Non-intrusive three-level inverter phase current reconstruction method based on fixed sampling time
By adopting a non-invasive phase current reconstruction method with fixed sampling time in a three-level inverter, using redundant small vector and zero vector groups, the current reconstruction blind spot is eliminated, efficient current reconstruction is achieved, harmonics and switching losses are reduced, and bus utilization and dynamic response capabilities are improved.
Patent Information
- Application Number
- CN202510522108.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-24
- Publication Date
- 2025-07-25
AI Technical Summary
In the current reconstruction method of three-level inverter, the sampling time is not fixed, and there are current reconstruction blind spots in high and low modulation zones and sector boundaries, and the existing blind spot compensation scheme increases current harmonics and switching losses.
A non-invasive phase current reconstruction method at fixed sampling time is used, and a redundant small vector at the start of the carrier and the center time is used as the effective current reconstruction vector to design a single sensor topology that couples the phase current and the midpoint capacitor branch, and a zero vector group is used instead of the redundant small vector group in the low-modulation region to realize non-invasive current reconstruction.
The full-region current reconstruction is realized in the hexagonal spatial voltage vector profile, with the reconstruction error being less than 5%, and no complex compensation algorithm is required, which reduces harmonics and switching losses, and improves bus utilization and dynamic response capabilities.
Smart Images

Figure CN120377730A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of permanent magnet synchronous motor control, and relates to a non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants. Background Art
[0002] Compared with two-level drivers, T-type three-level drivers have higher conversion efficiency and stronger fault tolerance, and are widely used in medium and large power application scenarios. In an electric drive control system, current signals are crucial for monitoring the system operation status and implementing closed-loop control. Three-level three-phase drivers are commonly used in high-value equipment with high power, so stronger fault redundancy is required. The single-sensor current reconstruction technology, that is, using a single sensor to sample the instantaneous current at specific positions of the inverter at different times to achieve current reconstruction, on the one hand, can be used as a redundant current sensor when the phase current sensor fails and eliminate the influence of sensor inconsistency, and on the other hand, can be used for low-cost and fast fault diagnosis applications, which is very suitable for the application requirements of three-level drivers.
[0003] Existing three-level schemes still mainly adopt topologies that couple the midpoint branch of the bus. The sampling instants are not fixed, the current sampling points need to be corrected cycle by cycle, and there are problems such as the current reconstruction blind areas at the high and low modulation regions and sector boundaries that cannot be eliminated. And existing current reconstruction blind area compensation schemes are mostly invasive schemes that need to modify the PWM, which will increase current harmonics, switching losses, and even noise. In addition, there is no perfect and simple multi-branch coupling scheme in three-level systems like that in two-level systems. In view of this, a non-invasive phase current reconstruction method based on fixed-time sampling is proposed. Summary of the Invention
[0004] In view of this, the purpose of the present invention is to provide a non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants. Traditional three-level inverter current reconstruction methods need to sample the current at uncertain instants and rely on PWM correction algorithms to compensate for the periodic current reconstruction blind areas (current reconstruction blind areas) at the high and low modulation regions and sector boundaries. This method not only increases the algorithm complexity, but also increases current harmonics due to asymmetric PWM. In view of this, a non-invasive phase current reconstruction method based on fixed-time sampling is proposed in this paper.
[0005] First, the method uses a pair of redundant small vectors output at the starting and central moments of the space vector pulse width modulation (SVPWM) period as the effective current reconstruction vectors, thus simply fixing the sampling at the starting and central moments of the carrier wave. Secondly, a single current sensor topology that couples the phase current branch and the midpoint capacitor branch is designed to eliminate the current reconstruction blind area at the sector boundary. And in the low modulation region, the zero vector group is used to replace the redundant small vector group for sampling, thereby eliminating the current reconstruction blind area in the low modulation region. Finally, non-invasive phase current reconstruction without modifying the PWM is achieved in all regions within the hexagon space voltage vector profile.
[0006] The proposed method has good dynamic response ability. Under rated conditions, the error of the reconstructed current is less than 5%.
[0007] To achieve the above object, the present invention provides the following technical solutions:
[0008] A non-invasive three-level inverter phase current reconstruction method based on fixed sampling moments, comprising the following steps:
[0009] S1. According to the periodic characteristics of space vector pulse width modulation (SVPWM), select the first group of voltage vectors and the second group of voltage vectors corresponding to the starting moment and the central moment of the carrier wave respectively as the effective vectors for current reconstruction;
[0010] S2. Construct a single current sensor topology that couples the phase current branch of the three-level inverter and the DC side midpoint capacitor branch, and fix the sampling of the instantaneous current at the starting moment and the central moment of the carrier wave through the single current sensor;
[0011] S3. In the low modulation region, replace the first group of voltage vectors and the second group of voltage vectors with a zero vector group for current sampling to eliminate the current reconstruction blind area in the low modulation region;
[0012] S4. Based on the current signals at the fixed sampling moments, combined with the current path determination conditions of the single current sensor topology, reconstruct the phase current of the three-level inverter.
[0013] Furthermore, the first group of voltage vectors and the second group of voltage vectors are a redundant small vector group, and the redundant small vector group is fixed as the starting vector and the central vector within the SVPWM period.
[0014] Furthermore, the total voltage potential difference between the starting vector and the central vector remains fixed, and the starting vector and the central vector form a symmetric distribution within the hexagon space voltage vector profile.
[0015] Furthermore, the single current sensor topology determines the conduction state of the current path by collecting the current flowing through the midpoint capacitor branch and combining the switching states of the phase current branches to decouple each phase current.
[0016] Furthermore, the low modulation region extends the duration of the effective vector to meet the minimum sampling time constraint by additionally dividing the sector boundary and replacing the redundant small vector group with a zero vector group as the effective vector for current reconstruction.
[0017] Furthermore, the current path determination condition is based on the switching state combination of the three-level inverter, and determines the current flow direction and amplitude relationship of each phase through logical operations.
[0018] Furthermore, the interval of the fixed sampling moment is synchronized with the SVPWM carrier period, and at least two samplings are included in each carrier period.
[0019] Furthermore, the error of the reconstructed phase current is less than 5% under rated conditions, and covers all working regions except the overmodulation region within the hexagonal space voltage vector profile.
[0020] A three-level inverter system includes:
[0021] A DC side bus capacitor, composed of an upper bus capacitor and a lower bus capacitor connected in series, provides a midpoint potential;
[0022] A three-phase bridge arm circuit, and each phase bridge arm outputs three level states of P, O, and N;
[0023] A single current sensor, coupled between the phase current branch and the midpoint capacitor branch, is used to collect current signals at a fixed sampling moment;
[0024] A controller, configured to execute the method according to any one of claims 1 to 8 to reconstruct the phase current and output a control signal to the three-phase bridge arm circuit.
[0025] Furthermore, the installation position of the single current sensor makes its instantaneous sampling current expression satisfy:
[0026] i smp =Z c i c +i a
[0027] where Z c is a determination variable, representing the path state of the current flowing through the midpoint capacitor branch, and i c and i a are the phase currents of the C-phase and A-phase windings respectively.
[0028] The beneficial effects of the present invention are as follows: The present invention uses redundant small vectors and zero vectors as effective vectors for current reconstruction, and establishes the simplest current reconstruction topology for the coupled phase current branch and the neutral point capacitor branch. Compared with the traditional single-branch coupling method, the proposed method eliminates the current reconstruction blind areas at low modulation and sector boundaries, has the advantage of fixed current sampling moments, and does not require a complex blind area compensation algorithm. The area where current reconstruction can be performed covers all inverter operating areas except the overmodulation area, the bus utilization rate can reach 95%, and it has good dynamic response capabilities. Under rated conditions, the reconstructed current error is less than 5%.
[0029] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail preferably with reference to the accompanying drawings, where:
[0031] Figure 1 are sector divisions and their starting and center vectors;
[0032] Figure 2 is the current reconstruction topology;
[0033] Figure 3 are comparisons of space vector pulse width modulation waveforms for normal and compensation in the low modulation region; (a) is the normal space vector pulse width modulation waveform; (b) is the space vector pulse width modulation waveform for compensation in the low modulation region.
[0034] Figure 4 are experimental results when operating under rated conditions (load 4 Nm, speed 500 rpm); (a) are the a and b phase currents; (b) is the reconstruction error of the a and b phase currents; (c) is the current update moment;
[0035] Figure 5 are variable load process curves at 500 rpm; (a) is the torque tracking curve, (b) are the original phase currents and reconstructed phase currents, (c) is the current reconstruction error;
[0036] Figure 6 are variable speed process curves with a load of 4 Nm; (a) is the torque tracking curve, (b) are the original phase currents and reconstructed phase currents, (c) is the current reconstruction error. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0037] The following specific examples illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concept of the present invention. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.
[0038] Among them, the drawings are only for illustrative purposes, showing only schematic diagrams rather than physical diagrams, and should not be construed as a limitation to the present invention; in order to better illustrate the embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, which does not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.
[0039] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "rear", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and should not be construed as a limitation to the present invention. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.
[0040] To achieve fixed-time current sampling and single-cycle phase current reconstruction, it is necessary to reasonably design the installation position of a single current sensor and the effective vector of the reconstructed current.
[0041] 1. Selection of the effective vector for current reconstruction
[0042] According to the principle of space vector pulse width modulation, the reference vector is synthesized by the three voltage vectors in its triangular sector, satisfying the constraints:
[0043] 1) The switching state is only allowed to change according to the principle of N→O→P, P→O→N, which can minimize the electromagnetic interference caused by du / dt.
[0044] 2) Each switch transition should preferably involve only one phase, avoiding switching two or three phases simultaneously.
[0045] According to the above principle, the total voltage potential of the starting vector of the seven-segment space vector pulse width modulation is the smallest, while the potential value of the total voltage of the vector at the PWM center is the largest, and the total voltage potential between the center and the starting vector maintains a fixed difference of 3. Therefore, the starting vector and the center vector of the space vector pulse width modulation in the sector are composed of a group of fixed vectors. According to the vector group, the space voltage vectors are divided into Figure 1 as shown.
[0046] Since the starting and center moments of the space vector pulse width modulation are fixed as a pair of redundant small vectors, they are used as effective current reconstruction vectors for current reconstruction, which can realize current sampling at a fixed time and simplify the algorithm.
[0047] 2. Phase current reconstruction topology design
[0048] Figure 2 is the proposed single current sensor phase current reconstruction topology. In the figure: C1 and C2 respectively represent the upper and lower bus capacitors on the DC side, C = C1 = C2; o is the midpoint of the capacitor, and U dc is the DC side bus voltage, and i abc is the three-phase current. The positive direction of the current is defined as the direction flowing into the motor. The input voltage on the DC power supply side is U dc , and the midpoint voltage of U dc / 2 is provided by two identical capacitors connected in series. Therefore, the level states S x (x = a to c) that each phase can output have three types: [P], [O], and [N], which respectively represent 0.5U dc , 0, and -0.5U dc . The current expression sampled instantaneously at the installation position of the sensor is
[0049] i smp = Z c i c + i a (1)
[0050] where Z c is the decision variable, which characterizes the path of the current flow. Being equal to 1 (0) respectively characterizes that the current flows through (does not flow through) the midpoint branch of the power transistor in the c phase at this time. The decision condition of Z c is
[0051]
[0052] Substituting the vectors at the starting and center moments in Figure 1 into equation (1), the phase current collected by the redundant small vector group can be obtained, and the collection results are shown in Table 1.
[0053] Table 1 Current reconstruction results in different sectors
[0054] Sector Starting vector Read current Central vector Read current Ⅰ <![CDATA[V0(NNN)]]> <![CDATA[i smp1 = i a > <![CDATA[V1(OOO)]]> <![CDATA[i smp2 = i c + i a = -i b > Ⅱ <![CDATA[V0(OOO)]]> <![CDATA[i smp2 = i c + i a = -i b > <![CDATA[V3(PPP)]]> <![CDATA[i smp2 = i a > Ⅲ <![CDATA[V4(ONN)]]> <![CDATA[i smp1 = i a > <![CDATA[V3(POO)]]> <![CDATA[i smp2 = i c + i a = -i b > Ⅳ <![CDATA[V6(OON)]]> <![CDATA[i smp1 = i a > <![CDATA[V5(PPO)]]> <![CDATA[i smp2 = i c + i a = -i b > Ⅴ <![CDATA[V8(NON)]]> <![CDATA[i smp1 = i a > <![CDATA[V7(OPO)]]> <![CDATA[i smp2 = i c + i a = -i b > Ⅵ <![CDATA[V 10 (NOO)]]> <![CDATA[i smp1 = i c + i a = -i b > <![CDATA[V9(OPP)]]> <![CDATA[i smp2 = i a > Ⅶ <![CDATA[V 12 (NNO)]]> <![CDATA[i smp1 = i c + i a = -i b > <![CDATA[V 11 (OOP)]]> <![CDATA[i smp2 = i a > Ⅷ <![CDATA[V 14 (ONO)]]> <![CDATA[i smp1 = i c + i a = -i b > <![CDATA[V 13 (POP)]]> <![CDATA[i smp2 = i a >
[0055] 3. Dead Zone Optimization
[0056] The actual current signal is a non-ideal step signal, so there is a minimum value T for the actual time of current acquisition. min . In the low modulation region, the action time of the redundant small vectors is short and does not satisfy T min constraint, resulting in a current reconstruction blind area. Therefore, it is necessary to compensate for the current reconstruction blind area in the low modulation region. To further optimize the current reconstruction blind area in the low modulation region, it is additionally divided into Sector I and Sector II, as Figure 1 shown. In this region, taking the zero vector groups of NNN and OOO, and OOO and PPP as u0 to synthesize u s , that is, space vector pulse width modulation takes the zero vector groups of NNN and OOO, and OOO and PPP as the starting and center vectors respectively. Figure 3 shows the PWM waveforms before and after improvement with OOO and NNN as the starting and center vectors.
[0057] After improvement, the effective vectors for current reconstruction in the blind area of the low modulation region are replaced by NNN and OOO for the redundant small vector groups in the normal region, significantly extending the duration of the effective current reconstruction vectors, and the current reconstruction blind area is eliminated.
[0058] Through the verification experiment of fault-tolerant operation under rated steady-state and dynamic conditions on the T-type three-level back-to-back experimental platform with a rated load (4 Nm), the following experimental effect diagrams under each condition are finally obtained, indicating that the proposed method can achieve non-invasive phase current reconstruction without modifying the PWM.
[0059] Figure 4 are the experimental results of operating at the rated condition (load 4 Nm, speed 500 rpm); (a) are the a and b phase currents; (b) is the reconstruction error of the a and b phase currents; (c) is the current update moment;
[0060] Figure 5 are the variable load process curves at 500 rpm; (a) is the torque tracking curve, (b) are the original phase current and reconstructed phase current, (c) is the current reconstruction error;
[0061] Figure 6 are the variable speed process curves with a load of 4 Nm; (a) is the torque tracking curve, (b) are the original phase current and reconstructed phase current, (c) is the current reconstruction error.
[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the present technical solution, and they should all be covered within the scope of the claims of the present invention.
Claims
1. A non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants, characterized in that: It includes the following steps: S1. According to the periodic characteristics of space vector pulse width modulation (SVPWM), select the first group of voltage vectors and the second group of voltage vectors corresponding to the starting moment and the central moment of the carrier wave respectively as the effective vectors for current reconstruction; S2. Construct a single current sensor topology for the phase current branch and the midpoint capacitor branch of the DC side of the coupled three-level inverter, and perform fixed sampling of the instantaneous current at the starting moment and the central moment of the carrier wave through the single current sensor; S3. In the low modulation region, replace the first group of voltage vectors and the second group of voltage vectors with a zero vector group for current sampling to eliminate the current reconstruction blind area in the low modulation region; S4. Based on the current signals at the fixed sampling moments, and combined with the current path determination conditions of the single current sensor topology, reconstruct the phase currents of the three-level inverter.
2. The non-invasive three-level inverter phase current reconstruction method based on a fixed sampling time according to claim 1, wherein: The first group of voltage vectors and the second group of voltage vectors are redundant small vector groups, and the redundant small vector groups are fixed as the starting vector and the central vector within the SVPWM period.
3. The non-invasive three-level inverter phase current reconstruction method based on a fixed sampling moment according to claim 2, characterized in that: The total voltage potential difference between the starting vector and the central vector remains fixed, and the starting vector and the central vector form a symmetric distribution within the hexagonal space voltage vector contour.
4. The non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants according to claim 1, wherein: The single current sensor topology determines the conduction state of the current path by collecting the current flowing through the midpoint capacitor branch and combining the switching states of the phase current branches to decouple each phase current.
5. The non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants according to claim 1, wherein: In the low modulation region, by additionally dividing the sector boundaries and replacing the redundant small vector group with a zero vector group as the effective vector for current reconstruction, the duration of the effective vector is extended to meet the minimum sampling time constraint.
6. The non-invasive three-level inverter phase current reconstruction method based on a fixed sampling time according to claim 1, wherein: The current path determination conditions are based on the switching state combinations of the three-level inverter, and determine the flow direction and amplitude relationship of each phase current through logical operations.
7. The non-invasive three-level inverter phase current reconstruction method based on fixed sampling instants according to claim 1, characterized in that: The interval of the fixed sampling moments is synchronized with the SVPWM carrier period, and there are at least two samplings within each carrier period.
8. The non-invasive three-level inverter phase current reconstruction method based on a fixed sampling time according to claim 1, characterized in that: The error of the reconstructed phase current is less than 5% under rated conditions, and covers all working regions within the hexagonal space voltage vector contour except the overmodulation region.
9. A three-level inverter system, characterized in that: It includes: A DC side bus capacitor, composed of an upper bus capacitor and a lower bus capacitor connected in series, providing a midpoint potential; A three-phase bridge arm circuit, and each phase bridge arm outputs three level states of P, O, and N; A single current sensor, coupled between the phase current branch and the midpoint capacitor branch, for collecting current signals at fixed sampling moments; A controller, configured to execute the method according to any one of claims 1 to 8 to reconstruct the phase current and output a control signal to the three-phase bridge arm circuit.
10. The three-level inverter system according to claim 9, wherein: The installation position of the single current sensor is such that its instantaneous sampling current expression satisfies: i smp = Z c i c + i a Among them, Z c is a judgment variable, representing the path state of the current flowing through the midpoint capacitor branch, i c and i a are the phase currents of the C-phase and A-phase windings respectively.
Citation Information
Cited By
Electromechanical transducer drive control system for precision instrument
CN122292844A