A T-type three-level energy storage converter hybrid model predictive control method and device

Abstract

This invention discloses a hybrid model predictive control method and device for a three-level energy storage converter. The method determines candidate voltage vectors based on voltage vector transformation rules; calculates the cost function value of each candidate voltage vector using a cost function; and selects the voltage vector corresponding to the minimum cost function value as the first... k The pre-optimal voltage vector for the first control cycle; based on the first... k The preliminary optimal voltage vector and midpoint voltage of the first control cycle are combined to determine the first control cycle. k The optimal voltage vector in the first control cycle; in the second... k Calculate the control cycle. k -1 and k The duty cycle of the optimal voltage vector is determined; the gate drive signal corresponding to the optimal voltage vector and its duration is applied to the power electronic semiconductor devices of the T-type three-level energy storage converter. This invention combines the advantages of intuitive and simple finite set model predictive control and high current quality of continuous set model predictive control, enabling current ripple attenuation and current quality optimization of the T-type three-level energy storage converter.

Description

A hybrid model predictive control method and device for a T-type three-level energy storage converter Technical Field

[0001] This invention belongs to the field of three-level inverter control, and particularly relates to a hybrid model predictive control method and device for a T-type three-level energy storage converter. Background Technology

[0002] As the proportion of new energy sources in the power system increases, the inertia of the power system decreases, and its stability declines. Therefore, additional energy storage devices are needed to improve system stability and compensate for the intermittent and uncertain nature of new energy power generation. Power electronic converters, as the interface between photovoltaic, wind power, and energy storage batteries and the traditional power grid, play a crucial role in the efficiency and stability of the entire system. Compared to two-level converters, three-level converters have advantages such as larger capacity, higher voltage levels, and better output current quality, and are widely used in medium-voltage, high-power applications.

[0003] Model predictive control (MDC) offers numerous advantages, including fast dynamic response, high flexibility, robustness, ease of design and tuning, and the ability to handle multivariable constraints and implement multivariable control within a single control loop. MDC can be categorized into finite-set MDC and continuous-set MDC. Finite-set MDC eliminates the need for modulation, offering advantages such as computational simplicity and ease of implementation. However, it suffers from variable switching frequencies and relatively large output current ripple. Continuous-set MDC features a fixed switching frequency and high current quality. However, it requires additional modulation elements and incurs significant computational overhead.

[0004] To address the aforementioned issues, there is an urgent need for a novel hybrid model predictive control method that combines the advantages of finite set model predictive control and continuous set model predictive control. Summary of the Invention

[0005] The purpose of this invention is to provide a hybrid model predictive control method and device for a T-type three-level energy storage converter. This addresses the problems of high algorithm complexity and poor current quality in traditional methods. This invention combines the advantages of finite set model predictive control (which is intuitive, simple, and easy to implement) with continuous set model predictive control (which offers high current quality), enabling the attenuation of current ripple and optimization of current quality in T-type three-level energy storage converters.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] This invention provides a hybrid model predictive control method for a T-type three-level energy storage converter, the method comprising:

[0008] The state variables of the T-type three-level energy storage converter at time k are collected; the state variables include grid current, grid-side voltage and capacitor voltage.

[0009] A discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system is established, and the reference value of the grid current at time k+1 is obtained according to the Lagrange extrapolation theorem.

[0010] Candidate voltage vectors are determined based on voltage vector transformation rules;

[0011] The cost function value of each candidate voltage vector is calculated by the cost function, and the voltage vector corresponding to the minimum cost function value is selected as the preliminary optimal voltage vector for the kth control cycle.

[0012] The optimal voltage vector for the kth control cycle is determined by combining the preliminary optimal voltage vector and the midpoint voltage for the kth control cycle.

[0013] In the k-th control cycle, calculate the duty cycle of the optimal voltage vector in the (k-1)-th and k-th cycles;

[0014] The optimal voltage vector and the gate drive signal corresponding to its action time are applied to the power electronic semiconductor device of the T-type three-level energy storage converter.

[0015] In one embodiment, the discrete mathematical model of the T-type three-level energy storage converter in the two-phase stationary α-β coordinate system is as follows:

[0016]

[0017] In the formula, L and R are the inductance of the filter inductor and its equivalent resistance, respectively, and T is the value of the filter inductor. s i is the sampling period; α (k), i β (k) represents the sampled value of the grid current in the α-β coordinate system at time k; i α (k+1), i β (k+1) represents the predicted grid current in the α-β coordinate system at time k+1; e α (k), e β (k) represents the sampled value of the grid voltage in the α-β coordinate system at time k; u α (k), u β (k) represents the AC output voltage of the converter in the α-β coordinate system at time k, u o (k), u o (k+1) represent the DC side midpoint voltage values ​​at times k and k+1, respectively; i o (k) represents the DC side midpoint current value at time k; C1 and C2 are the capacitance values ​​of the upper and lower DC side bus capacitors, respectively; T s This refers to the control cycle of the control system.

[0018] In one implementation, the reference value of the grid current at time k+1 calculated according to the Lagrange extrapolation theorem is:

[0019]

[0020] In the formula, These are the reference values ​​for the α-axis grid current at times k+1, k, k-1, and k-2, respectively. These are the reference values ​​for the β-axis grid current at times k+1, k, k-1, and k-2, respectively.

[0021] In one implementation, the voltage vector conversion rule is as follows:

[0022] Only select the voltage vector adjacent to the optimal voltage vector at the previous moment, and ensure that the changes in phase voltage and line voltage of the three-level energy storage converter do not exceed u. dc The voltage vector required by the voltage change of / 2 is used as the candidate voltage vector, and the zero voltage vector is only considered in the "000" state.

[0023] In one implementation, the candidate voltage vector determined based on the voltage vector transformation rule is:

[0024] The optimal voltage vector for period k-1 is V1, and the corresponding candidate voltage vectors for period k are: V1, V4, V5, V6, V7, V8, V9, V 10 V 11 V 12 V 13 V 14 V 15 ;

[0025] The optimal voltage vector for period k-1 is V2, and the corresponding candidate voltage vector for period k is none.

[0026] The optimal voltage vector for period k-1 is V3, and the corresponding candidate voltage vector for period k is none.

[0027] The optimal voltage vector for period k-1 is V4, and the corresponding candidate voltage vectors for period k are V1, V4, V6, V7, and V4. 14 V 15 V 16 V 21 V 22 ;

[0028] The optimal voltage vector for period k-1 is V5, and the corresponding candidate voltage vectors for period k are V1, V5, V6, and V5. 14 V 16 V 21 V 22 ;

[0029] The optimal voltage vector for period k-1 is V6, and the corresponding candidate voltage vectors for period k are: V1, V4, V5, V6, V8, V9, V6. 16 V 17 V 23 ;

[0030] The optimal voltage vector for period k-1 is V7, and the corresponding candidate voltage vectors for period k are V1, V4, V7, V8, and V7. 16 V 17 V 23 ;

[0031] The optimal voltage vector for period k-1 is V8, and the corresponding candidate voltage vectors for period k are V1, V6, V7, V8, and V8. 10 V 11 V 17 V 18 V 24 ;

[0032] The optimal voltage vector for period k-1 is V9, and the corresponding candidate voltage vectors for period k are V1, V6, V9, and V9. 10 V 17 V 18 V 24 ;

[0033] The optimal voltage vector for period k-1 is: V 10 The corresponding candidate voltage vectors for period k are: V1, V8, V9, V 10 V 12 V 13 V 18 V 19 V 25 ;

[0034] The optimal voltage vector for period k-1 is: V 11 The corresponding candidate voltage vectors for period k are: V1, V8, V 11 V 12 V 18 V 19 V 25 ;

[0035] The optimal voltage vector for period k-1 is: V 12 The corresponding candidate voltage vectors for period k are: V1, V 10 V 11 V 12 V 14 V 15 V 19 V 20 V 26 ;

[0036] The optimal voltage vector for period k-1 is: V 13 The corresponding candidate voltage vectors for period k are: V1, V 10 V 13 V 14 V 19 V 20 V 26 ;

[0037] The optimal voltage vector for period k-1 is: V 14 The corresponding candidate voltage vectors for period k are: V1, V4, V5, V 12 V 13 V 14 V 20 V 21 V 27 ;

[0038] The optimal voltage vector for period k-1 is: V 15 The corresponding candidate voltage vectors for period k are: V1, V4, V 12 V 15 V 20 V 21 V 27 ;

[0039] The optimal voltage vector for period k-1 is: V 16 The corresponding candidate voltage vectors for period k are: V4, V5, V6, V7, V... 16 V 22 V 23 ;

[0040] The optimal voltage vector for period k-1 is: V 17 The corresponding candidate voltage vectors for period k are: V6, V7, V8, V9, V 17 V 23 V 24 ;

[0041] The optimal voltage vector for period k-1 is: V 18 The corresponding candidate voltage vectors for period k are: V8, V9, V 10 V 11 V 18 V 24 V 25 ;

[0042] The optimal voltage vector for period k-1 is: V 19 The corresponding candidate voltage vector for period k is: V 10 V 11 V 12 V 13 V 19 V25 V 26 ;

[0043] The optimal voltage vector for period k-1 is: V 20 The corresponding candidate voltage vector for period k is: V 12 V 13 V 14 V 15 V 20 V 26 V 27 ;

[0044] The optimal voltage vector for period k-1 is: V 21 The corresponding candidate voltage vectors for period k are: V4, V5, V 14 V 15 V 21 V 22 V 27 ;

[0045] The optimal voltage vector for period k-1 is: V 22 The corresponding candidate voltage vectors for period k are: V4, V5, V 16 V 21 V 22 ;

[0046] The optimal voltage vector for period k-1 is: V 23 The corresponding candidate voltage vectors for period k are: V6, V7, V... 16 V 17 V 23 ;

[0047] The optimal voltage vector for period k-1 is: V 24 The corresponding candidate voltage vectors for period k are: V8, V9, V 17 V 18 V 24 ;

[0048] The optimal voltage vector for period k-1 is: V 25 The corresponding candidate voltage vector for period k is: V 10 V 11 V 18 V 19 V 25 ;

[0049] The optimal voltage vector for period k-1 is: V 26 The corresponding candidate voltage vector for period k is: V 12 V 13 V 19 V 20 V 26 ;

[0050] The optimal voltage vector for period k-1 is: V 27 The corresponding candidate voltage vector for period k is: V 13 V 14 V 20 V 21 V 27 .

[0051] In one implementation, the cost function includes only the grid current control objective, specifically:

[0052]

[0053] in, Let i be the reference value of the grid current in the α-β coordinate system at time k+1. α (k+1),i β (k+1) represents the predicted value of the grid current in the α-β coordinate system at time k+1.

[0054] In one implementation, when the pre-selected optimal voltage vector is a zero, medium, or large voltage vector, the pre-selected optimal voltage vector is the optimal voltage vector; when the pre-selected optimal voltage vector is a small voltage vector, it is necessary to select the optimal voltage vector from the pre-selected optimal voltage vectors by combining the midpoint voltage.

[0055] The process of determining the optimal voltage vector based on the relationship between the small voltage vector and the midpoint current includes:

[0056] The small voltage vector is divided into P-type voltage vector and N-type voltage vector. The P-type voltage vector is the small voltage vector when the midpoint current is less than zero, and the N-type voltage vector is the small voltage vector when the midpoint current is greater than zero.

[0057] When the midpoint voltage is greater than zero, the N-type voltage vector is selected as the optimal voltage vector; when the midpoint voltage is less than zero, the P-type voltage vector is selected as the optimal voltage vector.

[0058] In one embodiment, the duty cycle of the voltage vector is:

[0059]

[0060] Where J1 and J2 are the cost function values ​​of the optimal voltage vector for periods k-1 and k, respectively, and d1 and d2 are the duty cycles of the optimal voltage vector for periods k-1 and k, respectively.

[0061] The present invention also provides a hybrid model predictive control device for a T-type three-level energy storage converter, the device comprising:

[0062] The acquisition module is used to acquire the state variables of the T-type three-level energy storage converter at time k; the state variables include grid current, grid-side voltage and capacitor voltage;

[0063] The discrete mathematical model module is used to establish a discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system, and obtain the reference value of the grid current at time k+1 based on the Lagrange extrapolation theorem.

[0064] The candidate voltage vector selection module is used to determine candidate voltage vectors based on voltage vector transformation rules;

[0065] The preliminary voltage vector selection module is used to calculate the cost function value of each candidate voltage vector through the cost function, and select the voltage vector corresponding to the minimum cost function value as the preliminary optimal voltage vector for the kth control cycle.

[0066] The optimal voltage vector determination module is used to determine the optimal voltage vector for the kth control cycle by combining the preliminary optimal voltage vector and the midpoint voltage for the kth control cycle.

[0067] The voltage vector duty cycle calculation module is used to calculate the duty cycle of the optimal voltage vector in the (k-1)th and kth cycles during the kth control cycle.

[0068] The action module is used to apply the gate drive signal corresponding to the optimal voltage vector and its action time to the power electronic semiconductor device of the T-type three-level energy storage converter.

[0069] The technical solutions provided in this application embodiment may include the following beneficial effects:

[0070] This invention reduces current ripple and improves current quality while achieving voltage balance and stable operation at the midpoint of a three-level energy storage converter. The proposed hybrid model predictive control differs from traditional model predictive control with discrete output characteristics, where the optimal voltage vector is executed at the start of the control cycle. Instead, it switches at specific moments within the control cycle through calculation, resulting in a time-continuous output characteristic and improved current quality. Attached Figure Description

[0071] The accompanying drawings, as part of this invention, are provided to further illustrate the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention, but do not constitute an undue limitation thereof. Clearly, the drawings described below are merely some embodiments, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0072] Figure 1 is a topology diagram of a T-type three-level energy storage converter provided in an embodiment of the present invention;

[0073] Figure 2 is a voltage vector distribution diagram of the T-type three-level energy storage converter provided in the embodiment of the present invention;

[0074] Figure 3 is a flowchart of a hybrid model predictive control method for a T-type three-level energy storage converter provided in an embodiment of the present invention;

[0075] Figure 4 is a voltage vector switching rule diagram of the T-type three-level energy storage converter provided in the embodiment of the present invention;

[0076] Figure 5 is a comparison of the optimal voltage vector used by finite set model predictive control and hybrid model predictive control provided in the embodiments of the present invention. In Figure 5(a), the optimal voltage vector is used by the traditional finite set model predictive control at the beginning of the control cycle, and in Figure 5(b), the optimal voltage vector is used by the hybrid model predictive control proposed in the present invention.

[0077] Figure 6 is a comparison of the steady-state simulation waveforms of finite set model predictive control and hybrid model predictive control provided in the embodiments of the present invention. In Figure 6(a), the steady-state waveform of finite set model predictive control is shown, and in Figure 6(b), the steady-state waveform of hybrid model predictive control is shown.

[0078] Figure 7 is a comparison of the dynamic simulation waveforms of finite set model predictive control and hybrid model predictive control provided in the embodiments of the present invention. Figure 7(a) shows the dynamic waveform of finite set model predictive control, and Figure 7(b) shows the dynamic waveform of hybrid model predictive control. Detailed Implementation

[0079] To enhance understanding of the present invention, the technical solution of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.

[0080] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0081] Figure 1 shows the topology of a T-type three-level energy storage converter. Each phase arm consists of four power electronic semiconductor devices. The DC side is composed of capacitors C1 and C2 connected in series, with identical parameters for C1 and C2. The midpoint of the DC side is defined as o, with zero potential. The grid-side filtering stage consists of a single L-type filter, with resistor R being the equivalent resistance of the filter. x i x Let x be the grid voltage and current, respectively (x = a, b, c). c1 and u c2 The voltages of capacitors C1 and C2 are respectively, udc It is DC voltage. Each arm of the T-type three-level energy storage converter has three effective switching states: S x ∈{1, 0, -1}. Assuming the capacitor voltages are balanced, i.e., the voltage across each capacitor is equal to half the DC voltage, then the output voltage generated by these three switching states is u. xo ∈{u dc / 2,0,-u dc Table 1 summarizes the relationship between the output voltage, switching state, and gate signal generated by the converter.

[0082] Table 1 shows the relationship between output voltage, switching state, and gate signal, where x = (a, b, c).

[0083]

[0084] To establish a mathematical model for a T-type three-level energy storage converter, it is assumed that all power electronic semiconductor devices are ideal switching devices, and nonlinear effects such as on-state voltage drop and switching delay of power electronic semiconductor devices are ignored.

[0085] According to Kirchhoff's voltage law, the dynamic response of the grid current in a three-phase stationary coordinate system is as follows:

[0086]

[0087] The T-type three-level energy storage converter has a total of 27 (3) switching states. 3 There are 19 non-redundant voltage vectors and 8 redundant voltage vectors in the α-β coordinate system, as shown in Figure 2. Based on the magnitude of the output voltage, the voltage vectors are divided into zero vectors (V1-V3), small vectors (V4-V5), and large vectors (V6-V7). 15 ), medium vector (V) 16 -V 21 ) and large vector (V) 22 -V 27 There is only one type of zero vector, with two redundant vectors; there are six types of small vectors, each with one redundant vector; there are six types of medium and large vectors, with no redundant vectors. The various voltage vectors are shown in Table 2.

[0088] Table 2 Voltage Vector Table

[0089]

[0090] As shown in Figure 3, the disclosed embodiment provides a hybrid model predictive control method for a T-type three-level energy storage converter, which specifically includes the following steps:

[0091] Step S100: Collect the state variables of the T-type three-level energy storage converter at time k.

[0092] Specifically, the grid current i at time k is collected using voltage and current sensors. a (k), i b (k), i c (k), grid voltage e a (k), e b (k), e c (k), and capacitor voltage u c1 (k), u c2 (k).

[0093] Step S200: Establish a discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system, and obtain the reference value of the grid current at time k+1 according to the Lagrange extrapolation theorem.

[0094] A three-phase stationary coordinate system can be converted to a two-phase stationary coordinate system using the Clarke transformation. The Clarke transformation is as follows:

[0095]

[0096] The mathematical model of the three-level energy storage converter in the two-phase stationary coordinate system is as follows:

[0097]

[0098] Using the forward Euler method, the discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary coordinate system is as follows:

[0099]

[0100] In the formula, L and R are the inductance of the filter inductor and its equivalent resistance, respectively, and T is the value of the filter inductor. s i is the sampling period; α (k), i β (k) represents the sampled value of the grid current in the α-β coordinate system at time k; i α (k+1), i β (k+1) represents the predicted grid current in the α-β coordinate system at time k+1; e α (k), e β (k) represents the sampled value of the grid voltage in the α-β coordinate system at time k; u α (k), u β (k) represents the AC output voltage of the converter in the α-β coordinate system at time k.

[0101]

[0102] The DC side midpoint voltage is:

[0103] u o =u c2 -u c1

[0104] Assuming equal capacitor voltages and balanced grid currents, the dynamic process of the DC-side midpoint voltage can be modeled as follows:

[0105]

[0106] i o The DC-side midpoint current is given by the following formula:

[0107] i o =|S abc |i abc

[0108] In the formula, |S abc |=[1-|S a |,1-|S b |,1-|S c |],i abc =[i a i b i c ].

[0109] The predicted value of the midpoint voltage is:

[0110]

[0111] In the formula, u o (k), u o (k+1) represent the DC side midpoint voltage values ​​at times k and k+1, respectively; i o (k) represents the DC side midpoint current value at time k; C1 and C2 are the capacitance values ​​of the upper and lower DC side bus capacitors, respectively; T s This refers to the control cycle of the control system.

[0112] According to the Lagrange extrapolation theorem, in the α-β coordinate system, the reference value of the grid current at time k+1 is:

[0113]

[0114] In the formula, in the formula, These are the reference values ​​for the α-axis grid current at times k+1, k, k-1, and k-2, respectively. These are the reference values ​​for the β-axis grid current at times k+1, k, k-1, and k-2, respectively.

[0115] Step S300: Determine candidate voltage vectors based on voltage vector transformation rules.

[0116] In the application embodiment, considering the characteristics of the three-level energy storage converter itself, reducing the damage to power electronic semiconductor devices and the impact on the power grid, and reducing the complexity and computation of the algorithm, voltage vector switching rules were formulated.

[0117] The changes in phase voltage and line voltage of a three-level energy storage converter must not exceed u. dc / 2, because the former may damage the switching devices of the converter, and the latter will have a significant impact on the power grid. Therefore, for three-level energy storage converters, not all voltage vectors can be directly converted. Furthermore, to reduce the computational load of the algorithm, the candidate voltage vectors for each cycle should be optimized while meeting the above requirements.

[0118] The voltage vector switching rule in this invention is as follows: only voltage vectors adjacent to the optimal voltage vector at the previous moment and satisfying the voltage change requirements in the above analysis are selected as candidate voltage vectors. The voltage vector switching rule of this invention is shown in Figure 4, wherein, in order to reduce common-mode voltage and switching frequency, the zero voltage vector only considers the "000" state.

[0119] Specifically, the candidate voltage vectors corresponding to each voltage vector are shown in Table 3.

[0120] Table 3 Candidate Voltage Vector Switching Rules

[0121]

[0122]

[0123] Step S400: Calculate the cost function value of each candidate voltage vector using the cost function, and select the voltage vector corresponding to the minimum cost function value as the preliminary optimal voltage vector for the kth control cycle.

[0124] Specifically, based on the candidate voltage vectors determined in Figure 4 and Table 3, the cost function value of each candidate voltage vector is calculated in conjunction with the cost function.

[0125] The cost function of a three-level energy storage converter typically includes grid current and neutral point voltage, and the cost function is as follows:

[0126]

[0127] Where k uThese are weighting factors used to adjust the control priority of grid current and midpoint voltage. However, these weighting factors involve large computational loads and cumbersome adjustments. Since small voltage vectors at the same location have the same current control capability and opposite midpoint voltage control effects, they can be used to adjust the midpoint voltage.

[0128] The improved cost function only includes the grid current control objective:

[0129]

[0130] Based on the cost function described above, the cost function values ​​for all candidate voltage vectors can be calculated. The voltage vector corresponding to the minimum cost function value is selected as the preliminary minimum voltage vector.

[0131] Step S500: Based on the combination of the preliminary optimal voltage vector and the midpoint voltage of the kth control cycle, determine the optimal voltage vector of the kth control cycle;

[0132] If the calculated preliminary optimal voltage vector is a small voltage vector, due to the redundancy of small voltage vectors, the cost function values ​​of two small voltage vectors at the same location are the same, thus there are two preliminary optimal voltage vectors. However, small voltage vectors at the same location have opposite effects on the midpoint voltage. Therefore, it is necessary to combine the midpoint voltage to further determine the optimal voltage vector. When the preliminary optimal voltage vector is a zero, medium, or large voltage vector, it is unique because it does not have redundancy. Therefore, the preliminary optimal voltage vector is the optimal voltage vector.

[0133] Based on the dynamic mathematical model of the DC-side midpoint voltage derived in step S200, the effect of the small vector on the midpoint voltage is related to the midpoint current i. o Relevant. Specifically, the midpoint current i corresponding to the small voltage vector. o As shown in Table 4.

[0134] Table 4 Relationship between voltage vector and midpoint current

[0135] Small voltage vector i o Small voltage vector i o V4-i a V5i a V6-i c V7i c V8-i b V9i b V 10 -i a V 11 i a V 12 -i c V 13 ic V 14 -i b V 15 i b surface

[0136] Small voltage vectors can be divided into P-type voltage vectors and N-type voltage vectors. P-type voltage vectors can increase the midpoint voltage, while N-type voltage vectors can decrease it. Combining this with the dynamic mathematical model of the DC-side midpoint voltage derived in step S200, it can be seen that when i o When i is greater than zero, the voltage vector is an N-type voltage vector; when i o When the voltage is less than zero, the voltage vector is a P-type voltage vector. Therefore, when the midpoint voltage is greater than zero, an N-type voltage vector is selected; when the midpoint voltage is less than zero, a P-type voltage vector is selected.

[0137] Based on the above analysis, when the pre-optimal voltage vector is zero, medium, or large, the pre-optimal voltage vector is the optimal voltage vector. When the pre-optimal voltage vector is small, the optimal voltage vector needs to be selected from the pre-optimal voltage vectors in conjunction with the midpoint voltage.

[0138] Step S600: In the k-th control cycle, calculate the duty cycle of the optimal voltage vector in the (k-1)-k cycle.

[0139] Assume the optimal voltage vectors for periods k-1 and k are V4 and V7, respectively. Traditional finite set model predictive control uses the optimal voltage vector at the beginning of the control period, as shown in Figure 5(a). The voltage vector remains constant throughout the control period. Therefore, finite set model predictive control has discrete output characteristics.

[0140] The hybrid model predictive control proposed in this invention still applies the optimal voltage vector of the previous control cycle at the beginning of the control cycle, and switches the optimal voltage vector of the current cycle at a specific time, as shown in Figure 5(b). The application time of the two voltage vectors can be adjusted within a control cycle. Therefore, the hybrid model predictive control can obtain a time-continuous output, with better current ripple suppression and current quality.

[0141] Furthermore, the duty cycle of the voltage vector is inversely proportional to the square of the cost function value, and the duty cycle of the voltage vector is:

[0142]

[0143] J1 and J2 are the cost function values ​​of the optimal voltage vector for periods k-1 and k, respectively. d1 and d2 are the duty cycles of the optimal voltage vector for periods k-1 and k, respectively.

[0144] Step S700: Apply the gate drive signal corresponding to the optimal voltage vector and its duration to the power electronic semiconductor device of the T-type three-level energy storage converter.

[0145] After a series of calculations in steps S100-S600, the control cycle value, comparison value and other information are transmitted to the FPGA through the DSP's communication module. The FPGA is then used to implement functions such as pulse distribution and dead-time delay for the power devices.

[0146] The following is an embodiment of the hybrid model predictive control device for a three-level energy storage converter according to the present invention, which can be used to execute the embodiment of the hybrid model predictive control method for a three-level energy storage converter according to the present invention. For details not disclosed in the embodiment of the hybrid model predictive control device for a three-level energy storage converter according to the present invention, please refer to the embodiment of the hybrid model predictive control method for a three-level energy storage converter according to the present invention.

[0147] In one embodiment, a hybrid model predictive control device for a three-level energy storage converter is proposed, the device comprising:

[0148] The acquisition module is used to acquire the state variables of the T-type three-level energy storage converter at time k; the state variables include grid current, grid-side voltage and capacitor voltage;

[0149] The discrete mathematical model module is used to establish a discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system, and obtain the reference value of the grid current at time k+1 based on the Lagrange extrapolation theorem.

[0150] The candidate voltage vector selection module is used to determine candidate voltage vectors based on voltage vector transformation rules;

[0151] The preliminary voltage vector selection module is used to calculate the cost function value of each candidate voltage vector through the cost function, and select the voltage vector corresponding to the minimum cost function value as the preliminary optimal voltage vector for the kth control cycle.

[0152] The optimal voltage vector determination module is used to determine the optimal voltage vector for the kth control cycle by combining the preliminary optimal voltage vector and the midpoint voltage for the kth control cycle.

[0153] The voltage vector duty cycle calculation module is used to calculate the duty cycle of the optimal voltage vector in the (k-1)th and kth cycles during the kth control cycle.

[0154] The action module is used to apply the gate drive signal corresponding to the optimal voltage vector and its action time to the power electronic semiconductor device of the T-type three-level energy storage converter.

[0155] The functional modules in this embodiment of the invention can be integrated into one processing module, or each unit can exist as a separate physical entity, or two or more units can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module.

[0156] As a specific example of this invention, a three-level energy storage converter simulation model was built in MATLAB / Simulink to test and verify the invention. The three-level energy storage converter is connected to a 220V AC grid on the AC side and powered by a battery on the DC side at 600V. The DC side capacitor is 2200μF, and the filter inductor is 4mH. The sampling and control period is 100μs. To highlight the advantages of this invention in improving current quality, the control performance of traditional finite control set model predictive control and the hybrid model predictive control proposed in this invention were compared.

[0157] Figure 6 compares the steady-state simulation results of traditional finite set model predictive control and hybrid model predictive control when the grid reference current is 50A. The simulation waveforms include the three-phase grid current and the DC side upper and lower bus capacitor voltages. Figure 6(a) shows the steady-state waveform of the finite set model predictive control, and Figure 6(b) shows the steady-state waveform of the hybrid model predictive control. The simulation waveforms show that both control strategies can achieve stable control of the three-level energy storage converter. The current THD of the finite set model predictive control and the hybrid model predictive control are 3.14% and 2.56%, respectively, with the hybrid model predictive control exhibiting better current quality. Both finite set model predictive control and hybrid model predictive control can keep the midpoint voltage below 5V, demonstrating good capacitor voltage control capability.

[0158] Figure 7 compares the steady-state dynamic simulation results of traditional finite set model predictive control and hybrid model predictive control. The simulation waveforms include the phase a grid current and the DC side upper and lower bus capacitor voltages. Figure 7(a) shows the dynamic waveform of finite set model predictive control, and Figure 7(b) shows the dynamic waveform of hybrid model predictive control. Both control strategies exhibit similar dynamic performance. When the current reference value jumps from 25A to 50A, the dynamic response time of both control strategies is approximately 2ms. When the current reference value jumps from 50A to 25A, the dynamic response time of both control strategies is approximately 1ms. Simulation results show that regardless of the current change trend, the output current can quickly track the current command change, indicating that the present invention has a fast dynamic response. During the dynamic process, both control strategies exhibit good capacitor voltage balance capability.

[0159] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-described technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A hybrid model predictive control method for a T-type three-level energy storage converter, characterized in that, The method includes: acquiring the state variables of the T-type three-level energy storage converter at time k; the state variables include grid current, grid-side voltage, and capacitor voltage; establishing a discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system, and obtaining the reference value of the grid current at time k+1 according to the Lagrange extrapolation theorem; the discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system is as follows: In the formula, L and R are the inductance of the filter inductor and its equivalent resistance, respectively, and T is the value of the filter inductor. s i is the sampling period; α (k), i β (k) represents the sampled value of the grid current in the α-β coordinate system at time k; i α (k+1), i β (k+1) represents the predicted grid current in the α-β coordinate system at time k+1; e α (k), e β (k) represents the sampled value of the grid voltage in the α-β coordinate system at time k; u α (k), u β (k) represents the AC output voltage of the converter in the α-β coordinate system at time k, u o (k), u o (k+1) represent the DC side midpoint voltage values ​​at times k and k+1, respectively; i o (k) represents the DC-side midpoint current value at time k; C1 and C2 are the capacitance values ​​of the upper and lower bus capacitors on the DC side, respectively; based on the voltage vector conversion rule, candidate voltage vectors are determined; the voltage vector conversion rule is: only select the voltage vector adjacent to the optimal voltage vector at the previous time, and satisfy the condition that the change in phase voltage and line voltage of the three-level energy storage converter cannot exceed u. dc The voltage vector required by the voltage change of / 2 is used as the candidate voltage vector, and the zero voltage vector is only considered in the "000" state. The cost function value of each candidate voltage vector is calculated by the cost function, and the voltage vector corresponding to the minimum cost function value is selected as the preliminary optimal voltage vector for the k-th control cycle. Based on the combination of the preliminary optimal voltage vector and the midpoint voltage for the k-th control cycle, the optimal voltage vector for the k-th control cycle is determined. In the k-th control cycle, the duty cycle of the optimal voltage vector for the (k-1)-th and k-th cycles is calculated. The optimal voltage vector and the gate drive signal corresponding to its action time are applied to the power electronic semiconductor device of the T-type three-level energy storage converter.

2. The hybrid model predictive control method for a T-type three-level energy storage converter according to claim 1, characterized in that, According to the Lagrange extrapolation theorem, in the α-β coordinate system, the reference value of the grid current at time k+1 is: In the formula, ， ， ， These are the reference values ​​of the α-axis grid current at times k+1, k, k-1, and k-2, respectively. ， ， ， These are the reference values ​​for the β-axis grid current at times k+1, k, k-1, and k-2, respectively.

3. The hybrid model predictive control method for a T-type three-level energy storage converter according to claim 2, characterized in that, Based on the voltage vector transformation rule, the candidate voltage vectors are determined as follows: the optimal voltage vector for period k-1 is V1, and the corresponding candidate voltage vectors for period k are: V1, V4, V5, V6, V7, V8, V9, V... 10 V 11 V 12 V 13 V 14 V 15 The optimal voltage vector for period k-1 is V2, and the corresponding candidate voltage vector for period k is none; the optimal voltage vector for period k-1 is V3, and the corresponding candidate voltage vector for period k is none; the optimal voltage vector for period k-1 is V4, and the corresponding candidate voltage vectors for period k are V1, V4, V6, V7, and V... 14 V 15 V 16 V 21 V 22 The optimal voltage vector for period k-1 is V5, and the corresponding candidate voltage vectors for period k are V1, V5, V6, and V5. 14 V 16 V 21 V 22 The optimal voltage vector for period k-1 is V6, and the corresponding candidate voltage vectors for period k are V1, V4, V5, V6, V8, V9, and V6. 16 V 17 V 23 The optimal voltage vector for period k-1 is V7, and the corresponding candidate voltage vectors for period k are V1, V4, V7, V8, and V7. 16 V 17 V 23 The optimal voltage vector for period k-1 is V8, and the corresponding candidate voltage vectors for period k are V1, V6, V7, V8, and V8. 10 V 11 V 17 V 18 V 24 The optimal voltage vector for period k-1 is V9, and the corresponding candidate voltage vectors for period k are V1, V6, V9, and V. 10 V 17 V 18 V 24 The optimal voltage vector for period k-1 is: V 10 The corresponding candidate voltage vectors for period k are: V1, V8, V9, V 10 V 12 V 13 V 18 V 19 V 25 The optimal voltage vector for period k-1 is: V 11 The corresponding candidate voltage vectors for period k are: V1, V8, V 11 V 12 V 18 V 19 V 25 The optimal voltage vector for period k-1 is: V 12 The corresponding candidate voltage vectors for period k are: V1, V 10 V 11 V 12 V 14 V 15 V 19 V 20 V 26 The optimal voltage vector for period k-1 is: V 13 The corresponding candidate voltage vectors for period k are: V1, V 10 V 13 V 14 V 19 V 20 V 26 The optimal voltage vector for period k-1 is: V 14 The corresponding candidate voltage vectors for period k are: V1, V4, V5, V 12 V 13 V 14 V 20 V 21 V 27 The optimal voltage vector for period k-1 is: V 15 The corresponding candidate voltage vectors for period k are: V1, V4, V 12 V 15 V 20 V 21 V 27 The optimal voltage vector for period k-1 is: V 16 The corresponding candidate voltage vectors for period k are: V4, V5, V6, V7, V... 16 V 22 V 23 The optimal voltage vector for period k-1 is: V 17 The corresponding candidate voltage vectors for period k are: V6, V7, V8, V9, V 17 V 23 V 24 The optimal voltage vector for period k-1 is: V 18 The corresponding candidate voltage vectors for period k are: V8, V9, V 10 V 11 V 18 V 24 V 25 The optimal voltage vector for period k-1 is: V 19 The corresponding candidate voltage vector for period k is: V 10 V 11 V 12 V 13 V 19 V 25 V 26 The optimal voltage vector for period k-1 is: V 20 The corresponding candidate voltage vector for period k is: V 12 V 13 V 14 V 15 V 20 V 26 V 27 The optimal voltage vector for period k-1 is: V 21 The corresponding candidate voltage vectors for period k are: V4, V5, V 14 V 15 V 21 V 22 V 27 The optimal voltage vector for period k-1 is: V 22 The corresponding candidate voltage vectors for period k are: V4, V5, V 16 V 21 V 22 The optimal voltage vector for period k-1 is: V 23 The corresponding candidate voltage vectors for period k are: V6, V7, V... 16 V 17 V 23 The optimal voltage vector for period k-1 is: V 24 The corresponding candidate voltage vectors for period k are: V8, V9, V 17 V 18 V 24 The optimal voltage vector for period k-1 is: V 25 The corresponding candidate voltage vector for period k is: V 10 V 11 V 18 V 19 V 25 The optimal voltage vector for period k-1 is: V 26 The corresponding candidate voltage vector for period k is: V 12 V 13 V 19 V 20 V 26 The optimal voltage vector for period k-1 is: V 27 The corresponding candidate voltage vector for period k is: V 13 V 14 V 20 V 21 V 27 。 4. The hybrid model predictive control method for a T-type three-level energy storage converter according to claim 1 or 3, characterized in that, The cost function contains only the grid current control objective, specifically: in, 、 This represents the reference value of the grid current in the α-β coordinate system at time k+1. 、 The value is the predicted value of the power grid current in the α-β coordinate system at time k+1.

5. The hybrid model predictive control method for a T-type three-level energy storage converter according to claim 1, wherein the duty cycle of the voltage vector is: in, J1 and J2 are the cost function values ​​of the optimal voltage vector for periods k-1 and k, respectively, and d1 and d2 are the duty cycles of the optimal voltage vector for periods k-1 and k, respectively.

6. A hybrid model predictive control device for a T-type three-level energy storage converter, characterized in that, The device includes: a data acquisition module for acquiring the state variables of the T-type three-level energy storage converter at time k; the state variables include grid current, grid-side voltage, and capacitor voltage; and a discrete mathematical model module for establishing a discrete mathematical model of the T-type three-level energy storage converter in a two-phase stationary α-β coordinate system, and obtaining the reference value of the grid current at time k+1 according to the Lagrange extrapolation theorem; the discrete mathematical model of the T-type three-level energy storage converter in the two-phase stationary α-β coordinate system is as follows: In the formula, L and R are the inductance of the filter inductor and its equivalent resistance, respectively, and T is the value of the filter inductor. s i is the sampling period; α (k), i β (k) represents the sampled value of the grid current in the α-β coordinate system at time k; i α (k+1), i β (k+1) represents the predicted grid current in the α-β coordinate system at time k+1; e α (k), e β (k) represents the sampled value of the grid voltage in the α-β coordinate system at time k; u α (k), u β (k) represents the AC output voltage of the converter in the α-β coordinate system at time k, u o (k), u o (k+1) represent the DC side midpoint voltage values ​​at times k and k+1, respectively; i o (k) represents the DC-side midpoint current value at time k; C1 and C2 are the capacitance values ​​of the upper and lower bus capacitors on the DC side, respectively; the candidate voltage vector selection module is used to determine candidate voltage vectors based on voltage vector conversion rules; the voltage vector conversion rules are: only select the voltage vector adjacent to the optimal voltage vector at the previous time, and satisfy the condition that the change in phase voltage and line voltage of the three-level energy storage converter cannot exceed u. dc The voltage vector required by the voltage change of / 2 is used as the candidate voltage vector, and the zero voltage vector is only considered in the "000" state; the preliminary voltage vector selection module is used to calculate the cost function value of each candidate voltage vector through the cost function, and select the voltage vector corresponding to the minimum cost function value as the preliminary optimal voltage vector for the k-th control cycle; the optimal voltage vector determination module is used to determine the optimal voltage vector for the k-th control cycle based on the combination of the preliminary optimal voltage vector and the midpoint voltage; the voltage vector duty cycle calculation module is used to calculate the duty cycle of the optimal voltage vector for the (k-1)-th and k-th cycles in the k-th control cycle; the action module is used to apply the optimal voltage vector and the gate drive signal corresponding to its action time to the power electronic semiconductor device of the T-type three-level energy storage converter.

Images