Cascade converter cell for medium voltage dc chain power supply network and control method thereof

Abstract

The application relates to a chain converter cabin for a medium-voltage direct-current chain power supply network and a control method thereof, wherein the chain converter cabin comprises a three-phase charging circuit for receiving alternating current input and completing primary rectification and energy collection; a plurality of power conversion modules, each of which comprises a plurality of single-phase power conversion branches; the single-phase power conversion branch comprises a filter circuit for suppressing high-frequency harmonic components generated by the primary rectification and energy collection to form stable alternating current; an ANPC single-phase H-bridge power unit adopting a three-level active neutral point clamped topological structure for completing energy conversion from stable alternating current to direct current; the output ends of each single-phase power conversion branch are connected in series, and the output ends of each power conversion module are connected in series. The application can simplify the main circuit and transformer structure, reduce control and communication dependence, and improve engineering deployment adaptability and operation safety under the premise of realizing equivalent voltage and power levels.

Description

Technical Field

[0001] This invention relates to the field of power electronic conversion equipment and flexible DC power supply and distribution technology, and in particular to a chain converter cabin and its control method for medium-voltage DC chain power supply networks. Background Technology

[0002] With the rapid development of high-power power electronics technology, transportation electrification, and distributed new energy industries, constructing medium-voltage DC chain power supply networks along transportation corridors has become an important technical direction for supporting continuous power supply to road loads, local consumption of new energy, and flexible interconnection of multiple power grids. In typical scenarios such as electrified highways, road-regional source-grid-load-storage coordinated systems, and flexible DC power distribution systems, multiple power conversion nodes need to be deployed along the lines to efficiently convert 10kV medium-voltage AC power into medium-voltage DC power, supplying power to loads along the lines in a segmented cascade manner. At the same time, it is also necessary to have distributed power source access capabilities, rapid fault isolation capabilities, and modular expansion capabilities to adapt to the engineering characteristics of decentralized deployment and segmented construction along the lines.

[0003] As the core conversion equipment in chain-type power supply networks, the topology, control performance, and system integration of medium-voltage AC / DC converters directly determine the reliability, construction cost, and operation and maintenance efficiency of the entire power supply system. Currently, the mainstream high-power medium-voltage AC / DC conversion solutions in the industry mainly include two types: the cascaded H-Bridge (CHB) solution and the modular multilevel converter (MMC) solution. However, both types of solutions have obvious technical limitations when applied to chain-type power supply scenarios.

[0004] The first type is a three-phase bridge cascaded converter scheme, whose basic structure consists of a multi-winding phase-shifting transformer and multiple bridge power units. In this scheme, the primary side of the transformer is connected to a 10kV AC grid, and the secondary side has multiple electrically isolated three-phase windings, each of which independently powers a power unit. Each power unit integrates a three-phase rectifier circuit, a DC support capacitor, and a power conversion unit. Multiple power units are connected in series phase-wise, and medium-voltage DC output is achieved through voltage superposition. Although this scheme can adapt to different voltage levels by adjusting the number of cascaded units, it has the following inherent drawbacks: complex transformer structure and high manufacturing cost; dispersed power supply structure, resulting in insufficient system reliability and maintainability; poor expansion flexibility and high cost of capacity expansion and modification.

[0005] The second approach is a modular multilevel converter (MLBC) solution. Its topology consists of three-phase upper and lower arms, with each arm containing several sub-modules connected in series and equipped with arm inductors. Each sub-module employs a half-bridge or full-bridge structure, incorporating IGBT switching devices and energy storage capacitors. By controlling the activation and bypassing of these sub-modules, the output voltage is progressively superimposed, thus achieving AC / DC energy conversion. This solution offers strong voltage expansion capabilities, but it suffers from the following drawbacks when applied to chain-type converter pods: high component quantity, leading to higher equipment cost and size; complex control system, making engineering implementation difficult; and low structural integration, making it unsuitable for pod-type deployments.

[0006] Besides the defects in the main circuit structure, existing solutions also suffer from insufficient adaptability in terms of control and protection. On the one hand, three-level power modules generally suffer from DC side midpoint potential offset, which easily leads to increased output harmonics and uneven device stress. Conventional control strategies cannot simultaneously achieve both current tracking accuracy and midpoint potential balance. On the other hand, multi-module cascaded systems often adopt centralized voltage equalization strategies that rely on high-speed communication, which pose a risk of single-point failure, have low tolerance for communication delays, and are difficult to adapt to the long-distance, decentralized operating environment of chain power supply networks. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a chain converter pod and its control method for medium-voltage DC chain power supply networks, which can simplify the main circuit and transformer structure, reduce control and communication dependence, and improve engineering deployment adaptability and operational safety while achieving the same voltage and power level.

[0008] The technical solution adopted by this invention to solve its technical problem is: to provide a chain converter compartment for medium-voltage DC chain power supply networks, comprising:

[0009] The three-phase charging circuit is used to receive AC input and complete primary rectification and energy collection.

[0010] Multiple power conversion modules, each including multiple single-phase power conversion branches; each single-phase power conversion branch includes a filter circuit and an ANPC (Active Neutral Point Clamped) single-phase H-bridge power unit;

[0011] The filter circuit is used to suppress high-frequency harmonic components generated by the primary rectification and energy collection to form stable AC;

[0012] The ANPC single-phase H-bridge power unit adopts a three-level active midpoint clamp topology to complete the stable AC to DC energy conversion.

[0013] The output terminals of each single-phase power conversion branch are connected in series, and the output terminals of each power conversion module are connected in series.

[0014] The single-phase power conversion branch also includes:

[0015] A solid-state DC circuit breaker, located at the rear end of the ANPC single-phase H-bridge power unit, is used to cut off current in the event of a fault.

[0016] A DC current-limiting reactor is also connected in series between the output terminals of each power conversion module. The DC current-limiting reactor is used to suppress the rising slope of the fault current by introducing inductive characteristics.

[0017] The controller in the ANPC single-phase H-bridge power unit uses a model predictive control strategy to control the switching state of the three-level active neutral clamp topology.

[0018] The model predictive control strategy specifically includes:

[0019] Establish a discrete-domain mathematical model of the system, input the current state information into the discrete-domain mathematical model, and obtain the values ​​of the controlled variables at the next moment under various switching states.

[0020] The value of the controlled variable at the next moment under various switching states is used as the input to the value function to obtain the value function value under various switching states.

[0021] The switching state corresponding to the minimum value function is selected as the control command to control the switching state of the three-level active midpoint clamping topology.

[0022] The discrete-domain mathematical model is as follows: ,in, and for The sampled current value of the rectifier grid side in the α-β coordinate system at time 1. and for Predicted current value of the rectifier grid side in the α-β coordinate system at any given time. and for The voltage sampling value of the rectifier grid side in the α-β coordinate system at any given time. and The voltage values ​​of the rectifier AC side input voltage in the α-β coordinate system are calculated from different switching states. The sampling period is The equivalent resistance on the grid side of the rectifier. This is the value of the filter inductance on the grid side of the rectifier. for The predicted value of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. for The measured value of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. This refers to the capacitance value of the DC-side voltage divider capacitor in the rectifier. , and For the state variables of a three-phase switch, , and This represents the current on the three-phase grid side.

[0023] The value function is expressed as: ,in, For the value function value, and for Predicted current value of the rectifier grid side in the α-β coordinate system at any given time. and for Reference current value of the rectifier grid side in the α-β coordinate system at any given time. for The predicted value of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. These are the weighting coefficients.

[0024] The technical solution adopted by this invention to solve its technical problem is as follows: A control method for the aforementioned chain converter pod for medium-voltage DC chain power supply networks is provided, employing a voltage equalization droop control strategy. Specifically, this includes a local control sub-strategy and a global coordination sub-strategy. The local control sub-strategy refers to the dynamic adjustment of the DC voltage reference value by each ANPC single-phase H-bridge power unit according to its local input power, and the use of a quasi-proportional resonant controller to achieve rapid tracking of the grid current. The global coordination sub-strategy refers to monitoring the total DC voltage of the system, generating a compensation amount through a PI controller, and superimposing the compensation amount onto the local reference voltage of each ANPC single-phase H-bridge power unit.

[0025] The DC voltage reference value is obtained through The calculation yielded that, This is the DC voltage reference value. As a global voltage reference, The droop coefficient is... For input power, This is the rated power.

[0026] The compensation amount is through Calculations show that For compensation amount, This is the reference value for the total DC voltage of the system. This represents the total DC voltage of the system.

[0027] Due to the adoption of the above-mentioned technical solutions, this invention has the following advantages and positive effects compared with the prior art: The chain converter compartment of this invention uses a three-phase charging circuit to uniformly complete the primary rectification and energy collection of the AC input, replacing the traditional cascaded scheme's distributed power supply architecture where a single winding corresponds to a single power unit. This significantly reduces the number of secondary windings in multi-winding transformers, lowers the complexity of transformer winding arrangement and insulation design, reduces manufacturing costs, and shrinks the overall size of the equipment. This invention uses an ANPC single-phase H-bridge three-level topology to construct the power conversion unit. A single module only requires low-voltage power devices to operate stably. Medium-voltage DC output is achieved through the series superposition of a single-phase branch and a power module, avoiding the high cost and high loss problems caused by high-voltage devices, effectively reducing output harmonic content, and improving power quality. Attached Figure Description

[0028] Figure 1 This is a topology diagram of a chain converter compartment for a medium-voltage DC chain power supply network according to an embodiment of the present invention.

[0029] Figure 2 This is a topology diagram of the ANPC single-phase H-bridge power unit in an embodiment of the present invention;

[0030] Figure 3 This is a control block diagram of an NPC-type three-level rectifier based on MPC in an embodiment of the present invention. Detailed Implementation

[0031] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0032] Embodiments of the present invention relate to a chain converter compartment for medium-voltage DC chain power supply networks, such as... Figure 1 As shown, it includes:

[0033] The three-phase charging circuit 1-1 is used to receive AC input and complete primary rectification and energy collection.

[0034] Multiple power conversion modules, each including multiple single-phase power conversion branches; each single-phase power conversion branch includes filter circuit 1-2 and ANPC single-phase H-bridge power unit 1-3;

[0035] Filter circuits 1-2 are used to suppress high-frequency harmonic components generated by the primary rectification and energy collection to form stable AC;

[0036] ANPC single-phase H-bridge power units 1-3 adopt a three-level active midpoint clamp topology to complete stable AC to DC energy conversion.

[0037] The output terminals of each single-phase power conversion branch are connected in series, and the output terminals of each power conversion module are connected in series.

[0038] In this embodiment, the chain converter compartment for medium-voltage DC chain power supply network uses a three-phase charging circuit 1-1 as the front-end interface to receive 10kV AC input and complete primary rectification and energy collection. It has strong input adaptability and power carrying capacity, and can realize unified power supply and energy distribution for multiple conversion units. In its subsequent stage, each branch is first configured with an LC filter circuit 1-2. The LC filter circuit 1-2 is composed of inductors and capacitors to suppress high-frequency harmonic components generated by rectification and power conversion, reduce input current distortion, and stabilize DC side voltage, providing a good electrical environment for subsequent power devices. Then, the power enters the ANPC single-phase H-bridge power unit 1-3. The ANPC single-phase H-bridge power unit 1-3 adopts a three-level active midpoint clamping structure, which has low switching loss and good voltage distribution characteristics. Through multi-module series connection, the voltage can be superimposed step by step. Its single-module DC voltage is about 1667V, which can be achieved by using IGBT devices with a nominal voltage of 1700V. It has a mature engineering application foundation and domestic substitution capability, and undertakes the core voltage regulation and energy conversion functions of the system.

[0039] At the DC output end, this embodiment can also include a protection and constraint unit consisting of a solid-state DC circuit breaker 1-4 and a DC current-limiting reactor 1-5. The solid-state DC circuit breaker 1-4 is located at the rear end of the ANPC single-phase H-bridge power unit 1-3, achieving rapid interruption capability based on power electronic devices. It can achieve millisecond-level or even faster current cut-off in the event of a fault, preventing fault propagation and protecting downstream equipment. The DC current-limiting reactor 1-5 is positioned between the output ends of the power conversion modules, introducing inductance characteristics to suppress the rise rate of the fault current, reducing short-circuit impact and coordinating with the circuit breaker for coordinated protection. The above units are connected step-by-step in a chain structure of rectified input—filtered stabilization—multi-level conversion—protection output. This achieves efficient conversion from 10kV AC to medium-voltage DC, and through a modular series-parallel structure, provides excellent voltage and capacity expansion capabilities. Simultaneously, it considers power quality improvement and system safety, providing a stable and reliable core conversion and control node for the chain-type medium-voltage DC power supply network. Meanwhile, by optimizing the secondary structure of the multi-winding transformer, the original solution requiring multiple independent windings for power supply is reduced to a small number of open windings for output, combined with module cascading to achieve voltage expansion. This significantly reduces transformer size and copper-iron losses while meeting output specifications, lowering the number of main circuit components and system manufacturing complexity. Based on engineering experience, it is estimated that, with the same ±10kV output capacity, the number of main circuit components can be reduced by approximately 20%–30%, and transformer capacity utilization can be increased by approximately 10%–15%, resulting in a simultaneous decrease in initial equipment investment and operation and maintenance costs.

[0040] The topology of ANPC single-phase H-bridge power units 1-3 in this embodiment is as follows: Figure 2 As shown, the core of the circuit consists of the grid-side voltage. Filter inductor and equivalent resistance Load resistance and four power switching transistors on each phase bridge arm. With two clamping diodes , Composition, for each phase Bridge arms (identified as follows) Mutually, Harmony The circuit is configured with phases; each power switch is connected in parallel with a freewheeling diode, improving circuit stability and efficiency. Additionally, the DC side consists of two capacitors of identical value. and capacitor The capacitors are connected in series, with point O serving as the potential reference point. Under steady-state operating conditions, the voltages across the two capacitors are assumed to be equal, enabling the rectifier to output three different levels: positive and negative values ​​of the DC bus voltage UDC / 2, -UDC / 2, and 0, where UDC is the DC bus voltage.

[0041] In this embodiment, the ANPC single-phase H-bridge power unit 1-3 is equipped with a main power circuit, an analog signal acquisition circuit, and a controller. The controller employs a model predictive control (MPC) strategy to control the switching state of the three-level active neutral-point clamping topology. Specifically, the model predictive control strategy includes:

[0042] Establish a discrete-domain mathematical model of the system, input the current state information into the discrete-domain mathematical model, and obtain the values ​​of the controlled variables at the next moment under various switching states.

[0043] The value of the controlled variable at the next moment under various switching states is used as the input to the value function to obtain the value function value under various switching states.

[0044] The switching state corresponding to the minimum value function is selected as the control command to control the switching state of the three-level active midpoint clamping topology.

[0045] like Figure 3 As shown, the current state information of the system is first sampled and then input into the prediction model (i.e., the discrete domain mathematical model). The value function values ​​of 27 spatial voltage vectors are calculated in a rolling manner. The voltage vector with the smallest value function value is applied to the rectifier. The above steps are repeated continuously in each control cycle so that the rectifier operates in an ideal working state.

[0046] Predictive models are the foundation of predictive control strategies. By substituting the sampled current state information into the predictive model, the state information of the system at the next moment can be predicted. The forms of predictive models are quite diverse, including parametric or nonparametric models such as differential equations, state equations, and fuzzy models.

[0047] Predictive control is generally based on a high sampling frequency and a small sampling period. Using the forward Euler method to discretize the AC-side mathematical model, we can obtain:

[0048] ;

[0049] In the formula, This represents the system's sampling period. and for The sampled current value of the rectifier grid side in the α-β coordinate system at time 1. and for The predicted current value of the rectifier grid side in the α-β coordinate system at any given time.

[0050] The predicted current value at time k+1 can be expressed as:

[0051] ;

[0052] in, and for The voltage sampling value of the rectifier grid side in the α-β coordinate system at any given time. The equivalent resistance on the grid side of the rectifier. This is the value of the filter inductance on the grid side of the rectifier. and The voltage value of the rectifier AC side input voltage in the α-β coordinate system, calculated from different switching states, is expressed as follows:

[0053] ;

[0054] In the formula, This is the DC bus voltage. , and Three-phase switch state variables

[0055] It is known that due to its inherent topology, the ANPC type three-level rectifier has a problem of unbalanced midpoint potential. A predictive model of the DC side of the system can be established to predict the fluctuation of the midpoint potential. Whether the current of the upper and lower capacitors is equal can determine whether there is current flowing through the midpoint O, that is, whether the midpoint potential has shifted.

[0056] The currents of the upper and lower capacitors can be expressed as:

[0057] ;

[0058] In the formula, and These are the currents of the upper and lower capacitors, respectively. This refers to the capacitance value of the DC-side voltage divider capacitor in the rectifier. and These are the upper voltage divider capacitors. and lower voltage divider capacitor The voltage.

[0059] After processing the above equation, we get:

[0060] ;

[0061] Set the voltage divider capacitor on the DC side and lower voltage divider capacitor voltage difference We can obtain:

[0062] ;

[0063] Discretizing the derivative of the voltage deviation yields:

[0064] ;

[0065] The prediction model for the DC side upper and lower capacitor voltage deviation at time k+1 can be obtained as follows:

[0066] .

[0067] The value function, serving as the basis for optimization selection in predictive control, is the core of the MPC control strategy. Different value functions have different impacts on the control effect. As a control algorithm that can simultaneously achieve multiple objectives, MPC is commonly used in the control of NPC three-level rectifiers for objectives such as power control, grid-side current tracking control, switching frequency control, and DC-side midpoint potential balance control.

[0068] In this embodiment, a weighted method is used to achieve grid-side current tracking control and DC-side midpoint potential balance control. A value function in the form of the sum of squares of absolute errors is designed, which represents the error between the system reference value and the predicted value under the action of a single voltage vector.

[0069] ;

[0070] In the formula, and for Reference current value of the rectifier grid side in the α-β coordinate system at any time. The weighting coefficient is the weighting coefficient of the midpoint potential balance control term. Its value indicates the strength of the constraint on the control objective term. By changing the weighting coefficient, multiple control objectives can be controlled.

[0071] Grid-side current The reference value is obtained from the DC-side voltage feedback. In traditional MPC control strategies, since its sampling frequency is much higher than the grid frequency, it is usually assumed that... The time reference value is approximately equal to Reference value for a given moment. That is:

[0072] ;

[0073] in, , Obtained from DC-side voltage feedback. and This represents the current reference value at time k+1. This future reference value, obtained through approximate estimation, has a certain degree of deviation, which will affect the accuracy of the objective function calculation, and thus impact the system control performance to some extent.

[0074] Let the voltage amplitude be Vector angle is The vector rotation angular velocity is ,but:

[0075] ;

[0076] Furthermore, the phase angle of the reference voltage vector at time k+1 can be derived as follows:

[0077] ;

[0078] It can be known that:

[0079] ;

[0080] Calculate the reference current at time k+1 using the vector angle compensation method. and for:

[0081] ;

[0082] In this implementation, when the MPC strategy is applied to an ANPC-type three-level rectifier, the predicted values ​​of 27 space voltage vectors are substituted into the calculation to obtain the corresponding value function value g. All value function values ​​are compared one by one, and finally the switching state corresponding to the space voltage vector that minimizes the combined error values ​​of the system grid-side current and the midpoint potential deviation is applied to the rectifier to achieve optimal control.

[0083] It is easy to see that the chain converter compartment of this invention uses a three-phase charging circuit to uniformly complete the primary rectification and energy collection of AC input, replacing the traditional cascaded scheme with a single winding corresponding to a single power unit's distributed power supply architecture. This significantly reduces the number of secondary windings in multi-winding transformers, lowers the complexity of transformer winding layout and insulation design, reduces manufacturing costs, and shrinks the overall size of the equipment. This invention uses an ANPC single-phase H-bridge three-level topology to construct the power conversion unit. A single module only requires low-voltage power devices to operate stably. Medium-voltage DC output is achieved by superimposing a single-phase branch and a power module in series. This avoids the high cost and high loss problems caused by high-voltage devices, effectively reduces output harmonic content, and improves power quality.

[0084] In topologies with multiple cascaded NPC rectifiers, the issue of DC-side capacitor voltage balancing is particularly prominent. On the one hand, differences in parameters between modules, uneven power distribution, and grid disturbances can easily lead to voltage imbalances in each sub-module, posing a risk of device overvoltage. On the other hand, bidirectional power flow characteristics (such as regenerative braking energy feedback) further exacerbate voltage fluctuations under dynamic operating conditions. While traditional centralized master-slave control can achieve precise voltage balancing, its reliance on high-speed communication architecture presents a single point of failure risk, making it difficult to meet the requirements of high-reliability scenarios.

[0085] Therefore, this embodiment also proposes a control method for the aforementioned chain converter pod for medium-voltage DC chain power supply networks. This method adopts a voltage equalization droop control strategy, which is a two-layer architecture. The two-layer architecture includes a local control sub-strategy and a global coordination sub-strategy, which takes into account both autonomy and overall stability.

[0086] The specific local control sub-policies are as follows:

[0087] Each ANPC single-phase H-bridge power unit is based on the locally measured input power. The DC voltage reference value is dynamically adjusted using the following method: ,in, This is the DC voltage reference value. As a global voltage reference, It is the droop factor (inversely proportional to the capacity of the ANPC single-phase H-bridge power unit). This is the rated power.

[0088] A quasi-proportional resonant controller is used to achieve rapid tracking of the grid current, ensuring unity power factor operation. The transfer function of the quasi-proportional resonant controller can be expressed as: ,in, The transfer function of the quasi-proportional resonant controller. This is the proportionality coefficient. The resonance coefficient, The fundamental angular frequency, To achieve the resonant bandwidth, suppress the 6k±1st harmonic. For the Laplace operator.

[0089] The specific global coordination sub-strategy is as follows:

[0090] Monitoring system total DC voltage The compensation amount is generated through the PI controller. To eliminate steady-state deviation, compensation amount The generation method is as follows: ,in, This is the reference value for the total DC voltage of the system.

[0091] compensation amount The local reference voltage superimposed on each ANPC single-phase H-bridge power unit achieves a global voltage equalization closed loop, i.e.: ,in, Set the voltage control loop value. This is the local reference voltage for the ANPC single-phase H-bridge power unit.

[0092] Therefore, the control method of this embodiment adopts a two-layer voltage-equalizing droop control architecture that combines local autonomous adjustment with global voltage compensation. Each ANPC power unit can dynamically adjust the DC voltage reference based solely on local input power information, eliminating the need for high-speed communication between modules. This effectively reduces system communication complexity, avoids the single-point failure risk of centralized control, and improves the robustness and scalability of the cascaded system. Combined with a quasi-proportional resonant controller, it enables rapid tracking of grid current, ensuring high power factor operation. The global coordination layer generates a unified voltage compensation amount through a PI controller, eliminating overall steady-state voltage deviation. It maintains stable voltage equalization even under scenarios with discrete module parameters, load fluctuations, and bidirectional power flow, adapting to the operational requirements of distributed deployments in chain-type power supply networks.

[0093] The present invention will be further illustrated by a specific embodiment below.

[0094] The hardware of the chain converter compartment in this embodiment mainly includes an AC circuit breaker, an isolation transformer, an ANPC single-phase converter module, and a control system. The chain converter compartment features constant DC voltage, PQ control, power factor control, harmonic and reactive power compensation functions. It communicates with the central controller via RS485 and Ethernet to receive control commands. Simultaneously, the chain converter compartment has a complete protection scheme, including overcurrent protection, over / undervoltage protection, over / underfrequency protection, AC input phase sequence error protection, cooling system fault protection, communication fault protection, surge overvoltage and lightning overvoltage protection, and battery over / undervoltage protection. In case of a fault, microsecond-level protection is locked out; in severe faults, the switch can be quickly disconnected for fault isolation.

[0095] In addition to the main AC winding, the primary winding of the multi-winding power frequency isolation transformer also includes at least one additional primary winding. This additional primary winding is connected to a photovoltaic power generation unit, energy storage unit, or distribution power supply in the transformer substation via an LC filter unit and a power interface, thereby expanding the chain-type converter compartment into a three-port power routing node. This three-port extended structure enables flexible exchange and energy routing of main grid power, distributed power sources in the region, and energy storage devices within the same compartment, improving the green electricity absorption capacity and enhancing the power supply reliability and dispatch flexibility of key nodes in the chain-type power supply network.

[0096] This chain converter compartment preferably adopts an integrated container layout. The compartment integrates 10kV AC incoming switchgear, 10kV metering cabinet, 10kV AC outgoing switchgear, 10kV AC load cabinet, starter cabinet, multi-winding transformer cabinet, AC / DC converter cabinet, air-cooled system cabinet, solid-state switchgear, and control and protection cabinet. The compartment dimensions are approximately 15000mm × 3200mm × 3200mm, with a protection rating of IP54 for outdoor use or IP21 for indoor use. Cooling methods can employ intelligent air cooling, water cooling, or a hybrid water-air heat exchange system. The containerized layout allows for factory prefabrication, on-site hoisting, rapid commissioning, and convenient maintenance.

[0097] The chain converter pod has a rated output power of 10MW. The AC side wiring is a three-phase, three-wire isolated type, with a transformer ratio of 5.77:1, short-circuit impedance of 5%–8%, AC filter inductor of 500μH, and filter capacitor of 250μF. The DC side output range is 16kVdc–24kVdc, corresponding to bipolar ±8kV to ±12kV; the preferred rated output is ±10kV. The maximum DC current is rated at 750A and maximum at 825A. Voltage and current regulation accuracy can be controlled within ±2%, total current harmonic distortion is no greater than 3%, and the power factor can reach 0.99. The equipment can operate in ambient temperatures of -30℃ to 50℃ and relative humidity of 0%RH to 95%RH. The control system communicates via RS485, CAN, or Ethernet, supporting Modbus protocol or other industrial communication protocols.

[0098] It is easy to see that, in terms of structure and engineering implementation, this invention adopts a modular cascaded converter structure. Each converter unit is designed according to standardized dimensions, enabling factory prefabrication, rapid on-site assembly and replacement, significantly improving system deployment efficiency and maintainability. Compared with traditional centralized substation solutions, this invention supports distributed layout and containerized deployment along the line, reducing civil engineering work by more than 30%, and supporting on-demand expansion, avoiding large-scale one-time investment. Simultaneously, the control method based on the ANPC three-level topology achieves high-precision tracking of AC side current (current THD can be controlled within 5%) while effectively suppressing DC side midpoint potential deviation, keeping capacitor voltage imbalance within ±5%, significantly improving rectifier operation stability and output power quality. Combined with the proposed cascaded module voltage equalization control method, adaptive voltage and power balancing between modules is achieved without the need for high-speed centralized communication, reducing system communication complexity by more than 40%, improving the system's tolerance to communication delays and local faults, thereby enhancing overall robustness and engineering adaptability.

[0099] The overall modular cascaded architecture facilitates standardized production, on-site assembly, and fault replacement, supports segmented capacity expansion along the line, and is suitable for the decentralized deployment and phased construction requirements of medium-voltage DC chain power supply networks. Based on engineering experience, it is estimated that, with the same ±10kV output capacity, the number of main circuit components can be reduced by approximately 20% to 30%, and the transformer capacity utilization rate can be increased by approximately 10% to 15%, thereby resulting in a simultaneous reduction in initial equipment investment and operation and maintenance costs.

[0100] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A chain converter compartment for medium-voltage DC chain power supply networks, characterized in that, include: The three-phase charging circuit is used to receive AC input and complete primary rectification and energy collection. Multiple power conversion modules, each including multiple single-phase power conversion branches; each single-phase power conversion branch includes a filter circuit and an ANPC single-phase H-bridge power unit; The filter circuit is used to suppress high-frequency harmonic components generated by the primary rectification and energy collection to form stable AC; The ANPC single-phase H-bridge power unit adopts a three-level active midpoint clamp topology to complete the stable AC to DC energy conversion. The output terminals of each single-phase power conversion branch are connected in series, and the output terminals of each power conversion module are connected in series.

2. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 1, characterized in that, The single-phase power conversion branch also includes: A solid-state DC circuit breaker, located at the rear end of the ANPC single-phase H-bridge power unit, is used to cut off current in the event of a fault.

3. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 1, characterized in that, A DC current-limiting reactor is also connected in series between the output terminals of each power conversion module. The DC current-limiting reactor is used to suppress the rising slope of the fault current by introducing inductive characteristics.

4. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 1, characterized in that, The controller in the ANPC single-phase H-bridge power unit uses a model predictive control strategy to control the switching state of the three-level active neutral clamp topology.

5. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 4, characterized in that, The model predictive control strategy specifically includes: Establish a discrete-domain mathematical model of the system, input the current state information into the discrete-domain mathematical model, and obtain the values ​​of the controlled variables at the next moment under various switching states. The value of the controlled variable at the next moment under various switching states is used as the input to the value function to obtain the value function value under various switching states. The switching state corresponding to the minimum value function is selected as the control command to control the switching state of the three-level active midpoint clamping topology.

6. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 5, characterized in that, The discrete-domain mathematical model is as follows: ,in, and for The sampled current value of the rectifier grid side in the α-β coordinate system at time 1. and for Predicted current value of the rectifier grid side in the α-β coordinate system at any given time. and for The voltage sampling value of the rectifier grid side in the α-β coordinate system at any given time. and The voltage values ​​of the rectifier AC side input voltage in the α-β coordinate system are calculated from different switching states. The sampling period is The equivalent resistance on the grid side of the rectifier. This is the value of the filter inductance on the rectifier grid side. for The predicted value of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. for The measured values ​​of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. This refers to the capacitance value of the DC-side voltage divider capacitor in the rectifier. , and For the state variables of a three-phase switch, , and This represents the current on the three-phase grid side.

7. The chain converter compartment for medium-voltage DC chain power supply networks according to claim 5, characterized in that, The value function is expressed as: ,in, For the value function value, and for Predicted current value of the rectifier grid side in the α-β coordinate system at any given time. and for Reference current value of the rectifier grid side in the α-β coordinate system at any given time. for The predicted value of the voltage deviation between the upper and lower capacitors on the DC side of the rectifier at any given time. These are the weighting coefficients.

8. A control method for a chain converter pod for a medium-voltage DC chain power supply network as described in any one of claims 1-7, characterized in that, A voltage equalization droop control strategy is adopted for control, specifically including a local control sub-strategy and a global coordination sub-strategy. The local control sub-strategy refers to the dynamic adjustment of the DC voltage reference value of each ANPC single-phase H-bridge power unit according to the local input power, and the use of a quasi-proportional resonant controller to achieve rapid tracking of the grid current. The global coordination sub-strategy refers to monitoring the total DC voltage of the system, generating a compensation amount through a PI controller, and superimposing the compensation amount onto the local reference voltage of each ANPC single-phase H-bridge power unit.

9. The control method according to claim 8, characterized in that, The DC voltage reference value is obtained through The calculation yielded that, This is the DC voltage reference value. As a global voltage reference, The droop coefficient is... For input power, This is the rated power.

10. The control method according to claim 8, characterized in that, The compensation amount is through Calculations show that For compensation amount, This is the reference value for the total DC voltage of the system. This represents the total DC voltage of the system.

Images