Fault detection and fault-tolerant control strategy for modular multilevel converter in MVDC ship

Abstract

The invention discloses a fault detection and fault-tolerant control strategy for a modular multilevel converter in an MVDC ship, and belongs to the technical field of power electronics, and the fault detection and fault-tolerant control strategy comprises the following steps: S1, carrying out capacitor voltage estimation based on a Luenberger observer; s2, performing fault detection on each sub-module by using an SM fault detection method; and S3, a balanced line voltage is obtained by injecting zero sequence voltage, so that an overmodulation phenomenon of a fault phase is avoided, and equalization control is performed on the energy of each bridge arm by adjusting circulating current. According to the fault detection and fault-tolerant control strategy of the modular multilevel converter in the MVDC ship, the fault detection process can be simplified, accurate fault diagnosis can be achieved only by setting a fixed algorithm gain coefficient, the complexity of an algorithm can be greatly simplified, the calculated amount is reduced, and the fault-tolerant operation capability is improved.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and particularly to a fault detection and fault-tolerant control strategy for a modular multilevel converter in an MVDC ship. Background Art

[0002] In the prior art, the modular multilevel converter (MMC) has been favored in high-voltage and high-power scenarios due to its excellent output voltage waveform and extremely high transmission efficiency. The MMC contains a large number of sub-modules (SMs), and each SM contains several insulated gate bipolar transistors (IGBTs). However, the use of a large number of power electronic devices increases the probability of system failures and affects the stable operation of the MMC. Therefore, it is very necessary to quickly detect IGBT faults in the SMs.

[0003] Currently, the SM fault diagnosis is integrated into each SM control unit, and the SM capacitor is used to obtain energy in common with multiple power supplies such as the operating power supply. Although this SM fault detection method is simple and fast, the structure of single optical fiber channel transmission of information for each SM and the common loop of multiple power supplies easily causes problems of false transmission and misjudgment of SM faults.

[0004] In addition, after detecting and locating the sub-module faults, it is necessary to ensure that the MMC can continue to operate before maintenance and replacement of the faulty modules. Therefore, in order to improve the operation reliability of the MMC, it is urgent to conduct in-depth research on its fault-tolerant control method. Summary of the Invention

[0005] The object of the present invention is to provide a fault detection and fault-tolerant control strategy for a modular multilevel converter in an MVDC ship, which can simplify the fault detection process, achieve accurate diagnosis of faults only by setting fixed algorithm gain coefficients, greatly simplify the complexity of the algorithm, reduce the calculation amount, and is also beneficial to improving the fault-tolerant operation ability.

[0006] To achieve the above object, the present invention provides a fault detection and fault-tolerant control strategy for a modular multilevel converter in an MVDC ship, including the following steps:

[0007] S1. Estimate the capacitor voltage based on the Luenberger observer;

[0008] S2. Detect faults in each sub-module using the SM fault detection method;

[0009] S3. Obtain balanced line voltages by injecting zero-sequence voltages to avoid over-modulation phenomena in the faulty phase, and then balance the energies of each bridge arm by adjusting the circulating current.

[0010] Preferably, in S1, the SM capacitor voltage u is obtained from the MMC system circuit c as follows:

[0011]

[0012] where u c (t - 1) and u c (t) are the capacitor voltage values at time t - 1 and time t respectively; s(t) is the switch function value at time t; i arm (t) is the arm current value at time t;

[0013] A Luenberger observer including a time update equation and a measurement update equation is constructed from the above formula.

[0014] Preferably, the method for constructing the time update equation is: discretize the calculation formula of the SM capacitor voltage u c to obtain the predicted state estimate value of the capacitor voltage:

[0015]

[0016] where u c_c (k - 1) represents the estimated value of the capacitor voltage at time k - 1; u c_p (k) represents the predicted value of the capacitor voltage at time k; ΔT represents the sampling time; s(k) represents the switch function value at time k; i arm (k) represents the arm current value at time k;

[0017] The method for constructing the measurement update equation is: according to the time update equation, the expression of the estimated value of the capacitor voltage based on the Luenberger observer is:

[0018] u c_c (k) = u c_p (k) + K(u c (k) - u c_p (k));

[0019] where u c_c (k) represents the estimated value of the capacitor voltage at time k; K represents the gain coefficient of the Luenberger observer; u c (k) represents the measured value of the capacitor voltage at time k.

[0020] Preferably, in S2, the specific process is as follows:

[0021] First, collect the capacitor voltage, arm current, and switch state of the i-th module, calculate the estimated value u c_c (k) of the module capacitor voltage at time k according to the measurement update equation, and calculate the predicted value u c_p (k) of the capacitor voltage at time k according to the time update equation;

[0022] Then, compare the measured value u of the capacitor voltage c (k) with the estimated value u c_c (k). Let Δu t be the error threshold of the capacitor voltage. If |u c_c (k) - u c (k)| > Δu t , and it lasts for at least a duration of Δt, then a fault is considered to have occurred; otherwise, normal operation is considered, and this detection algorithm is continued to be executed.

[0023] Preferably, in S3, it specifically includes the following steps:

[0024] S31. Adjust the total energy. By adjusting the total energy reference value, control the sum of the stored energies of all sub-module capacitors of the MMC;

[0025] S32. Inject the fundamental frequency zero-sequence voltage to ensure the balance of the output line voltages;

[0026] S33. Adjust the arm energy to balance the energy and capacitor voltages between the arms.

[0027] Preferably, in S31, assume that there are N f faulty sub-modules in the upper arm. Then, the expressions for the maximum value of the output voltage of the faulty arm and the maximum value of the output voltage of the faulty phase are:

[0028]

[0029] where u f_max is the maximum value of the output voltage of the faulty phase, and u uf_max is the maximum value of the output voltage of the faulty arm;

[0030] To expand the phase voltage output range of the MMC and at the same time increase the capacitor voltage reference value of the three-phase non-faulty sub-modules, there is:

[0031]

[0032] where λ is the boosting coefficient of the sub-module capacitor voltage;

[0033] The DC bus voltage is increased to:

[0034]

[0035] where is the DC bus voltage reference value under fault-tolerant operation;

[0036] The maximum value of the output voltage of the faulty arm and the maximum value of the output voltage of the faulty phase are further expressed as follows:

[0037]

[0038] Among them, u uFT_max is the maximum value of the output voltage of the faulty bridge arm under the proposed fault-tolerant control method; u FT_max is the maximum value of the output voltage of the faulty phase under the proposed fault-tolerant control method;

[0039] Based on the DC bus voltage reference value, the reference value of the total energy stored in all sub-module capacitors is obtained as follows:

[0040]

[0041] Preferably, in S32, the injected fundamental-frequency zero-sequence voltage should satisfy the following conditions: the equivalent output range of the a-phase voltage is the same before and after the fault, that is, MA′ = AO; let the number of faulty sub-modules be N f , then the amplitude of the zero-sequence voltage to be injected is expressed as:

[0042]

[0043] After injecting the zero-sequence voltage, the output amplitude of the non-faulty phase voltage also increases accordingly, up to:

[0044]

[0045] The output voltage amplitudes of other non-faulty phases need to satisfy:

[0046]

[0047] The voltage boost coefficient λ under different numbers of faulty sub-modules should satisfy:

[0048]

[0049] Substitute the calculation formula of λ into the calculation formula of V b , to obtain the amplitude of the zero-sequence voltage to be injected, and express the zero-sequence voltage reference value as:

[0050] u0 = V0cos(ωt);

[0051] Among them, ω is the grid angular frequency.

[0052] Preferably, in S33, set the capacitor storage energy of the faulty sub-module to the rated value to ensure that the number of modules participating in the balancing control remains unchanged, and the virtual bridge arm energy is:

[0053]

[0054] Among them, E uj_vir is the virtual energy of the upper bridge arm, Ω uj is the set of serial numbers of the faulty sub-modules of the upper bridge arm, E lj_viris the virtual energy of the lower arm, Ω lj is the set of serial numbers of the faulty sub - modules of the lower arm;

[0055] Define the total virtual energy E stored in the capacitor as: vir Define as:

[0056]

[0057] The MMC adjusts the active power between three - phases by injecting DC circulating current components, and adjusts the active power between the upper arm and the lower arm by injecting fundamental - frequency circulating current components. The circulating current reference value of the MMC is:

[0058]

[0059] Among them, K p1 is the control parameter of the arm energy and the controller, K p2 is the control parameter of the arm energy difference controller, E vir is the total virtual energy stored in the capacitors in the MMC, Φ is the intermediate parameter of the transfer matrix, and its calculation formula is:

[0060]

[0061] By adjusting the reference value of the circulating current, the balanced control of the energy between each arm is realized.

[0062] Therefore, by adopting the above - mentioned fault detection and fault - tolerant control strategy of the modular multilevel converter in the MVDC ship, the invention can simplify the fault detection process. Only by setting a fixed algorithm gain coefficient can accurate fault diagnosis be realized, which can greatly simplify the complexity of the algorithm, reduce the calculation amount, and is also beneficial to improving the fault - tolerant operation ability.

[0063] Next, through the drawings and embodiments, the technical solution of the present invention will be further described in detail. Description of the Drawings

[0064] Figure 1 is the step - flow chart of the embodiment of the fault detection and fault - tolerant control strategy of the modular multilevel converter in the MVDC ship of the present invention;

[0065] Figure 2 is the schematic diagram of the estimation algorithm of the SM capacitor voltage of the present invention. Detailed Embodiment

[0066] The technical solution of the present invention will be further described below through the drawings and embodiments.

[0067] Unless otherwise defined, the technical terms or scientific terms used in this invention shall have the ordinary meanings as understood by those of ordinary skill in the field to which this invention pertains. The "first", "second" and similar terms used in this invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0068] Embodiment 1

[0069] As Figure 1 shown, the present invention provides a fault detection and fault-tolerant control strategy for a modular multilevel converter in an MVDC ship, including the following steps:

[0070] S1. Estimate the capacitor voltage based on the Luenberger observer.

[0071] From the MMC system circuit, obtain the SM capacitor voltage u c as:

[0072]

[0073] wherein, u c (t - 1), u c (t) are the capacitor voltage values at time t - 1 and time t respectively; s(t) is the switching function value at time t; i arm (t) is the arm current value at time t.

[0074] Construct a Luenberger observer for capacitor voltage observation from the above formula. Among them, the Luenberger observer includes a time update equation and a measurement update equation.

[0075] The construction method of the time update equation is: discretize the calculation formula of the SM capacitor voltage u c to obtain the predicted state estimate value of the capacitor voltage as:

[0076]

[0077] wherein, u c_c (k - 1) represents the estimated value of the capacitor voltage at time k - 1; u c_p (k) represents the predicted value of the capacitor voltage at time k; ΔT represents the sampling time; s(k) represents the switching function value at time k; iarm The arm current value at time k is represented by (k).

[0078] The method for constructing the measurement update equation is as follows: According to the time update equation, the estimated value of the capacitor voltage based on the Luenberger observer is expressed as:

[0079] u c_c (k) = u c_p (k) + K(u c (k) - u c_p (k))

[0080] Among them, u c_c (k) represents the estimated value of the capacitor voltage at time k; K represents the gain coefficient of the Luenberger observer; u c (k) represents the measured value of the capacitor voltage at time k.

[0081] Based on the time update equation and the measurement update equation, the capacitor voltage estimation algorithm based on the Luenberger observer can be obtained, as Figure 2 shown, where Z -1 is one control cycle behind.

[0082] S2. Use the SM fault detection method to detect faults in each sub-module.

[0083] First, collect the capacitor voltage, arm current, and switch state of the i-th module. Calculate the estimated value uc_c(k) of the module capacitor voltage at time k according to the measurement update equation, and calculate the predicted value uc_p(k) of the capacitor voltage at time k according to the time update equation.

[0084] Then, compare the measured value uc(k) and the estimated value uc_c(k) of the capacitor voltage. Let Δut be the capacitor voltage error threshold. If |uc_c(k) - uc(k)| > Δut and lasts for at least a duration of Δt, it is considered that a fault has occurred; otherwise, it is considered normal operation, and this detection algorithm is continued.

[0085] S3. Obtain balanced line voltages by injecting zero-sequence voltages to avoid overmodulation in the faulty phase, and then balance the energy of each arm by adjusting the circulating current.

[0086] Specifically, it includes the following steps:

[0087] S31. Adjust the total energy to control the total stored energy of all the capacitors in the MMC by adjusting the total energy reference value.

[0088] Assume that there are N f faulty sub-modules in the upper arm. Then the maximum value of the output voltage of the faulty arm and the maximum value of the output voltage of the faulty phase are expressed as:

[0089]

[0090] Among them, u f_max is the maximum value of the output voltage of the faulty phase, and u uf_max is the maximum value of the output voltage of the faulty arm.

[0091] To expand the phase voltage output range of the MMC and at the same time increase the reference value of the capacitor voltage of the three-phase non-faulty sub-modules, there is:

[0092]

[0093] Among them, λ is the boosting coefficient of the sub-module capacitor voltage.

[0094] Correspondingly, the DC bus voltage will increase to:

[0095]

[0096] Among them, is the reference value of the DC bus voltage under fault-tolerant operation. The maximum value of the output voltage of the faulty arm and the maximum value of the output voltage of the faulty phase are further expressed as follows:

[0097]

[0098] Among them, u uFT_max is the maximum value of the output voltage of the faulty arm under the proposed fault-tolerant control method; u FT_max is the maximum value of the output voltage of the faulty phase under the proposed fault-tolerant control method.

[0099] Based on the reference value of the DC bus voltage, the reference value of the total energy stored in all sub-module capacitors is obtained as follows:

[0100]

[0101] S32. Inject the fundamental-frequency zero-sequence voltage to ensure the balance of the output line voltage.

[0102] The injected fundamental-frequency zero-sequence voltage needs to meet the following conditions: The injected fundamental-frequency zero-sequence voltage needs to meet the following conditions, that is, MA′ = AO; Let the number of faulty sub-modules be N f , then the amplitude of the zero-sequence voltage to be injected is expressed as:

[0103]

[0104] After injecting the zero-sequence voltage, the output amplitude of the non-faulty phase voltage also increases accordingly and will increase to:

[0105]

[0106] To avoid overmodulation, the output voltage amplitudes of other non-fault phases need to satisfy:

[0107]

[0108] The voltage boost coefficient λ under different numbers of faulty sub-modules needs to satisfy:

[0109]

[0110] Substitute the calculation formula of λ into the calculation formula of V b to obtain the amplitude of the required injected zero-sequence voltage, and then express the zero-sequence voltage reference value as:

[0111] u0 = V0 c os(ωt) where ω is the grid angular frequency.

[0112] S33. Adjust the arm energy to balance the energy and capacitor voltage between each arm.

[0113] Set the capacitor stored energy of the faulty sub-module to the rated value, ensure that the number of modules participating in the balancing control remains unchanged, and the virtual arm energy is:

[0114]

[0115] where E uj_vir is the virtual energy of the upper arm, Ω uj is the sequence set of the faulty sub-modules of the upper arm, E lj_vir is the virtual energy of the lower arm, Ω lj is the sequence set of the faulty sub-modules of the lower arm. Define the total virtual energy E vir stored in the capacitor as:

[0116]

[0117] The MMC adjusts the active power between three phases by injecting a DC circulating current component, and adjusts the active power between the upper arm and the lower arm by injecting a fundamental frequency circulating current component. The circulating current reference value of the MMC is:

[0118]

[0119] where K p1 is the control parameter of the arm energy and the controller, K p2 is the control parameter of the arm energy difference controller, E vir is the total virtual energy stored in the capacitors in the MMC, Φ is the intermediate parameter of the transfer matrix, and its calculation formula is:

[0120]

[0121] By adjusting the reference value of the circulating current, the balanced control of the energy among each bridge arm is realized.

[0122] Therefore, by adopting the above-mentioned fault detection and fault-tolerant control strategy for the modular multilevel converter in an MVDC ship, the fault detection process can be simplified. Only by setting a fixed algorithm gain coefficient can the accurate diagnosis of faults be achieved, which can greatly simplify the complexity of the algorithm, reduce the calculation amount, and is also beneficial to improving the fault-tolerant operation ability.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended 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 they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. Fault detection and fault tolerance control strategy for modular multilevel converters in MVDC ships, characterized in that: It includes the following steps: S1. Estimate the capacitor voltage based on the Luenberger observer; S2. Use the SM fault detection method to detect faults in each sub-module; S3. Obtain balanced line voltages by injecting zero-sequence voltages to avoid over-modulation in the faulty phase, and then balance the energies of each arm by adjusting the circulating current.

2. The fault detection and fault-tolerant control strategy for the modular multilevel converter in the MVDC ship according to claim 1, characterized in that: In S1, the voltage u of the SM capacitor is obtained by the MMC system circuit c It is: where, u c (t - 1), u c (t) are the capacitor voltage values at time t - 1 and time t respectively; s(t) is the switch function value at time t; i arm (t) is the arm current value at time t; Construct a Luenberger observer including a time update equation and a measurement update equation from the above formula.

3. The fault detection and fault tolerance control strategy for a modular multilevel converter in an MVDC ship according to claim 2, characterized in that: The construction method of the time update equation is as follows: discretize the calculation formula of the SM capacitor voltage u c to obtain the predicted state estimate value of the capacitor voltage: Among them, u c_c (k - 1) represents the estimated value of the capacitor voltage at time k - 1; u c_p (k) represents the predicted value of the capacitor voltage at time k; ΔT represents the sampling time; s(k) represents the switching function value at time k; i arm (k) represents the arm current value at time k; The construction method of the measurement update equation is as follows: According to the time update equation, the expression for the estimated value of the capacitor voltage based on the Luenberger observer is: u c_c y(k) = u c_p (k) + K(u c (k) - u c_p (k)); where, u c_c (k) represents the estimated value of the capacitor voltage at time k; K represents the gain coefficient of the Luenberger observer; u c (k) represents the measured value of the capacitor voltage at time k.

4. The fault detection and fault tolerance control strategy of the modular multilevel converter in the MVDC ship according to claim 3, characterized in that: In S2, the specific process is as follows: First, collect the capacitor voltage, arm current, and switch state of the i-th module, and calculate the estimated value u of the module capacitor voltage at time k according to the measurement update equation c_c (k), and calculate the predicted value u of the capacitor voltage at time k according to the time update equation c_p (k); Then, compare the measured value u c (k) of the capacitor voltage with the estimated value u c_c (k). Let Δu t be the error threshold of the capacitor voltage. If |u c_c (k) - u c (k)| > Δu t , and it lasts for at least a duration of Δt, then a fault is considered to have occurred; otherwise, it is considered to be operating normally, and this detection algorithm is continued to be executed.

5. The fault detection and fault tolerance control strategy of the modular multilevel converter in the MVDC ship according to claim 2, characterized in that: In S3, it specifically includes the following steps: S31. Adjust the total energy, and control the total stored energy of all sub-module capacitors in the MMC by adjusting the total energy reference value; S32. Inject the fundamental zero-sequence voltage to ensure the balance of the output line voltages; S33. Adjust the energies of the arms to balance the energies and capacitor voltages between each arm.

6. The fault detection and fault-tolerant control strategy of the modular multilevel converter in the MVDC ship according to claim 5, characterized in that: In S31, assume that there are N f faulty sub-modules in the upper bridge arm. Then the expressions for the maximum value of the output voltage of the faulty bridge arm and the maximum value of the output voltage of the faulty phase are as follows: where, u f_max is the maximum value of the output voltage of the faulty phase, and u uf_max is the maximum value of the output voltage of the faulty bridge arm; To expand the phase voltage output range of the MMC and simultaneously increase the capacitor voltage reference values of the three-phase non-faulty sub-modules, there is: where λ is the boosting coefficient of the sub-module capacitor voltage; The DC bus voltage increases to: Among them, is the DC bus voltage reference value under fault-tolerant operation; The maximum values of the output voltage of the faulty arm and the maximum value of the output voltage of the faulty phase are further expressed as follows: Among them, u uFT_max is the maximum value of the output voltage of the faulty bridge arm under the proposed fault-tolerant control method; u FT_max is the maximum value of the output voltage of the faulty phase under the proposed fault-tolerant control method; Based on the DC bus voltage reference value, the reference value of the total stored energy in all sub-module capacitors is obtained as follows:

7. The fault detection and fault tolerance control strategy of the modular multilevel converter in the MVDC ship according to claim 6, characterized in that: In S32, the injected fundamental zero-sequence voltage should meet the following conditions: the equivalent output range of the phase-a voltage is consistent before and after the fault, that is, MA′ = AO; let the number of faulty sub-modules be N f , then the amplitude of the injected zero-sequence voltage is expressed as: After injecting the zero-sequence voltage, the output amplitude of the non-faulty phase voltage also increases correspondingly, up to: The output voltage amplitudes of other non-faulty phases need to satisfy: The boosting coefficient λ under different numbers of faulty sub-modules needs to satisfy: Substitute the calculation formula of λ into the calculation formula of V b to obtain the amplitude of the required injected zero-sequence voltage, and then express the zero-sequence voltage reference value as: u0 = V0cos(ωt); where ω is the grid angular frequency.

8. The fault detection and fault-tolerant control strategy of the modular multilevel converter in the MVDC ship according to claim 7, characterized in that: In S33, set the stored energy of the capacitor of the faulty sub-module to the rated value to ensure that the number of modules participating in the balance control remains unchanged, and the virtual arm energy is: Among them, E uj_vir is the virtual energy of the upper arm, Ω uj is the set of serial numbers of the faulty sub-modules of the upper arm, E lj_vir is the virtual energy of the lower arm, Ω lj is the set of serial numbers of the faulty sub-modules of the lower arm; Define the total virtual energy E stored in the capacitor as: vir ​ The MMC adjusts the active power between the three phases by injecting a DC circulating current component, and adjusts the active power between the upper and lower arms by injecting a fundamental frequency circulating current component. The circulating current reference value of the MMC is: Among them, K p1 is the control parameter of the arm energy and the controller, and K p2 is the control parameter of the arm energy difference controller. E vir is the total virtual energy stored in the capacitors in the MMC. Φ is the intermediate parameter of the transfer matrix, and its calculation formula is: Realize the balanced control of the energies between each arm by adjusting the reference value of the circulating current.

Images