Distributed energy storage system and its control method for front-end charging of power devices at the power level of 10 watts.

Abstract

This invention discloses a distributed energy storage system and its control method for front-end charging of a power-level pulsed power device. The system includes an AC distribution unit, a DC bus, M energy storage charging modules, an output-side DC / DC boost converter, a system controller, and an energy management unit. Each energy storage charging module comprises an AC / DC conversion unit, a supercapacitor bank, and a DC / DC conversion unit. During the energy storage charging phase, the system controller controls each AC / DC conversion unit to draw power from the grid at a first power level to charge the supercapacitor bank. Upon receiving a discharge command, the system controller synchronously controls each DC / DC conversion unit, causing all supercapacitor banks to simultaneously release energy to the DC bus, which is then boosted by the output-side DC / DC boost converter and supplied to the pulsed power device. This invention decouples the slow, stable power draw from the grid side from the transient, high-power discharge on the load side, effectively reducing the impact of the power-level pulsed power device operation on the grid.

Description

Technical Field

[0001] This invention belongs to the fields of high-power pulse power technology, power electronics technology, and energy storage technology. Specifically, it relates to a distributed energy storage system for front-end charging of large-scale pulse power devices at the power level (e.g., watts) and its control method in transient high-power charging applications. Background Technology

[0002] Power-level pulsed power sources (such as particle collisions, high-power microwave sources, fusion experimental devices, etc.)

[0003] These devices are characterized by extremely high instantaneous power (reaching hundreds of MW or even GW levels), extremely short single-operation time (μs to ms levels), and low operating frequency (intervals of several seconds to several minutes). The power supply system of such devices typically consists of two stages: a front-end charging system that stores grid energy in intermediate energy storage elements (capacitor banks, inductors, etc.), and a back-end system that releases the stored energy to the load in a very short time through fast switching or converters.

[0004] Existing front-end charging systems mainly adopt the following technical approaches: 1. Direct grid rectification power supply scheme: This scheme directly converts three-phase AC power into DC power through a high-power rectifier to charge the pulse capacitor bank. This scheme has significant drawbacks: the power drawn from the grid during charging can reach several MW to tens of MW, causing power quality problems such as grid voltage drops, flicker, and excessive harmonics; simultaneously, the power devices of the charging equipment must withstand extremely high instantaneous current stress, reducing system reliability.

[0005] 2. Single Centralized Energy Storage Buffer Scheme: This scheme involves installing a centralized energy storage device (such as a battery bank or supercapacitor bank) between the power grid and the charging equipment to smooth out charging power fluctuations. While this scheme can mitigate grid impacts, it has the following drawbacks: High risk of single point of failure: Energy storage systems are centrally located, and failure of any critical component (such as DC / DC converter or energy storage element) will paralyze the entire system; Poor scalability: As pulse power devices develop towards higher energy levels, it is difficult to expand centralized systems, requiring the entire energy storage system to be redesigned and replaced. Thermal management is challenging: centralized energy storage systems have high power density and concentrated heat dissipation, making thermal design a major challenge. High cost of redundancy design: To achieve system reliability, centralized solutions usually adopt full redundancy configuration, which causes a sharp increase in cost.

[0006] 3. Lack of friendly interaction with the power grid: Existing solutions rarely consider the active support function for the power grid's power quality during the design phase, and cannot provide auxiliary services such as reactive power support or harmonic control to the power grid while meeting charging needs. Summary of the Invention

[0007] The purpose of this invention is to propose a distributed energy storage system and its control method for front-end charging of power devices at the power level of watts. By adopting a distributed energy storage architecture, it achieves grid-friendly access and fast and reliable charging of power devices.

[0008] To achieve the above objectives, in a first aspect, the present invention proposes a distributed energy storage system for front-end charging of a power-level pulsed power device, comprising: An AC power distribution unit, the input of which is used to connect to a three-phase AC power grid; DC busbar; There are M energy storage and charging modules, where M ≥ 2. Each energy storage and charging module includes an AC / DC conversion unit, a supercapacitor bank, and a DC / DC conversion unit. The input terminal of the AC / DC conversion unit is connected to the output terminal of the AC power distribution unit. The supercapacitor bank is connected to both the output terminal of the AC / DC conversion unit and the first terminal of the DC / DC conversion unit. The second terminal of the DC / DC conversion unit is connected to the DC busbar through an isolation device. The output-side DC / DC boost converter has its input terminal connected to the DC bus and its output terminal used to connect to a pulse power device. The system controller is communicatively connected to each of the energy storage charging modules and the output-side DC / DC boost converter, and is used to coordinate and control the charging and discharging process. The energy management unit is communicatively connected to the upper-level monitoring system and the system controller, and is used to receive charging and discharging commands issued by the upper-level monitoring system and transmit corresponding control commands to the system controller. The system controller is configured as follows: During the energy storage and charging phase, each of the AC / DC conversion units is controlled to draw power from the three-phase AC grid at a first power to charge the supercapacitor bank. Upon receiving the discharge command transmitted by the energy management unit, the DC / DC conversion units are synchronously controlled to enable all the supercapacitor banks to release energy to the DC bus simultaneously, and the energy is then boosted by the output-side DC / DC boost converter and output to the pulse power device. Wherein, the first power is less than the discharge power of the supercapacitor bank releasing energy to the DC bus.

[0009] Optionally, the AC power distribution unit includes an AC EMI filter, an AC contactor, and surge protection devices; The AC / DC conversion unit adopts a three-phase PWM rectifier topology, which is a Vienna rectifier, a two-level active front end, or a three-level active front end. The power factor of the AC / DC conversion unit is not less than 0.99, and the total harmonic distortion rate of the current is not higher than 5%. The DC / DC conversion unit adopts a bidirectional resonant converter or an interleaved parallel converter topology and has a high-frequency isolation transformer.

[0010] Optionally, the output-side DC / DC boost converter adopts a multi-stage interleaved parallel BOOST topology, with a switching frequency of 16kHz to 20kHz, an adjustable output voltage range of 15kV to 50kV, and a maximum output power of not less than 1.2 times the peak power of the pulse power device.

[0011] Optionally, the supercapacitor bank adopts a multi-series and multi-parallel structure, the rated voltage of the supercapacitor bank is 600V to 900V, the energy storage capacity of each energy storage and charging module is 0.5kWh to 5kWh, and the energy storage capacity of the supercapacitor bank has a design margin of not less than 20% relative to the maximum single energy demand of the pulse power device. The number M of the energy storage charging modules is 4 to 24, the rated charging power of each energy storage charging module is 20kW to 50kW, the rated discharging power is 100kW to 300kW, and the rated discharging power of each energy storage charging module is 5 to 10 times its rated charging power.

[0012] Optionally, the system controller adopts an architecture combining dual DSPs and an FPGA; The dual DSPs include a master DSP and a slave DSP. The master DSP is responsible for task scheduling, status monitoring, and communication with the energy management unit. The slave DSP is responsible for the coordinated control of each energy storage charging module. The FPGA is responsible for high-speed PWM pulse generation, fault protection, and the allocation of synchronization trigger signals.

[0013] Optionally, each of the energy storage charging modules is also provided with a local protection unit for monitoring the overvoltage, overcurrent and overtemperature status of the module, and blocking the DC / DC conversion unit of the module when a fault is detected. The isolation device is an isolation diode or a DC contactor; the second terminal of the DC / DC conversion unit of each energy storage charging module is connected to the DC busbar through the isolation device and the fuse. The system controller is connected to each of the energy storage and charging modules via a CAN bus or an optical fiber communication bus. The system controller is also used to adjust the charging power of each energy storage charging module according to the demand response command received by the energy management unit from the upper-level monitoring system, and to control the AC / DC conversion unit to output reactive power to the three-phase AC grid.

[0014] Optionally, the rated voltage of the DC busbar is 800V to 1500V; The system is also equipped with protection functions, including: AC input over / under voltage protection, AC input over / under frequency protection, module over-temperature protection, supercapacitor over-voltage protection, supercapacitor under-voltage protection, supercapacitor overcurrent protection, busbar short-circuit protection, insulation monitoring and grounding protection, communication fault protection, and circulating current protection between modules.

[0015] Secondly, the present invention also proposes a control method for a distributed energy storage system as described in the first aspect, wherein the system controller performs the following state control: Standby mode: The system performs a power-on self-test, and each energy storage charging module does not output power, waiting for a charging request; Energy storage charging status: After receiving the charge preparation command, each energy storage charging module is started one by one in an interleaved start mode. Each energy storage charging module is controlled to charge its respective supercapacitor bank with constant current at the first power. After the voltage reaches the preset value, it switches to constant voltage float charging. Standby hold state: After each supercapacitor bank reaches the preset voltage, each energy storage charging module is controlled to maintain the voltage with float charging power, waiting for the discharge command; Pulse discharge state: Upon receiving a discharge command, the DC / DC conversion units of all energy storage charging modules are synchronously triggered, causing each supercapacitor group to simultaneously release energy to the DC bus, which is then boosted and supplied to the pulse power device. Recovery state: After the discharge is completed, the output of each DC / DC converter unit is blocked, and the system returns to the energy storage charging state.

[0016] Optionally, the staggered startup method is as follows: Each energy storage charging module is started one by one at a preset time interval, so that the total power on the grid side increases in a stepwise manner; the first power is 30% to 50% of the rated charging power of each module; the total response time of the system from receiving the discharge command to outputting the rated value is less than 1ms.

[0017] Optionally, the system controller also performs at least one of the following state controls: Fault-tolerant state: When any energy storage charging module fails, the module is controlled to disconnect and operate at reduced capacity; Online expansion status: When a new energy storage charging module is connected, the module is automatically identified and included in the scheduling; Grid-friendly interaction status: Reduce the charging power of each energy storage charging module during peak grid load periods, and control the AC / DC conversion unit to inject reactive power into the grid when the grid is abnormal.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) Significantly reduce grid impact: Through the distributed energy storage architecture, “slow charging and fast discharging” is achieved, and the grid side only needs to provide average power instead of peak power. Previously, 10MW of transient power needed to be directly drawn from the grid, but now only 1MW of continuous power needs to be drawn from the grid to slowly charge each energy storage module. The grid impact is reduced by 90%, and voltage fluctuations and harmonic pollution are greatly reduced.

[0019] (2) Significantly improved system reliability: Adopting the “NM” redundancy design concept, even if some modules among the M modules fail, the system can still operate at reduced capacity without affecting the core task of the pulse power device. A single point of failure will not cause the entire system to fail.

[0020] (3) Strong scalability and smooth capacity upgrade: As the energy demand of pulse power devices increases, the system can be expanded simply by increasing the number of energy storage charging modules, without having to redesign the entire charging system. It truly realizes "configuration on demand and modular construction".

[0021] (4) Decentralized thermal management reduces design difficulty: The total power is distributed to M independent modules, and the power density and heat flux density of each module are significantly reduced. The heat dissipation design is simpler and more reliable, and various methods such as air cooling or liquid cooling can be flexibly combined.

[0022] (5) Power quality friendly on the grid side: The AC / DC rectifier unit of each module adopts PWM rectification technology, which can achieve unity power factor operation and low harmonic grid connection, and can even provide reactive power support for the grid, turning "grid pollution source" into "grid friendly load".

[0023] (6) Lifespan advantage of supercapacitors: The cycle life of supercapacitors can reach more than 500,000 times, which is far superior to that of lithium batteries (~thousands of times). It is highly matched with the frequent charging and discharging characteristics of pulse power devices, and the system has a lower total lifespan cost.

[0024] (7) Extremely fast response speed: Each module adopts an independent DC / DC converter, which can synchronously respond to the discharge command within a ms time, meeting the stringent requirements of pulse power devices for power supply timeliness.

[0025] (8) Low operating voltage of energy storage system: Low-voltage energy storage is adopted, and DC-DC boost is performed during discharge operation, which greatly reduces the operating voltage of energy storage system, reduces the high voltage dwell time, and extends life.

[0026] The present invention has other features and advantages, which will be apparent from or will be set forth in detail in the accompanying drawings and the following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0027] The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0028] Figure 1 This is a schematic diagram of the overall architecture of a distributed energy storage system for front-end charging of a power-level pulsed power device according to Embodiment 1 of the present invention.

[0029] Figure 2 This is a schematic diagram of the internal circuit framework of a single energy storage charging module in Embodiment 1 of the present invention.

[0030] Figure 3 This is a schematic diagram of the topology of the output-side DC / DC boost converter in Embodiment 1 of the present invention.

[0031] Figure 4 This is a schematic diagram of the hardware architecture and communication connection relationship of the system controller in Embodiment 1 of the present invention.

[0032] Figure 5 This is a schematic diagram of the system working state transition in Embodiment 2 of the present invention.

[0033] Figure 6 This is a schematic diagram of the grid-side power smoothing effect under the staggered start-up charging strategy in Embodiment 2 of the present invention.

[0034] Figure 7 This is a schematic diagram comparing the time waveforms of the power drawn from the grid side and the power discharged from the load side in Embodiment 2 of the present invention. Detailed Implementation

[0035] To overcome the shortcomings of existing front-end charging technologies for power watt-level pulse power drives, such as large grid impact, low system reliability, poor scalability, and difficult thermal management, this invention provides a distributed energy storage system and its control method for front-end charging of power watt-level pulse power devices. This system achieves decoupling between slow and stable power extraction from the grid side and transient high-power discharge from the load side through a distributed energy storage architecture. This ensures the energy demand of the pulse power device, effectively reduces the impact of power watt-level pulse power drive operation on the grid, and ensures stable grid operation.

[0036] The distributed energy storage technology solution of this invention has advantages such as high modularity, independent control of multiple energy storage units, strong redundancy, and easy expansion. By setting up the energy storage system in a distributed manner, it can achieve friendly access to the power grid and reliable power supply to the load.

[0037] The invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0038] Example 1

[0039] like Figure 1 As shown, this embodiment provides a distributed energy storage system for front-end charging of a power-level pulsed power device. The system includes: an AC power distribution unit, a DC bus, M energy storage charging modules, an output-side DC / DC boost converter, a system controller, and an energy management unit.

[0040] The input terminal of the AC power distribution unit is used to connect to the three-phase AC power grid. The AC power distribution unit includes an AC EMI filter, an AC contactor, and surge protection devices, and its output terminals are connected to the AC input terminals of each energy storage charging module.

[0041] The DC bus is used to collect the DC power output from each energy storage charging module, and its rated voltage is 800V to 1500V. In this embodiment, the DC bus operates stably at 1000V.

[0042] The system comprises M energy storage charging modules (M≥2), optionally M being 4 to 24, flexibly configurable according to system capacity. This embodiment uses M=12 as an example. Each energy storage charging module has a rated charging power of 20kW to 50kW and a rated discharging power of 100kW to 300kW, with the rated discharging power of each module being 5 to 10 times its rated charging power. In this embodiment, a single module has a rated charging power of 30kW and a rated discharging power of 300kW, with the discharging power being 10 times the charging power.

[0043] Each energy storage charging module includes an AC / DC converter, a supercapacitor bank, and a DC / DC converter. The input terminal of the AC / DC converter is connected to the output terminal of the AC power distribution unit. The supercapacitor bank is connected to both the output terminal of the AC / DC converter and the first terminal of the DC / DC converter. The second terminal of the DC / DC converter is connected to the DC bus via an isolation device, which can be an isolation diode or a DC contactor. The second terminal of the DC / DC converter of each energy storage charging module is connected to the DC bus via an isolation device and a fuse. In this embodiment, a DC contactor is used as the isolation device, and a fuse is connected in series to provide overcurrent protection.

[0044] The system controller is communicatively connected to each energy storage charging module and the output-side DC / DC boost converter to coordinate and control the charging and discharging process. The system controller connects to each energy storage charging module via a CAN bus or a fiber optic communication bus. This embodiment uses CAN bus communication.

[0045] The Energy Management Unit (EMU) is communicatively connected to the upper-level monitoring system and the system controller. It receives charging and discharging commands from the upper-level monitoring system and transmits corresponding control commands to the system controller. In this embodiment, the EMU communicates with the upper-level monitoring system and the control system of the pulse power device via Ethernet, monitoring the status of each module and reporting system health information.

[0046] like Figure 2 As shown, the internal circuit framework of a single energy storage charging module is as follows: The three-phase AC input (AC 380V) first enters the AC EMI filter (containing surge protection device) to filter out high-frequency interference and absorb voltage spikes. The filtered AC power then enters the AC / DC conversion unit.

[0047] The AC / DC converter unit employs a three-phase PWM rectifier topology, which can be a Vienna rectifier, a two-level active front-end, or a three-level active front-end, with a power factor not lower than 0.99 and a total harmonic distortion (THD) not higher than 5%. In this embodiment, the AC / DC converter unit specifically uses a Vienna rectifier topology with a rated input power of 30kW, an input voltage of AC380V, a power factor >0.99, and THD <3%. The Vienna rectifier is a three-phase, three-level PWM rectifier topology, which has advantages such as fewer power devices, no need for shoot-through protection, and a high power factor, making it suitable for medium power density energy storage and charging applications. The output of the AC / DC converter unit is connected to a supercapacitor bank.

[0048] The supercapacitor bank adopts a multi-series, multi-parallel structure with a rated voltage of 600V to 900V. Each energy storage and charging module has an energy storage capacity of 0.5kWh to 5kWh. The energy storage capacity of the supercapacitor bank has a design margin of no less than 20% relative to the maximum energy demand of the pulse power device in a single operation. In this embodiment, the supercapacitor bank is composed of multiple 3000F / 2.7V cells connected in series and parallel, with a rated voltage of 800V and a single module energy storage capacity of 0.6kWh.

[0049] The output of the supercapacitor bank is connected to the first terminal of the DC / DC converter unit.

[0050] The DC / DC converter unit employs a bidirectional resonant converter or an interleaved parallel converter topology and includes a high-frequency isolation transformer. In this embodiment, the DC / DC converter unit employs a bidirectional CLLC resonant converter topology and includes a high-frequency isolation transformer.

[0051] Each energy storage charging module is also equipped with a local protection unit to monitor the module's overvoltage, overcurrent, and overtemperature conditions, and to shut down the module's DC / DC converter unit when a fault is detected. The local protection unit monitors the module status in real time, and once a fault is detected, it immediately shuts down the drive of the DC / DC converter unit to ensure that the faulty module can achieve thermal isolation and not affect other modules.

[0052] Each energy storage charging module is also equipped with a local controller (MCU / DSP) to control the AC / DC conversion unit.

[0053] like Figure 3 As shown, the output-side DC / DC boost converter adopts a multi-stage interleaved parallel BOOST topology. The DC bus input (rated voltage 800~1500V) is connected to node A and distributed to the parallel boost branches. The figure shows N BOOST boost branches (N=6 in this embodiment, i.e., 6 interleaved parallel branches). The physical structure of each branch is exactly the same: it contains an energy storage inductor (L1~LN) connected in series, followed by a switching transistor (S1~SN). The switching transistor is a high-power device IGBT / SiCMOSFET, with its source / emitter connected to ground. The system controller sends interleaved phase drive signals (360° / N) to each branch, so that the conduction phases of each branch are uniformly staggered, thereby significantly reducing the ripple current at the output. The diodes D1~DN of each branch converge at node B, forming a convergence point for the multi-path boost energy, and then the voltage waveform is smoothed by the output filter capacitor.

[0054] The output-side DC / DC boost converter adopts a multi-stage interleaved parallel BOOST topology, with a switching frequency of 16kHz to 20kHz and an adjustable output voltage range of 15kV to 50kV (adjusted according to the requirements of the pulse power device). The maximum output power is not less than 1.2 times the peak power of the pulse power device. In this embodiment, the switching frequency of the output-side DC / DC boost converter is 16kHz, the rated output power is 3.6MW (12 modules × 300kW discharge power), and the output voltage is 25kV.

[0055] like Figure 4 As shown, the system controller hardware architecture and communication connections are as follows: The system controller adopts an architecture combining dual DSPs and an FPGA. The dual DSPs include a master DSP and a slave DSP. The master DSP is responsible for task scheduling, status monitoring, and communication with the energy management unit, while the slave DSP is responsible for the coordinated control of each energy storage and charging module. The FPGA is responsible for high-speed PWM pulse generation, fault protection, and the distribution of synchronization trigger signals.

[0056] In this embodiment, the system controller adopts an architecture combining dual DSPs (TMS320F28377D) and an FPGA (10M08SAU169). The EMU receives grid dispatch / charging commands and sends them to the main DSP. The main DSP is responsible for task scheduling, status monitoring, and communication with the EMU. The slave DSPs are responsible for the coordinated control algorithms, charge / discharge control algorithms, and voltage / current closed-loop regulation of each energy storage charging module. The FPGA is responsible for high-speed PWM pulse generation, synchronous trigger signal allocation (synchronization accuracy <100ns), and hardware-level fault blocking. External hardware protection trigger signals (overcurrent / short circuit / overtemperature) are directly sent to the FPGA for ultra-fast response.

[0057] The drive commands issued by the FPGA are transmitted through multiple independent drive isolation circuits (corresponding to each energy storage charging module and the output-side DC / DC boost converter), achieving high-voltage and low-voltage isolation and precisely controlling the energy storage charging module and the output-side DC / DC boost converter. The real-time voltage, current, and temperature of the controlled object are acquired in real time through a high-speed ADC sampling module and the data is directly fed back to the slave DSP. The slave DSP performs closed-loop calculations based on the feedback data, and the calculation results are then sent back to the FPGA to adjust the output, forming a complete high-speed mixed-signal control closed loop.

[0058] The system controller is configured to: during the energy storage charging phase, control each AC / DC conversion unit to draw power from the three-phase AC grid at a first power to charge the supercapacitor bank; upon receiving the discharge command transmitted by the EMU, synchronously control each DC / DC conversion unit to enable all supercapacitor banks to simultaneously release energy to the DC bus, and output the energy to the pulse power device after being boosted by the output-side DC / DC boost converter; wherein, the first power is less than the discharge power of the supercapacitor bank releasing energy to the DC bus.

[0059] The system controller is also used to adjust the charging power of each energy storage charging module according to the demand response instructions received by the EMU from the upper-level monitoring system, and to control the AC / DC conversion unit to output reactive power to the three-phase AC grid.

[0060] The system also includes protection functions, including: AC input over / under voltage protection, AC input over / under frequency protection, module over-temperature protection, supercapacitor over-voltage protection, supercapacitor under-voltage protection, supercapacitor overcurrent protection, busbar short-circuit protection, insulation monitoring and grounding protection, communication fault protection, and circulating current protection between modules. These protection functions are implemented collaboratively by local protection units, the system controller, the FPGA, and isolation devices. Specifically, the local protection unit, in conjunction with the local controller, implements module-level protection (including supercapacitor over-voltage protection, supercapacitor under-voltage protection, supercapacitor overcurrent protection, and module over-temperature protection); the FPGA implements hardware-level high-speed protection (including busbar short-circuit protection, with external hardware protection trigger signals directly fed into the FPGA for μs-level response); the system controller implements system-level protection logic (including AC input over / under voltage protection, AC input over / under frequency protection, communication fault protection, circulating current protection between modules, and insulation monitoring and grounding protection); and isolation devices and fuses provide final disconnection and backup protection for faulty modules. These protection levels work together to form a complete system protection architecture.

[0061] Example 2

[0062] This embodiment provides a control method for the distributed energy storage system described in Embodiment 1, wherein the system controller controls the switching of the operating state of the distributed energy storage system.

[0063] Figure 5 The complete transition process of the system's operating states is demonstrated: After power-on, the system enters standby mode; upon receiving a charging preparation command, it switches to energy storage charging mode; once the supercapacitor bank reaches the preset voltage, it switches to standby hold mode; upon receiving a discharge command, it switches to pulse discharge mode; after discharge, it switches to recovery mode; and after recovery, it automatically returns to energy storage charging mode, forming a cycle. The following provides a detailed explanation of each state.

[0064] (1) Standby state

[0065] The system controller performs a power-on self-test on the system, and each energy storage charging module does not output power, waiting for a charging request.

[0066] Specifically, after the system is powered on, the EMU performs a self-test to confirm that the voltage of the supercapacitor bank of all energy storage charging modules (such as 12 energy storage charging modules) is not lower than the safety threshold (>100V), communication is normal, and there are no fault alarms. The system then enters standby mode, waiting for the "ready to charge" command from the pulse power device.

[0067] (2) Energy storage charging status

[0068] Upon receiving a charging preparation command, the system controller sequentially starts each energy storage charging module in an interleaved start-up mode. Each module charges its respective supercapacitor bank at a constant current with a first power output. Once the voltage reaches a preset value, it switches to constant voltage float charging. The interleaved start-up mode involves starting each energy storage charging module sequentially at preset time intervals, resulting in a step-like increase in the total power on the grid side.

[0069] Specifically, after the pulse power device completes its previous discharge, it sends a "request charging" signal to the higher-level monitoring system. Upon receiving this signal, the EMU issues an energy storage command to the system controller. The system controller then starts each energy storage charging module sequentially using an interleaved start-up method. For example... Figure 6 As shown, the AC / DC conversion units of the 12 modules are started one by one at 500ms intervals.

[0070] The initial power is 30% to 50% of the rated charging power of each module. In this embodiment, each module charges the supercapacitor bank with a constant current at 50% of its rated power (i.e., 15kW), and the charging current is set to 50A. When the voltage of the supercapacitor bank reaches the rated value of 800V, it switches to constant voltage float charging mode, and the charging current gradually decreases to near zero.

[0071] Because of the staggered start-up, the total power demand on the grid side does not jump instantaneously to 180kW (12×15kW), but increases gradually in a stepped manner, significantly reducing the grid inrush current. At the same time, the AC / DC conversion units of each module all adopt unity power factor control, and the grid-side current is sinusoidal and in phase with the voltage.

[0072] The energy storage charging phase lasts approximately 120 seconds, during which all modules' supercapacitor banks reach full charge (800V). The system then reports "Charging complete, ready to go" to the EMU.

[0073] (3) Standby status

[0074] After each supercapacitor bank reaches its preset voltage, the system controller controls each energy storage charging module to maintain the voltage with float charging power, waiting for the discharge command.

[0075] Specifically, each supercapacitor bank enters a hot standby state after reaching a preset voltage. The AC / DC conversion unit of each module switches to float charging mode to compensate for the self-discharge loss of the supercapacitor with minimal power (approximately 100W) and maintain voltage stability. The output-side DC / DC boost converter is in a ready-to-trigger state, with its power devices already biased and awaiting an external trigger signal.

[0076] (4) Pulse discharge state

[0077] Upon receiving the discharge command, the system controller synchronously triggers the DC / DC conversion units of all energy storage and charging modules, enabling each supercapacitor bank to simultaneously release energy to the DC bus, which is then boosted and supplied to the pulse power device.

[0078] The total response time of the system from receiving the discharge command to reaching the rated value is less than 1ms.

[0079] Specifically, once the pulsed power device is ready, it sends a "request discharge" command to the higher-level monitoring system. After confirmation, the EMU sends a "discharge trigger" command to the system controller. Within 50μs of receiving the trigger signal, the FPGA of the system controller simultaneously sends synchronous drive pulses to the DC / DC converter units of the 12 modules and the output-side DC / DC boost converter. All power devices turn on approximately simultaneously, and the supercapacitor banks of the 12 modules simultaneously release energy to the DC bus through their respective DC / DC converter units, stabilizing the bus voltage at 1000V. The output-side DC / DC boost converter boosts the 1000V DC to 25kV to charge the main energy storage capacitor bank of the pulsed power device.

[0080] During discharge, the instantaneous discharge power of a single module can reach 300kW (375A current), and the total discharge power of the 12 modules is 3.6MW. The discharge duration is approximately 200ms, and the total released energy is approximately 0.72kWh (considering a conversion efficiency of approximately 90%), which meets all the energy requirements for a single operation of the pulse power device. During this phase, the grid side only needs to provide approximately 1kW of maintenance power to the auxiliary power supply of each module, and is completely unaffected by the transient discharge power.

[0081] like Figure 7 As shown, the power drawn from the grid (slow charging) and the power discharged from the load (fast discharging) exhibit significant differences in their time waveforms: the power is stable and continuous during the charging phase (slow charging), while the instantaneous power is extremely high during the discharging phase (fast discharging) but the duration is extremely short (ms-level), achieving power decoupling between the grid side and the load side. Previously, 10MW of transient power needed to be drawn directly from the grid; now, only about 1MW of continuous power needs to be drawn from the grid for slow charging of each energy storage module.

[0082] (5) Recovering state

[0083] After the discharge is complete, the system controller blocks the output of each DC / DC converter unit and returns to the energy storage charging state.

[0084] Specifically, after the discharge ends, the FPGA blocks the drive pulses of all modules' DC / DC conversion units and the output-side DC / DC boost converter. The system automatically returns to the energy storage charging state and begins a new round of energy storage charging, preparing for the next pulse operation.

[0085] The entire "charging-standby-discharging-recovery" cycle takes about 125 seconds, during which the grid side spends more than 95% of the time in a stable, slow charging state, with only a very small amount of float charging power in standby mode, achieving a "nearly transparent" and friendly connection to the grid.

[0086] (6) Fault-tolerant status

[0087] When any energy storage charging module fails, the system controller will disconnect the module and degrade its operation.

[0088] Specifically, assuming that during operation, the AC / DC converter unit of module 7 malfunctions, its local protection unit detects overvoltage, overcurrent or overtemperature conditions and immediately shuts down the DC / DC converter unit of that module. At the same time, the local controller reports a fault code to the system controller and disconnects the faulty module from the DC bus via the internal isolation contactor.

[0089] Upon receiving the fault information, the system controller marked module 7 as "unavailable" and recalculated the charging schedule for the remaining 11 modules. In subsequent cycles, the system continued to operate with 11 modules, reducing the total energy storage to 91.7% of the rated value and the total discharge power to 3.3MW, which still met the operating requirements of the pulse power device (if a redundancy margin of more than 12% was considered in the design). The system sent a "derating operation" alarm to the EMU and the upper-level monitoring system, prompting the operators to arrange maintenance, but without interrupting the experimental plan of the pulse power device.

[0090] (7) Online expansion status

[0091] When a new energy storage charging module is connected, the system controller automatically identifies the module and includes it in the scheduling.

[0092] Specifically, when it is necessary to upgrade the energy level of the pulse power device, new energy storage and charging modules can be added online without shutting down the system. When a new energy storage and charging module is connected to the DC bus, the system controller automatically identifies the newly connected module and adds it to the scheduling queue, realizing "hot-swappable" capacity expansion.

[0093] (8) Power grid friendly interaction status

[0094] The system controller reduces the charging power of each energy storage charging module during peak grid load periods and controls the AC / DC conversion unit to inject reactive power into the grid when the grid is abnormal.

[0095] Specifically, during peak grid load periods, the EMU receives a "demand response" instruction from the upper-level grid dispatch system, requesting a reduction in grid power extraction. Based on the demand response instruction received by the EMU from the upper-level monitoring system, the system controller reduces the slow-charging power of each module from 15kW to 7.5kW (25% of rated power), correspondingly doubling the charging time. This adjustment process requires no hardware modifications; it can be achieved simply by modifying the power command value in the DSP, flexibly reducing the grid-side power demand from 180kW to 90kW.

[0096] When the grid voltage drops or frequency fluctuates, the AC / DC conversion units of each module can quickly switch to reactive power output mode (while ensuring the energy storage charging task), injecting inductive or capacitive reactive power into the grid to support grid voltage recovery. This function upgrades the system from a "simple electrical load" to a "grid-schedulable resource," improving the grid-friendliness of the pulse power device.

[0097] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A distributed energy storage system for front-end charging of a power-level pulsed power device, characterized in that, include: An AC power distribution unit, the input of which is used to connect to a three-phase AC power grid; DC busbar; There are M energy storage and charging modules, where M ≥ 2. Each energy storage and charging module includes an AC / DC conversion unit, a supercapacitor bank, and a DC / DC conversion unit. The input terminal of the AC / DC conversion unit is connected to the output terminal of the AC power distribution unit. The supercapacitor bank is connected to both the output terminal of the AC / DC conversion unit and the first terminal of the DC / DC conversion unit. The second terminal of the DC / DC conversion unit is connected to the DC busbar through an isolation device. The output-side DC / DC boost converter has its input terminal connected to the DC bus and its output terminal used to connect to a pulse power device. The system controller is communicatively connected to each of the energy storage charging modules and the output-side DC / DC boost converter, and is used to coordinate and control the charging and discharging process. The energy management unit is communicatively connected to the upper-level monitoring system and the system controller, and is used to receive charging and discharging commands issued by the upper-level monitoring system and transmit corresponding control commands to the system controller. The system controller is configured as follows: During the energy storage and charging phase, each of the AC / DC conversion units is controlled to draw power from the three-phase AC grid at a first power to charge the supercapacitor bank. Upon receiving the discharge command transmitted by the energy management unit, the DC / DC conversion units are synchronously controlled to enable all the supercapacitor banks to release energy to the DC bus simultaneously, and the energy is then boosted by the output-side DC / DC boost converter and output to the pulse power device. Wherein, the first power is less than the discharge power of the supercapacitor bank releasing energy to the DC bus.

2. The distributed energy storage system according to claim 1, characterized in that, The AC power distribution unit includes an AC EMI filter, an AC contactor, and surge protection devices; The AC / DC conversion unit adopts a three-phase PWM rectifier topology, which is a Vienna rectifier, a two-level active front end, or a three-level active front end. The power factor of the AC / DC conversion unit is not less than 0.99, and the total harmonic distortion rate of the current is not higher than 5%. The DC / DC conversion unit adopts a bidirectional resonant converter or an interleaved parallel converter topology and has a high-frequency isolation transformer.

3. The distributed energy storage system according to claim 1, characterized in that, The output-side DC / DC boost converter adopts a multi-stage interleaved parallel BOOST topology, with a switching frequency of 16kHz to 20kHz, an adjustable output voltage range of 15kV to 50kV, and a maximum output power of not less than 1.2 times the peak power of the pulse power device.

4. The distributed energy storage system according to claim 1, characterized in that, The supercapacitor bank adopts a multi-series and multi-parallel structure. The rated voltage of the supercapacitor bank is 600V to 900V. The energy storage capacity of each energy storage and charging module is 0.5kWh to 5kWh. The energy storage capacity of the supercapacitor bank has a design margin of not less than 20% relative to the maximum single energy demand of the pulse power device. The number M of the energy storage charging modules is 4 to 24, the rated charging power of each energy storage charging module is 20kW to 50kW, the rated discharging power is 100kW to 300kW, and the rated discharging power of each energy storage charging module is 5 to 10 times its rated charging power.

5. The distributed energy storage system according to claim 1, characterized in that, The system controller adopts an architecture combining dual DSPs and FPGA; The dual DSPs include a master DSP and a slave DSP. The master DSP is responsible for task scheduling, status monitoring, and communication with the energy management unit. The slave DSP is responsible for the coordinated control of each energy storage charging module. The FPGA is responsible for high-speed PWM pulse generation, fault protection, and the allocation of synchronization trigger signals.

6. The distributed energy storage system according to claim 1, characterized in that, Each of the energy storage charging modules is also equipped with a local protection unit for monitoring the overvoltage, overcurrent and overtemperature status of the module, and blocking the DC / DC conversion unit of the module when a fault is detected. The isolation device is an isolation diode or a DC contactor; the second terminal of the DC / DC conversion unit of each energy storage charging module is connected to the DC busbar through the isolation device and the fuse. The system controller is connected to each of the energy storage and charging modules via a CAN bus or an optical fiber communication bus. The system controller is also used to adjust the charging power of each energy storage charging module according to the demand response command received by the energy management unit from the upper-level monitoring system, and to control the AC / DC conversion unit to output reactive power to the three-phase AC grid.

7. The distributed energy storage system according to claim 1, characterized in that, The rated voltage of the DC busbar is 800V to 1500V; The system is also equipped with protection functions, including: AC input over / under voltage protection, AC input over / under frequency protection, module over-temperature protection, supercapacitor over-voltage protection, supercapacitor under-voltage protection, supercapacitor overcurrent protection, busbar short-circuit protection, insulation monitoring and grounding protection, communication fault protection, and circulating current protection between modules.

8. A control method for a distributed energy storage system as described in any one of claims 1 to 7, characterized in that, The system controller performs the following state control: Standby mode: The system performs a power-on self-test, and each energy storage charging module does not output power, waiting for a charging request; Energy storage charging status: After receiving the charge preparation command, each energy storage charging module is started one by one in an interleaved start mode. Each energy storage charging module is controlled to charge its respective supercapacitor bank with constant current at the first power. After the voltage reaches the preset value, it switches to constant voltage float charging. Standby hold state: After each supercapacitor bank reaches the preset voltage, each energy storage charging module is controlled to maintain the voltage with float charging power, waiting for the discharge command; Pulse discharge state: Upon receiving a discharge command, the DC / DC conversion units of all energy storage charging modules are synchronously triggered, causing each supercapacitor group to simultaneously release energy to the DC bus, which is then boosted and supplied to the pulse power device. Recovery state: After the discharge is completed, the output of each DC / DC converter unit is blocked, and the system returns to the energy storage charging state.

9. The control method according to claim 8, characterized in that, The staggered startup method is as follows: Each energy storage charging module is started one by one at a preset time interval, so that the total power on the grid side increases in a stepwise manner; the first power is 30% to 50% of the rated charging power of each module; the total response time of the system from receiving the discharge command to outputting the rated value is less than 1ms.

10. The control method according to claim 8, characterized in that, The system controller also performs at least one of the following state controls: Fault-tolerant state: When any energy storage charging module fails, the module is controlled to disconnect and operate at reduced capacity; Online expansion status: When a new energy storage charging module is connected, the module is automatically identified and included in the scheduling; Grid-friendly interaction status: Reduce the charging power of each energy storage charging module during peak grid load periods, and control the AC / DC conversion unit to inject reactive power into the grid when the grid is abnormal.

Images