Low-voltage power generation vehicle non-inductive grid-connected control system and oscillation suppression method

Abstract

The application relates to a low-voltage power generation vehicle non-inductive grid-connected control system and a shock suppression method. The system comprises a power grid / low-voltage power generation vehicle, a control system and an executing mechanism. The power grid / low-voltage power generation vehicle is connected with the control system to communicate with the control system. The control system is connected with the executing mechanism to control the non-inductive grid connection of the low-voltage power generation vehicle. The application realizes more accurate parameter calculation, avoids multiple protocol compatibility, greatly reduces the manual operation time, and avoids system shock and other problems.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of new power systems, in particular to a low-voltage power car non-inductive grid-connected control system and oscillation suppression method. BACKGROUND

[0002] New power systems have higher requirements for the rapid access capability of distributed emergency power supplies. As the core equipment for emergency power supply of distribution networks, the grid-connection efficiency and reliability of low-voltage power cars directly affect the quality of people's livelihood power supply. However, the industry has been plagued by the following problems for a long time: first, interface protocol fragmentation. 38% of existing power cars use old protocols such as Modbus RTU, 45% support the new IEC 61850 standard, and the rest use private protocols, resulting in a high protocol parsing failure rate of up to 12.7% during grid-connection; second, system oscillation caused by grid-connection impact. Traditional closing relies on manual phase adjustment, with a phase difference tolerance of only ±5°, and is prone to low-frequency oscillation when the grid frequency fluctuates; third, lack of multi-car coordinated control. Existing systems cannot achieve wireless connection of more than 3 kilometers, and power distribution relies on physical cables, with an emergency response time of more than 15 minutes.

[0003] Existing industry solutions have obvious limitations: for example, the puncture access box proposed by the State Grid Shandong patent CN202311694075A simplifies physical connections but does not solve the protocol compatibility problem; the dynamic frequency tracking algorithm of the Guangzhou patent CN119921389A of the Southern Power Grid improves synchronization accuracy but does not eliminate the risk of oscillation caused by power backflow. With the strengthening of non-inductive grid-connection index requirements in the "Power Emergency Equipment Technical Guidelines" (GB / T 36549-2023), it is necessary to develop a system that combines multi-protocol adaptation, zero-impact closing, and active oscillation suppression. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a low-voltage power car non-inductive grid-connection control system and oscillation suppression method, which uses a hardware platform to support long-distance wireless multi-car coordinated control, effectively solving the key problems of non-uniform grid-connection protocols of traditional power cars, long time-consuming manual operation, and power backflow-induced oscillation, providing core equipment support for building a new power system with fast response, non-inductive switching, and safety and stability, and promoting the progress of non-inductive grid-connection technology for emergency power cars.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] In a first aspect, the embodiments of the present application provide a low-voltage power car non-inductive grid-connection control system, comprising a power grid / low-voltage power car, a control system, and an execution mechanism,

[0007] The power grid / low-voltage power car is connected to the control system for communication with the control system;

[0008] The control system is connected to an actuator to control the non-inductive grid connection of the low-voltage power car.

[0009] The control system comprises a Hall current / voltage sensor, a high-speed ADC, an FPGA core controller, a multi-standard physical layer interface, a wireless communication module, an excitation / inverter drive interface, and a hardware direct connection tripping path interface. The Hall current / voltage sensor collects real-time current and voltage signals of the power grid, and transmits the analog signals to the FPGA core controller after being converted into digital signals by the high-speed ADC. The FPGA core controller is also connected to the multi-standard physical layer interface, which is connected to the power grid / low-voltage power car. The FPGA core controller is also connected to the wireless communication module, the excitation / inverter drive interface, and the hardware direct connection tripping path interface. The wireless communication module communicates with the power grid / low-voltage power car. The excitation / inverter drive interface and the hardware direct connection tripping path interface are connected to the actuator to realize the control of the actuator by the FPGA core controller.

[0010] A non-inductive grid connection control method for a low-voltage power car, comprising the following specific steps:

[0011] Protocol non-inductive compatibility, capturing and analyzing power car and grid data from multiple heterogeneous protocols, and outputting them as unified standardized data streams;

[0012] Microsecond-level grid connection detection, uninterrupted high-speed sampling of three-phase current of the power grid, accurate capture of the starting time of grid connection or disturbance by calculating the instantaneous change rate of current modulus;

[0013] Zero phase difference closing control, in the state of waiting for grid connection, generating excitation adjustment signal by Lyapunov stability control algorithm according to the real-time detected voltage phase difference between the power grid and the power car, driving the phase difference to converge to zero, and executing closing when the preset conditions are met;

[0014] Multi-dimensional oscillation suppression, after closing, real-time monitoring of system frequency, activation of virtual impedance damping and power feedforward compensation in stages according to the size and duration of frequency deviation, to maintain the stability of the system after grid connection.

[0015] The protocol non-inductive compatibility specifically comprises,

[0016] The multi-standard physical layer interface is connected to the low-voltage power car, and the FPGA core controller captures the corresponding protocol bottom bit stream / frame through internal logic;

[0017] The key information of different protocols is located at different offset positions of the message. According to the preset or dynamically issued configuration, the mask is set;

[0018] The mask performs a bit-level AND operation with the data stream, stripping out the key fields and extracting the core information used to identify the protocol identity;

[0019] The function code and device address key fields stripped out in the first stage are concatenated together as input to a dedicated hash operation unit;

[0020] The hash operation unit is implemented in hardware logic inside the FPGA core controller, and uses a hardware-optimized variant of the FNV-1a algorithm to generate a fixed-length hash value for the key field combination. This hash value is the protocol's fingerprint;

[0021] A pre-compiled protocol fingerprint-resolver address mapping table is integrated inside the FPGA core controller, stored in Block RAM and organized in a balanced binary tree or content-addressable memory structure. The generated protocol fingerprint is used as a key to search this table;

[0022] The search result is a pointer or ID that points to the corresponding complete protocol resolver in memory. Based on this pointer or ID, the resolver is dynamically loaded and bound;

[0023] The bound resolver is then responsible for parsing the complete raw packet into a standardized internal data object.

[0024] The microsecond-level grid detection specifically involves,

[0025] The real-time grid current signal is obtained through a closed-loop Hall effect current sensor, and after being conditioned by the front-end conditioning circuit, the three-phase current is converted from the stationary a-b-c coordinate system to the stationary two-phase , coordinate system through Clarke transformation;

[0026] The spatial vector information and zero sequence component of the current are separated out, and the current modulus is calculated;

[0027] The time rate of change of the current modulus dI / dt is calculated,

[0028] The mean and variance of the recent current modulus are continuously calculated through a moving average filter to assess the current noise level,

[0029] (1),

[0030] (2),

[0031] To eliminate interference from individual glitch pulses, a valid grid-connected event must satisfy the condition that the dI / dt values ​​at N consecutive sampling points all exceed the dynamic threshold. An improved multi-point summation filter is used to eliminate high-frequency noise in the sampling circuit.

[0032] (3),

[0033] in The average modulus is N, the size of the filter window is N, and k represents the current sampling time. A dynamic threshold is set, and a detection signal is sent when the rate of change of the current modulus is greater than or equal to the threshold.

[0034] The zero-phase-difference closing control is specifically as follows:

[0035] A real-time dynamic model of the phase difference between the power grid and the generator is established based on event-triggered signals.

[0036] Real-time phase difference:

[0037] (4),

[0038] in: The phase angle of the grid voltage is obtained through a 100kHz high-speed sampling circuit. The phase angle of the generator output voltage is adjusted by the excitation controller;

[0039] Constructing the Lyapunov energy function:

[0040] Design a quadratic energy function It needs to meet the following requirements:

[0041] 1. 2. ,

[0042] (5),

[0043] First item The kinetic energy corresponding to the phase deviation, i.e., dynamically adjusting the excitation voltage to correct the rotor magnetic field, the second term The potential energy corresponding to the accumulated phase error, i.e., the integral term eliminating the steady-state error,

[0044] Differentiation of Lyapunov functions:

[0045] (6),

[0046] Introducing a control objective, by adjusting the excitation voltage make Substituting into the derivative:

[0047] (7),

[0048] To meet , eliminate cross effects, let

[0049] (8),

[0050] Where is the proportional gain fast response phase mutation, is the integral gain to eliminate steady-state error, eliminate cross term effect;

[0051] The control equation is:

[0052] (9),

[0053] When > threshold value, Increase the excitation current to accelerate the generator; when , reduce the excitation current to slow down, so that Always shrink to the threshold value until the closing command is met.

[0054] The multi-dimensional oscillation suppression is specifically,

[0055] When the absolute value of the frequency deviation exceeds the threshold value, the virtual impedance layer is activated first, the preset impedance is injected at the grid connection point to change the network damping characteristics, and the low-frequency oscillation energy is absorbed specifically by modifying the voltage command reference value of the inverter PWM controller in real time, thereby equivalently changing the output impedance characteristics of the grid connection point on the electrical level.

[0056] Feedforward power compensation P comp The calculation result is directly added to the active power reference set value of the power control loop of the generator car as a direct power regulation amount, so as to directly offset the power fluctuation caused by frequency disturbance from the power source.

[0057] Compared with the prior art, the beneficial effects of the present application are:

[0058] By innovating the multi-protocol adaptive interface to realize the plug and play of new and old equipment, using high-speed sampling of three-phase current and Lyapunov stability algorithm to achieve microsecond-level grid connection detection and zero-phase difference closing control, and adopting virtual impedance injection, power feedforward compensation and other triple protection mechanisms to greatly shorten the oscillation suppression time. The special hardware platform supports long-distance wireless multi-car cooperative control, effectively solves the key problems of non-uniform grid connection protocol of traditional generator cars, long time of manual operation, power recoil induced oscillation, etc., provides core equipment support for building a new type of power system with fast response, no-sense switching and safety and stability, and promotes the progress of emergency generator car no-sense grid connection technology. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0060] Figure 1 System block diagram of the present application.

[0061] Figure 2 System diagram of the embodiments of the present application.

[0062] Figure 3 Protocol compatible implementation flowchart of the present application.

[0063] Figure 4 Microsecond-level grid connection detection flowchart of the present application.

[0064] Figure 5 Zero-phase difference closing control flowchart of the present application.

[0065] Figure 6 Module integration system flowchart of the present application. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. It should be noted that similar reference numerals and letters in the following drawings represent similar items, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0067] The term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article or equipment including the element.

[0068] The terms "first", "second" and the like are only used to distinguish one entity or operation from another entity or operation, and cannot be understood as indicating or implying relative importance, and cannot be understood as requiring or implying any such actual relationship or order between the entities or operations.

[0069] As Figure 1 and Figure 2As shown, a low-voltage power generation vehicle non-inductive grid-connected control system includes a power grid / low-voltage power generation vehicle, a control system, and an actuator,

[0070] The power grid / low-voltage power generation vehicle 1 is connected to the control system for communication with the control system.

[0071] The control system is connected to the actuator 9 for controlling the low-voltage power generation vehicle non-inductive grid connection.

[0072] The control system includes a Hall current / voltage sensor 2, a high-speed ADC 3, a FPGA core controller 4, a multi-standard physical layer interface 5, a wireless communication module 6, an excitation / inverter drive interface 7, a hardware direct connection trip path interface 8, the Hall current / voltage sensor 2 collects real-time current and voltage signals of the power grid, and transmits the analog signals to the FPGA core controller 4 after converting them into digital signals by the high-speed ADC 3, the FPGA core controller 4 is also connected to the multi-standard physical layer interface 5, which is connected to the power grid / low-voltage power generation vehicle 1, the FPGA core controller 4 is also connected to the wireless communication module 6, the excitation / inverter drive interface 7, and the hardware direct connection trip path interface 8, the wireless communication module 6 communicates with the power grid / low-voltage power generation vehicle 1, the excitation / inverter drive interface 7 and the hardware direct connection trip path interface 8 are connected to the actuator 9 for realizing the control of the actuator 9 by the FPGA core controller 4.

[0073] Step 1: Protocol non-inductive compatibility

[0074] As Figure 3As shown, step 1 addresses the difficulty of power car protocol fragmentation by a hardware-accelerated multi-standard physical layer interface and a three-stage pipeline mechanism. First, physical layer signal capture and key field hardware stripping. A multi-standard physical layer interface compatible with RS-485, CAN, Ethernet, etc. is constructed using a field programmable gate array (FPGA) and a set of pluggable or on-board physical layer transceivers (PHYs). The FPGA is connected to these PHYs through its flexible I / O pins and implements the capture of the corresponding protocol bit stream / frame through internal logic, such as NRZ encoding and decoding, Manchester encoding and decoding, CAN bit stuffing / despilling, etc. The key information of different protocols is located at different offset positions in the message. The system can set a mask according to the preset or dynamically issued configuration. When the data stream enters the FPGA, this mask performs a bit-level "and" operation with the data stream, retaining only the key fields we need and masking out all other data, accurately extracting the core information for identifying protocol identity within nanoseconds. Then, protocol fingerprint generation and efficient matching. The function code + device address key field stripped in the first stage is concatenated to form an input to a dedicated hash operation unit (Hash Core). This unit is implemented through hardware logic inside the FPGA, using a non-encrypted hash algorithm optimized for hardware. The invention preferably uses a hardware-optimized variant of the FNV-1a algorithm. The reason for choosing this algorithm is that it has excellent avalanche effect and low collision rate, and the calculation process does not involve complex multiplication or lookup table operations, making it easy to implement efficiently in FPGA through shift and XOR logic, thereby ensuring that the generation delay of the protocol fingerprint is on the order of nanoseconds. A fixed-length hash value is generated for the combination of key fields, which is the fingerprint of the protocol. Finally, dynamic binding of the parser and standardized data stream output. A pre-compiled protocol fingerprint-parser address mapping table is integrated inside the FPGA. This table can be efficiently stored in Block RAM and organized in the structure of a balanced binary tree or content addressable memory (CAM). The protocol fingerprint generated in the second stage is used as a key to search this table. The search operation is completed in hardware, with a time complexity of O(log n) or O(1). The search result is a pointer or ID pointing to the corresponding complete protocol parser (a microcode or software module) in the memory. The system controller dynamically loads and binds the parser according to the ID. The bound parser is then responsible for parsing the complete raw message into a standardized internal data object. For example, whether the raw message is a Modbus 0x03 read register instruction or an MMS Read service, it will ultimately be converted into a standardized structure such as {device_id: X, variable: "VA", value:220.1}. Meanwhile, the modular design supports plug-and-play deployment, laying a stable data foundation for subsequent microsecond-level grid detection.

[0075] Step 2: Microsecond-level grid-connection detection

[0076] As shown in Figure 4 , Step 1 eliminates protocol parsing delay and outputs standardized data stream, avoiding detection failure in Step 2 due to data format error. In order to calculate three-phase current in real time and accurately, we must convert it from complex three-phase time-varying coordinate system (a-b-c) to static orthogonal coordinate system (α-β) which is convenient for microprocessor analysis and control. Step 2 obtains real-time current signal of power grid through closed-loop Hall effect current sensor, and after conditioning by front-end conditioning circuit, three-phase current is converted from static a-b-c coordinate system to static two-phase , coordinate system through Clarke transformation. The key of this transformation is that it can separate the spatial vector information and zero sequence component of current, as shown in equation 1, and the current modulus is calculated through equation 2. Instead of using traditional fixed amplitude threshold, the system calculates the time rate of change of current modulus (dI / dt). At the same time, through a moving average filter, the mean and variance of recent current modulus are continuously calculated to evaluate the current noise level.

[0077] (1),

[0078] (2),

[0079] In order to eliminate the interference of single glitch pulse, an effective grid-connection event must satisfy that the dI / dt value of continuous N sampling points exceeds the dynamic threshold. An improved multi-point summation filter is used to eliminate high-frequency noise of sampling circuit,

[0080] (3),

[0081] where is the average modulus, N is the size of filter window, and k represents the current sampling time. The dynamic threshold is set, and the detection signal is sent when the current modulus rate of change ≥ threshold.

[0082] Dynamic differential calculation: calculate the rate of change of current modulus every 0.01ms.

[0083] Condition 1: rate of change ≥ threshold

[0084] Condition 2: continuous 3 sampling points satisfy condition 1

[0085] Output logic: after both conditions are met, send detection signal to closing controller within 1ms. ​

[0086] Step 3: Zero-phase closing control

[0087] As Figure 5 the grid-connected moment has been accurately captured, the phase difference, frequency difference, and voltage difference between the generator and the grid are simultaneously driven to zero in the shortest time and in the most stable manner to achieve ideal closing. The core of Step 3 is to solve the closing inrush problem caused by the insufficient accuracy of traditional manual phase adjustment. The event trigger signal is sent out in Step 2; the real-time high-precision measurement of the phase and amplitude of the grid and generator voltage. The zero-phase closing control technology studied achieves precise phase synchronization through a rigorous mathematical framework. The core is to establish a real-time phase difference dynamic model (Formula 4) between the grid and the generator. This model relies on the microsecond-level current modulus detection capability of Step 2 to refresh data at high speed, ensuring that the resolution of Δθ reaches 0.01°. On this basis, a Lyapunov stability control algorithm (Formulas 5-9) is innovatively designed: an energy function is constructed, and its physical meaning is that the first term (proportional term ) uses a coefficient of 0.8 to quickly respond to phase deviation, and the second term (integral term ) accumulates historical errors through a coefficient of 0.2 to eliminate steady-state deviation. The proportional coefficient and the integral coefficient are not randomly set. They are the results of iterative optimization in the MATLAB / Simulink simulation environment for a typical distribution network parameter model and a preset step disturbance signal. The optimization goal is to achieve the shortest phase difference convergence time and the smallest overshoot while ensuring system stability. The fast response to phase deviation is ensured, and is used to eliminate long-term existing steady-state errors, and the combination of the two ensures the rapidity and accuracy of the control. By taking the derivative and substituting it into the frequency difference dynamic equation, the control equation (Formula 9) is derived. The controller dynamically adjusts the generator excitation voltage accordingly, and the output phase is corrected in real time by changing the rotor magnetic field. When the phase difference converges to ±0.5° and the frequency difference ≤0.05Hz, this threshold is established by GB / T 36549-2023, and when the above conditions are met, the closing is triggered, and the inrush current is suppressed within 5% of the rated current. This process forms a closed-loop logic: high-precision sensing provides real-time Δθ, Lyapunov algorithm generates instructions to achieve exponential synchronization, and strict threshold guarantees non-impact closing.

[0088] Real-time phase difference:

[0089] (4),

[0090] Where: ​For grid voltage phase angle (acquired by 100 kHz high-speed sampling circuit), For generator output voltage phase angle (adjusted by excitation controller).

[0091] Construct Lyapunov energy function:

[0092] Design quadratic energy function , need to meet: 1. >0 (positive definite), 2. <0 (negative definite, to ensure asymptotic stability)

[0093] (5),

[0094] The first term Corresponding to the kinetic energy of phase deviation (dynamic adjustment of excitation voltage to correct the rotor magnetic field) The second term Corresponding to the potential energy of cumulative phase error (integral term to eliminate steady-state error).

[0095] Derivation of Lyapunov function:

[0096] (6),

[0097] Introduce control target, adjust excitation voltage Make , into the derivative formula:

[0098] (7),

[0099] To meet <0, eliminate cross effects, let

[0100] (8),

[0101] Where Proportional gain to quickly respond to phase mutations, Integral gain to eliminate steady-state error. Eliminate cross term effects

[0102] Control equation:

[0103] (9),

[0104] When Threshold value, Increase the excitation current to speed up the generator; when , reduce the excitation current to slow down, so that Δθ always shrinks to the threshold value until the closing command is met.

[0105] Step 4: Multi-dimensional oscillation suppression

[0106] The core of the multi-dimensional oscillation active suppression system is to build a hierarchical progressive protection system. The design logic starts from real-time monitoring of the system frequency after grid connection. When the absolute value of the frequency deviation in step 3 exceeds the threshold value, the system first activates the virtual impedance layer, changes the network damping characteristics by injecting a preset impedance at the grid connection point. The impedance value is optimized through electromagnetic transient simulation, which can selectively absorb low-frequency oscillation energy. Specifically, it is realized by modifying the voltage command reference value of the inverter PWM (pulse width modulation) controller in real time, thereby equivalently changing the output impedance characteristics of the grid connection point in the electrical layer. Feedforward power compensation P comp The calculation result is directly added to the active power reference set value of the power control loop as a direct power adjustment amount, which directly offsets the power fluctuation caused by frequency disturbance from the power source. This double dynamic suppression forms a closed-loop control: the virtual impedance provides fast damping, and the power feedforward realizes accurate compensation. The two work together to compress the traditional suppression time from 10 seconds to sub-second.

[0107] As the final protection means, the system synchronously executes the cumulative risk judgment: the judgment logic is executed by the independent safety monitoring module in the FPGA. Once triggered, it will bypass the main processor and send a high-priority trip signal directly to the trip coil of the circuit breaker through a dedicated hardware I / O path. When the frequency fluctuation lasts more than the set threshold, the judgment system enters the instability critical state and immediately triggers the fast tripping protection. This protection uses a hardware direct tripping mechanism to avoid oscillation spread by physically isolating the fault unit. The whole suppression process follows the progressive principle of adjusting parameters first and then compensating, and cutting off locally first and then cutting off, which maximizes the maintenance of grid connection and reduces the cascading failure caused by the failure of a single measure in traditional solutions.

[0108] Step 5: Module integration application

[0109] As Figure 6 shown, step 5 integrates the hardware of the first four steps and realizes multi-car collaborative control. The outputs of steps 1-4 are integrated through a wireless communication network. The protocol compatible module of step 1 and the oscillation suppression data of step 4 are transmitted wirelessly for multi-car power coordination; the current sampling of step 2 and the control command of step 3 are embedded in the FPGA hardware. The integration of the controller starts with the construction of the hardware platform: FPGA is used as the core to handle the protocol conversion task, and multi-protocol dynamic loading is realized through hardware acceleration; the three-phase sampling circuit collects current modulus at a high rate to provide real-time data basis for power distribution. The cooperative work of the two chips compresses the signal processing delay to milliseconds.

[0110] To realize multi-car coordination in a long distance range, the system uses a wireless communication architecture, and the physical layer parameters ensure that the bit error rate is less than 10⁻ 6When the master controller receives the data and feedback signals Si of each generator car that has passed step 4 through the wireless link, it starts the dynamic power distribution algorithm: first, it calculates the total demand power of the system The ratio of the total capacity of the generator car ∑S is generated, and the basic distribution amount is generated; then the proportional term is introduced Real-time compensation of individual differences, so that the distribution error is strictly controlled within 3%. Multi-car power distribution algorithm, based on the consistency protocol design dynamic allocation logic:

[0111] (10),

[0112] The closed-loop control characteristics of this process are reflected in the continuous data interaction - after the generator car executes the power instruction, its actual output capacity Si is returned to the master controller through the wireless network. When it is detected that the load rate of a vehicle deviates from the set value by 0.5%, the algorithm immediately triggers incremental adjustment ( ), forming a negative feedback loop of instruction issuance → execution feedback → error correction.

[0113] Specific implementation case:

[0114] The multi-standard physical layer interface receives the communication protocol of the low-voltage generator car and transmits it to the FPGA core controller, and the controller transmits the processed standard data to the low-voltage generator car through the wireless communication module.

[0115] Voltage / current sensor, real-time acquisition of voltage and current signals of power grid and low-voltage generator car, high-speed ADC converts the collected analog signals into digital signals and transmits them to the FPGA core controller;

[0116] First stage: protocol non-inductive compatibility and standardized access (step 1)

[0117] Physical connection: repair personnel connect the output cables of the three generator cars to the respective grid connection points, and connect the data interfaces equipped with the control system of the invention to the switch (Ethernet), RS-485 terminal and CAN interface.

[0118] Protocol adaptive identification: after the control system is powered on, its core FPGA module captures the data stream from the grid interface in real time through the multi-standard physical layer transceiver (PHY).

[0119] Car A: the system detects the MMS message on the Ethernet, performs "and" operation and mask extraction on the function code and device address and other key fields through hardware logic, and sends them to the FNV-1a hash operation unit, which generates the "fingerprint" of the IEC 61850 protocol within nanoseconds.

[0120] B car: The system captures serial data frames from the RS-485 bus, and extracts the key fields to generate a Modbus RTU protocol "fingerprint".

[0121] C car: The same operation is performed on the data frames on the CAN bus.

[0122] Parser dynamic loading and data standardization: The "protocol fingerprint-parser address" mapping table is stored in the Block RAM inside the FPGA. The system instantaneously retrieves and dynamically loads the corresponding complete protocol parser (microcode module) according to the generated fingerprint. Whether it is a complex MMS service or a simple Modbus read-write instruction, it is ultimately parsed and converted into a unified internal standardized data structure: {device_id: "Hospital_Grid", variable: "Va", value:220.1}. This process takes less than 10 microseconds, completely eliminating the obstacle of protocol incompatibility and providing standardized data input for subsequent control.

[0123] Second stage: Microsecond-level grid connection detection and zero-phase difference closing (steps 2, 3)

[0124] Precise capture of grid connection opportunity: The power generation car is ready to be connected to the grid. Its control system samples the three-phase current of the grid at a high speed of 100 kHz through a Hall sensor. At the moment of closing the grid connection switch, the current modulus changes abruptly. The system calculates the rate of change of the current modulus (dI / dt) in real time and combines a judgment logic that all three consecutive sampling points exceed the dynamic threshold to accurately identify the starting time of the grid connection event within 1 millisecond, triggering the closing control program immediately.

[0125] Phase synchronization driven by Lyapunov algorithm: After the trigger signal is sent, the system measures the initial phase difference Δθ between the power generation car and the grid to be -12.5°. At this time, the zero-phase difference closing control algorithm starts:

[0126] The controller generates the adjustment command of the excitation voltage V f according to the pre-set Lyapunov energy function .

[0127] The proportional term (k p =0.8) responds quickly and corrects the phase deviation with great torque; the integral term (k i =0.2) is responsible for eliminating historical cumulative errors and ensuring no steady-state residual error.

[0128] Under the algorithm-driven, the phase difference Δθ converges exponentially. Within 300 milliseconds, the phase difference is accurately controlled within ±0.5°, and the frequency difference is less than 0.05 Hz.

[0129] No impact closing: When all conditions are met, the system automatically sends closing instructions. Field monitoring oscilloscope display, the inrush current at the moment of closing is successfully suppressed within 5% of the rated current, the grid voltage fluctuation is less than 0.3%, achieving "no sense of grid" of zero interference to precision equipment.

[0130] Third stage: multi-dimensional oscillation suppression and multi-vehicle cooperative control (steps 4, 5)

[0131] System stability and oscillation suppression: After grid connection, a large equipment starts, causing the system frequency to drop by 0.2 Hz.

[0132] Fast damping: The control system detects that the frequency deviation overshoots, immediately activates the first defense - virtual damping injection. By modifying the voltage reference command of the inverter PWM in real time, a preset virtual impedance is equivalent in the electrical layer, effectively absorbing the low-frequency oscillation energy caused by the disturbance.

[0133] Accurate compensation: At the same time, the second defense - power feedforward compensation starts. The system calculates the compensation power P comp based on the frequency change rate dΔf / dt, and directly superimposes it on the active power set value of the generating vehicle, actively offsetting the disturbance from the power source.

[0134] Under the synergistic action of the double mechanism, the system frequency is restored to stability within sub-second (about 750 milliseconds), avoiding the power oscillation that may last for several seconds or even tens of seconds in traditional schemes.

[0135] Multi-vehicle wireless cooperation and dynamic power distribution:

[0136] The controllers of the three generating vehicles automatically form an Ad-hoc network through the built-in wireless module. Vehicle A is dynamically elected as the master controller.

[0137] The master controller aggregates real-time load data from each grid connection point and calculates the total demand power P total of 550kW. Based on the capacity of each vehicle and the preset weight, the initial power command is issued through the consensus protocol algorithm: vehicle A (300kW) bears the main load, vehicle B (150kW) bears the medium load, and vehicle C (100kW) bears the light load.

[0138] During operation, the IT load in the B vehicle power supply area decreases at night, and the B vehicle load rate deviates from the set value by more than 0.5%. Its controller returns the state information Si to the master controller through the wireless network. The master controller algorithm immediately triggers the incremental adjustment ΔP adjust , recalculates and issues a new power command to smoothly transfer the excess generating capacity of vehicle B to vehicle A with heavier load. The entire closed-loop negative feedback regulation process takes less than 100 milliseconds, achieving the optimization of the overall system operation efficiency.

[0139] The above merely provides an example of the present application, but is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A low-voltage power generation vehicle non-inductive grid-connected control method, characterized by, The method comprises the following specific steps: Protocol agnostic compatibility, capturing and parsing power car and grid data from multiple heterogeneous protocols, and outputting them as a unified standardized data stream; Microsecond-level grid connection detection, uninterrupted high-speed sampling of three-phase current of the grid, accurate capture of the starting time of grid connection or disturbance by calculating the instantaneous change rate of current modulus; Zero-phase difference closing control, in the state of waiting to connect to the grid, generating excitation regulation signal by Liapunov stability control algorithm according to the real-time detected phase difference between the grid and the power car voltage, driving the phase difference to converge to zero, and executing closing when the preset condition is met; Multi-dimensional oscillation suppression, after closing, real-time monitoring of system frequency, according to the size and duration of frequency deviation, activating virtual impedance damping and power feedforward compensation in stages to maintain the stability of the system after grid connection; The protocol agnostic compatibility specifically is, Multi-standard physical layer interface connects low-voltage power car, and the FPGA core controller realizes the capture of the corresponding protocol bottom bit stream through internal logic; The key information of different protocols is located at different offset positions of the message, and the mask is set according to the preset or dynamically issued configuration; The mask and the data stream are operated at the bit level, the key fields are stripped, and the core information for identifying the protocol identity is extracted; The function code and the device address key field stripped in the first stage are concatenated as input and sent to a dedicated hash operation unit; The hash operation unit is realized by hardware logic in the FPGA core controller, and a fixed-length hash value is generated for the key field combination by using the hardware optimization variant of FNV-1a algorithm, and the hash value is the fingerprint of the protocol; A pre-compiled protocol fingerprint-parser address mapping table is integrated in the FPGA core controller, the mapping table is stored in Block RAM, and is organized in the structure of balanced binary tree or content addressable memory, and the generated protocol fingerprint is used as a key to search in the table; The search result is a pointer or ID pointing to the corresponding complete protocol parser in the memory, and the parser is dynamically loaded and bound according to the pointer or ID; The bound parser is then responsible for parsing the complete original message into a standardized internal data object; The microsecond-level grid connection detection specifically is, The real-time current signal of the power grid is acquired by a closed-loop Hall effect current sensor, and after being conditioned by a front-end conditioning circuit, three-phase currents are converted from the stationary a-b-c coordinate system to the stationary two-phase , coordinate system through a Clarke transformation. Separating the spatial vector information of the current and the zero sequence component and calculating the current modulus; The time rate of change of the current modulus dI / dt is calculated The time rate of change of the current modulus dI / dt is calculated The mean and variance of the recent current modulus are continuously calculated through a moving average filter to evaluate the current noise level, ， ， In order to eliminate the interference of single glitch pulse, an effective grid connection event must satisfy that the dI / dt value of N consecutive sampling points exceeds the dynamic threshold, an improved multi-point summation filter is used to eliminate high-frequency noise of the sampling circuit, ， wherein is the average modulus, N is the size of the filter window, k represents the current sampling time, and the dynamic threshold is set, and the detection signal is sent when the current modulus change rate is greater than or equal to the threshold. The zero-phase difference closing control specifically is, Based on the event trigger signal, a real-time phase difference dynamic model of the grid and the generator is established; Real-time phase difference: ， wherein: is the grid voltage phase angle, which is obtained by a 100 kHz high-speed sampling circuit, is the generator output voltage phase angle, which is adjusted by the excitation controller; A Liapunov energy function is constructed: Designing a quadratic energy function , which needs to satisfy: , , ， The first term The kinetic energy corresponding to the phase deviation, i.e. the dynamic adjustment of the field voltage to correct the rotor field, the second term The potential energy corresponding to the accumulated phase error, i.e. the integral term eliminates the steady-state error, Derivation of the Liapunov energy function: ， The control target is introduced by adjusting the field voltage to make , the derivative formula is brought in: ， To meet , eliminate cross-influence, let ， wherein Kp is a proportional gain fast response phase jump, Ki is an integral gain to eliminate steady state error, to eliminate the cross term effect; The control equation is: ， When the threshold value, increasing the field current to accelerate the generator; when decreasing the field current to decelerate, so that always converging to the threshold value until the closing command is met.

2. The low voltage traction vehicle gridless control method of claim 1, wherein, The multi-dimensional oscillation suppression specifically is, When the absolute value of the frequency deviation exceeds the threshold value, the virtual impedance layer is first activated, the network damping characteristics are changed by injecting a preset impedance at the grid connection point, the low-frequency oscillation energy is absorbed, and the voltage command reference value of the inverter PWM controller is modified in real time to change the output impedance characteristics of the grid connection point at the electrical level, thereby equivalently changing the output impedance characteristics of the grid connection point at the electrical level. Feed forward power compensation P comp The calculation result is added to the active power reference setting value of the power car power control loop as a direct power adjustment amount, so as to directly offset the power fluctuation caused by the frequency disturbance from the power source.

3. A low voltage generator car non-inductive grid connection control system for implementing the method according to claim 1 or 2, characterized in that The low-voltage power generation vehicle, a control system, and an actuator, The low-voltage power generation vehicle is connected with the control system to communicate with the control system. The control system is connected with the actuator to control the non-inductive grid connection of the low-voltage power generation vehicle.

4. The low voltage traction vehicle non-inductive grid tie control system of claim 3, wherein, The control system includes a Hall current / voltage sensor, a high-speed ADC, an FPGA core controller, a multi-standard physical layer interface, a wireless communication module, an excitation inverter drive interface, and a hardware direct connection tripping path interface. The Hall current / voltage sensor collects real-time current and voltage signals of the power grid, and transmits the analog signals to the FPGA core controller after converting them into digital signals by the high-speed ADC. The FPGA core controller is also connected with the multi-standard physical layer interface, which is connected with the low-voltage power generation vehicle. The FPGA core controller is also connected with the wireless communication module, the excitation inverter drive interface, and the hardware direct connection tripping path interface. The wireless communication module communicates with the low-voltage power generation vehicle. The excitation inverter drive interface and the hardware direct connection tripping path interface are connected with the actuator to realize the control of the actuator by the FPGA core controller.

Images