13 ~V 18 ).

[0054] In a two-parallel converter, 64 voltage vector combinations correspond to only 19 different equivalent output voltages in terms of output voltage vector amplitude and direction. However, different vector combinations exhibit different characteristics in terms of single-phase circulating current, common-mode voltage, and zero-sequence circulating current. Based on Kirchhoff's laws, the expression for the rate of change of single-phase circulating current can be derived as follows:

[0055] (2)

[0056] Where iipcc,a i ipcc,b i ipcc,c This represents a single-phase circulating current in the three phases A, B, and C. Simultaneously, the zero-sequence circulating current (i...) zscc The sum of the three-phase single-phase circulating currents is equal to the sum of the three-phase single-phase circulating currents, expressed

[0057] (3)

[0058] According to the above formula, the rate of change of single-phase circulating current is related to the vector combination. Table 2 shows the rate of change of single-phase circulating current corresponding to 64 vector combinations. In the table, V... xyz This table represents the output voltage vector of a single transformer. x, y, and z represent the switching states of phases A, B, and C, respectively. x = 1 indicates that the upper transistor of phase A is on and the lower transistor is off; x = 0 indicates that the upper transistor of phase A is off and the lower transistor is on. The "+", "-", and "0" in the table represent the rate of change of the single-phase circulating current. When converter #1 uses vector V... 100 #2 converter adopts vector V 000 The corresponding single-phase circulating current change rate "+ / 0 / 0" indicates that the circulating current change rate of phase A is +, while the circulating current change rates of phases B and C are 0. In the table, "+3", "-3", "+2", "-2", "+1", "-1", and "0" represent the zero-sequence circulating current change rate corresponding to different vector combinations. For example, when converter #1 uses vector V... 100 #2 converter adopts vector V 000 This corresponds to a zero-sequence circulation change rate of "+1".

[0059] Table 2. Mapping relationship between switch state combinations and single-phase circulating current and zero-sequence circulating current.

[0060]

[0061] Based on the above differences, while keeping the output voltage constant, the performance indicators can be optimized by selecting different vector combinations. For example, vector V 13 It can correspond to V 100 / V 000 V 000 / V 100 V 110 / V 101 V 101 / V 110 V 111 / V 100 V 100 / V 111 Six vector combinations. Using V... 110 / V 101 For example, where V 110 Indicates the three-phase switch status of the first converter, V 101This indicates the three-phase switching status of the second converter. Although the above six vector combinations have the same output voltage, they differ in characteristics such as single-phase circulating current and zero-sequence circulating current, as shown in the switching combination V. 100 / V 000 The corresponding A-phase circulating current change rate is "+", the B-phase circulating current change rate is "0", and the C-phase circulating current change rate is "0". This vector combination only introduces a single-phase circulating current in phase A; while the switch combination V 111 / V 100 The corresponding A-phase circulating current change rate is "0", the B-phase circulating current change rate is "+", and the C-phase circulating current change rate is "+". This vector combination introduces single-phase circulating current in both phases B and C. Regarding the zero-sequence circulating current, the switch combination V... 100 / V 000 The corresponding zero-sequence circulating current change rate is "+1"; while the switching combination V 100 / V 111 The corresponding zero-sequence circulating current change rate is "-2" compared to the switching combination V. 100 / V 000 Larger.

[0062] This method divides the complete equivalent voltage vector plane into 12 basic sectors. The 12 basic sectors correspond to the following angle ranges: basic sector 1 is 0°~30°, basic sector 2 is 30°~60°, basic sector 3 is 60°~90°, basic sector 4 is 90°~120°, basic sector 5 is 120°~150°, basic sector 6 is 150°~180°, basic sector 7 is 180°~210°, basic sector 8 is 210°~240°, basic sector 9 is 240°~270°, basic sector 10 is 270°~300°, basic sector 11 is 300°~330°, and basic sector 12 is 330°~360°.

[0063] Each basic sector contains four basic vectors and their corresponding redundant vectors, specifically: Basic sector 1 contains the basic vectors {V0, V...} 13 The basic sector 2 contains the basic vectors {V0, V1, V7}. 14 The basic sector 3 contains the basic vectors {V0, V2, V7}. 14 The basic sector 4 contains the basic vectors {V0, V2, V8}. 15 The basic sector 5 contains the basic vectors {V0, V3, V8}. 15 The basic sector 6 contains the basic vectors {V0, V3, V9}. 16 The basic sector 7 contains the basic vectors {V0, V4, V9}. 16 V4, V 10}, basic sector 8 contains basic vectors {V0, V... 17V5, V 10}, basic sector 9 contains basic vectors {V0, V... 17 V5, V 11}, basic sector 10 contains basic vectors {V0, V... 18 V6, V 11}, basic sector 11 contains basic vectors {V0, V... 18 V6, V 12}, basic sector 12 contains basic vectors {V0, V... 13 V1, V 12 Each basic vector may correspond to one or more redundant vectors, as shown in Table 1, which illustrates the correspondence between basic voltage vectors and switch combinations. The complete vector plane is divided into 12 basic sectors, as follows: Figure 2 As shown.

[0064] For the 12 basic sectors mentioned above, each basic sector can be divided into two sub-sectors, such as... Figure 3 The details are as follows:

[0065] Basic sector 1 with {V 13 The region formed by the three vectors {V1, V7} as vertices is sub-sector 1-1, and {V} is defined as sub-sector 1-1. 13 The region formed by the three vectors {V0, V7} as vertices is sub-sector 1-2;

[0066] Basic sector 2 with {V 14 The region formed by the three vectors {V0, V2, V7} as vertices is sub-sector 2-1, with {V0, V1, V2, V7} as vertices. 14 The region formed by the three vectors V7} as vertices is sub-sector 2-2;

[0067] Basic sector 3 with {V 14 The region formed by the three vectors {V8, V2} as vertices is sub-sector 3-1, with {V... 14 The region formed by the three vectors {V0, V8} as vertices is a sub-sector 3-2;

[0068] Basic sector 4 with {V 15 The region formed by the three vectors {V0, V3, V8} as vertices is sub-sector 4-1, with {V0, V... 15 The region formed by the three vectors V8 as vertices is sub-sector 4-2;

[0069] Basic sector 5 with {V 15 The region formed by the three vectors {V9, V3} as vertices is sub-sector 5-1, with {V 15 The region formed by the three vectors {V0, V9} as vertices is a sub-sector 5-2;

[0070] Basic sector 6 uses {V16 The region formed by the three vectors {V0, V4, V9} as vertices is sub-sector 6-1, with {V0, V...} as the vertices. 16 The region formed by the three vectors V9} as vertices is sub-sector 6-2;

[0071] Basic sector 7 with {V 16 V4, V 10 The region formed by the three vectors as vertices is sub-sector 7-1, with {V} 16 V0, V 10 The region formed by the three vectors as vertices is sub-sector 7-2;

[0072] Basic sector 8 uses {V 17 V5, V 10 The region formed by the three vectors as vertices is sub-sector 8-1, with {V0, V...} 17 V 10 The region formed by the three vectors as vertices is sub-sector 8-2;

[0073] Basic sector 9 uses {V 17 V5, V 11 The region formed by the three vectors as vertices is sub-sector 9-1, with {V} 17 V0, V 11 The region formed by the three vectors as vertices is sub-sector 9-2;

[0074] Basic sector 10 with {V 11 V 18 The region formed by the three vectors {V0, V6} as vertices is sub-sector 10-1, with {V0, V6} as vertices. 18 V 11 The region formed by the three vectors as vertices is a sub-sector 10-2;

[0075] Basic sector 11 with {V 18 V 12 The region formed by the three vectors {V6} as vertices is sub-sector 11-1, with {V 18 V0, V 12 The region formed by the three vectors as vertices is sub-sector 11-2;

[0076] Basic sector 12 with {V 13 V1, V 12 The region formed by the three vectors as vertices is sub-sector 12-1, with {V0, V...} 12 V 13 The region formed by the three vectors as vertices is sub-sector 12-2;

[0077] (2) Based on the 12 basic sectors divided in step 1 and their contained basic and redundant vectors, according to the different performance characteristics corresponding to different vector combinations, four basic vectors and their corresponding redundant vectors are selected for each basic sector. Taking the medium vector as the starting vector, the corresponding vector timing sequence composed of the four vectors is constructed through the arrangement of "medium vector-long vector-short vector-zero vector-short vector-long vector-medium vector". The specific timing sequence constructed for each basic sector is as follows:

[0078] The vector timing sequence constructed in basic sector 1 is as follows:

[0079] In the vector sequence constructed in basic sector 1, vector V 13 The vector V1 has the same angle but a different magnitude. V0 is a zero vector. The vector V7 has a different angle and magnitude from the other vectors.

[0080] The vector timing sequence constructed in basic sector 2 is as follows:

[0081] Where vector V 14 The vector V2 has the same angle but a different magnitude. V0 is a zero vector. The vector V7 has a different angle and magnitude from the other vectors.

[0082] The vector sequence constructed in basic sector 3 is as follows:

[0083] Where vector V 14 The vector V2 has the same angle but a different magnitude. V0 is a zero vector. The vector V8 has a different angle and magnitude from the other vectors.

[0084] The vector sequence constructed in basic sector 4 is as follows:

[0085] Where vector V 15 The vector V0 has the same angle as V3 but a different magnitude. V0 is a zero vector. The vector V8 has a different angle and magnitude than the other vectors.

[0086] The vector sequence constructed in basic sector 5 is as follows:

[0087] Where vector V 15 The vector V9 has the same angle as V3 but a different magnitude. V0 is a zero vector. The vector V9 has a different angle and magnitude than the other vectors.

[0088] The vector sequence constructed in basic sector 6 is as follows:

[0089] Where vector V 16The vector V4 has the same angle but a different magnitude. V0 is a zero vector. Vector V9 has a different angle and magnitude from the other vectors.

[0090] The vector sequence constructed in basic sector 7 is as follows:

[0091] Where vector V 16 The vector V0 has the same angle as V4 but a different magnitude. V0 is a zero vector. 10 It differs from other vectors in both angle and magnitude.

[0092] The vector sequence constructed in basic sector 8 is as follows:

[0093] Where vector V 17 The vector angle is the same as V5, but the magnitude is different. V0 is a zero vector, and the vector V... 10 It differs from other vectors in both angle and magnitude.

[0094] The vector sequence constructed in basic sector 9 is as follows:

[0095] Where vector V 17 The vector angle is the same as V5, but the magnitude is different. V0 is a zero vector, and the vector V... 11 It differs from other vectors in both angle and magnitude.

[0096] The vector sequence constructed in basic sector 10 is as follows:

[0097] Where vector V 18 The vector angle is the same as V6, but the magnitude is different. V0 is a zero vector, and the vector V... 11 It differs from other vectors in both angle and magnitude.

[0098] The vector sequence constructed in basic sector 11 is as follows:

[0099] Where vector V 17 The vector angle is the same as V5, but the magnitude is different. V0 is a zero vector, and the vector V... 12 It differs from other vectors in both angle and magnitude.

[0100] The vector sequence constructed in basic sector 12 is as follows:

[0101] Where vector V 18 The vector angle is the same as V6, but the magnitude is different. V0 is a zero vector, and the vector V... 12 It differs from other vectors in both angle and magnitude.

[0102] (3) Based on the vector time series constructed for each basic sector in step 2 and the characteristics of the basic vectors included in the vector time series, set the duty cycle distribution coefficient k, and perform duty cycle distribution on the vectors with the same angle but different modulus values in each vector time series. The specific distribution method is as follows:

[0103] For basic sectors 1 and 12, both basic sectors contain the basic vectors V 13 and V1, whose modulus values are different but the corresponding vector angles are both 0°. Introduce the duty cycle distribution coefficient k between the basic vectors V 13 and V1. If k = 0, then in the constructed basic time series, the vector V1 does not participate in the action of the vector time series; if k = 1, then in the constructed basic time series, the vector V 13 does not participate in the action of the vector time series; if 0 < k < 1, then in the constructed basic time series, the vectors V 13 and V1 act together.

[0104] For basic sectors 2 and 3, both basic sectors contain the basic vectors V 14 and V2, whose modulus values are different but the corresponding vector angles are both 60°. Introduce the duty cycle distribution coefficient k between the basic vectors V 14 and V2. If k = 0, then in the constructed basic time series, the vector V2 does not participate in the action of the vector time series; if k = 1, then in the constructed basic time series, the vector V 14 does not participate in the action of the vector time series; if 0 < k < 1, then in the constructed basic time series, the vectors V 14 and V2 act together.

[0105] For basic sectors 4 and 5, both basic sectors contain the basic vectors V 15 and V3, whose modulus values are different but the corresponding vector angles are both 120°. Introduce the duty cycle distribution coefficient k between the basic vectors V 15 and V3. If k = 0, then in the constructed basic time series, the vector V3 does not participate in the action of the vector time series; if k = 1, then in the constructed basic time series, the vector V 15 does not participate in the action of the vector time series; if 0 < k < 1, then in the constructed basic time series, the vectors V 15 and V3 act together.

[0106] For basic sectors 6 and 7, both basic sectors contain the basic vectors V 16 and V4, whose modulus values are different but the corresponding vector angles are both 180°. Introduce the duty cycle distribution coefficient k between the basic vectors V 16 and V4. If k = 0, then in the constructed basic time series, the vector V4 does not participate in the action of the vector time series; if k = 1, then in the constructed basic time series, the vector V 16Does not participate in the action of the vector timing; if 0 < k < 1, then in the constructed basic timing, the vector V 16 and V4 act together.

[0107] For basic sectors 8 and 9, both basic sectors contain the basic vectors V 17 and V5, whose magnitudes are different but the corresponding vector angles are both 240°. A duty ratio distribution coefficient k is introduced between the basic vectors V 17 and V5. If k = 0, then in the constructed basic timing, the vector V5 does not participate in the action of the vector timing; if k = 1, then in the constructed basic timing, the vector V 17 does not participate in the action of the vector timing; if 0 < k < 1, then in the constructed basic timing, the vectors V 17 and V5 act together.

[0108] For basic sectors 10 and 11, both basic sectors contain the basic vectors V 18 and V6, whose magnitudes are different but the corresponding vector angles are both 300°. A duty ratio distribution coefficient k is introduced between the basic vectors V 18 and V6. If k = 0, then in the constructed basic timing, the vector V6 does not participate in the action of the vector timing; if k = 1, then in the constructed basic timing, the vector V 18 does not participate in the action of the vector timing; if 0 < k < 1, then in the constructed basic timing, the vectors V 18 and V6 act together.

[0109] In different basic sectors, the duty ratio distribution between two basic vectors with the same vector angle but different magnitudes is achieved through the introduced duty ratio distribution coefficient k. At the same time, considering that the action time of each vector must be greater than zero, the introduced duty ratio distribution coefficient k needs to satisfy a certain range of variation limits. Taking the first basic sector as an example, specifically:

[0110] According to Figure 4 the shown boundary, when the reference voltage is in sub-sector 1 - 2 of the first basic vector, the distribution coefficient k can take any value between 0 and 1, and the action time of any vector is greater than or equal to 0.

[0111] When the reference voltage is in sub-sector 1 - 1, to meet the above requirements, based on the defined variation method of the duty ratio constraint, according to the volt-second balance:

[0112] (4)

[0113] where, V 13α 、V 7α 、V 1α represent the components of the voltage vectors V 13 、V7、V1 on the α-axis, V 7βu represents the component of vector V7 on the β axis. rα u rβ The components of the reference voltage in the two-dimensional equivalent voltage plane are d7, d1, and d2. 13 And the corresponding synthesized voltage vectors V7, V1, V0. 13 The duty cycle of V0.

[0114] Solving for the given information yields:

[0115] (5)

[0116] The calculated duty cycle result is based on the boundary conditions of the first basic sector, and since the duty cycle allocation coefficient k always satisfies 0... <k<1,d7、d1、d 13 It is always greater than or equal to 0. However, the duration of action of vector V0 may be less than zero. Therefore, by solving for the boundary values ​​of d0, we can obtain the minimum value k required by the distribution coefficient k within this region. min1 Furthermore, based on the conversion relationship between the two-dimensional equivalent output voltage vector and the three-phase voltage, the expression for the minimum value of the distribution coefficient can be transformed into an expression for the three-phase voltage:

[0117] (6)

[0118] Where u α u β u represents the component of the equivalent output voltage in a two-phase stationary coordinate system. * b u * c This is the per-unit value of the reference voltage for phases B and C, with a base value of V. DC .

[0119] For the remaining 11 basic sectors, according to Figure 3 The same derivation and calculation can be performed on the divided regions. Based on the calculation results, the required range of the reference voltage distribution coefficient k in different sub-regions is expressed in a unified manner, as follows:

[0120] By comparing the magnitudes of the three-phase reference vector voltages in each basic sector, new variables a and b are introduced. These new variables satisfy the following expression:

[0121] (7)

[0122] in , , , These are the per-unit values ​​of the three-phase reference voltage vectors, with a per-unit reference value of V. DC .

[0123] When a < 0 in the basic sector, if b ≥ 0.5 - 3|a|, the range of the allocation coefficient k is: -1 + 6a / (2b - 1) ≤ k ≤ 1; if b < 0.5 - 3|a|, the range of the allocation coefficient k is: 0 ≤ k ≤ 1.

[0124] When a>0 in the basic sector, if b≥0.5-3|a|, the distribution coefficient k varies in the range of -1-6a / (2b-1)≤k≤1; if b<0.5-3|a|, the distribution coefficient k varies in the range of 0≤k≤1.

[0125] (4) Based on the vector sequence constructed for each basic sector in step 2, and combined with the duty cycle allocation coefficient k introduced in step 3 and its corresponding regional variation range requirements, the comprehensive optimization of the ripple, zero-sequence circulating current and single-phase circulating current performance of the two parallel converters can be achieved, as follows:

[0126] The ripple expression for two parallel converters is defined as follows:

[0127] (8)

[0128] Where, Δi ripa , Δi ripb , Δi ripc Δi represents the rate of change of the three-phase current ripple of phases A, B, and C. ripα , Δi ripβ Δu represents the rate of change of current ripple on the two-dimensional equivalent plane. α and Δu β The voltage error vector, formed by the difference between the actual output voltage vector and the reference voltage vector, is represented by its components on the α and β axes, and Δt represents the time period during which a certain vector acts.

[0129] According to equation (8), the output current ripple is proportional to the integral of the voltage error vector with respect to time. Therefore, this method dynamically adjusts the duty cycle allocation coefficient k according to the different positions of the reference voltage, increases the action time of the basic vector with a smaller deviation from the reference voltage, and correspondingly shortens the action time of the basic vector with a larger deviation from the reference voltage, thereby achieving the purpose of optimizing the current ripple. Taking the sequence SQ1 constructed in the first basic sector as an example, the specific calculation and derivation process is as follows:

[0130] (9)

[0131] The voltage difference and time intervals in each segment satisfy the following expression:

[0132] (10)

[0133] Among them, V 0α V 13α V 7α, V 1α denote the components of voltage vectors V0, V 13 , V7, V1 on the α-axis, V 0β , V 13β , V 7β , V 1β denote the components of vectors V0, V 13 , V7, V1 on the β-axis, d7, d1, d 13 and the action times of d0 corresponding to the synthesized voltage vectors V7, V1, V 13 , V0 are the same as the specific expression in Equation (5). Where u α and u β are the components of the equivalent output voltage in the two-phase stationary coordinate system, and are related to the three-phase voltages u a , u b , u c satisfy the following relationship:

[0134] (11)

[0135] Substituting Equation (5) can convert the duty cycle of each vector action into an expression about k and the three-phase voltages u a , u b , u c :

[0136] (12)

[0137] Substituting Equations (9) to (12) into Equation (8) can obtain the expression of the effective value of the ripple of sequence SQ1:

[0138] (13)

[0139] By solving the minimum value of Equation (13), the expression of the duty cycle distribution coefficient k1 that theoretically optimizes the ripple performance can be obtained:

[0140] (14)

[0141] However, this expression is derived without considering the range limitation of the distribution coefficient change. According to the analysis in Step 3, when the reference voltage is in different regions, the actual value range of the distribution coefficient needs to meet the corresponding requirements. Therefore, to determine the distribution coefficient with the optimal ripple at any point in the first basic sector, it is necessary to compare the boundaries of the distribution coefficient change range required for different regions in each basic sector calculated in Step 3 with k1. The specific comparison process is as follows:

[0142] When the reference voltage is in Region 1-2, if k1 < 0, then k opt takes 0, if k1 > 1, then k opt takes 1. When 0 < k1 < 1, kopt = k1; When the reference voltage is in Region 1-1, if k1 < 0, then k opt Take k min1 , if k1 > 1, then k opt Take 1. If 0 < k1 < 1 and k1 < k min1 , then k opt = k min1 obtains the minimum value; if k min1 < k1 < 1, then k opt = k1 takes the minimum. Where k min1 represents the minimum value that the k value of the corresponding region obtained in Step 3 needs to satisfy.

[0143] According to the same method, the same derivation is performed on the sequence SQ2 constructed for the second basic sector, and the corresponding ripple effective value results for this region are as follows:

[0144] (15)

[0145] Solving Equation (15), the duty ratio distribution coefficient expression k2 that theoretically optimizes the ripple performance of the second basic sector can also be obtained:

[0146] (16)

[0147] Using a comparison process similar to that of the first basic sector, the distribution coefficient value that optimizes the ripple at any point within the second basic sector can also be determined. Combining the comparison results of the first and second basic sectors in the 0~60° range of the complete vector plane, the distribution of the duty ratio distribution coefficient k expression that optimizes the ripple at any point within the region is as shown in Figure 5 (a). Among them, the optimal distribution coefficients for sub-sectors 1-1 and 2-1 will be determined by the expressions k min1 and k min2 , and their values will change according to the change of the reference voltage position. Within sub-sectors 1-2 and 2-2, the optimal duty ratio distribution coefficient is always 0.

[0148] Figure 5 (b) in further shows the specific numerical change trend of the optimal distribution coefficient within each sub-sector. It can be seen from the figure that when the reference voltage is in sub-regions 1-2 and 2-2, the difference between the reference voltage and the short vector is the smallest, and the k value is always 0; when the reference voltage is in sub-regions 1-1 and 2-1, when the reference voltage is closer to the short vector, the optimal k value is smaller, that is, the proportion of the short vector action time is larger; on the contrary, when the reference voltage is closer to the long vector, the optimal k value is larger, that is, the proportion of the long vector action time is larger. This distribution result is consistent with the characteristics shown in the ripple definition formula in Equation (8), verifying that the method proposed in this invention can achieve optimization for ripple performance.

[0149] Meanwhile, based on the aforementioned analysis, the value of the duty cycle allocation coefficient k not only affects the converter's output current ripple performance but also influences the peak values ​​of single-phase circulating current and zero-sequence circulating current. Selecting a k value solely for ripple optimization may lead to a deterioration in circulating current peak values. To avoid sacrificing circulating current performance while optimizing ripple, a multi-objective collaborative optimization strategy needs to be further derived. Therefore, this method, based on ripple optimization, further introduces circulating current constraints to derive the optimal distribution of the duty cycle allocation coefficient for achieving comprehensive optimization of both ripple and circulating current.

[0150] First, the relationship between the zero-sequence circulation peak value and the distribution coefficient k is derived. Taking the sequence SQ1 constructed in the first basic sector as an example, this sequence satisfies T s / 4 symmetry, in front T s The zero-sequence circulation expression of / 4 is as follows:

[0151] (17)

[0152] Where t represents the time corresponding to the current moment, t7, t1, t... 13 And t0 represents the voltage vectors V7, V1, and V. 13 The duration of V0. According to equation (17), there are two possible zero-sequence circulation peaks in the first basic sector, at t=t7 / 4 or t=(t1+t7+t). 13 The value is obtained at d7 / 4, and according to its corresponding zero-sequence circulation expression, the peak value of the zero-sequence circulation of the SQ1 sequence can be obtained by comparing d7 / 4 and d7 / 4. 13 The expression / 4-d7 / 4 is determined. Therefore, the following two expressions are introduced:

[0153] (18)

[0154] (19)

[0155] i zscc-peak1 and i zscc-peak2 The two introduced expressions are used to compare the maximum value of the zero-sequence circulating current within different ranges of allocation coefficients; meanwhile, M and θ in the formulas represent the modulation index and the vector angle of the target voltage in the equivalent vector plane diagram, and their relationship with the target voltage vector in the two-phase stationary coordinate system is as follows:

[0156] (20)

[0157] For equation (18), the peak value is obtained when θ = π / 6; while for equation (19), the peak value is obtained when θ = 0. The expressions for the two possible zero-sequence circulation peak values ​​of this sequence with respect to the modulation index M are respectively given by equations (21) and (22):

[0158] (twenty one)

[0159] (twenty two)

[0160] Based on equations (21) and (22), we can obtain the following: Figure 6 The figure shows the variation of the zero-sequence circulation peak value of the sequence SQ1. Curve AB corresponds to equation (21) and is independent of the value of the distribution coefficient k. The figure also shows the curves of equation (22) under specific modulation conditions such as 0.2, 0.7, and 0.9. It can be seen from the figure that the zero-sequence circulation peak value of the sequence SQ1 is determined by the larger of equations (21) and (22). For example, when k=0.2, the curve AC corresponding to equation (22) is greater than the curve AB, and the circulation peak value is determined by the curve AC. When k=0.7 and 0.9, the curves AD and AE corresponding to equation (22) are both less than the curve AB, and the circulation peak value is determined by the curve AB. At the same time, no matter how the value of k changes, the zero-sequence circulation peak value curve of this sequence will not be lower than the curve AB. Based on this analysis, the boundary of equations (21) and (22) can be calculated, and the following piecewise expression of the zero-sequence circulation peak value of the sequence SQ1 can be obtained:

[0161] (twenty three)

[0162] The above analysis shows that when The expression for the zero-sequence circulation peak value of the time series SQ1 is determined by equation (22), and the corresponding curve is always greater than AB. To optimize the zero-sequence circulation peak value, when timing k= At this point, equations (21) and (22) are the same expression. Therefore, within this range, the zero-sequence circulation peak value also becomes equation (21). Consequently, at any point, the zero-sequence circulation peak value corresponding to this sequence is the theoretical minimum value. Based on this, the remaining 11 basic sectors can be derived to obtain the same circulation optimization boundary value throughout the entire vector plane. According to this boundary condition constraint, this method can achieve the optimization of the zero-sequence circulation.

[0163] Next, the relationship between the peak value of the single-phase circulating current and the distribution coefficient k is derived. Again, taking the sequence SQ1 constructed within the first basic sector as an example, in the first T... s The expression for a three-phase single-phase circulating current is as follows:

[0164] (twenty four)

[0165] Where t represents the time corresponding to the current moment, t7, t1, t... 13 And t0 represents the voltage vectors V7, V1, and V. 13The duration of V0. According to equation (24), there are two possible single-phase circulation peaks in the first basic sector, which may be obtained in phase B at t=t7 / 4 and at t=(t1+t7+t0). 13 The value of d7 / 4 is obtained from the expression for A, and according to its corresponding zero-sequence circulation expression, the peak value of the single-phase circulation of the SQ1 sequence can be obtained by comparing d7 / 4 and d7 / 4. 13 The expression / 4 is determined. Therefore, the following two expressions are introduced:

[0166] (25)

[0167] (26)

[0168] i ipcc,a-peak and i ipcc,b-peak Two expressions are introduced to compare the maximum single-phase circulation values ​​within different ranges of distribution coefficients. For equation (25), the peak value is obtained when θ=0; while for equation (26), the peak value is obtained when θ=π / 6. The expressions for the two possible single-phase circulation peak values ​​of this sequence with respect to the modulation index M are respectively expressed as equations (27) and (28):

[0169] (27)

[0170] (28)

[0171] Based on equations (27) and (28), we can obtain the following: Figure 6 The figure shows the peak value variation of the single-phase circulating current of sequence SQ1. Curve AB corresponds to equation (28) and is independent of the value of the distribution coefficient k. The figure also shows the curves of equation (27) under specific modulation conditions such as 0.2, 0.7, and 0.9. It can be seen from the figure that the peak value of the single-phase circulating current of sequence SQ1 is determined by the larger of equations (27) and (28). For example, when k=0.2, the curve AC corresponding to equation (27) is greater than the curve AB, and the peak value of the circulating current is determined by the curve AC. When k=0.7 and 0.9, the curves AD and AE corresponding to equation (27) are less than the curve AB, and the peak value of the circulating current is determined by the curve AB. At the same time, no matter how the value of k changes, the peak value curve of the single-phase circulating current of this sequence will not be lower than the curve AB. Based on this analysis, the boundary of equations (27) and (28) can be calculated, and the following piecewise expression of the peak value of the single-phase circulating current of sequence SQ1 can be obtained:

[0172] (29)

[0173] The above analysis shows that when The expression for the peak value of the single-phase circulating current in time series SQ1 is determined by equation (27), and the corresponding curve is always greater than AB. To optimize the peak value of the single-phase circulating current, when timing k= At this point, equations (27) and (28) are the same expression. Therefore, within this range, the expression for the single-phase circulating peak value also becomes equation (28). Consequently, at any point, the single-phase circulating peak value corresponding to this sequence is the theoretical minimum value. Based on this, the remaining 11 basic sectors are derived to show that the same single-phase circulating peak value boundary value is followed throughout the entire vector plane. According to this boundary condition constraint, this method can also achieve optimization of single-phase circulating flow.

[0174] The above analysis determined both the optimal distribution of the allocation coefficient k for ripple and the boundary conditions for suppressing the peak values ​​of zero-sequence and single-phase circulating currents. To ensure that the allocation coefficient at any point satisfies the comprehensive optimization of ripple, single-phase circulating current, and zero-sequence circulating current, taking the first and second basic sectors (0-60°) in the complete vector plane diagram as an example, it is necessary to... Figure 5 The boundary map showing the optimal allocation coefficient values ​​for ripple optimization is used for correction. Specifically, the correction method is as follows: when the reference voltage is at a certain point, the optimal allocation coefficient value for ripple is less than... Let k opt = When the reference voltage at a certain point makes the optimal ripple distribution factor greater than or equal to... Then k opt The optimal value remains unchanged.

[0175] The optimal allocation coefficient partitioning can achieve comprehensive optimization of converter ripple, zero-sequence circulating current and single-phase circulating current, such as... Figure 7 As shown, the optimal vector sequences for the first and second basic sectors are SQ1 and SQ2, respectively. However, within the same basic sector, the reference voltage is located at different points, and the duration of each vector action varies according to the distribution boundary of the optimal k value. Based on the new variables a and b introduced in equation (7), their corresponding boundaries and the boundary expressions of each sub-region are as follows:

[0176] (30)

[0177] (31)

[0178] Based on the derived expressions for the ranges of each sub-sector within the basic sector I, the expression can be extended to the full vector plane diagram by comparing the magnitudes of the three-phase reference vector voltages and using the introduced new variables a and b. When a < 0 in the basic sector, the expression is as follows:

[0179] (32)

[0180] When a > 0 in the basic sector, the expression is as follows:

[0181] (33)

[0182] (5) Based on the distribution boundary of the optimal allocation coefficient k value for the comprehensive optimization of ripple, zero-sequence circulating current and single-phase circulating current described in step 4, calculate the modulation wave expression corresponding to each switching sequence, determine the switching action mode corresponding to the modulation wave, and generate a carrier modulation scheme. Specifically, this includes:

[0183] Taking the first and second basic sectors as an example, based on the basic vector combination of timing constructed in step 2, the duty cycle of each vector in the first basic sector is:

[0184] (34)

[0185] The duty cycles of each vector in the second basic sector are:

[0186] (35)

[0187] in, These are the per-unit values ​​of the three-phase reference voltage vectors, with a per-unit reference value of V. DC .

[0188] Based on the duty cycle expressions of the aforementioned vector sets, it is necessary to design a magnitude relationship between the modulation wave and the carrier wave, and to design corresponding switching operation modes to generate the target timing sequence. To generate the target timing sequence, this invention first proposes five switching operation modes, such as... Figure 7 As shown.

[0189] For a three-phase voltage that switches twice within one cycle, two modulated waves are compared simultaneously with a carrier wave. The expressions for the two modulated waves satisfy 0. mx1 <1 and 0 mx2 <1andu mx1 >u mx2 It has four action modes, as follows:

[0190] ActionA: Compared with the falling edge of the carrier wave, the modulated wave u mx1 ≥ carrier and u mx2 When the carrier wave is ≤, the drive signal outputs a high level; otherwise, it outputs a low level.

[0191] ActionB: Compared with the falling edge of the carrier wave, the modulated wave u mx1 ≥ carrier and u mx2 When the carrier wave is ≤, the drive signal outputs a low level; otherwise, it outputs a high level.

[0192] ActionC: Compared with the rising edge of the carrier wave, the modulated wave u mx1 ≥ carrier and u mx2 When the carrier wave is ≤, the drive signal outputs a high level; otherwise, it outputs a low level.

[0193] ​​ActionD: Compared with the rising edge of the carrier wave, the modulated wave u mx1 ≥ carrier and u mx2 When the carrier wave is ≤, the drive signal outputs a low level; otherwise, it outputs a high level.

[0194] For a phase in a three-phase voltage cycle that switches once, a modulation wave and a carrier wave are compared. The expression of the modulation wave satisfies 0. mx <1, the carrier corresponds to an inverted triangular carrier with an amplitude range of 0 to 1, and it has one operating mode, as follows:

[0195] ActionE: If the modulated wave u mx If the carrier wave is ≥, the drive signal outputs a high level; otherwise, it outputs a low level.

[0196] To generate the optimal switching timing derived above, based on the proposed switching action modes and boundary line expressions for different allocation coefficient values, this invention further derives the modulation wave expression and the corresponding action modes.

[0197] For the sequence used in the first basic sector, the modulation wave expression and corresponding switching action mode of the timing sequence can be calculated by combining equation (34):

[0198] (36)

[0199] For the sequence used in the second basic sector, the modulation wave expression and corresponding switching action mode of the timing sequence can be calculated by combining equation (35):

[0200] (37)

[0201] Combining equations (36) and (37), the results can be extended to the entire vector plane. The modulation wave expression and the corresponding action modes are summarized as follows:

[0202] When u * mid <0 o'clock:

[0203] (38)

[0204] When u * mid When ≥0:

[0205] (39)

[0206] Among them, u mmax1 and u mmax2 This represents the two modulating waves of the maximum phase of the three-phase voltage; u mmin1 and u mmin2 This represents the two modulating waves of the least phase of a three-phase voltage; u*​max u* mid and u* min They represent V respectively DC The per-unit values ​​of the three-phase voltages, representing the maximum, intermediate, and minimum phases, are used as the reference; ActionA~ActionE represent the switching operation modes. The complete implementation flow of this method is as follows: Figure 8 As shown.

[0207] (6) Finally, the effectiveness of the proposed vector timing construction method for two parallel converters with duty cycle allocation degrees of freedom is verified by simulation. The simulation parameters are shown in Table 3 below:

[0208] Table 3 Simulation Parameters

[0209]

[0210] The simulation mainly demonstrates the optimization of different performance indicators of this scheme, ISVM, and MDPWM within the full modulation range. Figure 9 Figures (a), (b), and (c) show the three-phase current waveform, zero-sequence circulating current waveform, and single-phase circulating current waveform of ISVM, MDPWM, and the present invention scheme, respectively, when the modulation index is 0.7. Simulation results show that when the modulation index is 0.7, the ripple suppression effect of the present invention scheme is obvious, and the zero-sequence circulating current and single-phase circulating current indices are almost consistent with the MDPWM method and are better than the ISVM method. Figure 10 (a) further compares the RMS ripple values ​​of the three modulation algorithms across the entire modulation range. It can be seen that the modulation method proposed in this invention maintains the minimum RMS ripple value across the entire modulation range. Figure 10 Figure (b) shows a comparison of the total harmonic distortion (THD) of the current under various modulation indices for the three modulation algorithms. It can be seen that the THD of the output current proposed in this invention is superior to the other two methods across the entire modulation range. Furthermore... Figure 11 (a) and (b) in the figure compare the zero-sequence circulating current peak value and single-phase circulating current peak value of the three modulation algorithms in the full modulation range. Obviously, in the full modulation range, the zero-sequence circulating current and single-phase circulating current peak values ​​of the method proposed in this invention are almost consistent with the MDPWM method, and are always better than the ISVM method.

[0211] This embodiment also provides a vector timing construction system for two parallel converters with duty cycle allocation degrees of freedom, including:

[0212] The sector partitioning unit evenly divides the equivalent vector plane of the two parallel converters into 12 basic sectors. Each basic sector contains four basic vectors and their corresponding redundant vectors.

[0213] The vector timing construction unit selects four basic vectors and their redundant vectors in each basic sector to construct the vector timing. The four basic vectors include a zero vector, two non-zero vectors with the same angle but different magnitudes, and a third non-zero vector with different magnitudes and angles from the other vectors.

[0214] The duty cycle allocation coefficient solving unit introduces a duty cycle allocation coefficient to allocate the duty cycle for two vectors with the same angle but different magnitudes in each vector timing sequence; and adjusts the duty cycle allocation coefficient according to the position of the reference voltage vector in the basic sector to obtain the optimal duty cycle allocation coefficient, thereby optimizing single-phase circulating current, zero-sequence circulating current or output current ripple, or achieving comprehensive optimization of single-phase circulating current, zero-sequence circulating current and output current ripple;

[0215] The carrier modulation scheme generation unit calculates the modulation wave based on the optimal duty cycle allocation coefficient, determines the corresponding carrier comparison relationship and switching operation mode, and generates the carrier modulation scheme.

[0216] This embodiment also provides a vector timing construction device for two parallel converters, including a processor and a memory. The memory stores a program, which, when executed by the processor, implements the vector timing construction method for two parallel converters.

[0217] This embodiment also provides a storage medium on which a computer program is stored, which, when executed by a processor, implements the vector timing construction method for two parallel converters.

[0218] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and incorporate common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A method for constructing vector timing sequences for two parallel converters with duty cycle allocation degrees of freedom, characterized in that: including: The equivalent vector plane of two parallel converters is evenly divided into 12 basic sectors, and each basic sector contains four basic vectors and their corresponding redundant vectors; Four basic vectors and their redundant vectors are selected in each basic sector to construct a vector time sequence, where the four basic vectors include a zero vector, two non-zero vectors with the same angle but different magnitudes, and a third non-zero vector with different magnitude and angle from other vectors; A duty ratio distribution coefficient is introduced to distribute the duty ratio of two vectors with the same angle but different magnitudes in each vector time sequence; and the duty ratio distribution coefficient is adjusted according to the position of the reference voltage vector in the basic sector to obtain the optimal duty ratio distribution coefficient, so as to optimize the single-phase circulating current, zero-sequence circulating current or output current ripple, or achieve the comprehensive optimization of the single-phase circulating current, zero-sequence circulating current and output current ripple; The modulation wave is calculated according to the optimal duty ratio distribution coefficient, and the corresponding carrier comparison relationship and switching operation mode are determined to generate a carrier modulation scheme.

2. The vector timing construction method for two parallel converters according to claim 1, characterized in that: Based on their magnitudes, the four basic vectors are classified into long vectors, medium vectors, short vectors, and zero vectors. The magnitude of a long vector is 2 / 3 of the DC bus voltage, and the magnitude of a medium vector is [missing value]. The magnitude of the short vector is 1 / 3 of the bus voltage, and the magnitude of the zero vector is zero.

3. The vector timing construction method for two parallel converters according to claim 2, characterized in that: Constructing the vector time sequence specifically includes: in each basic sector, four basic vectors and their redundant vectors are selected. First, a 4-segment sequence is formed according to the arrangement of "medium vector - long vector - short vector - zero vector", and the last zero vector in this 4-segment sequence is used as the mirror point, and the redundant vector with the opposite change rate of zero-sequence circulating current to that in the 4-segment sequence is correspondingly selected. Finally, the vector time sequence is constructed in the arrangement of "medium vector - long vector - short vector - zero vector - short vector - long vector - medium vector".

4. The method for constructing vector timing of two parallel converters according to claim 1, characterized in that: The duty ratio distribution of two vectors with the same angle but different magnitudes in each vector time sequence specifically includes: For any two adjacent basic sectors, both include two basic vectors with the same angle but different magnitudes. A duty ratio distribution coefficient k is introduced between the two basic vectors. If k = 0, then in the constructed basic time sequence, one of the basic vectors does not participate in the action of the vector time sequence; if k = 1, then in the constructed basic time sequence, the other basic vector does not participate in the action of the vector time sequence; if 0 < k < 1, then in the constructed basic time sequence, the two common basic vectors act together.

5. The vector timing construction method for two parallel converters according to claim 4, characterized in that: The value range of the duty ratio distribution coefficient k satisfies: When a < 0 in the basic sector, if b ≥ 0.5 - 3|a|, the change range of the duty ratio distribution coefficient k is: -1 + 6a / (2b - 1) ≤ k ≤ 1; if b < 0.5 - 3|a|, then the change range of the duty ratio distribution coefficient k is: 0 ≤ k ≤ 1; When a > 0 in the basic sector, if b ≥ 0.5 - 3|a|, the change range of the duty ratio distribution coefficient k is: -1 - 6a / (2b - 1) ≤ k ≤ 1; if b < 0.5 - 3|a|, then the change range of the duty ratio distribution coefficient k is: 0 ≤ k ≤ 1; where the variables a and b satisfy: Where u* max u* mid and u* min These represent the DC bus voltage V. DC The per-unit values ​​of the maximum, intermediate, and minimum three-phase voltages are used as a reference. These are the per-unit values ​​of the three-phase reference voltage vectors.

6. The method for constructing vector timing of two parallel converters according to claim 1, characterized in that: The duty cycle allocation coefficient is adjusted according to the position of the reference voltage vector in the basic sector. Specifically, this includes: establishing a time-domain mathematical model of output current ripple, single-phase circulating current, and zero-sequence circulating current, and introducing the duty cycle allocation coefficient into the time-domain mathematical model; obtaining the optimal duty cycle allocation coefficient under different optimization requirements by differentiating or finding the maximum boundary of the time-domain mathematical model with the duty cycle allocation coefficient introduced, and the optimal duty cycle allocation coefficient satisfies the constraint that the action time of the four vectors is greater than or equal to zero.

7. The method for constructing vector timing of two parallel converters according to claim 1, characterized in that: The carrier comparison relationship and switching operation mode are as follows: When u * mid <0 o'clock: When u * mid When ≥0: Among them, u mmax1 and u mmax2 This represents the two modulating waves of the maximum phase of the three-phase voltage; u mmin1 and u mmin2 This represents the two modulating waves of the least phase of a three-phase voltage; u* max u* mid and u* min These represent the DC bus voltage V. DC The per-unit values ​​of the three-phase voltages, including the maximum, intermediate, and minimum phases, are used as the reference; ActionA~ActionE represent the switching operation modes, and k is the duty cycle allocation coefficient.

8. A vector timing construction system for two parallel converters with duty cycle allocation degrees of freedom, characterized in that: include: The sector partitioning unit evenly divides the equivalent vector plane of the two parallel converters into 12 basic sectors. Each basic sector contains four basic vectors and their corresponding redundant vectors. The vector timing construction unit selects four basic vectors and their redundant vectors in each basic sector to construct the vector timing. The four basic vectors include a zero vector, two non-zero vectors with the same angle but different magnitudes, and a third non-zero vector with different magnitudes and angles from the other vectors. The duty cycle allocation coefficient solving unit introduces a duty cycle allocation coefficient to allocate the duty cycle for two vectors with the same angle but different magnitudes in each vector timing sequence; and adjusts the duty cycle allocation coefficient according to the position of the reference voltage vector in the basic sector to obtain the optimal duty cycle allocation coefficient, thereby optimizing single-phase circulating current, zero-sequence circulating current or output current ripple, or achieving comprehensive optimization of single-phase circulating current, zero-sequence circulating current and output current ripple; The carrier modulation scheme generation unit calculates the modulation wave based on the optimal duty cycle allocation coefficient, determines the corresponding carrier comparison relationship and switching operation mode, and generates the carrier modulation scheme.

9. A vector timing construction device for two parallel converters, characterized in that: It includes a processor and a memory, wherein the memory stores a program that, when executed by the processor, implements the vector timing construction method for two parallel converters as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: It stores a computer program, which, when executed by a processor, implements the vector timing construction method for two parallel converters as described in any one of claims 1 to 7.

Images