Active Fault Current Suppression Method Applicable to LCC / MMC Hybrid Cascaded DC Networks
By changing the order of submodule deployment in an LCC/MMC hybrid cascaded DC network and combining it with other current limiting methods, fault current is effectively suppressed. This solves the problem of high fault current in MMC and the submodule overvoltage problem existing in traditional methods, and achieves high-precision fault current analysis and suppression.
Patent Information
- Application Number
- CN202310488104.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-04
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-05-04
AI Technical Summary
Existing technologies struggle to effectively analyze and suppress fault currents in LCC/MMC hybrid cascaded DC networks, especially since MMC fault currents can be tens of times higher than the rated current. Furthermore, traditional current limiting methods suffer from submodule overvoltage and system stability issues.
An active fault current suppression method based on submodule ordering (SMS-FCL) is proposed. By changing the order in which submodules are put into operation, combined with traditional methods (such as N-FCL), fault current is suppressed, submodule overvoltage is avoided, and fault current is effectively suppressed.
It achieves high-precision fault current analysis and suppression for LCC/MMC hybrid cascaded DC networks, avoids submodule overvoltage problems, reduces energy loss during faults, and can be used in conjunction with other current limiting methods, with good scalability.
Smart Images

Figure CN116505736B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of DC power grid protection and control technology, specifically relating to an active fault current suppression method suitable for LCC / MMC hybrid cascaded DC networks. Background Technology
[0002] High-voltage direct current (HVDC) transmission technology based on line commutated converters (LCC-HVDC) is widely used in DC transmission due to its advantages of large capacity and mature technology. However, problems such as commutation failure and excessive reactive power consumption have restricted its development. Modular multilevel converter (MMC-HVDC) technology offers strong controllability and eliminates commutation failure issues, thus overcoming the shortcomings of LCC. However, it has higher costs and operating losses, and a smaller transmission capacity.
[0003] To fully leverage the advantages of LCC and MMC, researchers have proposed various novel hybrid DC transmission systems. Among them, the Baihetan-Jiangsu HVDC transmission project, currently under construction, has attracted widespread attention. This project employs an LCC at the sending end and a topology of an LCC cascaded with three parallel MMCs at the receiving end. This type of HVDC transmission system integrates the advantages of traditional DC transmission, such as high transmission capacity, strong controllability, and no commutation failures. However, the receiving-end grid contains two different types of converter stations, resulting in a complex system structure and fault characteristics that differ from those of single-type converter stations. Therefore, studying the fault characteristics of hybrid cascaded LCC / MMC-HVDC (HC-HVDC) systems, particularly solving the mathematical expressions for the fault currents of the receiving-end LCC and MMC during DC-side faults, and proposing corresponding short-circuit current suppression methods, has significant theoretical and engineering implications for the research of hybrid multi-terminal HVDC systems.
[0004] Researchers have conducted in-depth studies and achieved fruitful results regarding the fault characteristics of traditional LCC-HVDC systems. Some researchers have constructed small-signal models of the LCC and calculated in detail the parameters of the LCC DC and AC equivalent circuits considering the inverter commutation process and transformer leakage reactance. Others have used the state-space method to analyze the DC fault characteristics when the constant current command value of the sending-end LCC converter station experiences a 5% step change. Still others have used the least squares method to linearize the firing angle of the sending-end LCC during DC faults, and analyzed the fault current expression considering the controller response characteristics. These numerical calculation methods can provide relatively accurate fault current expressions. However, these methods involve large computational loads, and the short-circuit current expression has limitations in fault characteristic analysis. Similarly, researchers have proposed numerous methods for analyzing DC fault currents in MMC transmission systems: some scholars have equated the MMC discharge circuit during a fault to an RLC circuit, thus analyzing the DC fault current expression for the MMC; others have further considered the influence of the AC side on the DC short-circuit current, providing a more accurate expression for the MMC DC fault current; still others have used a decoupling method based on the network inductance ratio to simplify the complex discharge circuit of multi-terminal MMCs into a superposition of multiple second-order RLC circuits, thus analyzing the DC side fault current expression for multi-terminal MMCs. Regarding the analysis of DC fault currents in HC-HVDC systems, some scholars have constructed a mathematical model of the DC line current when the receiving-end LCC experiences a commutation failure fault using state equations; some researchers have fully considered the response characteristics of the LCC and MMC controllers and the DC line parameters, constructing a small-signal model of the LCC / MMC hybrid cascade system. However, none of these methods provide a method for calculating the analytical DC fault current. Therefore, this patent constructs an equivalent discharge circuit for DC faults in an HC-HVDC system and uses the Laplace transform method to analyze the analytical expression of the short-circuit current when a DC fault occurs at different locations in the system, which can provide a certain theoretical basis for the protection of hybrid multi-terminal high-voltage DC transmission lines.
[0005] Similarly, DC fault characteristic analysis can provide a theoretical basis for fault current limiting technology. The clamping effect of the thyristors inside the LCC converter station will force the DC fault current of the LCC and LCC / MMC cascaded inverters to rapidly decay from the rated value to 0; while the discharge of the submodule (SM) capacitor can cause the MMC fault current to be as high as ten times the rated current, so it is necessary to study the active suppression method of MMC fault current. Some scholars have introduced the submodule duty cycle D, and achieved fault current limiting by reducing D during a fault; some scholars have proposed an adaptive fault current limiting control method, which suppresses the discharge of submodule capacitors by adaptively reducing the number of submodules in operation during a fault, thereby achieving fault current limiting; some scholars indirectly reduce the number of submodules by changing the MMC outer loop controller reference value (Udc reference based fault current limiting strategy, U-FCL) to achieve fault current limiting. While these active current limiting methods based on reducing the number of submodules (N-FCL) offer good current limiting performance, they suffer from severe submodule capacitor overvoltage problems. Furthermore, they can cause sudden changes in AC and DC voltages, threatening the stability of AC / DC systems. Therefore, this current limiting method has certain limitations. To limit short-circuit current while avoiding these problems, a different approach can be adopted: changing the way submodules are connected without altering the number of submodules.
[0006] Based on this, this patent proposes an active current limiting method based on submodule sorting (SMS-FCL), which achieves MMC fault current suppression by changing the order in which submodules are put into operation. Summary of the Invention
[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an active fault current suppression method suitable for LCC / MMC hybrid cascaded DC networks, providing a certain theoretical basis for the protection and control of LCC / MMC hybrid cascaded systems.
[0008] The technical problem solved by this invention is achieved through the following technical solution:
[0009] An active fault current suppression method applicable to LCC / MMC hybrid cascaded DC networks, characterized in that: the method comprises the following steps:
[0010] S1. System Topology Introduction and Analytical Expression of DC Side Short-Circuit Fault Current and Equivalent Discharge Circuit of LCC and MMC Converter Stations
[0011] The hybrid cascaded DC network consists of an LCC rectifier station and an LCC / MMC hybrid cascaded inverter station. The LCC / MMC hybrid cascaded inverter station is composed of one LCC and three parallel MMCs connected in series. The rectifier-side LCC adopts constant current control to maintain the DC line current at the rated value; the inverter-side LCC adopts constant voltage control to support the DC voltage; among the three parallel MMC converter stations, MMC1 adopts constant DC voltage and constant reactive power control to provide DC voltage together with the inverter-side LCC, while MMC2 and MMC3 adopt constant active power and constant reactive power control to achieve reasonable power distribution among the MMC stations.
[0012] The possible fault points in this DC network are f1, f2 and f3, where f1 is a short circuit fault on the LCC output side of the receiving end; f2 is a ground fault on the positive terminal of the MMC of the receiving end, which causes a short circuit fault at the output of the MMC; f3 is a ground fault on the positive terminal of the DC line, which causes a short circuit fault in the LCC / MMC hybrid cascaded inverter.
[0013] Based on the above analysis, the analytical expressions for DC-side short-circuit fault currents and equivalent discharge circuits of LCC and MMC converter stations are as follows:
[0014] An LCC converter station consists of multiple 6-pulse LCC converters. To simplify the analysis, we will take one 6-pulse inverter as an example. The DC side voltage u dc1 It can be represented as:
[0015]
[0016] Where: U is the effective value of the AC line voltage on the secondary side of the transformer;
[0017] X T =ωL T For the leakage reactance of the LCC transformer;
[0018] β is the lead-out firing angle;
[0019] μ is the commutation overlap angle, calculated as follows:
[0020]
[0021] Based on the block diagram of the constant voltage PI controller of the LCC inverter, β is written as:
[0022]
[0023] Where: u dc1ref This is the reference value for the DC voltage of the LCC converter station on the inverter side;
[0024] K p K i These are the proportional and integral coefficients of the PI controller, respectively.
[0025] When a fault occurs on the DC side of the LCC, β will rapidly decrease to its minimum value β under the action of the PI controller. min To maintain DC voltage, after a DC fault occurs, due to the rapid change rate of the inverter angle, and the relationship between β0 and β... min Since the phase difference is small, the change in β can be ignored, meaning that β is assumed to remain constant at β0 after the fault. Under this assumption, the Laplace equivalent circuit for a DC fault in an LCC can be obtained, where the short-circuit current in the complex frequency domain can be expressed as:
[0026]
[0027] Where: β0 is the inverter angle before the fault, μ0 can be calculated by equation (2), and the equivalent parameter is calculated as follows:
[0028]
[0029] Combining equation (4), its equivalent discharge circuit can be obtained. Performing an inverse Laplace transform on equation (4), the expression for the LCC short-circuit current in the time domain can be obtained:
[0030]
[0031] Among them: U eq10 and I dc10 The steady-state voltage and current values before the fault are given, and τ is a time constant, calculated as follows:
[0032]
[0033] Differentiating equation (6), we get:
[0034]
[0035] Within a very short time after the fault occurs (within 5ms), if τ is large, then t / τ≈0, and equation (8) can be further simplified to:
[0036]
[0037] As can be seen from equation (9), after the fault occurs, the LCC short-circuit current exhibits an approximately linear trend. When i dc1 After the current drops to 0, due to the clamping effect of the thyristor, the short-circuit current will remain constant at 0. Therefore, the LCC short-circuit fault can be divided into two stages: transient and steady state. Let i dc1 =0, we can obtain the transient time t. p1 :
[0038]
[0039] In the structure of a half-bridge MMC, udc2 and i dc2 These represent the DC voltage and current of the MMC, R0 and L0 represent the bridge arm resistance and reactance, respectively, C0 is the submodule capacitor, and L... dc2 For smoothing reactor, v pj and v nj Let be the upper and lower arm voltages of phase j (j = a, b, c), respectively. From KCL and KVL, we can obtain:
[0040]
[0041]
[0042] Summing the three-phase differential equations and substituting them into (11), we get:
[0043]
[0044] When a fault occurs on the DC side, the DC voltage can be expressed as:
[0045]
[0046] Combining equations (13) and (14), we can obtain:
[0047]
[0048] in:
[0049]
[0050] Within a very short time (within 7ms) after a fault occurs, the discharge time of the submodule capacitor is very short, so its voltage change can be ignored, and the bridge arm voltage v can be approximated. ij It is a constant voltage source; and at the initial moment of the fault, there is U dc20 =v pj +v nj Therefore, according to equation (15), the analytical expression for the MMC short-circuit current can be obtained as follows:
[0051]
[0052] The calculation method for τ is the same as that for equation (7). According to equation (17), the MMC equivalent discharge circuit can be obtained.
[0053] S2. Derive the short-circuit current expression and analyze the fault characteristics of LCC / MMC hybrid cascaded DC networks.
[0054] Using the equivalent discharge circuits of LCC and MMC derived in step S1 above, the equivalent discharge circuit for DC faults in the LCC / MMC hybrid cascaded system is obtained, and the system equivalent parameters are obtained:
[0055]
[0056] The equivalent discharge circuit of the HC-HVDC system is a first-order RL circuit. The short-circuit current of the LCC / MMC hybrid cascaded inverter is expressed as:
[0057]
[0058] Where τ can be calculated by equation (7), due to the unidirectional conduction of the thyristors in the LCC station, after a DC fault occurs in the LCC / MMC hybrid cascade system, the short-circuit current will decrease from the rated value to 0, and the MMC submodule capacitors will not be able to discharge through the equivalent circuit, thus not causing system overcurrent problems. Let i dc3 =0, we can obtain the transient time t. p2 :
[0059]
[0060] During the transient phase, if t / τ≈0, the short-circuit current changes approximately linearly, therefore i can be... dc3 The rate of change is expressed as:
[0061]
[0062] By combining the calculated and simulated waveforms of short-circuit current in LCC, MMC, and LCC / MMC hybrid cascaded systems, it can be observed that when a DC fault occurs in an LCC or LCC / MMC cascaded converter, the clamping effect of the thyristor limits the fault current to near 0, and the system does not have an overcurrent problem; however, when a short-circuit fault occurs in an MMC, the fault current is as high as ten times the rated current, so it is necessary to suppress its short-circuit current.
[0063] S3. MMC Active Short-Circuit Current Suppression Strategy Based on Submodule Sorting Algorithm
[0064] Equation (15) gives the bridge arm voltage v ij With MMC short-circuit current i dc2 To study the relationship between the submodule capacitor voltage and the short-circuit current, further research on v is needed. ij with u c Given the relationship, let the time of the fault occurrence be t1, and consider the time after the fault occurs as t. i Within one IGBT switching cycle Δt, v ij It can be represented as:
[0065]
[0066] Wherein: S ijk The switching function for the k-th submodule on phase i of the bridge arm can be expressed as:
[0067]
[0068] Let t i -t i The number of sub-modules deployed on the i-th arm of phase j within the time interval +Δt is N. ij , i.e. ∑S ijk =N ij Then equation (22) can be written as:
[0069]
[0070] Substituting equation (24) into equation (15), we get:
[0071]
[0072] in: It depends only on the number of submodules deployed, and not on the initial voltage when the submodules are deployed; It is related to both the initial voltage of the submodule and the number of submodules.
[0073] Under normal operation, the capacitor voltage differences among the submodules are minimal under the capacitor voltage balancing strategy, all remaining near their rated values. However, when a fault occurs, the capacitor voltages of the submodules will exhibit some deviation. Therefore, by changing the submodule activation method during a fault, the voltage difference can be reduced. ij This enables the suppression of fault current;
[0074] In this invention, SMS-FCL establishes an x-axis representing the order of submodules and a y-axis representing the capacitor voltage of the submodules. SMS-FCL changes the traditional submodule sorting algorithm's order of submodule deployment: when P>0, the submodule with the highest voltage is deployed; when P<0, the submodule with the lowest voltage is deployed. Furthermore, to avoid submodule capacitor overvoltage issues, voltages exceeding the rated voltage U are... cn The submodules are configured to maintain the traditional submodule sorting algorithm.
[0075] Moreover, the bridge arm voltage in the SMS-FCL is related to both the method and number of sub-modules. The combination of SMS-FCL and N-FCL further reduces the number of sub-modules and further suppresses short-circuit current based on SMS-FCL. Suppose that the number of sub-modules becomes m times the original number after this current limiting method is implemented, then equation (24) becomes:
[0076]
[0077] The conditions under which SMS-FCL can suppress short-circuit current are that the values of each bridge arm submodule are not zero and the number of overvoltage submodules is less than the number of submodules put into operation in each bridge arm. Therefore, the conditions for current limiting when SMS-FCL is used in conjunction with N-FCL are expressed as follows:
[0078]
[0079] The advantages and beneficial effects of this invention are as follows:
[0080] 1. This invention is applicable to an active fault current suppression method for LCC / MMC hybrid cascaded DC networks. The proposed analytical expression for DC short-circuit current in LCC and LCC / MMC hybrid cascaded systems has high calculation accuracy when the transition resistance is low and the fault location is close to the receiving end; when the transition resistance is high or the fault is close to the sending end, it can also correctly reflect the trend of short-circuit current change.
[0081] 2) This invention is applicable to an active fault current suppression method for LCC / MMC hybrid cascaded DC networks. It can simultaneously achieve DC current suppression, DC voltage and reactive power support, and reduce the energy loss of submodule capacitors during faults, while avoiding the submodule overvoltage problem that exists in traditional fault current limiting methods.
[0082] 3) This invention is applicable to the active fault current suppression method of LCC / MMC hybrid cascaded DC network. It has good scalability and can be used in conjunction with various active current limiting methods to further suppress short-circuit current. Moreover, the conditions for cooperation are easy to meet. Attached Figure Description
[0083] Figure 1 This is a schematic diagram of the LCC-LCC / MMC system and its control system diagram of the present invention;
[0084] Figure 2 This is the basic topology and equivalent circuit diagram of the LCC-based inverter of this invention;
[0085] Figure 3 This is a diagram of the MMC structure and fault discharge equivalent circuit of the present invention.
[0086] Figure 4 This is the fault equivalent circuit diagram of the LCC / MMC hybrid cascaded converter station of the present invention;
[0087] Figure 5 The graph shows the calculated and simulated values of the short-circuit current in this invention.
[0088] Figure 6 This is a diagram of the sorting algorithm for a submodule of the present invention;
[0089] Figure 7 This is a schematic diagram of the rate limiting strategy proposed in this invention;
[0090] Figure 8 This is a diagram of the switching functions of a submodule of the present invention;
[0091] Figure 9This is a waveform diagram of the capacitor voltage of the submodule of the present invention;
[0092] Figure 10 This is the analytical calculation diagram of the LCC fault current in this invention;
[0093] Figure 11 This is a diagram showing the analytical calculation of LCC fault current under different controllers, smoothing reactors, and transition resistors according to the present invention.
[0094] Figure 12 This is a diagram illustrating the analytical calculation of fault current in the LCC / MMC hybrid cascaded system of this invention.
[0095] Figure 13 This is an analytical calculation diagram of the fault current at different fault locations in the LCC / MMC hybrid cascade system of the present invention;
[0096] Figure 14 This is an analytical calculation diagram of the fault current under different transition resistances in the LCC / MMC hybrid cascaded system of this invention;
[0097] Figure 15 This is a diagram showing the voltage and current of the MMC bridge arm in this invention.
[0098] Figure 16 This is the fault current limiting diagram for the MMC in this invention;
[0099] Figure 17 This is a diagram of MMC fault current under different transition resistances according to the present invention;
[0100] Figure 18 The diagram shows the fault current of MMC under different current limiting methods of this invention.
[0101] Figure 19 This is a comparison chart of the current limiting methods of the present invention: no action, N-FCL, U-FCL, and SMS-FCL.
[0102] Figure 20 This invention provides a current limiting diagram for N-FCL and U-FCL in conjunction with SMS-FCL.
[0103] Figure 21 This is a diagram illustrating the current limiting effect of the N-FCL and U-FCL combined with the SMS-FCL of the present invention.
[0104] Figure 22 This diagram illustrates the current limiting effect of N-FCL combined with SMS-FCL when m=0 according to the present invention. Detailed Implementation
[0105] The present invention will be further described in detail below through specific embodiments. The following embodiments are merely descriptive and not limiting, and should not be used to limit the scope of protection of the present invention.
[0106] An active fault current suppression method applicable to LCC / MMC hybrid cascaded DC networks includes:
[0107] 1. System topology introduction and analytical expressions for DC-side short-circuit fault currents in LCC and MMC converter stations, as well as equivalent discharge circuits.
[0108] The LCC-LCC / MMC hybrid cascaded high-voltage direct current transmission positive electrode topology based on the Baihetan Project is as follows: Figure 1 As shown in the diagram, the entire system consists of an LCC rectifier station and an LCC / MMC hybrid cascaded inverter station. The LCC / MMC hybrid cascaded inverter station comprises one LCC connected in series with three parallel MMCs. The rectifier-side LCC uses constant current control to maintain the DC line current at its rated value; the inverter-side LCC uses constant voltage control to support the DC voltage; among the three parallel MMC converter stations, MMC1 uses constant DC voltage and constant reactive power control to provide DC voltage together with the inverter-side LCC, while MMC2 and MMC3 use constant active power and constant reactive power control to achieve reasonable power distribution among the MMC stations.
[0109] The possible faults in this DC transmission system are as follows: Figure 1 As shown in f1, f2, and f3, f1 represents a short-circuit fault on the LCC output side at the receiving end; f2 represents a ground fault on the positive terminal of the MMC at the receiving end, causing a short-circuit fault at the MMC output; and f3 represents a ground fault on the positive terminal of the DC line, causing a short-circuit fault in the LCC / MMC hybrid cascaded inverter. The analytical expression for the short-circuit fault current on the DC side of the LCC and MMC converter stations, as well as the equivalent discharge circuit, will be analyzed below.
[0110] In actual engineering, an LCC converter station consists of multiple 6-pulse LCC converters. To simplify the analysis, we will take a single 6-pulse inverter as an example. Its topology is as follows: Figure 2 As shown in (a), the DC side voltage u dc1 It can be represented as:
[0111]
[0112] Where U is the effective value of the AC line voltage on the secondary side of the transformer, and X T =ωL T Let β be the leakage reactance of the LCC transformer, β be the leading trigger angle, and μ be the commutation overlap angle. The calculation is as follows:
[0113]
[0114] according to Figure 1 The block diagram of the constant voltage PI controller in the LCC inverter shows that β can be written as:
[0115]
[0116] Where: u dc1ref K is the reference value for the DC voltage of the inverter-side LCC converter station. p K i These are the proportional and integral coefficients of the PI controller, respectively. When a fault occurs on the DC side of the LCC, β will rapidly decrease to its minimum value β under the action of the PI controller. min To maintain DC voltage. Traditional methods obtain numerical solutions for short-circuit current by iteratively solving equations (1)-(3). However, this method involves a large amount of computation and cannot obtain analytical expressions for short-circuit current. Therefore, it is necessary to study analytical calculation methods for short-circuit current. After a DC fault occurs, due to the rapid change rate of the inverter angle and the relationship between β0 and β... min The difference is small, so the change in β can be ignored; that is, β is assumed to remain unchanged at β0 after the fault. Under this assumption, combined with... Figure 2 From equations (a) and (1), we can obtain the Laplace equivalent circuit for a DC fault in an LCC, such as... Figure 2 As shown in (b), the short-circuit current in the complex frequency domain can be expressed as:
[0117]
[0118] In the formula: β0 is the inverter angle before the fault, μ0 can be calculated by formula (2), and the equivalent parameter is calculated as follows:
[0119]
[0120] Combining formula (4) and Figure 2 (b) Its equivalent discharge circuit can be obtained, such as Figure 2 As shown in (c). Performing an inverse Laplace transform on equation (4), we can obtain the expression for the LCC short-circuit current in the time domain:
[0121]
[0122] U eq10 and I dc10 The steady-state voltage and current values before the fault are given, and τ is a time constant, calculated as follows:
[0123]
[0124] Differentiating equation (6), we get:
[0125]
[0126] Within a very short time after the fault occurs (within 5ms), if τ is large, then t / τ≈0, and equation (8) can be further simplified to:
[0127]
[0128] As can be seen from equation (9), after the fault occurs, the LCC short-circuit current exhibits an approximately linear trend. When i dc1 After the current drops to 0, due to the clamping effect of the thyristor, the short-circuit current will remain constant at 0. Therefore, the LCC short-circuit fault can be divided into two stages: transient and steady state. Let i dc1 =0, we can obtain the transient time t. p1 :
[0129]
[0130] The structure of a half-bridge MMC is as follows Figure 3 As shown in (a), where u dc2 and i dc2 These represent the DC voltage and current of the MMC, R0 and L0 represent the bridge arm resistance and reactance, respectively, C0 is the submodule capacitor, and L... dc2 For smoothing reactor, v pj and v nj These are the upper and lower bridge arm voltages of phase j (j = a, b, c), respectively. Figure 3 (a) From KCL and KVL, we can obtain:
[0131]
[0132]
[0133] Summing the three-phase differential equations and substituting them into (11), we get:
[0134]
[0135] When a fault occurs on the DC side, the DC voltage can be expressed as:
[0136]
[0137] Combining equations (13) and (14), we can obtain:
[0138]
[0139] in:
[0140]
[0141] Within a very short time (within 7ms) after a fault occurs, the discharge time of the submodule capacitor is very short, so its voltage change can be ignored, and the bridge arm voltage v can be approximated. ij It is a constant voltage source; and at the initial moment of the fault, there is U dc20 =v pj +v njTherefore, according to equation (15), the analytical expression for the MMC short-circuit current can be obtained as follows:
[0142]
[0143] The calculation method for τ is the same as that for equation (7). According to equation (17), the equivalent discharge circuit of MMC can be obtained, such as... Figure 3 As shown in (b).
[0144] 2. Short-circuit current expression and fault characteristic analysis of LCC / MMC hybrid cascaded system
[0145] Using the LCC and MMC equivalent discharge circuits derived above, we can obtain Figure 1 The equivalent discharge circuit for DC faults in the LCC / MMC hybrid cascaded system shown is as follows: Figure 4 As shown.
[0146] according to Figure 4 The equivalent parameters of the system can be obtained:
[0147]
[0148] Depend on Figure 4 It can be seen that the equivalent discharge circuit of the HC-HVDC system is a first-order RL circuit. Based on the previous analysis, the short-circuit current of the LCC / MMC hybrid cascaded inverter can be expressed as:
[0149]
[0150] τ can be calculated from equation (7).
[0151] Due to the unidirectional conduction of the thyristors within the LCC station, in the event of a DC fault in the LCC / MMC hybrid cascaded system, the short-circuit current will decrease from its rated value to 0. The MMC submodule capacitors will not be able to discharge through the equivalent circuit, thus preventing system overcurrent problems. Let i dc3 =0, we can obtain the transient time t. p2 :
[0152]
[0153] During the transient phase, if t / τ≈0, the short-circuit current changes approximately linearly, therefore i can be... dc3 The rate of change is expressed as:
[0154]
[0155] Figure 5 Calculated and simulated waveforms of short-circuit currents in LCC, MMC, and LCC / MMC hybrid cascade systems are presented. Observation shows that the calculated and simulated values of each short-circuit current fit well. Combined with the foregoing analysis... Figure 5 When a DC fault occurs in an LCC or LCC / MMC cascaded converter, the clamping effect of the thyristor limits the fault current to near 0, and the system does not have an overcurrent problem; however, when a short-circuit fault occurs in an MMC, the fault current is as high as ten times the rated current, so it is necessary to suppress its short-circuit current.
[0156] 3. MMC Active Short-Circuit Current Suppression Strategy Based on Submodule Sorting Algorithm
[0157] Equation (15) gives the bridge arm voltage v ij With MMC short-circuit current i dc2 To study the relationship between the submodule capacitor voltage and the short-circuit current, further research on v is needed. ij with u c The relationship. Let the time of the fault occurrence be t1, and consider the time after the fault occurs as t. i Within one IGBT switching cycle Δt, v ij It can be represented as:
[0158]
[0159] Where S ijk The switching function for the k-th submodule on phase i of the bridge arm can be expressed as:
[0160]
[0161] Let t i -t i The number of sub-modules deployed on the i-th arm of phase j within the time interval +Δt is N. ij , i.e. ∑S ijk =N ij Then equation (22) can be written as:
[0162]
[0163] Substituting equation (24) into equation (15), we get:
[0164]
[0165] in It depends only on the number of submodules deployed, and not on the initial voltage when the submodules are deployed; The voltage variation is related to both the initial voltage of the submodules and the number of submodules. Under normal operation, with the capacitor voltage balancing strategy, the difference in capacitor voltage among the submodules is minimal, all hovering near their rated values. However, when a fault occurs, the capacitor voltage of the submodules will exhibit a certain deviation. Therefore, the voltage variation can be reduced by changing the submodule activation method during a fault. ij This enables the suppression of fault current.
[0166] Based on the above analysis, this invention proposes an active rate limiting method based on submodule sorting. Traditional submodule sorting algorithms and the SMS-FCL proposed in this invention are as follows: Figure 6 As shown in the figure, the x-axis represents the submodule order, and the y-axis represents the submodule capacitor voltage. Traditional submodule sorting algorithms (as shown in region 1) activate the submodule with the lowest capacitor voltage when the power direction is positive, and the submodule with the highest voltage when the power direction is negative. To limit fault current, the SMS-FCL changes the submodule activation order (as shown in region 2): when P>0, the submodule with the highest voltage is activated; when P<0, the submodule with the lowest voltage is activated. Furthermore, to avoid submodule capacitor overvoltage problems, voltages exceeding the rated voltage U will be... cn The submodules are configured to maintain the traditional submodule sorting algorithm (as shown in region 3).
[0167] Depend on Figure 6 It is understood that the method of the present invention increases the bridge arm voltage during the charging stage of the submodule, suppresses the charging process of the submodule, and reduces the charging current of each bridge arm; it reduces the bridge arm voltage during the discharging process of the submodule, slows down the discharge process of the submodule to the fault circuit, and reduces the bridge arm current, thus suppressing the growth rate of the short-circuit current.
[0168] To further explain how to use SMS-FCL to limit fault current Figure 7 A schematic diagram of the current limiting strategy is provided. As shown in the diagram, the system employs a traditional submodule voltage sequencing algorithm during normal operation and fault detection phases. Once a fault is detected, the SMS-FCL is activated to limit the fault current. Since the SMS-FCL only operates after fault detection, it does not affect the normal operation of the system. After a fault occurs, the system switches back to the traditional voltage sequencing algorithm to rebalance the capacitor voltage and achieve fault recovery.
[0169] To analyze the impact of the proposed proactive rate limiting strategy on submodules Figure 8 The switching function diagram of the upper bridge arm submodule of phase a was drawn. (Comparison) Figure 8 As can be seen from (a) and (b), compared with no action, the switching function of each submodule changed significantly after the active current limiting method was applied. However, the total number of submodules applied in the same time period was the same, and the capacitor voltage of each submodule did not change. Therefore, compared with N-FCL, the method proposed in this invention will not cause submodule overvoltage problems.
[0170] Figure 9 The voltage waveform of the upper bridge arm submodule of phase a was plotted and compared. Figure 9As can be seen from (a) and (b), after the SMS-FCL is engaged, the capacitor voltages of the submodules exhibit an unbalanced state. The capacitor voltages of some submodules remain unchanged, while the capacitor voltages of the remaining submodules gradually decrease to 0. During the entire fault period, compared to no operation, the following are observed:
[0171]
[0172] Equation (26) shows that the proposed method can increase the sum of the capacitor voltages of the submodules, thus significantly increasing the capacitor energy of the submodules. This suppresses the capacitor discharge process during the fault and reduces the short-circuit current rise rate. On the other hand, it reduces the energy loss of the submodules during the fault, thereby enabling a faster fault recovery process.
[0173] 4. A current limiting method combining SMS-FCL and N-FCL
[0174] Equation (25) shows that the bridge arm voltage is related to both the submodule activation method and the number of submodules. However, SMS-FCL only changes the submodule sorting method and does not involve changing the number of submodules. If the number of submodules is further reduced on this basis, the short-circuit current can be further suppressed on the basis of SMS-FCL. N-FCL mainly achieves fault current limiting by directly or indirectly reducing the number of submodules during a fault. Suppose that after this current limiting method is activated, the number of submodules becomes m times the original number, then equation (24) becomes:
[0175]
[0176] according to Figure 6 It can be seen that the condition for SMS-FCL to suppress short-circuit current is that the submodules of each bridge arm are not zero and the number of overvoltage submodules is less than the number of submodules put into operation in each bridge arm. Therefore, the condition for current limiting by SMS-FCL in combination with N-FCL can be expressed as:
[0177]
[0178] To verify the correctness of the above fault current analysis and the effectiveness of the active current limiting method, this patent built a system in PSCAD / EMTDC software. Figure 1 The simulation model of the LCC-LCC / MMC DC transmission system is shown. In the simulation, a short-circuit fault occurs at the LCC converter station outlet at 1 second; a single-pole ground fault occurs on the DC line at 2 seconds, causing a fault in the LCC / MMC cascaded system; and a short-circuit fault occurs at the MMC converter station outlet at 3 seconds. Specific simulation parameters are shown in Tables 1-3. Table 1 shows the main simulation parameters for the LCC; Table 2 shows the main simulation parameters for the MMC; and Table 3 shows the main simulation parameters for the DC line.
[0179] Table 1
[0180]
[0181]
[0182] Table 2
[0183] parameter numerical values AC voltage 525kV Rated power 666.67MW DC voltage 400kV DC current 1.67kA Number of submodules N 218 <![CDATA[Bridge arm resistor R0]]> 0.5Ω <![CDATA[Arm inductor L0]]> 50mH <![CDATA[Submodule capacitor C0]]> 0.011F
[0184] Table 3
[0185] parameter numerical values length 2000km DC line resistance 0.00543Ω / km DC line inductance 0.80787mH / km DC line capacitor 0.01409μF / km
[0186] Figure 10 The figure presents a comparison of the simulated and calculated values of the LCC short-circuit current. Observation shows that the simulated and calculated values of the short-circuit current exhibit a generally consistent trend, with only a slight difference at the point where the current is zero. The simulated and calculated transient times are 4.50 ms and 4.56 ms, respectively, with an error of only 0.06 ms (1.33%). The simulated and calculated rates of change of the short-circuit current are -1108.9 kA / s and -1096.5 kA / s, respectively, with an error of 1.12%. This indicates that the proposed analytical expression for the short-circuit current can accurately reflect the changes in the LCC short-circuit current after a fault, and has high precision, providing a theoretical basis for fault detection and corresponding protection methods.
[0187] To further analyze the applicability of the proposed analytical formula for short-circuit current Figure 11 The simulated and calculated waveforms of the short-circuit current under different controllers, different smoothing reactors, and different transition resistors of the LCC inverter are presented respectively. The controller parameters are shown in Table 4.
[0188] Table 4
[0189] parameter <![CDATA[K p ]]> <![CDATA[T i ]]> PI controller 1 1 0.018 PI controller 2 1 0.2 PI controller 3 0.5 0.018
[0190] Figure 11 (a) shows that the short-circuit current waveforms under different PI controller parameters are very similar and are very close to those under constant β control. This indicates that although the inverter angle β changes slightly after a DC fault, this change has little impact on the short-circuit current. It can be approximated that the inverter angle remains at the pre-fault value β0, and the short-circuit current can be calculated accordingly. Furthermore, the proposed analytical expression can adapt to changes in PI controller parameters within a certain range and has strong resistance to parameter disturbances.
[0191] Figure 11 (b) The simulated and calculated waveforms of the short-circuit current under different smoothing reactors are given. It can be seen that when the smoothing reactor increases, the transient time t... p1 The proposed analytical expression for short-circuit current can accurately reflect the changes in short-circuit current under different smoothing reactor parameters, and the accuracy of each calculation result is similar. Figure 11 (c) The simulated and calculated waveforms of the short-circuit current under different transition resistances are presented. As shown in the figure, when the transition resistance is below 50Ω, the calculated and simulated waveforms of the short-circuit current match well. However, when the transition resistance is large, a certain difference appears between the calculated and simulated waveforms. This is mainly due to neglecting the influence of the DC line current on the transition resistance. Although a certain error appears between the simulated and calculated values when the transition resistance is large (the maximum error is 25.97%), the analytical expression can still correctly reflect the changing trend of the LCC short-circuit current, providing a new approach for high-resistance fault detection.
[0192] Figure 12 The simulation and calculated waveforms of the short-circuit current in the LCC / MMC hybrid cascaded system are presented. The calculated values show a linear trend, while the simulation values, although exhibiting slight fluctuations, generally agree with the calculated values. Regarding key parameters, the simulation and calculated transient times are 2.42 ms and 2.36 ms, respectively, with an error of 0.06 ms (2.48%). The simulation and calculated short-circuit current change rate are -2064.6 kA / s and -2118.6 kA / s, respectively, with an error of 2.62%. Therefore, the proposed analytical expression for the short-circuit current is relatively accurate. Figure 10 In comparison, the short-circuit current of the LCC / MMC hybrid cascade system changes faster and the transient time is shorter, which is also an important difference between the DC fault characteristics of the LCC / MMC hybrid cascade system and the DC fault characteristics of LCC.
[0193] In the above simulation and theoretical calculation waveforms, the fault point is located at the end of the DC line (x=1). In order to analyze the influence of the DC line on the analytical expression of the short-circuit current, Figure 13 The simulated and calculated waveforms of the short-circuit current in the LCC / MMC cascaded system under different fault locations are presented. A PI-type DC line model is used for the DC line, while an RL line is used for comparison. Figure 13 It can be seen that the calculated short-circuit current fits well with the simulated value under the RL line model, while the short-circuit current of the PI type DC line is higher than the calculated value. This is because the DC line is equivalent to an RL series circuit in the analytical calculation, and the contribution of distributed capacitance to the short-circuit current during a fault is ignored. At the same time, the closer the fault point is to the inverter side, the smaller the influence of distributed capacitance on the short-circuit current, and the smaller the difference between the two DC line models, and the smaller the error between the calculated value and the simulated value.
[0194] Figure 14The simulated and calculated waveforms of short-circuit current under different transition resistances are presented. The fault point is located at the end of the DC line. Under four different transition resistance values, the calculated values can accurately reflect the changing trend of the short-circuit current. Except for the case where the transition resistance is 100Ω, the error is relatively large. Therefore, the short-circuit current analysis method proposed in this invention can provide a theoretical reference for DC fault detection and fault location in LCC / MMC cascaded systems.
[0195] To verify the effectiveness of the proactive rate limiting method mentioned above Figure 15 The voltage and current waveforms of each arm of the MMC are presented when using the current limiting method (SMS-FCL) proposed in this invention and when there is no action. Figure 16 The waveforms of various parts of the MMC converter station under constant voltage control mode are presented. A bipolar permanent metallic short-circuit fault is set at 3s, with a fault detection time of 3ms. The circuit breaker operates 4ms after fault detection, interrupting the fault current. The DC circuit breaker (DCCB) adopts the hybrid DC circuit breaker topology proposed by ABB. For ease of comparison with the SMS-FCL, only... Figure 16 (a) shows the circuit breaker action, while the other images assume that the circuit breaker does not trip during a fault.
[0196] Depend on Figure 15 It can be seen that during active current limiting, the voltage and current of each bridge arm are positive, and the submodule is in a discharging state at this time. Figure 15 The bridge arm voltage under no-action and active current limiting conditions shows that the bridge arm voltage decreases under active current limiting. This is because the discharge process of the capacitors in each submodule is suppressed, thus reducing the bridge arm current. This is also reflected in... Figure 15 This is reflected in the current waveform of the middle bridge arm, proving the correctness of the previous analysis.
[0197] Under constant voltage control, without current limiting control, the short-circuit current of the MMC will rapidly rise to 27.77kA after a short-circuit fault, which will severely impact power electronic devices. Using SMS-FCL can reduce the short-circuit current by 2.22kA, alleviating the current stress on each submodule and reducing the switching capacity of the DCCB. Furthermore, within 4ms after current limiting control is activated, active control suppresses the rise of the short-circuit current by reducing the DC voltage. Afterward, the DC voltage under active control will be higher than the DC voltage without control. 10ms after the fault occurs, the active control method increases the DC voltage from 1.40kV to 24.79kV, an increase of 16.71 times, providing higher voltage support for the DC line and reducing the impact of the faulty MMC on other parallel healthy MMC converter stations.
[0198] Figure 16 (c) shows the AC voltage waveforms in two cases. It can be seen that the AC voltage waveforms are very different, indicating that although SMS-FCL will cause a certain imbalance in the voltage of the submodule capacitor, it will not inject high harmonics into the AC voltage. Figure 16 (d) shows the change in the number of sub-modules engaged in phase a after the fault. By comparing the waveforms, it can be seen that SMS-FCL does not have a significant impact on the number of sub-modules. The voltage borne by each sub-module is similar to that when it is not in operation, so it will not cause sub-module overvoltage problems. Figure 16 (e) and (f) respectively show the waveforms of the sum of capacitor voltages and the sum of capacitor energy of phase a submodule. It can be seen that the active current limiting method suppresses the discharge process of the submodule capacitor, increases the sum of capacitor voltages, and thus reduces the energy loss of the submodule during the fault, thereby enabling a faster fault recovery process.
[0199] To verify the effectiveness of SMS-FCL in nonmetallic short-circuit faults. Figure 17 The graphs of short-circuit current variation under different transition resistances with no operation and active current limiting are given. Table 5 shows the short-circuit current values and current limiting effect at the moment of circuit breaker disconnection.
[0200] Table 5
[0201]
[0202]
[0203] As can be seen from the figures and tables, under short-circuit faults with different transition resistances, the SMS-FCL can suppress the growth rate of short-circuit current to a certain extent, with current limiting effects ranging from 4.3% to 33.1%.
[0204] To compare SMS-FCL with other current limiting methods, this section selects N-FCL and U-FCL for comparison. Figure 18 The values of m = 0.5-0.9 and U are given. ref The short-circuit current image at 0 is shown, and short-circuit current waveforms under no control and SMS-FCL conditions are added for comparison. Figure 21 Figures (a)-(f) show the current limiting effects of different control methods. Combining the two figures, it can be seen that the current limiting effect of SMS-FCL is better than that of U-FCL, and its effect is similar to that of N-FCL, which reduces the number of submodules by 50%.
[0205] To further compare the impact of the three active current limiting methods on AC / DC systems Figure 19 The values of m = 0.5 and U are given. refThe waveforms of DC voltage, AC voltage, number of submodules, and active and reactive power are shown when the current limiting method is applied (=0). Regarding DC voltage support, the N-FCL exhibits a momentary drop in DC voltage, which undoubtedly impacts the DC grid. The U-FCL, by lowering the controller's DC voltage reference value, causes a continuous decrease in DC voltage after the current limiting method is activated. While the SMS-FCL initially reduces DC voltage, it can raise the DC voltage after 7ms of fault detection, providing some voltage support for the DC line. Regarding AC voltage, all three current limiting methods cause slight voltage distortion, and the total harmonic distortion is similar. However, the N-FCL and U-FCL also cause sudden AC voltage changes upon activation, leading to additional adverse effects.
[0206] Regarding the number of submodules, SMS-FCL has little impact on the number of submodules and hardly changes the number of submodules put into operation during a fault. N-FCL will reduce the number of submodules to m times the original number (m<1) at the moment of operation. At this time, the capacitor voltage will rise to 1 / m times the original number. Therefore, this current limiting method has certain drawbacks and is prone to causing overvoltage of the submodule capacitor. U-FCL does not change the total number of submodules put into operation per phase, but it will change the number of submodules in the upper and lower bridge arms, thus causing a sudden change in AC voltage.
[0207] Figure 19 (d) The waveforms of AC active and reactive power are given. As can be seen from the figure, N-FCL can reduce the reactive power absorbed during MMC faults; however, this also leads to a higher active power deficit. U-FCL can increase the output power of MMCs during faults, but cannot provide reactive power support for the AC system. SMS-FCL has the same reactive power support capability as N-FCL, while its active power support capability is superior to N-FCL. In summary... Figure 19 (a)-(d) The method proposed in this invention can simultaneously suppress DC short-circuit current and support DC voltage and AC reactive power, while avoiding problems such as sub-module overvoltage, DC and AC voltage mutations and high active power loss in N-FCL and U-FCL, and has certain application prospects.
[0208] Since the current limiting method proposed in this invention does not change the number of sub-modules or the controller control method during the fault period, it can be used in conjunction with various current limiting methods such as N-FCL and U-FCL to further reduce the short-circuit current. Figure 20 The short-circuit current waveforms after combining various control methods with SMS-FCL are presented. To facilitate a direct comparison of the effects of different methods, Figure 21 The rate limiting effects of each method are plotted. Figure 20As shown in (a)-(e), SMS-FCL can significantly improve the effect of N-FCL, with an improvement of about 9%. Furthermore, the combined effect of the two current limiting methods is better than the effect of each method acting alone. When combined with U-FCL, SMS-FCL can also slightly improve its current limiting effect.
[0209] To verify the prerequisites for SMS-FCL to work with other rate limiting methods, Figure 22 The short-circuit current waveform when m=0 is given. At this time, the number of sub-modules in each bridge arm is 0 after the fault, which does not meet the condition in equation (28). Therefore, the current limiting effect of SMS-FCL and N-FCL is the same as that of N-FCL alone. Although the current limiting method of reducing the number of sub-modules to 0 is more effective, it will cause the simultaneous occurrence of AC three-phase short-circuit fault and DC bipolar short-circuit fault. Therefore, this current limiting method is used less often and the duration is shorter. So the condition shown in equation (28) is easily met. Therefore, the method of the present invention has a wide range of applications and is less limited when combined with other current limiting methods.
[0210] Although embodiments and drawings of the present invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.
Claims
1. An active fault current suppression method suitable for LCC / MMC hybrid cascaded DC networks, characterized in that: The steps of the method are as follows: S1. System Topology Introduction and Analytical Expression of DC Side Short-Circuit Fault Current and Equivalent Discharge Circuit of LCC and MMC Converter Stations The hybrid cascaded DC network consists of an LCC rectifier station and an LCC / MMC hybrid cascaded inverter station. The LCC / MMC hybrid cascaded inverter station is composed of one LCC and three parallel MMCs connected in series. The rectifier-side LCC adopts constant current control to maintain the DC line current at the rated value; the inverter-side LCC adopts constant voltage control to support the DC voltage; among the three parallel MMC converter stations, MMC1 adopts constant DC voltage and constant reactive power control to provide DC voltage together with the inverter-side LCC, while MMC2 and MMC3 adopt constant active power and constant reactive power control to achieve reasonable power distribution among the MMC stations. The possible fault points in this DC network are f1, f2 and f3, where f1 is a short circuit fault on the LCC output side of the receiving end; f2 is a ground fault on the positive terminal of the MMC of the receiving end, which causes a short circuit fault at the output of the MMC; f3 is a ground fault on the positive terminal of the DC line, which causes a short circuit fault in the LCC / MMC hybrid cascaded inverter. Based on the above analysis, the analytical expressions for DC-side short-circuit fault currents and equivalent discharge circuits of LCC and MMC converter stations are as follows: An LCC converter station consists of multiple 6-pulse LCC converters. To simplify the analysis, we will take one 6-pulse inverter as an example. The DC side voltage u dc1 It can be represented as: Where: U is the effective value of the AC line voltage on the secondary side of the transformer; X T =ωL T For the leakage reactance of the LCC transformer; β is the lead-out firing angle; μ is the commutation overlap angle, calculated as follows: Based on the block diagram of the constant voltage PI controller of the LCC inverter, β is written as: Where: u dc1ref This is the reference value for the DC voltage of the LCC converter station on the inverter side; K p K i These are the proportional and integral coefficients of the PI controller, respectively. When a fault occurs on the DC side of the LCC, β will rapidly decrease to its minimum value β under the action of the PI controller. min To maintain DC voltage, after a DC fault occurs, due to the rapid change rate of the inverter angle, and the relationship between β0 and β... min Since the phase difference is small, the change in β can be ignored, meaning that β is assumed to remain constant at β0 after the fault. Under this assumption, the Laplace equivalent circuit for a DC fault in an LCC can be obtained, where the short-circuit current in the complex frequency domain can be expressed as: Where: β0 is the inverter angle before the fault, μ0 can be calculated by equation (2), and the equivalent parameter is calculated as follows: Combining equation (4), its equivalent discharge circuit can be obtained. Performing an inverse Laplace transform on equation (4), the expression for the LCC short-circuit current in the time domain can be obtained: Among them: U eq10 and I dc10 The steady-state voltage and current values before the fault are given, and τ is a time constant, calculated as follows: Differentiating equation (6), we get: Within a very short time after the fault occurs (within 5ms), if τ is large, then t / τ≈0, and equation (8) can be further simplified to: As can be seen from equation (9), after the fault occurs, the LCC short-circuit current exhibits an approximately linear trend. When i dc1 After the current drops to 0, due to the clamping effect of the thyristor, the short-circuit current will remain constant at 0. Therefore, the LCC short-circuit fault can be divided into two stages: transient and steady state. Let i dc1 =0, we can obtain the transient time t. p1 : In the structure of a half-bridge MMC, u dc2 and i dc2 These represent the DC voltage and current of the MMC, R0 and L0 represent the bridge arm resistance and reactance, respectively, C0 is the submodule capacitor, and L... dc2 For smoothing reactor, v pj and v nj Let be the upper and lower arm voltages of phase j (j = a, b, c), respectively. From KCL and KVL, we can obtain: Summing the three-phase differential equations and substituting them into (11), we get: When a fault occurs on the DC side, the DC voltage can be expressed as: Combining equations (13) and (14), we can obtain: in: Within 7ms after the fault occurs, the discharge time of the submodule capacitor is very short, so its voltage change can be ignored, and the bridge arm voltage v can be approximated. ij It is a constant voltage source; and at the initial moment of the fault, there is U dc20 =v pj +v nj Therefore, according to equation (15), the analytical expression for the MMC short-circuit current can be obtained as follows: The calculation method for τ is the same as that for equation (7). According to equation (17), the MMC equivalent discharge circuit can be obtained. S2. Deriving the short-circuit current expression and fault characteristic analysis of the LCC / MMC hybrid cascaded DC network. Using the equivalent discharge circuits of LCC and MMC derived in step S1 above, the equivalent discharge circuit for DC faults in the LCC / MMC hybrid cascaded system is obtained, and the system equivalent parameters are obtained: The equivalent discharge circuit of the HC-HVDC system is a first-order RL circuit. The short-circuit current of the LCC / MMC hybrid cascaded inverter is expressed as: Where τ can be calculated by equation (7), due to the unidirectional conduction of the thyristors in the LCC station, after a DC fault occurs in the LCC / MMC hybrid cascade system, the short-circuit current will decrease from the rated value to 0, and the MMC submodule capacitors will not be able to discharge through the equivalent circuit, thus not causing system overcurrent problems. Let i dc3 =0, we can obtain the transient time t. p2 : During the transient phase, if t / τ≈0, the short-circuit current changes approximately linearly, therefore i can be... dc3 The rate of change is expressed as: By combining the calculated and simulated waveforms of short-circuit current in LCC, MMC, and LCC / MMC hybrid cascaded systems, it can be observed that when a DC fault occurs in an LCC or LCC / MMC cascaded converter, the clamping effect of the thyristor limits the fault current to near 0, and the system does not have an overcurrent problem; however, when a short-circuit fault occurs in an MMC, the fault current is as high as ten times the rated current, so it is necessary to suppress its short-circuit current. S3. MMC Active Short-Circuit Current Suppression Strategy Based on Submodule Sorting Algorithm Equation (15) gives the bridge arm voltage v ij With MMC short-circuit current i dc2 To study the relationship between the submodule capacitor voltage and the short-circuit current, further research on v is needed. ij with u c Given the relationship, let the time of the fault occurrence be t1, and consider the time after the fault occurs as t. i Within one IGBT switching cycle Δt, v ij It can be represented as: Wherein: S ijk The switching function for the k-th submodule on phase i of the bridge arm can be expressed as: Let t i -t i The number of sub-modules deployed on the i-th arm of phase j within the time interval +Δt is N. ij , i.e. ∑S ijk =N ij Then equation (22) can be written as: Substituting equation (24) into equation (15), we get: in: It depends only on the number of submodules deployed, and not on the initial voltage when the submodules are deployed; It is related to both the initial voltage of the submodule and the number of submodules. Under normal operation, the capacitor voltage differences among the submodules are minimal under the capacitor voltage balancing strategy, all remaining near their rated values. However, when a fault occurs, the capacitor voltages of the submodules will exhibit some deviation. Therefore, by changing the submodule activation method during a fault, the voltage difference can be reduced. ij This enables the suppression of fault current; In this invention, SMS-FCL establishes an x-axis representing the order of submodules and a y-axis representing the capacitor voltage of the submodules. SMS-FCL changes the traditional submodule sorting algorithm's order of submodule deployment: when P>0, the submodule with the highest voltage is deployed; when P<0, the submodule with the lowest voltage is deployed. Furthermore, to avoid submodule capacitor overvoltage issues, voltages exceeding the rated voltage U are... cn The submodules are configured to maintain the traditional submodule sorting algorithm.
2. The active fault current suppression method for LCC / MMC hybrid cascaded DC networks according to claim 1, characterized in that: The bridge arm voltage in the SMS-FCL is related to both the submodule activation method and the number of submodules. The SMS-FCL, in conjunction with the N-FCL, further reduces the number of submodules and further suppresses short-circuit current based on the SMS-FCL. Suppose that after this current-limiting method is activated, the number of submodules becomes m times the original number, then equation (24) becomes: The conditions under which SMS-FCL can suppress short-circuit current are that the values of each bridge arm submodule are not zero and the number of overvoltage submodules is less than the number of submodules put into operation in each bridge arm. Therefore, the conditions for current limiting when SMS-FCL is used in conjunction with N-FCL are expressed as follows:
Citation Information
Patent Citations
Direct-current fault current suppression method based on sub-module two-stage active control
CN113036739A
Fault analysis and short-circuit current suppression method for hybrid DC network
CN114928089A
