A self-starting system and method for an energy storage converter in an off-grid condition without commercial power

Abstract

The application discloses a kind of energy storage converter autonomous black start system and method under no commercial power islanding condition, it is related to microgrid energy storage and novel power system technical field, wherein system includes vehicle-mounted energy management system, battery management system, auxiliary power supply system, energy storage battery pack, static transfer switch and energy storage converter.Auxiliary power supply system directly wakes up bottom control loop beyond conventional process.Vehicle-mounted energy management system controls energy storage converter active disengagement from grid tracking mechanism.Energy storage converter is independently carried out based on embedded high-speed operation chip AC fast voltage build-up.Amplitude is converted into off-grid constant voltage constant frequency operating mode after build-up, and provides initial starting power supply to paralyzed load by static transfer switch.The application solves the starting deadlock problem when losing external commercial power reference voltage, realizes the autonomous black start of microgrid and regional power supply recovery.

Description

Technical Field

[0001] This invention relates to the fields of microgrid energy storage and new power system technology, and in particular to an autonomous black start system and method for energy storage converters under islanded conditions without mains power. Background Technology

[0002] New power systems and microgrids play a crucial role in modern energy architectures, with plug-in hybrid electric vehicles (PHEVs) and other technologies often serving as core components. In conventional grid-connected operation, the power conversion equipment of such energy storage systems primarily relies on the AC voltage and frequency provided by the external power grid as a reference, achieving synchronization and energy exchange with the main grid through grid tracking mechanisms such as phase-locked loops (PLLs).

[0003] In extreme islanding scenarios where natural disasters or sudden faults cause a complete power outage from the external mains, microgrids need to perform a black start to restore power to the area. Conventional inverter equipment typically requires a certain period of time for grid voltage and frequency detection and phase-locked loop synchronization before starting and connecting to the grid. In an environment where the external mains reference voltage is completely lost, this conventional grid tracking mechanism, which relies on an external reference waveform, cannot effectively establish a synchronized phase and struggles to obtain the necessary startup reference signal. Consequently, the related hardware-level control loops often fail to wake up properly and complete rapid voltage build-up due to the lack of grid support.

[0004] During the initial stage of black start, paralyzed external power lines and various transformers and inductive loads will experience significant surge current impacts and transient voltage drops upon connection. When using energy storage devices to inject initial startup power into paralyzed distribution areas, it is necessary to independently achieve AC voltage build-up after disconnecting from grid tracking logic and maintain the stability of output AC voltage and frequency. This objectively poses specific engineering requirements on the high-current discharge capability of the system's energy base and the control response speed of the power conversion equipment. Summary of the Invention

[0005] The purpose of this invention is to provide an autonomous black-start system and method for energy storage converters under islanded conditions without mains power, so as to solve the problems pointed out in the background art.

[0006] In a first aspect, the present invention provides an autonomous black start system for an energy storage converter under islanded conditions without mains power, comprising: an on-board energy management system, a battery management system, an auxiliary power system, an energy storage battery pack, at least one static transfer switch (STS), and at least one energy storage converter (PCS). The energy storage battery pack is electrically connected to the at least one energy storage converter PCS to provide a black start energy base. The auxiliary power system includes a redundant power module that supports three power input sources, including AC and DC inputs. The third input source is connected to the high-voltage bus of the energy storage battery pack through an isolated converter that converts high-voltage DC to low-voltage DC. The auxiliary power system is used to, upon receiving a black start command, bypass AC voltage detection and directly distribute the stable output DC power to the main control board, drive board, and core module of the energy storage converter and the battery management system, so as to instantly wake up the underlying control loop. The on-board energy management system is used to control the energy storage converter PCS to actively disconnect from the grid tracking mechanism when a black start command is received. The energy storage converter PCS has an embedded computing chip, which is configured to perform AC voltage build-up calculation: actively bypassing the call entry of the software phase-locked loop function module, pointing the internal phase generation register to the fixed-step integrator driven by the internal timer to generate a reference phase angle independent of the external power grid; calling the soft-start ramp function to linearly increase the voltage command setpoint to the rated target value according to the preset single-step increment step size, so as to complete the AC voltage build-up. The energy storage converter PCS is also configured to continuously monitor the AC side output voltage. When it is detected that the effective value of the three-phase line voltage at the output is within ±2% of the rated value, and the total harmonic distortion of the voltage waveform is continuously less than 3%, and the above two conditions are met for a complete fundamental frequency cycle, a physical command is output to trigger the closure of the static transfer switch STS. Then, it switches to the off-grid constant voltage and constant frequency working mode to provide initial start-up power to the paralyzed load.

[0007] Optionally, the off-grid constant voltage and constant frequency working mode adopts virtual synchronous machine technology combined with dual closed-loop feedback control of voltage outer loop and current inner loop; The voltage outer loop employs a proportional resonant regulator, which includes a resonant control branch that uses a second-order bandpass filter logic composed of a set resonant gain coefficient, a control bandwidth cutoff frequency, and a fundamental resonant angular frequency to extract the fundamental error signal. The inner current loop uses a proportional-integral regulator to limit the inductor current and track its waveform.

[0008] Optionally, the main control module of the battery management system internally runs a state-of-charge estimation algorithm that combines the ampere-hour integration method and the open-circuit voltage calibration method; this algorithm introduces a closed-loop correction observer based on Kalman filtering, specifically configured as follows: Based on the predicted value of the state of charge at the previous moment, the predicted value of the observed open-circuit voltage is obtained. The deviation between the actual collected open-circuit voltage and the predicted value is calculated. This deviation is multiplied by the Kalman gain obtained by real-time iterative optimization and added to the prior predicted value as a correction term to achieve dynamic convergence correction.

[0009] Optionally, the energy storage converter PCS includes an LCL-type low-pass filter composed of a grid-side inductor, a converter-side inductor, and a filter capacitor, the parameter configuration of which satisfies the following physical constraints: The lower limit of the sum of the converter-side inductance and the grid-side inductance is set as eight times the minimum DC bus voltage divided by the product of the switching frequency and the maximum allowable ripple current; the upper limit is set as the square of the AC-side rated phase voltage multiplied by 10%, then divided by the product of the converter's rated capacity and the fundamental angular frequency; and the natural resonant frequency of the LCL-type low-pass filter is limited to a range greater than ten times the fundamental frequency and less than half the switching frequency.

[0010] Optionally, it also includes a thermal management unit, wherein the on-board energy management system is configured to use a fuzzy PID algorithm to adjust the compressor speed and water pump flow of the thermal management unit in real time; The input variables of the fuzzy PID algorithm are the difference between the current highest temperature of the battery node and the target set temperature, and the rate of change of the temperature difference. The mapping logic of its control rule table includes: when the temperature difference is positive and the rate of change of the temperature difference is positive, the output proportional coefficient correction is mapped to positive, the integral correction is mapped to negative, and the derivative correction is mapped to positive.

[0011] Optionally, the energy storage battery pack adopts a design architecture in which multiple basic battery nodes are connected in series to establish a high-voltage DC bus. Its high-voltage DC bus supports continuous high-current discharge and provides an energy buffer to resist the surge current impact caused by the external transformer excitation impedance and inductive load when the static transfer switch STS is closed.

[0012] Optionally, a multi-protocol communication system is also included, in which the inter-module communication within the battery management system and the high real-time status frames with the vehicle energy management system are transmitted via CAN bus. The smart meter and insulation tester are connected to the slave station polling network via RS485 bus; the high-throughput waveform data and control commands between the vehicle energy management system and the energy storage converter are communicated in full-duplex mode via Ethernet bus based on TCP / IP protocol.

[0013] Optionally, the execution logic of the fixed-step integrator in each discrete control cycle is as follows: Based on the reference phase angle of the previous control cycle, add the product of the equivalent angular velocity corresponding to the fixed rated frequency and the interrupt execution cycle of the underlying microprocessor, and then perform a modulo operation on twice pi using the accumulated result.

[0014] Optionally, the single-step increment of the soft-start ramp function is equal to the rated target voltage multiplied by the underlying interrupt control cycle, and then divided by the configured expected voltage build-up time.

[0015] Secondly, this invention provides a method for autonomous black start of an energy storage converter under islanded conditions without mains power, comprising the following steps: Receive external black boot commands; By using an auxiliary power supply system that includes three power input sources, DC power is directly distributed without AC voltage detection to wake up the underlying control loop; Actively disconnect from the grid tracking mechanism, bypass the software phase-locked loop function, and generate a reference phase angle independent of the external grid through an integrator driven by an internal timer; The soft-start ramp function is invoked to increment the voltage command setpoint according to a preset slope, thus completing the AC voltage build-up calculation and execution. The system continuously monitors the AC voltage at the outlet. When the effective value of the three-phase line voltage is found to be within ±2% of the rated value and the total harmonic distortion is less than 3%, and this condition is maintained for a complete fundamental frequency cycle, the system switches to the off-grid constant voltage and constant frequency operating mode, triggering the static transfer switch (STS) to close and inject initial power into the external paralyzed transformer area.

[0016] The present invention has achieved the following beneficial effects: The beneficial effects of this invention lie in providing a system architecture and method for achieving autonomous black start of a microgrid under conditions of loss of external mains reference voltage. By configuring an auxiliary power system to directly inject working voltage into the underlying control loop to achieve hardware-level wake-up, it bypasses the grid voltage detection and synchronization waiting process that conventional equipment relies on. Relying on the on-board energy management system and the high-speed computing chip embedded in the energy storage converter, the control program actively bypasses the software phase-locked loop module and completes AC voltage build-up calculation based on a locally generated reference phase angle independent of the external grid, avoiding startup logic deadlock caused by the lack of an external reference. Combined with the high-current continuous discharge capability of the physical energy storage battery pack as an energy buffer, when switching to off-grid constant voltage and constant frequency working mode and connecting paralyzed loads, it can effectively resist the surge current impact caused by external line excitation impedance and inductive loads and quickly compensate for the bus voltage drop, realizing high-quality initial power injection and regional power restoration for paralyzed areas.

[0017] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the structure of an energy storage converter autonomous black start system under islanded operation without mains power in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the steps of an autonomous black-start method for an energy storage converter under islanded conditions without mains power, as described in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] This invention discloses an autonomous black-start system and method for energy storage converters under islanded conditions without mains power. This system and method are integrated into the technical equipment of plug-in hybrid electric vehicles, serving as a core component of a novel energy storage and power system. It is configured to achieve autonomous black-start of the microgrid and restoration of regional power supply under islanded conditions where the external mains reference voltage is lost.

[0022] In this embodiment, refer to Figure 1 As shown, the autonomous black-start system for the energy storage converter under islanded operation without mains power mainly includes: an on-board energy management system, a battery management system, an auxiliary power system, an energy storage battery pack, at least one static transfer switch, and at least one energy storage converter. These subsystems are interconnected via an electrical power bus and a low-voltage communication bus, forming a complete closed-loop control and energy dispatch network.

[0023] Specifically, the energy storage battery pack is electrically connected to the DC-side terminals of at least one energy storage converter, configured to provide power for the black start process and subsequent off-grid operation. An auxiliary power system is electrically connected to the vehicle-mounted energy management system and the system's underlying control circuit via DC power cables. Upon receiving a black start command, this auxiliary power system bypasses the AC voltage detection and synchronous soft-start process relied upon by conventional grid-connected equipment, directly injecting operating voltage into the underlying control circuit to achieve hardware-level wake-up. The vehicle-mounted energy management system establishes a data link connection with the energy storage converter via a communication bus network, used to switch the global state machine mode upon receiving a black start command and issue control word commands to cause the energy storage converter to actively disconnect from the phase-locked loop grid tracking mechanism. A high-speed computing chip is embedded on the main control board inside the energy storage converter. This high-speed computing chip is configured to autonomously and independently perform discretization calculations of the AC voltage build-up algorithm and execute pulse width modulation instructions based on the local clock generated by its internal crystal oscillator and a preset reference waveform after disconnecting the grid voltage tracking logic. After completing the rapid AC voltage build-up, it reconstructs and switches to an off-grid constant voltage and constant frequency operating mode through a control algorithm. Finally, the energy storage converter outputs initial startup power to the external AC load bus that is in a paralyzed state by controlling the closure of at least one static transfer switch.

[0024] Furthermore, regarding the physical energy storage hardware base of the system, the total capacity of the energy storage battery pack reaches 352.08 kWh. In terms of the internal cell topology, the battery pack adopts a 2P204S design architecture, that is, two individual cells are connected in parallel to form a basic battery node, and then 204 basic battery nodes are connected in series to establish a high-voltage DC bus. The entire high-voltage energy storage battery pack is physically composed of four independent energy storage boxes, each with a capacity of 88.02 kWh. The four energy storage boxes are connected in parallel within the combiner box via DC contactors and a fuse array, outputting current to the system's DC bus.

[0025] To ensure energy availability in long-term standby mode, the battery cell chemical material system and its internal balancing circuit of the energy storage battery pack are configured to strictly control the monthly self-discharge rate of the battery to less than 5%. Based on this hardware, the battery management system includes a master control module, slave control modules, and a high-voltage sampling module. Slave control modules are distributed within each battery box, acquiring the voltage and surface temperature of each battery node in real time via a high-precision analog-to-digital converter, and uploading the sampled data to the master control module through an internal communication network. The master control module internally runs a state-of-charge (SOC) estimation algorithm that combines the ampere-hour integral method and the open-circuit voltage calibration method, while also introducing a Kalman filter-based closed-loop correction observer to stabilize the SOC estimation accuracy within a range of ±5%. Specifically, the Kalman filter-based closed-loop correction observer fuses the ampere-hour integral method and the open-circuit voltage method in real time. The specific process is as follows: First, based on the state of charge (SOC) of the previous moment, subtract the quotient of the product of the charging / discharging current and time within the current sampling period divided by the battery's rated capacity to obtain the prior prediction value of the SOC at the current moment. Simultaneously, substitute this prediction value into a pre-calibrated piecewise linear interpolation mapping table of SOC and open-circuit voltage stored in a read-only memory to obtain the predicted value of the observed open-circuit voltage. Then, calculate the deviation between the actual acquired open-circuit voltage and this predicted value, and multiply this deviation by the Kalman gain obtained through real-time iterative optimization of the covariance matrix. Finally, add this product as a correction term to the prior prediction value to complete the posterior state estimation and dynamic convergence correction of the accumulated error of the ampere-hour integral every 100 milliseconds. Regarding the ultimate physical output capability, the high-voltage DC bus, copper busbar cross-sectional area, and contactor contact capacity of the energy storage battery pack are designed with electromagnetic thermal coupling to support continuous output of a large current of 1072 amperes for physical discharge. Combining the discharge voltage platform of the large-capacity battery pack with the aforementioned high-current discharge capability, the system's maximum AC-side output power in pure energy storage discharge operation can reach 276 kW. This high-current output capability provides substantial physical support for resisting transient voltage drops caused by sudden load increases when the subsequent static transfer switch is instantly closed.

[0026] In the power conversion execution unit architecture, the system's power conversion system is configured with two energy storage converter devices, each with a rated power of 125 kW. These energy storage converters are of the EPCS125 series. The two energy storage converters are connected in parallel on the DC side to the high-voltage DC bus of the energy storage battery pack, and in parallel on the AC side to the microgrid AC output bus. The main power circuit of each energy storage converter includes a DC-side smoothing capacitor array, a three-phase full-bridge inverter topology module, and an AC-side passive filter assembly. The three-phase full-bridge inverter uses insulated-gate bipolar transistors (IGBTs) as the core power switching devices. The AC-side passive filter is configured as an LCL-type low-pass filter composed of grid-side inductors, converter-side inductors, and filter capacitors, used to suppress high-frequency harmonic currents generated by switching operations. The LCL low-pass filter's parameter configuration follows these physical constraints to avoid resonance and instability: the lower limit of the sum of the converter-side inductance and the grid-side inductance is set to eight times the product of the minimum DC bus voltage divided by the switching frequency and the maximum allowable ripple current; the upper limit is set to 10% of the square of the AC-side rated phase voltage, divided by the product of the converter's rated capacity and the fundamental angular frequency. The upper limit of the filter capacitor value is set to 5% of the converter's rated capacity divided by three times the product of the fundamental angular frequency and the square of the phase voltage. Based on this, the natural resonant frequency of the LCL filter is strictly limited to a range greater than ten times the fundamental frequency and less than half the switching frequency, to ensure high-frequency harmonic attenuation while avoiding resonance with the control loop.

[0027] The main control board inside the energy storage converter houses the aforementioned embedded high-speed computing chip. This high-speed computing chip employs a digital signal processor architecture with multi-core parallel processing and hardware floating-point arithmetic capabilities. Within this chip, instruction execution time and the interrupt cycle of the control algorithm are strictly limited and controlled to within a physical time range of less than 5 milliseconds. This microsecond-level interrupt response capability ensures that the control loop can perform extremely high-frequency sampling and closed-loop feedback regulation of transient fluctuations in AC side voltage and current. In the hardware sampling link, the main control board acquires analog signals of the converter's three-phase bridge arm midpoint voltage, filter capacitor voltage, and grid connection point current in real time through Hall effect voltage sensors and current transformers. These signals are then sent to a high-speed analog-to-digital converter module for digital conversion via signal conditioning circuitry, providing a high-precision data source for the underlying AC rapid voltage build-up calculation.

[0028] Regarding the system's control center wake-up mechanism and auxiliary power supply link, the auxiliary power system includes redundant power modules. This auxiliary power system supports three power input sources in its hardware circuitry: AC and DC. Specifically, the first input source connects to the external mains power grid port, the second input source connects to the output port of the on-board diesel generator, and the third input source connects to the high-voltage bus of the energy storage battery pack via an isolated converter that converts high-voltage DC to low-voltage DC. The internal power selection logic circuit is configured to prioritize monitoring and using the AC input source; in the absence of mains power and in an islanded environment where the generator is not running, the logic circuit automatically and seamlessly switches to the third DC input source from the energy storage battery pack. After filtering and voltage regulation, the output of the auxiliary power system stably outputs 24V DC power to the entire system. This redundant input structure and direct DC power supply design ensure that after receiving a black start command, the on-board energy management system can directly distribute 24V DC power to the main control board, drive board, and core module of the battery management system by closing the solid-state relay array, bypassing the conventional start-up process and directly waking up the underlying control loop.

[0029] In the operation mode control logic after rapid voltage build-up, the energy storage converter is configured to switch to an off-grid constant voltage and constant frequency operating mode. This off-grid constant voltage and constant frequency operating mode integrates virtual synchronous machine technology with a high-frequency closed-loop feedback control algorithm to maintain the stability of the output AC voltage and frequency. Specifically, in the control program of the high-speed computing chip, the virtual synchronous machine algorithm module calculates the phase and amplitude of the reference voltage by solving the rotor motion equation and reactive voltage droop characteristic equation of the simulated synchronous generator. Internally, the virtual synchronous machine algorithm module calculates the reference value by simulating rotor motion and reactive voltage droop characteristics. In the active power-frequency control stage, the given active power reference value is subtracted from the real-time sampled actual output electromagnetic active power, and then the product of the active power damping coefficient and the actual angular frequency deviation (i.e., the difference between the current frequency and the grid's rated angular frequency) is subtracted. The result is then divided by the product of the virtual moment of inertia coefficient and the rated angular frequency to obtain the differential integral rate of change of the angular frequency. This rate of change is then used to obtain the reference phase of the output voltage through time integration. In the reactive power-voltage control stage, the reference amplitude of the current output command is directly obtained by subtracting the product of the reactive power droop coefficient and the reactive power deviation (i.e., the difference between the real-time actual output reactive power and the given reactive power reference value) from the rated phase voltage amplitude. The rotor motion equation introduces virtual inertia parameters and damping coefficients, causing the system frequency response to exhibit a smooth transition characteristic similar to a rotating motor when external load changes abruptly, suppressing frequency step jumps. Simultaneously, the generated reference voltage phase and amplitude signals are sent to a high-frequency closed-loop feedback control module consisting of an outer voltage loop and an inner current loop. The outer voltage loop uses a proportional resonant regulator to perform steady-state error control on the three-phase AC voltage, while the inner current loop uses a proportional-integral regulator to perform rapid current limiting and waveform tracking on the inductor current. Specifically, the proportional resonant regulator in the outer voltage loop includes a proportional control branch and a resonant control branch. The proportional branch directly amplifies the voltage error through a set proportional coefficient to control the dynamic response. The resonant branch takes the aforementioned voltage error as input and extracts and amplifies the error signal at the fundamental frequency through a second-order bandpass filter logic composed of a set resonant gain coefficient, control bandwidth cutoff frequency, and fundamental resonant angular frequency, to eliminate steady-state errors at specific frequencies. After digital discretization, the proportional-integral regulator in the current inner loop executes the following logic: multiplying the current current tracking error by the proportional coefficient, adding the sum of all historical current tracking errors, multiplying it by the integral coefficient and the control period, and summing the two to generate the final duty cycle control command. This composite control strategy ensures the absolute stability of AC power output under islanded operating conditions at the physical level.

[0030] To ensure the physical integrity and continuous operation of the equipment under various natural disasters and extreme environmental conditions, the system is equipped with a comprehensive safety protection system in terms of mechanical packaging and environmental adaptability. The battery box casing of the energy storage battery pack adopts high-strength alloy materials and a multi-seal gasket design, and its physical structure has the highest waterproof and dustproof protection level of IP68. The selection and verification data of structural components and sealing materials show that it can be continuously submerged in water at a depth of 1.5 meters for 30 minutes without structural damage or internal water ingress short circuit. The core electrical modules of the system, including the power cabinet of the energy storage converter and the low-voltage control box, have IP67 and IP20 protection levels respectively, and differentiated physical isolation is implemented for water-sensitive and ventilation / heat dissipation areas.

[0031] In terms of temperature field control and heat distribution management, the system also includes an independently configured thermal management unit. This thermal management unit employs liquid-cooled thermal management technology. The liquid-cooled circulation loop is connected via pipes to the liquid-cooled plate inside the energy storage box and the heat dissipation base of the energy storage converter power module. The thermal management unit integrates a variable frequency compressor, a circulating water pump, an electronic expansion valve, and a condenser fan. The nominal cooling capacity of this unit is rated at 13 kW, and its coefficient of performance (COP) measured under rated operating conditions is greater than or equal to 2.0. The on-board energy management system collects temperature sensor data from each node and uses a fuzzy PID algorithm to adjust the compressor speed and water pump flow rate in real time. The input variables of the fuzzy PID algorithm are the difference between the current highest temperature of the battery node and the target set temperature, as well as the rate of change of this temperature difference. The controller internally defines seven fuzzy subsets: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. A Gaussian membership function is used to convert the precise temperature difference and rate of change into fuzzy quantities. The specific mapping logic of its control rule table includes: when the temperature difference is large (i.e., severely overheating) and the rate of change of the temperature difference is large (i.e., extremely rapid temperature rise), the output proportional coefficient correction is mapped to large, the integral correction to large, and the derivative correction to small. The fuzzy output obtained through inference is defuzzified using the centroid method to obtain specific numerical corrections, which are then superimposed on the basic proportional, integral, and derivative parameters. This outputs pulse-width modulation signals with a duty cycle of 0 to 100% to drive the variable frequency compressor and circulating water pump respectively. This enables the system to maintain stable operation of its internal core electrical components within a safe operating temperature range across a wide temperature spectrum from -35°C to 65°C. Furthermore, the electrical clearances of the core equipment, the creepage distances of the insulation materials, and the boiling point of the coolant have all been specifically calibrated, ensuring stable operation in environments with altitudes of 4000 meters or less.

[0032] At the level of data interaction between subsystems and external command issuance within the system, a 10-inch in-vehicle touchscreen serves as the human-machine interface terminal, and a multi-protocol communication system is built-in. The physical and data link layers of this communication system support CAN bus, RS485 serial communication bus, and Ethernet communication protocols. Specifically, inter-module communication within the battery management system and high-real-time status frames (including node voltage, temperature, and charge / discharge current limits) between the battery management system and the vehicle energy management system utilize CAN bus for periodic broadcasting and event-triggered transmission. Smart meters, insulation detectors, and environmental temperature and humidity acquisition modules distributed within the system access the vehicle energy management system's slave polling network via RS485 bus based on the Modbus-RTU protocol. High-throughput waveform data, operation logs, and millisecond-level control commands between the vehicle energy management system and the energy storage converter are communicated via a 100Mbps Ethernet bus based on the TCP / IP protocol in full-duplex mode, ensuring accurate and low-latency data flow under the complex system architecture.

[0033] Based on the aforementioned hardware topology, communication network, and environmentally adaptable architecture, this invention also discloses an autonomous black-start method for energy storage converters under islanded conditions without mains power. This method relies on the rigorous logical timing and algorithm flow of each module in the system, referring to... Figure 2 As shown, the specific operation steps include the following: Step S1: The system is in black boot standby mode and receives external black boot commands.

[0034] In extreme islanded operation with no external AC power input, the system defaults to a deep sleep black-start standby mode upon power-up. In this mode, both the DC and AC contactors of the energy storage converter remain normally open, the pulse-width modulation output port of the digital signal processor is in a high-impedance blocked state, and the thyristor trigger circuit of the static transfer switch is suppressed. Only the auxiliary power system draws microamplitude current from the energy storage battery pack to maintain standby power for the core wake-up circuits of the vehicle energy management system and battery management system. When the operator issues a black-start message via the vehicle touchscreen or injects it into the vehicle energy management system via a remote Ethernet interface, the data reception interrupt service routine of the vehicle energy management system is triggered. The message undergoes CRC verification and instruction word parsing; after confirmation, the system's state machine transitions to formally receive external black-start commands.

[0035] In step S2, the vehicle energy management system performs a global response and instantly wakes up the underlying control loop through the auxiliary power system.

[0036] After the on-board energy management system confirms the validity of the black start command, the microprocessor's processing core outputs a high-level drive signal to its internal general purpose input / output (GPIO) ports. This drive signal triggers the solid-state relays on the distribution board, activating the 24V DC mainline output from the auxiliary power system. The power supply bus instantly injects energy into the control board of the energy storage converter, the gate driver board of the insulated gate bipolar transistor, the Hall current and voltage sensors, and the trigger board of the static transfer switch. Upon power-up of the control board, the reset circuit releases a reset signal, and the built-in high-speed computing chip completes the execution of low-level hardware initialization code within microseconds, including clock tree configuration, register initial value loading, and analog-to-digital converter bias voltage calibration, thus instantly waking up the underlying control loop. This entire wake-up process bypasses the traditional tens of seconds of grid voltage and frequency detection and phase-locked loop synchronization waiting process before grid connection of the inverter.

[0037] Step S3: The at least one energy storage converter PCS actively disconnects from the grid tracking mechanism.

[0038] After the underlying control loop is woken up, the state machine within the high-speed computing chip enters the voltage build-up preparation phase. In conventional algorithm logic, the software phase-locked loop function continuously reads the analog-to-digital converter register of the grid voltage sampling port and uses orthogonal vector calculation to extract the grid phase angle. However, upon receiving the islanded black-start mode control word from the vehicle energy management system via Ethernet, the main control program triggers a conditional branch instruction, actively bypassing the call entry point of the software phase-locked loop function module. The phase generation register inside the chip is redirected to a fixed-step integrator driven by an internal timer. This integrator continuously accumulates angular velocity at a preset rated frequency (e.g., 50 Hz) to generate a reference phase angle independent of any external physical grid. The execution logic of this fixed-step integrator in each discrete control cycle is as follows: based on the reference phase angle of the previous control cycle, add the product of the equivalent angular velocity corresponding to the fixed rated frequency (e.g., 50 Hz) and the underlying microprocessor interrupt execution cycle, and use this as the reference phase angle for the current control cycle. After each accumulation, the result is modulo 2 (pi) to limit the reference phase angle value to always cycle within the equivalent radian range of 0 to 360 degrees. Through this proactive intervention at the command logic level, the energy storage converter actively breaks away from the grid tracking mechanism, eliminating the startup logic deadlock caused by the loss of external mains reference voltage.

[0039] In step S4, the energy storage converter PCS autonomously performs AC voltage building calculations and executes instructions based on the built-in high-speed computing chip, completing the rapid voltage building in less than or equal to 20 milliseconds.

[0040] As the reference phase is generated, the high-speed computing chip initiates voltage outer-loop control calculations within each control cycle (less than 5 milliseconds). The chip's memory stores a target variable for the AC output voltage rating (e.g., three-phase 380 volts). The voltage build-up algorithm calls a soft-start ramp function, causing the voltage command setpoint to start from zero volts and linearly increase to the rated target value in the time domain according to a preset fixed slope. The iterative logic of the soft-start ramp function is as follows: within each control cycle, the voltage command setpoint is increased by a fixed single-step increment from the previous cycle. This single-step increment is equal to the rated target voltage (e.g., 380 volts) multiplied by the underlying interrupt control cycle, then divided by the configured desired voltage build-up time (set to 10 milliseconds), thus achieving a linear and smooth increase of the voltage command within the set physical time. Subsequently, the sampled voltages in the three-phase stationary coordinate system are mapped to a synchronously rotating orthogonal dq coordinate system using the Park Transform. The Park Transform, employing amplitude-invariant constraints, calculates the corresponding sine and cosine values ​​using the reference phase angles generated in the previous steps, and constructs the three-phase transformation coefficients by combining the trigonometric function relationships of phases differing by 120 degrees. The instantaneous values ​​of the real-time sampled three-phase AC voltages are multiplied by their corresponding cosine transformation coefficients and summed. The sum is then multiplied by two-thirds to obtain the DC-quantized d-axis component. Similarly, the instantaneous values ​​of the three-phase AC voltages are multiplied by their corresponding negative sine transformation coefficients and summed. The sum is then multiplied by two-thirds to obtain the DC-quantized q-axis component. This difference is calculated with the given voltage command. The error value is then fed into the proportional-integral controller to calculate and generate the desired inner current loop command. The inner current loop then calculates the final three-phase modulation wave duty cycle command.

[0041] Because the instruction execution time of the high-speed computing chip is strictly controlled within an extremely short range of microseconds to milliseconds, it can complete multiple duty cycle updates within a quarter cycle of the AC waveform. After receiving these duty cycle instructions, the pulse width modulation module generates a PWM wave sequence with dead time, driving the high and low voltage side switches of the inverter bridge arm to alternately turn on and off. The DC power from the battery pack is converted into high-frequency pulsed AC power through the inverter bridge, and then, after integration and smoothing by the LCL passive filter, a standard sinusoidal AC voltage waveform is formed at the AC output of the inverter. Through the above calculation and instruction execution, which is entirely based on the chip's autonomous computing power and independent of external references, the amplitude of the output AC voltage completes the extremely rapid voltage build-up process from zero to the rated value within a physical time of less than or equal to 20 milliseconds.

[0042] Step S5: After the voltage is built up, the energy storage converter PCS switches to the off-grid constant voltage and constant frequency working mode. Combined with the high current discharge capability of the energy storage battery pack, the initial power is injected into the external paralyzed transformer area through at least one static transfer switch STS.

[0043] Once the high-speed sampling channel continuously identifies that the effective value of the AC side outlet voltage and its total harmonic distortion (THD) are both stable within the set tolerance band, specifically, the determination logic for the set tolerance band needs to simultaneously satisfy two absolute threshold conditions for a complete fundamental frequency cycle (i.e., 20 milliseconds): First, the real-time calculation result of the effective value of the outlet three-phase line voltage is within ±2% of the rated 380V; second, the THD value of the voltage waveform calculated based on the Fast Fourier Transform is continuously less than 3%. Only when both conditions are simultaneously satisfied can the microcontroller lower the level of the general-purpose input / output port to output a physical command to trigger the closure of the static transfer switch. The main control logic flag is set to confirm the voltage build-up is complete. At this time, the control loop's control strategy smoothly transitions and locks into the off-grid constant voltage and constant frequency operating mode based on the virtual synchronous machine algorithm, maintaining the rotation speed and amplitude of the AC voltage space vector constant.

[0044] Subsequently, the high-speed computing chip sends a closing trigger signal to the static transfer switch via its digital output pin. The anti-parallel thyristor assembly inside the static transfer switch conducts instantaneously upon detecting a zero-crossing of the phase voltage, directly bridging the stable AC bus already established by the energy storage converter to the power supply line of the external paralyzed transformer area. At the instant of conduction, the excitation impedance of the transformer and various inductive loads on the external line generate a huge surge current impact. During this physical process, the 352.08 kWh high-safety energy storage battery base acts as an energy buffer. The extremely low internal resistance and widened current-conducting busbar of the battery allow and support a continuous output of a large current of 1072 amps. Simultaneously, the converter's internal current loop, upon detecting a current spike, quickly compensates for the bus voltage drop by adjusting the duty cycle, preventing system undervoltage collapse. Ultimately, thanks to the stable control achieved through deep integration of hardware and software, the system successfully injects high-quality initial startup power into the external paralyzed transformer area.

[0045] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A self-starting black-start system for an energy storage converter under islanded operation without mains power, characterized in that, include: Onboard energy management system, battery management system, auxiliary power system, energy storage battery pack, at least one static transfer switch (STS) and at least one energy storage converter (PCS); The energy storage battery pack is electrically connected to the at least one energy storage converter PCS to provide a black start energy base. The auxiliary power system includes a redundant power module that supports three power input sources, including AC and DC inputs. The third input source is connected to the high-voltage bus of the energy storage battery pack through an isolated converter that converts high-voltage DC to low-voltage DC. The auxiliary power system is used to, upon receiving a black start command, bypass AC voltage detection and directly distribute the stable output DC power to the main control board, drive board, and core module of the energy storage converter and the battery management system, so as to instantly wake up the underlying control loop. The on-board energy management system is used to control the energy storage converter PCS to actively disconnect from the grid tracking mechanism when a black start command is received. The energy storage converter PCS has an embedded computing chip, which is configured to perform AC voltage build-up calculation: actively bypassing the call entry of the software phase-locked loop function module, pointing the internal phase generation register to the fixed-step integrator driven by the internal timer to generate a reference phase angle independent of the external power grid; calling the soft-start ramp function to linearly increase the voltage command setpoint to the rated target value according to the preset single-step increment step size, so as to complete the AC voltage build-up. The energy storage converter PCS is also configured to continuously monitor the AC side output voltage. When it is detected that the effective value of the three-phase line voltage at the output is within ±2% of the rated value, and the total harmonic distortion of the voltage waveform is continuously less than 3%, and the above two conditions are met for a complete fundamental frequency cycle, a physical command is output to trigger the closure of the static transfer switch STS. Then, it switches to the off-grid constant voltage and constant frequency working mode to provide initial start-up power to the paralyzed load.

2. The system according to claim 1, characterized in that, The off-grid constant voltage and constant frequency working mode adopts virtual synchronous machine technology combined with dual closed-loop feedback control of voltage outer loop and current inner loop; The voltage outer loop employs a proportional resonant regulator, which includes a resonant control branch that uses a second-order bandpass filter logic composed of a set resonant gain coefficient, a control bandwidth cutoff frequency, and a fundamental resonant angular frequency to extract the fundamental error signal. The inner current loop uses a proportional-integral regulator to limit the inductor current and track its waveform.

3. The system according to claim 1, characterized in that, The main control module of the battery management system internally runs a state-of-charge estimation algorithm that combines the ampere-hour integration method and the open-circuit voltage calibration method. This algorithm introduces a closed-loop correction observer based on Kalman filtering, specifically configured as follows: Based on the predicted value of the state of charge at the previous moment, the predicted value of the observed open-circuit voltage is obtained. The deviation between the actual collected open-circuit voltage and the predicted value is calculated. This deviation is multiplied by the Kalman gain obtained by real-time iterative optimization and added to the prior predicted value as a correction term to achieve dynamic convergence correction.

4. The system according to claim 1, characterized in that, The energy storage converter PCS includes an LCL-type low-pass filter composed of a grid-side inductor, a converter-side inductor, and a filter capacitor. Its parameter configuration satisfies the following physical constraints: The lower limit of the sum of the converter-side inductance and the grid-side inductance is set as eight times the minimum DC bus voltage divided by the product of the switching frequency and the maximum allowable ripple current; the upper limit is set as the square of the AC-side rated phase voltage multiplied by 10%, then divided by the product of the converter's rated capacity and the fundamental angular frequency; and the natural resonant frequency of the LCL-type low-pass filter is limited to a range greater than ten times the fundamental frequency and less than half the switching frequency.

5. The system according to claim 1, characterized in that, It also includes a thermal management unit, wherein the on-board energy management system is configured to use a fuzzy PID algorithm to adjust the compressor speed and water pump flow of the thermal management unit in real time; The input variables of the fuzzy PID algorithm are the difference between the current highest temperature of the battery node and the target set temperature, and the rate of change of the temperature difference. The mapping logic of its control rule table includes: when the temperature difference is positive and the rate of change of the temperature difference is positive, the output proportional coefficient correction is mapped to positive, the integral correction is mapped to negative, and the derivative correction is mapped to positive.

6. The system according to claim 1, characterized in that, The energy storage battery pack adopts a design architecture in which multiple basic battery nodes are connected in series to establish a high-voltage DC bus. Its high-voltage DC bus supports continuous high-current discharge and provides an energy buffer to resist the surge current impact caused by the external transformer excitation impedance and inductive load when the static transfer switch STS is closed.

7. The system according to claim 1, characterized in that, It also includes a multi-protocol communication system, in which the inter-module communication within the battery management system and the high real-time status frames with the vehicle energy management system are transmitted via CAN bus; The smart meter and insulation tester are connected to the slave station polling network via RS485 bus; The high-throughput waveform data and control commands between the vehicle energy management system and the energy storage converter are transmitted via Ethernet bus using the TCP / IP protocol in full-duplex communication.

8. The system according to claim 1, characterized in that, The execution logic of the fixed-step integrator in each discrete control cycle is as follows: Based on the reference phase angle of the previous control cycle, add the product of the equivalent angular velocity corresponding to the fixed rated frequency and the interrupt execution cycle of the underlying microprocessor, and then perform a modulo operation on twice pi using the accumulated result.

9. The system according to claim 1, characterized in that, The single-step increment of the soft-start ramp function is equal to the rated target voltage multiplied by the underlying interrupt control cycle, and then divided by the configured expected voltage build-up time.

10. A method for autonomous black-start of an energy storage converter under islanded operation without mains power, implemented based on the system described in any one of claims 1 to 9, characterized in that, Includes the following steps: Receive external black boot commands; By using an auxiliary power supply system that includes three power input sources, DC power is directly distributed without AC voltage detection to wake up the underlying control loop; Actively disconnect from the grid tracking mechanism, bypass the software phase-locked loop function, and generate a reference phase angle independent of the external grid through an integrator driven by an internal timer; The soft-start ramp function is invoked to increment the voltage command setpoint according to a preset slope, thus completing the AC voltage build-up calculation and execution. The system continuously monitors the AC voltage at the outlet. When the effective value of the three-phase line voltage is found to be within ±2% of the rated value and the total harmonic distortion is less than 3%, and this condition is maintained for a complete fundamental frequency cycle, the system switches to the off-grid constant voltage and constant frequency operating mode, triggering the static transfer switch (STS) to close and inject initial power into the external paralyzed transformer area.

Images