A gas-electric decoupling feedforward control method and edge collaborative gateway for gas generator sets

Abstract

This invention discloses a gas-electric decoupling feedforward control method and an edge collaborative gateway for gas generator sets, relating to the field of distributed generation control technology. The method acquires gas composition data streams and real-time gas volumetric flow rates from the upstream gas supply network through an edge collaborative gateway. When a step change in gas composition is detected, the remaining time for the abnormal gas mass to reach the cylinder is obtained by solving an integral equation based on the pipeline volume. Within the window period, the expected Wobbe number and expected methane number are extrapolated to generate a combustion strategy bias for the native engine management system, and the target value of the grid active power to be reduced is calculated. At the moment the remaining time reaches zero, the combustion strategy bias and the grid active power target value are simultaneously issued. The aim is to overcome the feedback control lag bottleneck without disrupting the native control system, achieving highly stable operation of the unit under extreme fluctuating gas sources.

Description

Technical Field

[0001] This invention relates to the field of distributed gas-fired power generation and industrial automation control technology, and in particular to a gas-electric decoupling feedforward control method and an edge collaborative gateway for gas generator sets. Background Technology

[0002] With the development of microelectronics technology and lean-burn technology in internal combustion engines, modern gas generator sets are generally equipped with powerful native engine management systems. These systems are internally calibrated with precise static ignition advance angles and throttle pulse patterns for standard gas sources, and rely on pressure, temperature, and oxygen sensors in the intake and exhaust pipes to form closed-loop feedback control logic. In engineering practices utilizing unconventional gas sources such as coal mine methane, ventilation methane, and agricultural biomass gasification syngas for power generation, the gas source composition is extremely unstable due to changes in underground geological stress, fluctuations in extraction negative pressure, or differences in gasification processes. This results in dramatic, second-level changes in the Wobbe number and methane number of the mixed gas entering the cylinder.

[0003] Existing feedback control architectures heavily rely on post-event power drop calculations or exhaust temperature alarms. Due to a lack of mathematical quantification of the upstream pipeline fluid dynamics transmission process, there is a significant time misalignment between the timing of control command issuance and the actual physical arrival time of abnormal gas masses in the cylinder, leading to destructive knocking or misfires already occurring. Furthermore, the underlying code of the native control systems in existing units is typically highly encrypted, forming a technical black box, preventing domestic integrators from modifying the underlying combustion strategy for complex gas sources. Simultaneously, the engine's thermal combustion control and the generator's grid-connected electrical control are disconnected. When faced with harsh gas conditions exceeding physical regulation limits, singular thermal-side regulation is highly susceptible to reverse power disconnection accidents due to the combined effects of sudden power drops and grid electromagnetic resistance.

[0004] Therefore, how to achieve precise feedforward intervention for fluctuating gas sources without damaging the original control system to prevent the unit from disconnecting from the grid has become an urgent technical problem to be solved. Summary of the Invention

[0005] The main objective of this invention is to provide a gas-electric decoupling feedforward control method and an edge collaborative gateway for gas generator sets, aiming to achieve precise feedforward intervention for fluctuating gas sources without disrupting the native control system to prevent the generator set from disconnecting from the grid.

[0006] To achieve the above objectives, this invention proposes a gas-electric decoupling feedforward control method for gas generator sets, which executes the following steps through an edge collaborative gateway independent of the native control system: Acquire the gas composition data stream and real-time gas volume flow rate collected by the front-end sensing unit. The front-end sensing unit is deployed at a distance of a predetermined physical pipe length from the air inlet of the generator set. On the upstream gas supply pipeline; When a step change in the gas composition is detected, mark the current time as the initial trigger time. And according to the physical tube length Determined pipe volume Solve the integral equation (1):

[0007] In the formula, It is a time variable; The remaining time for the abnormal air mass to reach the cylinder of the generator set is obtained. ; In the remaining time During the window period, the expected number of Waubais is extrapolated based on the gas component data stream. and expected methane number And according to the expected number of Wah-pale people The difference between the preset Wobbe number and the expected methane number The difference between the value and the preset methane number generates a combustion strategy bias for the native engine management system. The combustion strategy bias includes an ignition timing bias and an air-fuel ratio correction coefficient. By substituting the current active load of the generator set into the preset explosion-proof limit model, the minimum safe methane number allowed under this load is obtained. If determined Then calculate the target value of the active power of the power grid that needs to be reduced; Real-time monitoring of the remaining time The countdown, in the remaining time At the zero-point, the combustion strategy bias is sent to the native engine management system via the communication interface, and the target value of the grid active power is sent to the native synchronous grid-connected cabinet main control system simultaneously.

[0008] Preferably, the pipe volume The calculation formula (2) is:

[0009] in, The inner diameter is the upstream gas supply pipeline.

[0010] Preferably, the step "obtaining real-time gas volumetric flow rate" It also includes: Obtain the real-time pressure and real-time temperature within the upstream gas supply network; Using the thermodynamic equation of state, combined with the real-time pressure and the real-time temperature, the real-time gas volumetric flow rate is calculated. Density compensation correction is performed to obtain the flow data under standard conditions.

[0011] Preferably, the gas component data stream includes: volume percentage concentrations of methane, hydrogen, and carbon monoxide, respectively.

[0012] Preferably, the step "detecting a step change in the gas composition" specifically means that the rate of change of the concentration of the combustible component in the gas composition exceeds a preset change threshold.

[0013] Preferably, in the remaining time During the window period, the edge collaborative gateway maintains its computing state and does not output intervention commands to the native engine management system and the native synchronous grid-connected cabinet main control system until the remaining time. Reset to zero.

[0014] Preferably, for the native engine management system that does not support bus bias, the step of "issuing the combustion strategy bias amount" specifically means: The edge collaboration gateway outputs an analog voltage signal, which is then connected in series to the sensor input circuit of the native engine management system. This generates a voltage offset that induces the native engine management system to adjust combustion parameters.

[0015] Preferably, after the target value of active power in the power grid is issued, the method further includes a recovery mechanism: Once the upstream gas composition is detected to have returned to the preset range, a new start time is marked. ; Based on the integral equation (1), after the restored gas completely replaces the remaining abnormal gas mass in the pipeline, the combustion strategy bias is canceled, and an instruction to increase the load at a preset rate is issued to the native engine management system and the native synchronous grid-connected cabinet main control system.

[0016] This application also discloses an edge collaboration gateway, including a microprocessor core control board and a peripheral I / O isolation board, wherein the microprocessor in the microprocessor core control board is used to execute the gas-electric decoupling feedforward control method of the gas generator set as described above.

[0017] Preferably, it also includes a sensing input side interface, a native system takeover interface, and a grid-connected collaboration interface, wherein: The sensing input side interface is connected to the front-end sensing unit deployed in the upstream gas supply pipeline network; The native system takeover interface is physically connected to the native engine management system via a CAN bus or analog channel. The grid-connected collaborative interface is physically connected to the native synchronous grid-connected cabinet main control system via industrial Ethernet.

[0018] The above technical solution has the following advantages: By using an edge-coordinated gateway independent of the native control system to monitor upstream gas composition and flow in real time, and employing a fluid dynamic integral model to calculate the precise time delay of abnormal gas masses reaching the cylinder, the time-scale misalignment of traditional feedback control is eliminated, ensuring physical synchronization between the issuance of intervention commands and the entry of abnormal gas masses into the cylinder. The combustion strategy bias generated by combining expected component parameters enables non-intrusive flexible takeover of the closed native management system. Combined with the grid active power target value calculated based on the anti-knock limit model, cross-boundary coordination between thermal-side combustion intervention and electrical-side proactive load reduction is achieved. Without altering the native underlying code, the impact of drastic gas source changes is effectively mitigated, fundamentally curbing generator set knock damage and reverse power grid disconnection risks, and endowing standard units with flexible power generation capabilities that transcend the limitations of a single gas source. Attached Figure Description

[0019] The present invention will now be described in detail with reference to specific embodiments and accompanying drawings, wherein: Figure 1 A schematic diagram of the control system provided in an embodiment of the present invention.

[0020] Figure 2 A flowchart of a feedforward control method provided in an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram illustrating the linkage between control timing and spatial dynamics provided in an embodiment of the present invention.

[0022] Figure 4 This is a hardware interface block diagram of an edge collaboration gateway provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the following description, refer to the appendix. Figure 1 To be continued Figure 4 This description forms part of this document and illustrates, by way of illustration, specific aspects of embodiments of the invention or the context in which embodiments of the invention may be used. It should be understood that other aspects may be utilized and structural or logical changes may be made without departing from the scope of the invention. Therefore, the following detailed description should not be construed as limiting, and the scope of the invention is defined by the appended claims.

[0025] Example 1.

[0026] This embodiment provides a gas-electric decoupling feedforward control method for gas generator sets, designed for unconventional gas sources with drastic fluctuations in composition or calorific value. This method is based on fluid dynamic time delay prediction and cross-boundary coordination between the generator set's native engine management system and the generator grid-connected control system. Executed through an edge-coordinated gateway independent of the generator set's native control system, this method aims to overcome the physical lag bottleneck of existing feedback control without disrupting or rewriting the generator set's native underlying control system. It addresses the problem of poor adaptability of the natively fixed pulse spectrum to drastically changing gas sources, establishing a cross-boundary feedforward defense link that connects engine-level combustion intervention with proactive load reduction by the generator grid. This enables standard gas generator sets to operate with high stability under extreme fluctuating gas sources.

[0027] The physical topology of the hardware system involved in this embodiment mainly consists of a front-end sensing unit, an edge collaborative gateway, and a lower-level execution layer. The front-end sensing unit is deployed on the gas supply pipeline, at a distance of a predetermined physical pipe length L from the generator set's air inlet. The front-end sensing unit specifically includes an online gas analyzer, a pipeline flow meter, and pressure and temperature transmitters. The online gas analyzer collects the volume percentage of each component of the gas in the gas supply pipeline in real time, such as the volume percentage of methane, hydrogen, and carbon monoxide. The pipeline flow meter and pressure and temperature transmitters are used to acquire the flow rate, pressure, and temperature of the gas in the gas supply pipeline in real time. All of the above sensors are physically connected unidirectionally to the input end of the edge collaborative gateway through an industrial isolation safety barrier.

[0028] The edge collaboration gateway is an industrial edge computing hardware independent of the generator set's native control system. The gateway integrates a microprocessor, a real-time operating system, and a memory. The memory pre-programs the feedforward control algorithm of this invention, the volume parameters of the gas supply pipeline (i.e., pipe diameter D and physical pipe length L), and the power explosion-proof limit model calibrated for the target generator set. The microprocessor executes the core algorithms at high speed in a loop, acquiring real-time data streams of gas composition and real-time gas volumetric flow rate from the front-end sensing unit. To eliminate interference from environmental factors, the microprocessor also acquires real-time pressure and temperature within the gas supply pipeline and uses the thermodynamic equation of state to calculate the real-time gas volumetric flow rate. Density compensation correction is performed to obtain the flow data under standard conditions.

[0029] The detailed steps of this feedforward control method are described below.

[0030] Step S100 involves accurate calculation of dynamic time delay based on fluid dynamics integration. The microprocessor reads and monitors the gas component data stream in real time. When a step change in a key combustible component is detected, i.e., the concentration change rate exceeds a preset threshold, the current moment is marked as the initial trigger moment. At this point, the microprocessor determines the pipe volume based on the physical pipe length L. Solve the integral equation. Pipe volume. The calculation formula is

[0031] The integral equation is specifically as follows:

[0032] In the formula, t is the time variable. The calculus algorithm module inside the microprocessor uses... As the lower limit of the points, for traffic Perform real-time definite integral calculations. The accumulated fluid volume equals the preset physical pipe volume. When, the upper limit variable of the integral is solved This refers to the precise remaining time before the abnormal air mass reaches the lower engine cylinder, thus achieving microsecond-level countdown monitoring.

[0033] Step S200 involves advanced simulation of combustion characteristics and generation of nozzle offset. This is done after determining the remaining time. During the waiting window, the edge collaborative gateway remains in a silent calculation state, refraining from outputting any instructions to the lower-level machine to avoid disrupting the current stable combustion. Based on the acquired abnormal component data, the microprocessor uses its built-in thermodynamic formula to calculate the expected Wobbe number of the abnormal gas mass. and expected methane number Next, the microprocessor runs lookup and interpolation algorithms to compare the differences between normal and expected air source parameters, calculating the combustion strategy bias that needs to be applied to the native engine management system (EMS). This combustion strategy bias specifically includes ignition timing bias and air-fuel ratio correction coefficients, such as generating a retarded ignition angle bias coefficient for the native pulse spectrum and a throttle opening compensation ratio. Simultaneously, the microprocessor acquires the generator's current active load in real time. Substituting this into a preset anti-blast limit model, the minimum safe methane number allowed under this load is obtained. If the expected methane number is determined... Less than the minimum safe methane number If the underlying combustion parameters are adjusted, it will inevitably trigger detonation. Based on this, the microprocessor calculates the target value of the active power to be reduced from the power grid. .

[0034] Step S300 is the gas-electric bidirectional decoupling and synchronization transmission triggered by the time delay returning to zero. The real-time high-frequency clock inside the microprocessor continuously updates the remaining time. Count down. In the remaining time... At the absolute physical instant of zeroing, the edge collaborative gateway simultaneously sends commands to both sides via its output interface. For the air circuit soft-connection command, the gateway forcibly overwrites the generated ignition timing offset and air-fuel ratio correction coefficient to the native engine management system (EMS) via the CAN bus. If the native system does not support bus offset, the edge collaborative gateway outputs an analog voltage signal, which is connected in series to the sensor input circuit of the native system, artificially creating a voltage offset to induce the native system to adjust combustion parameters. For the grid decoupling command, the edge collaborative gateway simultaneously sends a command to the native synchronous grid-connected main control system (PLC) to reduce active power. The generator set completes electrical unloading at the same instant the abnormal air mass enters the cylinder, achieving perfect decoupling and smooth transition between mechanical power and electrical resistance, fundamentally preventing generator set reverse power grid disconnection accidents.

[0035] In the aforementioned control process, this embodiment also establishes a safety recovery mechanism to prevent blind operation. Once the upstream gas composition is detected to have returned to a preset normal range, a new start time is marked. The system does not immediately cancel the coordinated defense command. Instead, based on the new physical integral delay, it forcibly waits until high-quality gas completely replaces the residual abnormal gas mass in the pipeline and cylinder. After the delay returns to zero, the edge coordination gateway cancels the combustion strategy bias and issues a command to the synchronous grid-connected cabinet to smoothly ramp up the load at a preset rate, allowing the unit to return to its original high-load operating state. This logic, which follows the laws of physical evolution, ensures that the system has extremely high engineering robustness under harsh operating conditions of high-frequency, irregular gas oscillations.

[0036] Example 2.

[0037] This embodiment further illustrates the specific implementation logic of the edge collaborative gateway for advanced combustion characteristic extrapolation and mechanical limit determination. This logic is mainly for unconventional gas sources with drastic component fluctuations, such as biomass gasification syngas or multi-source blended gases containing multiple combustible components such as methane, hydrogen, and carbon monoxide.

[0038] The online gas analyzer in the front-end sensing unit continuously acquires a data stream of gas composition, specifically including the volume percentage concentrations of methane, hydrogen, and carbon monoxide. The microprocessor calculates the physical properties of the current gas mass in real time. In step S201, based on the acquired abnormal component data, the microprocessor uses a built-in thermodynamic formula to deduce the expected Wobbe number. and expected methane number The expected number of methanes This directly reflects the gas mass's anti-knock performance. Because hydrogen has an extremely high combustion rate and is prone to pre-ignition, if the proportion of hydrogen in the fuel gas suddenly increases dramatically, the expected methane count will... It will decrease significantly.

[0039] In step S202, the microprocessor establishes a multidimensional bias matrix. This matrix stores combustion intervention parameters corresponding to different methane and Wobbe number deviations. The microprocessor generates combustion strategy biases for the native engine management system (EMS) through table lookup and interpolation algorithms. These biases include not only an offset coefficient for ignition timing but also an air-fuel ratio correction coefficient, aiming to mitigate severe combustion fluctuations that may be caused by poor-quality air masses by delaying ignition timing or adjusting the air-fuel ratio.

[0040] In step S203, the microprocessor introduces a power explosion resistance limit model. This model pre-stores a nonlinear mapping relationship between different load points and the safe methane number. The microprocessor acquires the generator's current active load in real time. Substituting this into the model, we can derive the minimum safe methane number allowed under this load. When judged This means that even if the ignition timing is delayed to the mechanical safety boundary, the engine still cannot smoothly handle the poor-quality gas mass under the current load. At this point, the microprocessor calculates the target value of the active power to be reduced from the grid. It is also prepared to initiate an active load reduction command to the generator side the instant the time delay returns to zero. This cross-physical domain collaborative determination effectively solves the problem in traditional solutions where unilaterally closing the gas valve causes the unit to be dragged down by the electromagnetic resistance of the grid, resulting in reverse power disconnection.

[0041] Example 3.

[0042] This embodiment details the hardware interface and signal interaction mechanism of the edge collaborative gateway to achieve non-intrusive soft takeover. To ensure compatibility with gas generator sets from different technological backgrounds, the edge collaborative gateway is configured with multiple link output channels.

[0043] For high-end native systems with standard bus interfaces, the edge collaboration gateway includes a CAN 2.0B bus interface with error checking mechanisms. (The remaining time...) At the absolute physical moment of zeroing, the microprocessor sends an offset data frame containing a specific parameter group number (PGN) to the native engine management system (EMS) via the J1939 protocol. Upon receiving this high-priority external offset signal, the EMS adds the offset to its existing calculation logic, thereby achieving adaptive reconfiguration of the ignition advance angle and intake volume.

[0044] For native Engine Management Systems (EMS) that do not support communication buses or are in a low-level encrypted state, this embodiment provides a signal-biased soft-switch solution. The peripheral I / O isolation board of the edge collaboration gateway includes a high-precision analog voltage output channel. This channel is physically connected in series to the sensor circuit of the native EMS, such as the intake pressure sensor signal line, via signal lines. In the remaining time At the instant of zeroing, the edge collaborative gateway outputs a specific analog voltage signal, using the superposition principle to create a voltage offset. By simulating extremely high intake pressure, it induces the native engine management system (EMS) to automatically retard the ignition timing and reduce the throttle position. Alternatively, as an alternative, the edge collaborative gateway can also inject a simulated high-frequency AC signal into the knock sensor input to prematurely trigger the native system's built-in knock protection program.

[0045] On the electrical coordination side, the edge coordination gateway runs the Modbus TCP protocol via a standard RJ45 port and is physically connected to the PLC of the native grid-connected cabinet's main control system. At the instant the latency reaches zero, the gateway writes the load limit value to the grid-connected cabinet's register, achieving seamless integration. The gas generator set completes electrical unloading at the same moment the cylinder ingests a low-quality gas mass, achieving precise matching between mechanical power and electrical resistance.

[0046] Example 4.

[0047] This embodiment mainly illustrates the silent computation logic and security recovery mechanism adopted in this invention.

[0048] The remaining time calculated in step S103 During the window period, the edge collaborative gateway remains in a background computing state and is strictly prohibited from issuing any physical intervention commands to the lower-level actuators. This silent calculation phase is designed based on the physical laws of fluid transport; before the countdown reaches zero, the gas entering the cylinder remains the stable combustion gas from before the sensing moment. If intervention occurs prematurely at this point, it will disrupt the current combustion balance, causing violent fluctuations in the unit's speed.

[0049] This invention also includes a safety recovery mechanism based on spatiotemporal evolution. When the upstream sensing unit of the gas supply pipeline detects that the gas composition has recovered to a preset optimal range, this moment is marked as the start time. At this point, the system does not immediately release the load, but instead initiates a new round of integral delay calculation. The microprocessor calculates this based on the current real-time gas volumetric flow rate. The absolute time it takes for the high-quality gas mass to reach the cylinder is calculated. The edge coordination gateway only cancels the combustion strategy bias when it is determined that the abnormal gas mass remaining inside the gas supply pipeline has been completely replaced and expelled—that is, the physical instant when the high-quality gas truly enters the cylinder. Subsequently, the edge coordination gateway sends a smooth ramp-up command to the PLC of the native synchronous grid-connected cabinet's main control system. This command includes a preset load ramp rate, for example, at a rate of [per second]. The load is gradually increased until it returns to full load. This recovery logic effectively avoids the risk of secondary knocking caused by blindly increasing the load under conditions of frequent gas supply fluctuations.

[0050] Example 5.

[0051] This embodiment describes the internal hardware configuration of the edge collaboration gateway.

[0052] The edge collaboration gateway adopts an industrial-grade split architecture. Its core component is a microprocessor-based control board, equipped with a high-performance computing chip, such as an ARM architecture or NXP series processor, used to run fluid dynamic integration algorithms and limit discrimination models. The microprocessor-based control board connects to an external I / O isolation board via an internal bus. The external I / O isolation board features opto-isolation and includes multiple 4-20mA analog signal acquisition terminals for connecting to front-end sensing units. Furthermore, the gateway integrates an isolated power management subsystem, supporting a wide voltage input of AC 220V or DC 24V to ensure power supply stability in the complex electromagnetic environment of power plants. The gateway housing uses a standard DIN rail mounting design for easy deployment inside the gas generator control cabinet.

[0053] Example 6.

[0054] This embodiment further illustrates various equivalent replacement schemes for edge collaborative gateways in terms of algorithm prediction, front-end perception, execution takeover, and application scenarios.

[0055] Regarding methods for measuring fluid dynamic time delay, in addition to the aforementioned calculus model based on flow rate and volume, microprocessors can also employ machine learning prediction models based on time-series data. Specifically, lightweight long short-term memory networks (LSTM) or one-dimensional convolutional neural networks (1D-CNN) are deployed in edge collaborative gateways, using historical flow rate, pressure, and concentration sequences within the air supply pipeline as input features to directly output a countdown prediction value for the arrival of the abnormal gas mass at the cylinder inlet. Furthermore, time delay measurement can also be based on pressure wave or acoustic wave tracing technology. By capturing the propagation phase difference of pressure waves in variable-density fluids, the actual migration velocity of the abnormal gas mass can be inferred and calibrated, thereby determining the remaining time. .

[0056] Regarding hardware replacements for the front-end sensing unit, in addition to online gas analyzers, rapid Wobbe number analyzers or calorimeters can be used, directly using the step drop rate of calorific value or Wobbe number as the initial signal to trigger feedforward calculations. Alternatively, tunable semiconductor laser absorption spectroscopy (TDLAS) sensors can be used, installed on both sides of the pipeline to capture millisecond-level concentration abrupt changes.

[0057] Regarding the signal interface replacement for non-intrusive soft-pipe systems, in addition to the aforementioned methods, the edge collaboration gateway can also be connected in series to the intake air temperature sensor or exhaust oxygen sensor signal circuit of the native engine management system (EMS). At the instant the delay returns to zero, by outputting a spurious voltage signal of extremely high intake air temperature or extremely rich oxygen, the native system's built-in compensation logic induces it to automatically retard the ignition advance angle and close the throttle.

[0058] In cross-domain decoupling alternatives at the execution level, if the grid connection protocol is stringent, the load reduction command from the edge collaborative gateway is not directly issued to a single unit. Instead, it is uploaded via industrial Ethernet to the automatic generation control system (AGC) or microgrid energy management system (EMS) of the entire power station. The upper-level system coordinates the load increase of other units within the station at the instant the delay reaches zero, while unloading the affected units to maintain a constant total power. Alternatively, at time T1, a high-power resistive dummy load connected in parallel with the generator is rapidly introduced via thyristors, causing the electrical energy generated by the generator to be instantly consumed by the dummy load, effectively removing the mechanical resistance of the generator while maintaining grid connection.

[0059] The edge collaboration gateway and control method provided by this invention are widely applicable to unconventional gas source power generation scenarios such as coal mine gas extraction, ventilation gas extraction, agricultural biomass gasification syngas, coke oven gas, and natural gas hydrogen blending pipelines. By adding this gateway, domestically produced basic gas generator sets or imported bare-metal equipment can be given flexible power generation capabilities that transcend the limitations of a single gas source, achieving non-intrusive intelligent upgrades to the underlying control performance of generator sets of different brands and models.

[0060] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A gas-electric decoupling feedforward control method for a gas generator set, characterized in that, Perform the following steps via an edge collaboration gateway that is independent of the native control system: Acquire the gas composition data stream and real-time gas volume flow rate collected by the front-end sensing unit. The front-end sensing unit is deployed at a distance of a predetermined physical pipe length from the air inlet of the generator set. On the upstream gas supply pipeline; When a step change in the gas composition is detected, mark the current time as the initial trigger time. And according to the physical tube length Determined pipe volume Solve the integral equation (1): In the formula, It is a time variable; The remaining time for the abnormal air mass to reach the cylinder of the generator set is obtained. ; In the remaining time During the window period, the expected number of Waubais is extrapolated based on the gas component data stream. and expected methane number And according to the expected number of Wah-pale people The difference between the preset Wobbe number and the expected methane number The difference between the value and the preset methane number generates a combustion strategy bias for the native engine management system. The combustion strategy bias includes an ignition timing bias and an air-fuel ratio correction coefficient. By substituting the current active load of the generator set into the preset explosion-proof limit model, the minimum safe methane number allowed under this load is obtained. If determined Then calculate the target value of the active power of the power grid that needs to be reduced; Real-time monitoring of the remaining time The countdown, in the remaining time At the zero-point, the combustion strategy bias is sent to the native engine management system via the communication interface, and the target value of the grid active power is sent to the native synchronous grid-connected cabinet main control system simultaneously.

2. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, The pipe volume The calculation formula (2) is: in, The inner diameter is the upstream gas supply pipeline.

3. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, The step "obtaining real-time gas volume flow rate" It also includes: Obtain the real-time pressure and real-time temperature within the upstream gas supply network; Using the thermodynamic equation of state, combined with the real-time pressure and the real-time temperature, the real-time gas volumetric flow rate is calculated. Density compensation correction is performed to obtain the flow data under standard conditions.

4. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, The gas composition data stream includes the volume percentage concentrations of methane, hydrogen, and carbon monoxide, respectively.

5. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, The step "detecting a step change in the gas composition" specifically means that the rate of change of the concentration of the combustible component in the gas composition exceeds a preset change threshold.

6. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, In the remaining time During the window period, the edge collaborative gateway maintains its computing state and does not output intervention commands to the native engine management system and the native synchronous grid-connected cabinet main control system until the remaining time. Reset to zero.

7. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, For the native engine management system that does not support bus bias, the step of "issuing the combustion strategy bias amount" specifically means: The edge collaboration gateway outputs an analog voltage signal, which is then connected in series to the sensor input circuit of the native engine management system. This generates a voltage offset that induces the native engine management system to adjust combustion parameters.

8. The gas-electric decoupling feedforward control method for gas generator sets according to claim 1, characterized in that, After the target value for active power of the power grid is issued, the method further includes a recovery mechanism: Once the upstream gas composition is detected to have returned to the preset range, a new start time is marked. ; Based on the integral equation (1), after the restored gas completely replaces the remaining abnormal gas mass in the pipeline, the combustion strategy bias is canceled, and an instruction to increase the load at a preset rate is issued to the native engine management system and the native synchronous grid-connected cabinet main control system.

9. An edge collaboration gateway, characterized in that, It includes a microprocessor core control board and a peripheral I / O isolation board, wherein the microprocessor in the microprocessor core control board is used to execute the gas-electric decoupling feedforward control method of the gas generator set as described in any one of claims 1 to 8.

10. The edge collaboration gateway according to claim 9, characterized in that, It also includes the sensing input side interface, the native system takeover interface, and the grid-connected collaboration interface, among which: The sensing input side interface is connected to the front-end sensing unit deployed in the upstream gas supply pipeline network; The native system takeover interface is physically connected to the native engine management system via a CAN bus or analog channel. The grid-connected collaborative interface is physically connected to the native synchronous grid-connected cabinet main control system via industrial Ethernet.

Images