Hydrogen gas turbine combustion mode switching and control method, device, equipment and medium

Abstract

The embodiment of the invention discloses a hydrogen gas turbine combustion mode switching and control method, device and equipment and a medium. A specific embodiment of the method comprises the following steps: executing ignition operation in response to a situation that an acquired rotating speed signal reaches a first rotating speed threshold value; in response to the rotating speed signal reaching a second rotating speed threshold value, a first-stage fuel stop valve is controlled to introduce first-stage premixed fuel into the combustion chamber; the first adjusting valve is adjusted based on the flame temperature signal; in response to the rotating speed signal reaching a third rotating speed threshold value, a second-stage fuel stop valve and a second regulating valve are regulated and controlled; in response to the identified stable operation condition confirmation signal, controlling the on-duty acceleration level fuel stop valve to be closed; and in response to the flame temperature identification result, representing that the flame temperature is stable, generating a full premixed combustion mode confirmation signal. According to the implementation mode, switching of the combustion modes of the hydrogen gas turbine is achieved, and key technical support is provided for low-emission and high-stability operation of the hydrogen gas turbine within the wide working condition range.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and specifically to a method, apparatus, device, and medium for switching and controlling combustion modes in hydrogen gas turbines. Background Technology

[0002] As hydrogen gas turbines develop towards higher efficiency and lower emissions, staged combustion has become the mainstream technology for achieving stable combustion under wide operating conditions and low NOx emissions. During startup, acceleration, and low-load phases, diffusion combustion in a standby acceleration stage is typically used to ensure ignition reliability and combustion stability. Under medium- to high-load conditions, a premixed combustion mode is switched to reduce NOx emissions. Against this backdrop, existing technologies have developed control methods based on speed thresholds to trigger the introduction of fuel at each stage. This involves sequentially opening the first-stage and second-stage fuel shut-off valves at preset speed nodes, achieving a gradual transition from diffusion combustion to premixed combustion.

[0003] However, the above methods often present the following technical problems: First, after the primary premixed fuel is introduced, the flame characteristics in the combustion chamber change drastically. Existing technologies rely solely on indirect parameters such as engine speed thresholds or exhaust temperature for judgment, lacking direct monitoring and identification methods for the transient process of combustion mode switching. This makes it impossible to accurately determine whether primary premixed combustion has stabilized. If the switching process is not complete before acceleration continues, it may lead to combustion oscillations or even flameout. Second, after the standby accelerator fuel is shut off, existing technologies cannot promptly confirm whether combustion has fully transitioned to the fully premixed mode. Improper shut-off timing not only affects emission performance but may also trigger backfire risks. More importantly, the regulation of each stage of fuel in existing technologies is mostly open-loop control. When abnormal fluctuations occur in combustion, there is a lack of effective closed-loop adjustment mechanisms, making it difficult to adapt to the characteristics of hydrogen fuel's fast flame propagation speed and high reactivity. Therefore, there is an urgent need for a combustion mode switching and control method that can identify the switching status in real time based on directly measured parameters in the combustion chamber and actively adjust the fuel supply.

[0004] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of this disclosure provide methods, apparatus, electronic devices, and computer-readable media for switching and controlling combustion modes in hydrogen gas turbines to address one or more of the technical problems mentioned in the background section above.

[0007] In a first aspect, some embodiments of this disclosure provide a method for switching and controlling the combustion mode of a hydrogen gas turbine, including: real-time acquisition of the rotational speed signal of the hydrogen gas turbine rotor; ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold; and the following steps in response to the acquired rotational speed signal reaching a second rotational speed threshold: controlling a first-stage fuel shut-off valve to introduce first-stage premixed fuel into the combustion chamber; generating a flame temperature change rate signal based on the acquired flame temperature signal, and performing fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result; adjusting a first regulating valve in response to the obtained fluctuation identification result indicating that the flame temperature exhibits a characteristic fluctuation of first rising and then falling; and adjusting a second-stage fuel shut-off valve and a second regulating valve in response to the acquired rotational speed signal reaching a third rotational speed threshold; closing a shift acceleration stage fuel shut-off valve in response to the stable operating condition confirmation signal identified after adjusting the second-stage fuel shut-off valve and the second regulating valve; flame temperature identification on the acquired flame temperature decrease signal to obtain a flame temperature identification result; and generating a fully premixed combustion mode confirmation signal in response to the obtained flame temperature identification result indicating that the flame temperature is stable.

[0008] Secondly, some embodiments of this disclosure provide a data acquisition unit configured to acquire the rotational speed signal of a hydrogen gas turbine rotor in real time; an ignition unit configured to perform an ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold; and a first execution unit configured to perform the following steps in response to the acquired rotational speed signal reaching a second rotational speed threshold: controlling a first-stage fuel shut-off valve to introduce first-stage premixed fuel into the combustion chamber; generating a flame temperature change rate signal based on the acquired flame temperature signal, and performing fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result; and characterizing the obtained fluctuation identification result. When the flame temperature exhibits a characteristic fluctuation of first rising and then falling, the first regulating valve is adjusted. The second actuator is configured to, in response to the collected speed signal reaching a third speed threshold, perform the following steps: regulate the secondary fuel shut-off valve and the second regulating valve; in response to the identification of a stable operating condition confirmation signal after regulating the secondary fuel shut-off valve and the second regulating valve, control the duty acceleration stage fuel shut-off valve to close; perform flame temperature identification on the collected flame temperature drop signal to obtain a flame temperature identification result; in response to the obtained flame temperature identification result indicating that the flame temperature is stable, generate a fully premixed combustion mode confirmation signal.

[0009] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0010] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0011] Fifthly, some embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0012] The embodiments disclosed herein have the following beneficial effects: The hydrogen gas turbine combustion mode switching and control methods of some embodiments of this disclosure can achieve accurate identification and stable control of the combustion mode switching process, significantly improving the reliability and operational stability of the hydrogen gas turbine during the staged combustion transition phase. Specifically, traditional hydrogen gas turbine combustion mode switching technologies (such as existing schemes that rely on speed thresholds or exhaust temperature for judgment) may face problems such as the inability to accurately determine whether premixed combustion has been stably established after the first-stage premixed fuel is introduced, the lack of multi-dimensional verification when the shift acceleration stage fuel is shut off, and the lack of effective confirmation methods after the full premixed mode is established. If only a single parameter judgment or open-loop control is relied upon, there may be risks of combustion oscillation, flameout, or even backfire caused by improper switching timing, ultimately leading to the unit's inability to operate safely and stably. Based on this, the hydrogen gas turbine combustion mode switching and control methods of some embodiments of this disclosure firstly acquire the rotational speed signal of the hydrogen gas turbine rotor in real time, and then execute an ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold. This provides a clear rotational speed reference for subsequent staged fuel introduction, ensuring the safety and reliability of the ignition phase. Then, in response to the collected rotational speed signal reaching the second rotational speed threshold, the primary fuel shut-off valve is controlled to supply primary premixed fuel to the combustion chamber. A flame temperature change rate signal is generated based on the collected flame temperature signal, and fluctuation identification is performed on this signal to obtain the fluctuation identification result. Thus, by directly monitoring the flame temperature change in the combustion chamber, the dynamic response of the combustion state after the primary premixed fuel is introduced can be captured in real time, providing a direct physical basis for subsequent judgments. Next, in response to the obtained fluctuation identification result indicating a characteristic fluctuation of flame temperature that first rises and then falls, the first regulating valve is adjusted. Therefore, using the "rise then fall" characteristic fluctuation as a marker of stable premixed combustion, the first regulating valve is immediately closed-loop regulated upon identification of this characteristic fluctuation, ensuring rapid flame temperature stabilization and avoiding the risk of combustion oscillation or flameout due to inaccurate switching judgments. Then, in response to the collected rotational speed signal reaching the third rotational speed threshold, the secondary fuel shut-off valve and the second regulating valve are controlled. This enables on-demand introduction of secondary premixed fuel, providing a fuel base for full-speed no-load and loaded operation. Next, in response to the stable operating condition confirmation signal detected after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is closed. Thus, by confirming a stable operating condition through multi-parameter fusion judgment (speed, fuel flow, exhaust temperature) before shutting off the shift fuel, combustion is ensured to be in a steady state upon shift fuel withdrawal, achieving uninterrupted shift fuel withdrawal. Then, the collected flame temperature drop signal is analyzed for flame temperature identification, yielding the flame temperature identification result. Therefore, by directly monitoring the flame temperature change after shift fuel shutdown, real-time feedback is provided for establishing the full premixed mode.Finally, in response to the flame temperature identification result indicating that the flame temperature is stable, a confirmation signal for the fully premixed combustion mode is generated. Thus, by using a dual-condition judgment method—calculating the average temperature and standard deviation separately through sliding window segmentation—both the absolute temperature level meets the target and the convergence of temperature fluctuations is guaranteed, ultimately confirming that the fully premixed combustion mode has been stably established. Furthermore, because this method introduces a real-time identification and closed-loop adjustment mechanism based on directly measured parameters in the combustion chamber throughout the entire combustion mode switching process (first-stage input, second-stage input, shift shutdown, and full premix confirmation), it can effectively adapt to the characteristics of rapid flame propagation and high reactivity of hydrogen fuel, and possesses inherent suppression capabilities for transient disturbances during the switching process, thereby enhancing robustness and reliability under real-world complex operating conditions. Simultaneously, through the synergistic optimization of staged fuel input, characteristic fluctuation identification, multi-parameter stable operating condition judgment, and temperature drop confirmation, the entire switching process can be kept consistent with the final fully premixed operation target, ensuring combustion stability across the entire operating range. Thus, by combining characteristic fluctuation identification, closed-loop regulation, multi-parameter fusion judgment and temperature drop confirmation, the overall reliability, stability and safety of hydrogen gas turbine combustion mode switching process are improved to a certain extent, providing key technical support for hydrogen gas turbine to achieve low emission and high stability operation in a wide range of operating conditions. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0014] Figure 1 This is a flowchart of some embodiments of the hydrogen gas turbine combustion mode switching and control method according to the present disclosure; Figure 2 This is a schematic diagram of the structure of some embodiments of the hydrogen gas turbine combustion mode switching and control device according to the present disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure; Figure 4 This is an internal test image of the fuel control valve of a hydrogen gas turbine according to this disclosure. Detailed Implementation

[0015] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0016] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0017] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0018] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0019] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0020] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Figure 1 A flowchart 100 is shown, illustrating some embodiments of a hydrogen gas turbine combustion mode switching and control method according to the present disclosure. This hydrogen gas turbine combustion mode switching and control method includes the following steps: Step 101: Real-time acquisition of the rotational speed signal of the hydrogen gas turbine rotor.

[0022] In some embodiments, the actuator of the hydrogen gas turbine combustion mode switching and control method (e.g., the gas turbine control system) can acquire the rotational speed signal of the hydrogen gas turbine rotor in real time. The rotational speed signal is a physical quantity characterizing the rotational frequency of the gas turbine rotor, typically measured in revolutions per minute (rpm). It is a core parameter for determining the gas turbine's operating status, implementing staged fuel injection, and controlling rotational speed. In practice, the actuator can acquire the rotor's rotational speed signal in real time using an eddy current sensor installed on the rotor journal. This eddy current sensor utilizes the principle of electromagnetic induction, detecting changes in the gap between the rotor's tooth tip and the sensor probe during rotor rotation, and outputting a pulse frequency signal proportional to the rotational speed. The control system (e.g., a programmable logic controller or distributed control system) receives this pulse signal and converts it into an instantaneous rotational speed value expressed in engineering units (rpm) through frequency counting or period measurement. For example, during unit operation, the eddy current sensor continuously detects the rotor speed. The control system reads the pulse signal output by the sensor at a preset sampling period (such as 10 milliseconds). After signal conditioning and analog-to-digital conversion, the instantaneous speed value is updated in real time, providing basic data for subsequent speed judgment and fuel control.

[0023] Step 102: In response to the collected speed signal reaching the first speed threshold, ignition operation is performed.

[0024] In some embodiments, the aforementioned actuator can perform an ignition operation in response to the collected speed signal reaching a first speed threshold. The first speed threshold is a pre-set critical speed value used to trigger the ignition program, typically measured in revolutions per minute (rpm). This threshold is determined comprehensively based on the structural characteristics of the gas turbine, the starter motor's driving capacity, and ignition conditions; for example, it can be set as a specific percentage of the rated speed (e.g., 19.35% TNH). The ignition operation refers to a series of control actions that establish a stable flame in the combustion chamber by controlling the fuel supply and ignition device. In practice, the aforementioned actuator monitors the speed signal in real time. When the speed signal rises to the first speed threshold, it generates an ignition start command. Based on this ignition start command, the control system first opens the standby acceleration stage fuel shut-off valve and controls the first regulating valve to open to a first preset opening degree. The standby accelerator stage fuel shut-off valve is a shut-off valve installed on the standby accelerator stage fuel pipeline. It controls the supply and disconnection of fuel to the standby accelerator stage and is typically driven by a pneumatic or electric actuator. It has a fast response characteristic. After the valve opens during the ignition phase, hydrogen fuel enters the main nozzle of the combustion chamber through the standby accelerator stage pipeline. The first regulating valve is a flow regulating valve installed on the main fuel pipeline or the first-stage fuel pipeline. It controls the fuel flow rate entering the combustion chamber and is typically driven by an electric actuator. It can continuously adjust the valve opening according to control commands, thereby changing the fuel supply flow rate. The first preset opening degree refers to the preset opening value of the first regulating valve during the ignition phase, expressed as a percentage (%). It is used to provide a fuel flow rate suitable for ignition conditions, for example, it can be set to a valve opening degree of 8% to 10%. At the same time, the igniter is energized to generate an electric spark, igniting the hydrogen-air mixture in the combustion chamber. After successful ignition, the control system monitors the flame temperature signal through a thermocouple installed at the main nozzle outlet to confirm flame establishment and maintains the first regulating valve at the first preset opening degree (e.g., 8%) to execute the warm-up procedure. For example, during the gas turbine startup process, the starter motor increases the unit speed to the first speed threshold (e.g., 999 rpm). After the control system detects this speed signal, it immediately opens the shift acceleration stage fuel shut-off valve and sets the opening of the first regulating valve to 10%. After 2 seconds, it reduces the opening to 8%. At the same time, the igniter discharges, igniting the hydrogen-air mixture in the combustion chamber. After the control system monitors the rise and stabilization of the flame temperature through the thermocouple, it confirms successful ignition and then maintains the opening of the first regulating valve at 8% for 60 seconds to warm up the unit.

[0025] Step 103: In response to the acquired rotational speed signal reaching the second rotational speed threshold, the following steps are executed: Step 1031: Control the primary fuel shut-off valve to introduce primary premixed fuel into the combustion chamber.

[0026] In some embodiments, the aforementioned actuator can control the primary fuel shut-off valve to supply primary premixed fuel to the combustion chamber in response to the collected speed signal reaching a second speed threshold. The second speed threshold is a pre-set critical speed value used to trigger the introduction of primary premixed fuel, typically measured in revolutions per minute (rpm). This threshold is determined comprehensively based on the combustion chamber characteristics, fuel staging strategy, and turbine heat load capacity of the gas turbine; for example, it can be set as a specific percentage of the rated speed (e.g., 60% TNH). The primary fuel shut-off valve is a shut-off valve installed on the primary fuel pipeline, used to control the supply and disconnection of primary premixed fuel. It is typically driven by a pneumatic or electric actuator and has a fast response characteristic. When the valve is open, hydrogen fuel enters the main nozzle of the combustion chamber through the primary fuel pipeline, where it is premixed with compressed air discharged from the compressor in a micro-premixing pipe before being injected into the combustion zone. Primary premixed fuel refers to the fuel introduced into the combustion chamber. This fuel is fully premixed with air before entering the combustion chamber to form a homogeneous hydrogen-air mixture. Its equivalence ratio is typically designed in the lean premixed combustion range to reduce combustion temperature and nitrogen oxide formation. In practice, the aforementioned actuator monitors the engine speed signal in real time. When the engine speed signal rises to the second speed threshold, a primary fuel injection command is generated. Based on this command, the control system opens the primary fuel shut-off valve, allowing hydrogen fuel to flow through the primary fuel pipeline into the micro-premixing tube of the main nozzle. There, it rapidly mixes with compressed air from the compressor to form a homogeneous premixed combustible mixture. This mixture is then injected into the combustion chamber through the micro-premixing tube outlet, where it achieves stable combustion under the ignition of the booster flame. For example, during the acceleration of a gas turbine, when the control system detects that the speed signal has reached the second speed threshold (such as 3201 rpm), it immediately generates a first-stage fuel injection command, opens the first-stage fuel shut-off valve, and introduces the first-stage premixed fuel into the combustion chamber. At this time, the combustion mode gradually transitions from simple on-duty acceleration stage diffusion combustion to a "diffusion + premixed" mixed combustion mode, providing a fuel base for subsequent acceleration to full-speed no-load conditions.

[0027] In addressing the stability issues of combustion mode switching and the reliability issues of full premixing establishment in the aforementioned background technology using the aforementioned hydrogen gas turbine combustion mode switching and control methods, the following technical problem arises regarding the application scenario: After the hydrogen gas turbine successfully completes combustion mode switching and enters a stable operation phase of full-speed no-load or load power generation, the fuel supply system needs to continuously supply hydrogen to the combustion chamber to maintain unit operation. At this time, the fuel manifold pressure is a key parameter reflecting whether the fuel supply is sufficient. However, the following technical problem often arises: When the fuel supply system (such as a hydrogen tanker or hydrogen storage tank) has limited fuel reserves, as the unit continues to operate and fuel is continuously consumed, the fuel manifold pressure shows a continuous downward trend. If the pressure drops rapidly, it may quickly drop below the trip threshold without the operator's notice, triggering an emergency shutdown and causing unplanned unit shutdown, affecting the completion of the power generation task. Current technology lacks the ability to predict fuel pressure drop trends, relying solely on a single pressure threshold. When the pressure falls below the trip threshold, it directly triggers an emergency shutdown. It cannot provide early warnings before the pressure drops below the threshold, cannot predict remaining available operating time, and cannot proactively implement load reduction measures to slow the pressure drop. This results in a lack of operational window for operators, and after an emergency shutdown, there is a lack of pipeline purging synchronized with fuel cutoff, potentially leading to the accumulation of residual hydrogen and subsequent deflagration. To address the following requirements for this application scenario: during fuel pressure drop, it is necessary to quantify the pressure drop trend in real time, predict remaining available operating time, proactively issue warnings and reduce fuel flow to slow the pressure drop before it falls below the trip threshold, providing operators with time to respond or achieving a smooth shutdown. Simultaneously, when the pressure reaches the trip threshold, it must simultaneously execute emergency cutoff and purging, forming a complete chain of tiered protection from warning and proactive load reduction to emergency shutdown, ensuring operational safety under fuel shortage conditions. Therefore, we have decided to adopt the following solution: Optionally, the aforementioned implementing entity may also perform the following steps: The first step involves controlling the primary fuel shut-off valve to introduce primary premixed fuel into the combustion chamber and simultaneously acquiring the fuel main pressure signal in real time. The fuel main pressure signal refers to the pressure value acquired in real time by an absolute pressure transmitter installed on the hydrogen fuel main pipeline, typically expressed in megapascals (MPa). It reflects the pressure level of the fuel supply system and is a key parameter for determining whether the fuel supply is sufficient. In practice, after the primary fuel shut-off valve opens, the aforementioned actuator continuously acquires the fuel main pressure signal at a preset sampling frequency (e.g., 10Hz) to provide basic data for subsequent monitoring and judgment. For example, during gas turbine operation under load, the control system reads the fuel main pressure every 0.1 seconds and records it as P_fuel = 1.8 MPa.

[0028] The second step involves generating an emergency shutdown command in response to the current fuel main pressure falling below a preset low-pressure trip threshold. The low-pressure trip threshold is a pre-set lower limit of the fuel main pressure used to trigger an emergency shutdown, typically measured in megapascals (MPa), such as 1.25 MPa. When the fuel pressure falls below this threshold, it indicates a severe fuel shortage; continued operation may lead to backfire or flameout, necessitating immediate shutdown. The emergency shutdown command is the control signal used to execute the emergency shutdown procedure, including closing all fuel shut-off valves and initiating purging. In practice, the actuator compares the real-time fuel main pressure with the low-pressure trip threshold. When the fuel main pressure falls below the threshold, an emergency shutdown command is immediately generated. For example, when the fuel main pressure drops to 1.24 MPa, below the preset trip threshold of 1.25 MPa, the control system generates an emergency shutdown command.

[0029] The third step, based on the aforementioned emergency shutdown command, involves simultaneously closing all fuel shut-off valves and concurrently performing an emergency purging operation on the fuel lines. All fuel shut-off valves refer to the duty accelerator stage fuel shut-off valve, the primary fuel shut-off valve, and the secondary fuel shut-off valve. During an emergency shutdown, all three valves must be closed simultaneously to completely cut off the fuel supply. The emergency purging operation involves introducing an inert gas (such as nitrogen) into the fuel lines to purge any residual hydrogen, preventing hydrogen accumulation and potential deflagration. In practice, the aforementioned executing entity sends the emergency shutdown command to the actuators of all fuel shut-off valves, which simultaneously close the valves. Simultaneously, the purging system opens the purging valves, introducing nitrogen into the fuel lines for a preset time (e.g., 30 seconds) to ensure that any residual hydrogen in the lines is completely purged. For example, after the emergency shutdown command is generated, the control system immediately sends a closing signal to all three fuel shut-off valves simultaneously. All valves close within 0.5 seconds, and the nitrogen purging valves open simultaneously, remaining open for 30 seconds before closing.

[0030] The fourth step involves generating time-series pressure data on fuel pressure variations over time, based on the collected fuel manifold pressure signal. This time-series data refers to a sequence of fuel manifold pressure sampling values ​​arranged chronologically, with each sampling point including a timestamp and the corresponding pressure value. In practice, the aforementioned actuator continuously records fuel manifold pressure sampling values ​​during operation, forming continuous time-series pressure data to provide a basis for subsequent calculations of the pressure drop rate. For example, the control system records the pressure value every 0.1 seconds within 30 seconds, forming a time-series pressure data sequence containing 300 sampling points.

[0031] The fifth step involves generating the fuel pressure drop rate based on the aforementioned pressure time-series data. The fuel pressure drop rate refers to the first derivative of the fuel manifold pressure with respect to time, characterizing how quickly the pressure decreases over time, typically expressed in megapascals per second (MPa / s). In practice, the executing entity performs a first-order difference calculation on the pressure time-series data: for the current time t, the drop rate = (P(t) - P(t-Δt)) / Δt, where P(t) is the current pressure, P(t-Δt) is the pressure before time Δt, and Δt is the time interval. For example, if the pressure drops from 1.8 MPa to 1.7 MPa in 10 seconds, the drop rate is (1.7-1.8) / 10 = -0.01 MPa / s (the negative sign indicates a decrease).

[0032] Step 6: Based on the current fuel manifold pressure, the aforementioned fuel pressure drop rate, and the preset low-pressure trip threshold, generate the remaining available operating time. The remaining available operating time refers to the time required for the pressure to drop to the low-pressure trip threshold from the current moment, assuming the current pressure drop rate remains constant; it is typically measured in seconds (s). In practice, the executing entity calculates the remaining available operating time using the following formula: t_remaining = (P_current - P_trip) / |dP / dt|, where P_current is the current fuel manifold pressure, P_trip is the low-pressure trip threshold, and dP / dt is the pressure drop rate (absolute value). For example, if the current pressure is 1.6 MPa, the trip threshold is 1.25 MPa, and the pressure drop rate is 0.01 MPa / s, then the remaining available operating time is (1.6 - 1.25) / 0.01 = 35 seconds.

[0033] Step 7: In response to the generated remaining available operating time being less than a preset warning time and the current fuel main pressure not being less than the aforementioned preset low-pressure trip threshold, a fuel flow reduction command is generated. The preset warning time is a pre-set time threshold used to trigger active load reduction, typically measured in seconds (s), for example, 60 seconds. When the remaining available operating time is less than this threshold, it indicates that the fuel supply will be exhausted in a short time, requiring active reduction of fuel flow to lower the load and slow the pressure drop. The fuel flow reduction command is a control signal used to control the first or second regulating valve to reduce its opening. In practice, the aforementioned actuator compares the remaining available operating time with the warning time, while ensuring that the current pressure has not fallen below the trip threshold. When the remaining available operating time is less than the warning time and the current pressure is not less than the trip threshold, a fuel flow reduction command is generated. For example, if the remaining available operating time is 35 seconds, less than the warning time of 60 seconds, and the current pressure is 1.6 MPa, higher than the trip threshold of 1.25 MPa, the control system generates a fuel flow reduction command.

[0034] Step 8: Based on the aforementioned fuel flow reduction command, control the first or second regulating valve to reduce the fuel supply flow. In practice, the aforementioned actuator selects to reduce primary or secondary fuel flow according to the current operating conditions, sending the fuel flow reduction command to the corresponding regulating valve actuator. The actuator drives the valve to reduce its opening, reducing the fuel supply, thereby reducing the unit load, slowing down the rate of pressure drop in the fuel mains, and buying time for operators to handle the situation or creating conditions for a smooth shutdown. For example, under load operation, if the control system selects to reduce secondary fuel flow, reducing the opening of the second regulating valve from 26.8% to 24%, the total fuel flow will decrease by approximately 10%, the unit load will decrease accordingly, the rate of fuel pressure drop will slow from 0.01 MPa / s to 0.005 MPa / s, and the remaining usable operating time will be extended to 70 seconds.

[0035] Steps one through eight of this disclosure are an inventive point of this disclosure, solving the technical problem that "when the hydrogen fuel supply pressure of a pure hydrogen gas turbine continues to drop during operation, there is a lack of a graded protection method based on pressure trend prediction, resulting in a passive emergency shutdown that can only be triggered when the pressure is below the trip threshold, without early warning, without actively reducing the load to slow down the rate of pressure drop, and without synchronous purging after emergency shutdown." Existing technologies have the following shortcomings in fuel pressure protection: Firstly, they rely solely on a single pressure threshold for judgment, directly triggering an emergency shutdown when the pressure drops to the trip threshold, lacking the ability to predict the pressure drop trend and unable to provide early warning before the pressure falls below the threshold; secondly, when the fuel pressure drops rapidly, the system cannot predict the remaining available operating time, making it difficult for operators to determine when action is needed, resulting in a lack of preparation for passive shutdown; thirdly, even if the pressure drops but has not yet reached the trip threshold, the system cannot actively take load reduction measures to slow down the rate of pressure drop, and can only watch the pressure continue to drop until the trip; and fourthly, there is a lack of pipeline purging after emergency shutdown that is synchronized with fuel cutoff, which may lead to the accumulation of residual hydrogen and cause deflagration, posing a safety hazard. Solving the above problems would enable the prediction of remaining operating time and the issuance of early warnings in the initial stage of fuel pressure decline. It would also allow for proactive reduction of fuel flow to slow the pressure drop before the pressure falls below the trip threshold, providing operators with time to respond or achieving a smooth shutdown. Simultaneously, when the pressure reaches the trip threshold, emergency shut-off and purging would be executed concurrently, achieving a full-chain, tiered protection system from early warning and proactive load reduction to emergency shutdown, ensuring the safe operation of the hydrogen gas turbine under fuel shortage conditions. To achieve this, this disclosure proposes: First, in response to the completion of controlling the primary fuel shut-off valve to introduce primary premixed fuel into the combustion chamber, the fuel main pressure signal is collected in real time. Thus, continuous monitoring of the fuel main pressure begins immediately after the primary fuel is introduced, establishing a real-time sensing basis for subsequent pressure trend analysis and protection judgment, solving the problem of fuel pressure changes not being captured in a timely manner. Second, in response to the current fuel main pressure falling below the preset low-pressure trip threshold, an emergency shutdown command is generated. Therefore, when the fuel pressure drops directly below the safety threshold, an emergency shutdown is immediately triggered, ensuring that the unit can promptly cut off fuel supply in the worst-case scenario, avoiding the risk of backfire or flameout, and solving the safety fallback problem during sudden pressure drops. The third step, based on the aforementioned emergency shutdown command, simultaneously closes all fuel shut-off valves and performs an emergency purging operation on the fuel pipeline. This simultaneously cuts off all fuel supply and purges residual hydrogen in the pipeline during an emergency shutdown, eliminating the safety hazard of deflagration caused by hydrogen accumulation and solving the problem of residual fuel disposal after shutdown. The fourth step, based on the collected fuel mains pressure signal, generates pressure time-series data on fuel pressure changes over time.Thus, discrete pressure sampling values ​​are transformed into continuous pressure time-series data, providing a data foundation for subsequent calculations of pressure drop rate and remaining operating time, solving the problem of the inability to quantify and analyze pressure change trends. Fifth, based on the aforementioned pressure time-series data, the fuel pressure drop rate is generated. Thus, by calculating the first derivative of pressure with respect to time, the rate of pressure drop is quantified, solving the problem of lacking quantitative indicators for pressure change trends and the inability to predict future trends. Sixth, based on the current fuel manifold pressure, the aforementioned fuel pressure drop rate, and the aforementioned preset low-pressure trip threshold, the remaining available operating time is generated. Thus, assuming the current pressure drop rate remains constant, the time remaining from the current moment until the pressure drops to the trip threshold is calculated, solving the problem of operators being unable to determine when intervention is needed and lacking time expectations. Seventh, in response to the generated remaining available operating time being less than the preset warning time and the current fuel manifold pressure not being less than the aforementioned preset low-pressure trip threshold, a fuel flow reduction command is generated. Thus, when the pressure has not yet fallen below the trip threshold but the remaining time is insufficient, a load reduction operation is actively triggered, solving the problem of continuous pressure drop without active system intervention, leaving the system only able to passively wait for tripping. Step 8: Based on the aforementioned fuel flow reduction command, control the first or second regulating valve to reduce the fuel supply flow. This proactively reduces the unit load by decreasing the fuel flow, slows the rate of pressure drop in the fuel main, extends the remaining available operating time, buys time for operators to handle the situation, or creates conditions for a smooth shutdown, thus solving the problem of not being able to proactively intervene in the pressure drop process. In summary, steps 1 to 8 of this embodiment cooperate with each other, starting from the entire chain of "real-time pressure monitoring, trend quantification and prediction, remaining time calculation, and graded protection decision-making," and constructs a graded protection framework integrating "early warning, proactive load reduction, emergency shutdown, and purging," achieving full-process safety protection for the hydrogen gas turbine under fuel pressure drop conditions. This effectively solves the problems of traditional technologies that rely solely on a single pressure threshold judgment and lack trend prediction and proactive intervention capabilities, ensuring that in cases of insufficient fuel supply, early warnings can be issued to buy time for handling the situation, and that safe shutdown can be achieved when the risk cannot be eliminated, providing a complete and reliable fuel pressure protection solution for the safe and stable operation of pure hydrogen gas turbines.

[0036] Step 1032: Based on the collected flame temperature signal, generate a flame temperature change rate signal, and perform fluctuation identification on the generated flame temperature change rate signal to obtain the fluctuation identification result.

[0037] In some embodiments, the aforementioned executing entity can generate a flame temperature change rate signal based on the acquired flame temperature signal, and perform fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result. The flame temperature signal refers to the temperature value collected in real time by a thermocouple installed at the main nozzle outlet, typically in degrees Celsius (°C), reflecting the instantaneous temperature of the flame inside the combustion chamber. The flame temperature change rate signal is the first derivative of the flame temperature with respect to time, characterizing the rate of change of the flame temperature over time, typically in °C / s. Fluctuation identification refers to the process of performing feature analysis on the flame temperature change rate signal to identify specific fluctuation patterns (such as a characteristic fluctuation of first rising and then falling). The fluctuation identification result refers to the judgment result output after fluctuation identification, used to characterize whether a characteristic fluctuation of first rising and then falling in flame temperature has been identified.

[0038] In some optional implementations of certain embodiments, the aforementioned execution entity may generate a flame temperature change rate signal based on the collected flame temperature signal, and perform fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result: Step one: Based on the collected flame temperature signals, generate flame temperature time-series data. This flame temperature time-series data refers to a sequence of flame temperature sampling values ​​arranged chronologically, with each sampling point including a timestamp and the corresponding temperature value. In practice, the aforementioned execution entity continuously collects thermocouple signals from the main nozzle outlet of each combustion chamber at a preset sampling frequency (e.g., 100Hz). The collected analog quantities are then converted from analog to digital to generate continuous time-series data. Each combustion chamber's main nozzle is typically equipped with two K-type thermocouples, thus allowing for the generation of flame temperature time-series data from multiple channels. For example, during the gas turbine's acceleration process, the control system reads the measurements from each thermocouple every 10 milliseconds, forming a temperature sequence including time information, providing raw data for subsequent processing.

[0039] Step two involves performing differential processing on the aforementioned flame temperature time-series data to obtain flame temperature change rate time-series data. Differential processing refers to calculating the ratio of the difference between temperature values ​​at adjacent time points to the time interval, thereby obtaining the temperature change rate. Flame temperature change rate time-series data is a sequence composed of the temperature change rates at each time point, reflecting the trend of flame temperature change over time. In practice, the executing entity performs first-order differential calculation on the flame temperature time-series data generated in Step one. Specifically, for the i-th sampling point, the change rate ΔT / Δt = (T_i - T_{i-1}) / (t_i - t_{i-1}), where T_i is the temperature of the current sampling point, T_{i-1} is the temperature of the previous sampling point, and t_i and t_{i-1} are the corresponding sampling times. This calculation is repeated to obtain the change rates for all sampling points, thus forming the flame temperature change rate time-series data. For example, when the flame temperature rapidly rises from 150°C to 350°C in a short period of time, differential processing will capture a large positive rate of change; when the temperature subsequently falls back, differential processing will capture a negative rate of change, thus fully reflecting the dynamic process of temperature fluctuation.

[0040] Step 3 involves performing extreme point identification processing on the aforementioned flame temperature change rate time series data to obtain positive and negative extreme points. Extreme point identification processing refers to mathematical analysis of the change rate sequence to identify local maxima and minima. Positive extreme points are those corresponding to the local maximum values ​​in the change rate sequence, representing the moment when the temperature rise rate reaches its peak; negative extreme points are those corresponding to the local minimum values ​​in the change rate sequence, representing the moment when the temperature fall rate reaches its peak. In practice, the aforementioned execution entity performs sliding window analysis on the flame temperature change rate time series data, comparing the change rate values ​​of adjacent points within each window. Points with change rates greater than their immediate neighbors are marked as positive extreme points, and points with change rates less than their immediate neighbors are marked as negative extreme points. For example, immediately after the addition of primary premixed fuel, the flame temperature rises rapidly, resulting in a significant positive extreme point in the change rate sequence; subsequently, the temperature falls back, resulting in a significant negative extreme point. The order and magnitude relationship of these two extreme points together constitute the key identification basis for characteristic fluctuations.

[0041] Step four: Based on the obtained positive and negative extreme points, generate a fluctuation recognition result. The fluctuation recognition result refers to the logical signal output after comprehensively judging the occurrence order, time interval, and amplitude of the positive and negative extreme points. In practice, the execution entity first determines whether the positive extreme point appears before the negative extreme point, and whether the time interval between them is less than a preset first time threshold (e.g., 2 seconds); then it calculates the temperature difference ΔT between the temperature peak corresponding to the positive extreme point and the temperature valley corresponding to the negative extreme point, and determines whether ΔT exceeds the first temperature difference threshold (e.g., 200℃); if the above conditions are met simultaneously, it is determined that the characteristic fluctuation of "first rising and then falling" in the flame temperature has been identified, and a fluctuation recognition result representing the identification of the characteristic fluctuation of "first rising and then falling" in the flame temperature is generated; otherwise, a fluctuation recognition result representing the failure to identify the characteristic fluctuation of "first rising and then falling" in the flame temperature is generated. For example, after the first-stage premixed fuel is introduced, the control system monitors that the flame temperature rises from 150°C to 350°C (positive extreme point) within 0.5 seconds, and then falls back to 250°C (negative extreme point) within 1 second, with a temperature difference ΔT of 100°C. If the temperature difference threshold is set to 50°C and the time threshold is set to 2 seconds, the conditions are met, and a fluctuation recognition result is generated to characterize the characteristic fluctuation of the flame temperature that first rises and then falls, indicating that the first-stage premixed combustion has been stably established.

[0042] Step 1033: In response to the obtained fluctuation identification result indicating that the flame temperature first rises and then falls, the first regulating valve is adjusted.

[0043] In some embodiments, the aforementioned actuator can adjust the first regulating valve in response to the obtained fluctuation identification result indicating a characteristic fluctuation in flame temperature that first rises and then falls. Here, the characteristic fluctuation refers to the "rise and fall" pattern of flame temperature after the primary premixed fuel is introduced. Specifically, in the initial stage of primary premixed fuel introduction, the drastic change in the fuel-air mixing ratio causes a brief rise in flame temperature. After uniform mixing and premixed combustion becoming dominant, the temperature gradually decreases. Therefore, the identification result of this characteristic fluctuation is used as an indicator that primary premixed combustion has been stably established. The first regulating valve is a flow regulating valve installed on the main fuel line or primary fuel line, used to control the flow rate of primary premixed fuel entering the combustion chamber. It is typically driven by an electric actuator and can continuously adjust the valve opening according to control commands. Adjustment refers to changing the primary premixed fuel supply flow rate by changing the valve opening of the first regulating valve, thereby causing the flame temperature to approach and stabilize within a preset target temperature range. In practice, when the fluctuation identification result indicates that characteristic fluctuations have been detected, it signifies that primary premixed combustion has been initially established. At this point, the control system initiates closed-loop regulation, comparing the actual measured flame temperature with the preset target temperature. Based on the deviation, it calculates the adjustment amount of the opening of the first regulating valve, generates an opening adjustment command, and drives the actuator to operate, rapidly stabilizing the flame temperature near the target temperature, providing stable combustion conditions for subsequent acceleration and secondary fuel injection. If the fluctuation identification result does not detect characteristic fluctuations in the flame temperature, such as an initial rise followed by a fall, it indicates that stable premixed combustion has not been established within the expected time after the primary premixed fuel injection. In this case, the aforementioned actuator can perform at least one of the following operations: maintain the current opening of the first regulating valve and extend the waiting time; if no characteristic fluctuations are detected within the preset waiting time window, a primary premixing establishment failure signal is generated, triggering the unit's speed reduction or shutdown protection logic to prevent equipment damage due to unstable combustion. For example, after the first-stage premixed fuel is introduced, the control system detects the characteristic fluctuation of the flame temperature rapidly rising from 150°C to 350°C and then falling back to 250°C. At this time, the system compares the current flame temperature (250°C) with the preset target temperature (e.g., 300°C), and calculates the first regulating valve needs to be opened more to increase the fuel supply based on the deviation (-50°C), so that the flame temperature rises back to near the target value, thereby stabilizing the combustion state.

[0044] In some optional implementations of certain embodiments, the aforementioned execution entity may adjust the first regulating valve in response to the obtained fluctuation identification result characterizing a characteristic fluctuation in flame temperature that first rises and then falls: Step one: Based on the collected flame temperature signal and the preset target temperature, a temperature deviation signal is generated. The preset target temperature refers to the desired flame temperature value pre-set according to the gas turbine's operating conditions (such as acceleration phase, no-load phase, or loaded phase), usually expressed in degrees Celsius (°C). This value is determined by the combustion chamber design characteristics and fuel grading strategy. The temperature deviation signal is the difference between the actual measured flame temperature and the preset target temperature, usually denoted by e(t). A positive deviation occurs when the actual temperature is higher than the target temperature, and a negative deviation occurs otherwise. In practice, the aforementioned actuator reads the instantaneous flame temperature value at a preset sampling period, and simultaneously reads the target temperature value corresponding to the current operating condition from memory, calculating the difference between the two as the temperature deviation signal. For example, under full-speed no-load conditions, if the preset target temperature is 310°C and the actual flame temperature is 250°C, then the temperature deviation signal is -60°C.

[0045] Step two involves proportionally amplifying the temperature deviation signal to obtain a proportional control quantity. Proportional amplification involves multiplying the temperature deviation signal by a preset proportional coefficient Kp, making the control quantity proportional to the deviation. The proportional control quantity is the control component output after proportional amplification, with a value of Kp·e(t). This component reflects the magnitude of the current deviation; the larger the deviation, the larger the proportional control quantity and the faster the response. In practice, the actuator pre-sets the proportional coefficient Kp (e.g., Kp=0.5) based on the flame temperature response characteristics, and multiplies the temperature deviation signal by this coefficient to obtain the proportional control quantity. For example, when the temperature deviation is -60℃ and Kp=0.5, the proportional control quantity is -30, indicating an equivalent control quantity requiring a 30% reduction in the opening of the first regulating valve (actually constrained by valve position limits).

[0046] Step three involves performing time integration processing on the aforementioned temperature deviation signal to obtain the integral control quantity. Time integration processing refers to cumulatively integrating the temperature deviation signal over time to eliminate static deviation. The integral control quantity is the control component output after integration processing, with a value of Ki·∫e(t)dt, where Ki is the integral coefficient. This component reflects the cumulative effect of the deviation, enabling the flame temperature to ultimately and accurately reach the target value. In practice, the aforementioned actuator performs time integration calculations on the temperature deviation signal. The accumulated deviation gradually increases over time, and the integral control quantity increases accordingly until the deviation is eliminated. For example, when the flame temperature remains below the target value for an extended period, the integral control quantity gradually increases, pushing the first regulating valve to gradually open until the temperature rises back to the target value.

[0047] Step four involves performing time-differential processing on the temperature deviation signal to obtain the differential control quantity. Time-differential processing involves calculating the rate of change of the temperature deviation signal to predict its trend and apply control in advance. The differential control quantity is the control component output after differential processing, with a value of Kd·de(t) / dt, where Kd is the differential coefficient. This component reflects the rate of change of the deviation and can act in advance before significant changes in the deviation, suppressing overshoot. In practice, the actuator calculates the difference between the current temperature deviation and the deviation at the previous moment, divides it by the sampling period to obtain the rate of change of the deviation, and then multiplies it by the differential coefficient to obtain the differential control quantity. For example, when the flame temperature rises rapidly, the rate of change of the deviation is negative, and the differential control quantity is negative, allowing for an advance reduction in the opening of the first regulating valve to prevent temperature overshoot.

[0048] Step five involves superimposing the obtained proportional, integral, and derivative control quantities to obtain the PID control quantity of the first regulating valve. Superposition refers to adding the proportional, integral, and derivative control quantities together. The PID control quantity is the sum of the three control components, i.e., u(t) = Kp·e(t) + Ki·∫e(t)dt + Kd·de(t) / dt. This control quantity comprehensively reflects the current deviation, historical cumulative deviation, and future trend, and is used to generate the final control command. In practice, the executing entity adds the three control quantities calculated in steps two through four to obtain the PID control quantity of the first regulating valve. For example, when the proportional control quantity is -30, the integral control quantity is -10, and the derivative control quantity is -5, the PID control quantity is -45, indicating an equivalent control quantity that needs to reduce the opening of the first regulating valve by 45%.

[0049] Step Six: Based on the PID control amount of the first regulating valve, generate an opening adjustment command for the first regulating valve. The opening adjustment command refers to the electrical or pneumatic signal used to control the action of the first regulating valve actuator, including the target opening value (expressed as a percentage %) or the opening change. In practice, the actuator adds the PID control amount to the current valve opening to obtain the target valve opening, and after limiting processing (ensuring the target opening is within the range of 0% to 100%), generates an opening adjustment command including the target opening value. For example, if the current first regulating valve opening is 34.1% and the PID control amount is -45%, then the target opening is -10.9%, which is adjusted to 0% after limiting, generating an opening adjustment command to close the valve to the 0% position; if the PID control amount is +20%, then the target opening is 54.1%, generating an opening adjustment command to open the valve to the 54.1% position.

[0050] Step 7: Based on the aforementioned opening adjustment command, control the first regulating valve to change the supply flow rate of the primary premixed fuel. Changing the supply flow rate of the primary premixed fuel refers to altering the hydrogen flow rate through the primary fuel pipeline by adjusting the valve opening of the first regulating valve, thereby affecting the fuel equivalence ratio and flame temperature in the combustion chamber. In practice, the aforementioned actuator sends the opening adjustment command to the electric actuator of the first regulating valve. The actuator drives the valve core to actuate according to the command value, changing the valve orifice flow area, thereby adjusting the supply flow rate of the primary premixed fuel. The control system continuously monitors changes in flame temperature, forming a closed-loop control to stabilize the flame temperature within the preset target temperature range. For example, when the opening adjustment command is to open the first regulating valve from 34.1% to 54.1%, the electric actuator drives the valve core to open, increasing the primary premixed fuel flow rate. The flame temperature gradually rises from 250℃ to 310℃, the temperature deviation gradually decreases, and the PID adjustment value approaches zero, achieving stable control.

[0051] Step 104: In response to the acquired rotational speed signal reaching the third rotational speed threshold, the following steps are executed: Step 1041: Adjust the secondary fuel shut-off valve and the second regulating valve.

[0052] In some embodiments, the aforementioned actuator can regulate the secondary fuel shut-off valve and the second regulating valve. The third speed threshold refers to a pre-set critical speed value used to trigger the introduction of secondary premixed fuel, typically measured in revolutions per minute (rpm). This threshold is determined comprehensively based on the combustion chamber characteristics of the gas turbine, the fuel staging strategy, and the turbine's heat load capacity; for example, it can be set as a specific percentage of the rated speed (e.g., 70% TNH). The secondary fuel shut-off valve is a shut-off valve installed on the secondary fuel pipeline, used to control the supply and disconnection of secondary premixed fuel. It is typically driven by a pneumatic or electric actuator and has a fast response characteristic. When this valve is open, hydrogen fuel enters the main nozzle of the combustion chamber through the secondary fuel pipeline, where it is pre-mixed with compressed air discharged from the compressor in the micro-premixing tube before being injected into the combustion zone. The second regulating valve is a flow regulating valve installed on the secondary fuel pipeline, used to control the flow rate of secondary premixed fuel entering the combustion chamber. It is typically driven by an electric actuator and can continuously adjust the valve opening according to control commands. Secondary premixed fuel refers to the secondary fuel introduced into the combustion chamber. This fuel is fully premixed with air before entering the combustion chamber to form a homogeneous hydrogen-air mixture. Its equivalence ratio is typically designed within the lean premixed combustion range. It works in conjunction with primary premixed fuel to support fuel demands under full-speed no-load and loaded conditions. Regulation refers to the comprehensive control of the secondary fuel shut-off valve and the second regulating valve, specifically including a series of actions such as secondary premixed fuel injection, flow regulation, and combustion pulsation suppression. In practice, when the control system detects that the speed signal reaches the third speed threshold, it generates a secondary fuel injection command, executes the opening of the secondary fuel shut-off valve and the gradual opening control of the second regulating valve, and simultaneously monitors the combustion chamber pulsation pressure signal in real time. Based on the pulsation situation, it coordinates the adjustment of the first and second regulating valves to suppress combustion oscillations until combustion stabilizes, after which the second regulating valve gradually opens to the target opening degree. For example, during the acceleration of the gas turbine, when the control system detects that the speed signal has reached the third speed threshold (such as 3614 rpm), it immediately generates a secondary fuel injection command, opens the secondary fuel shut-off valve, and gradually increases the opening of the second regulating valve at a preset rate to introduce secondary premixed fuel into the combustion chamber. At the same time, the combustion stability is monitored in real time by the pulsating pressure sensor, and coordinated adjustment is performed when necessary to provide a fuel basis for subsequent full-speed no-load and loaded operation.

[0053] In some optional implementations of certain embodiments, the aforementioned executing entity can regulate the secondary fuel shut-off valve and the second regulating valve through the following steps: Step one: Control the secondary fuel shut-off valve to introduce secondary premixed fuel into the combustion chamber. The secondary premixed fuel refers to the fuel introduced into the combustion chamber after being fully premixed with air to form a homogeneous hydrogen-air mixture. In practice, the actuator responds to the speed signal reaching the third speed threshold by generating a secondary fuel shut-off valve opening command. This command is sent to the actuator of the secondary fuel shut-off valve, which drives the valve to open, allowing hydrogen fuel to flow through the secondary fuel pipeline into the micro-premixing tube of the main nozzle. There, it rapidly mixes with compressed air from the compressor, forming a homogeneous premixed combustible mixture. This mixture is then injected into the combustion chamber through the micro-premixing tube outlet, achieving stable combustion under the existing flame. For example, when the speed reaches 3614 rpm, the control system outputs an opening signal to the secondary fuel shut-off valve. After the valve opens, the pressure in the secondary fuel pipeline is established, fuel begins to flow into the combustion chamber, the total fuel flow in the combustion chamber increases, and the unit prepares to enter full-speed no-load operation.

[0054] Step two involves generating time-series data of pulsating pressure amplitude over time based on the acquired combustion chamber pulsating pressure signal. The combustion chamber pulsating pressure signal refers to the pressure fluctuation signal acquired in real time by a dynamic pressure sensor installed on the combustion chamber casing. It is typically measured in kilopascals (kPa) or megapascals (MPa) and reflects the high-frequency fluctuation characteristics of the pressure within the combustion chamber over time, serving as a key parameter characterizing combustion stability. The pulsating pressure amplitude refers to the peak value of the pulsating pressure signal, reflecting the intensity of combustion oscillations. The pulsating pressure time-series data is a sequence of sampled pulsating pressure amplitude values ​​arranged chronologically, with each sampling point including a timestamp and the corresponding amplitude. In practice, the aforementioned execution entity continuously acquires signals from the dynamic pressure sensors on each combustion chamber casing at a preset sampling frequency (e.g., 1000Hz). After signal conditioning (e.g., amplification, filtering) and analog-to-digital conversion, the pulsating pressure amplitude is extracted to generate continuous time-series data. For example, after the secondary premixed fuel is introduced, the control system reads the measurement value of the pulsating pressure sensor every 1 millisecond to form a pulsating pressure amplitude sequence including time information, providing raw data for subsequent peak detection.

[0055] Step three involves performing peak detection processing on the generated pulsating pressure time-series data to obtain a pulsating pressure peak sequence. Peak detection processing refers to the analysis of the pulsating pressure time-series data to identify the local maxima within each fluctuation cycle. The pulsating pressure peak sequence is a sequence composed of peak values ​​from each cycle, reflecting the extreme value distribution of the combustion chamber pulsating pressure. In practice, the aforementioned execution entity performs sliding window analysis on the pulsating pressure time-series data, identifying local maxima within each window, marking values ​​greater than adjacent data points as peak values, and recording these peak values ​​and their occurrence times to form the pulsating pressure peak sequence. For example, when high-frequency oscillations occur in the combustion chamber, periodic fluctuations will appear in the pulsating pressure time-series data. Peak detection processing will extract the amplitude of each peak, forming a sequence including multiple peaks, used to determine whether a preset pulsating pressure threshold is exceeded.

[0056] Step four: In response to any pulsating pressure peak exceeding a preset pulsating pressure threshold in the obtained pulsating pressure peak sequence, the following sub-steps are executed. The preset pulsating pressure threshold is a pre-set critical value for triggering combustion oscillation suppression, typically expressed in kilopascals (kPa). This threshold is determined based on the combustion chamber's vibration resistance characteristics and safety margin. When the pulsating pressure exceeds this threshold, it indicates significant combustion oscillation in the combustion chamber, requiring active adjustment to suppress the oscillation.

[0057] Sub-step one involves generating a first coordinated adjustment value for the first regulating valve and a second coordinated adjustment value for the second regulating valve. The first coordinated adjustment value refers to the compensation opening value used to adjust the first regulating valve, typically expressed as a percentage (%); the second coordinated adjustment value refers to the compensation opening value used to adjust the second regulating valve, also typically expressed as a percentage (%). In practice, the aforementioned actuator calculates the first and second coordinated adjustment values ​​based on the deviation between the peak pulsating pressure and a preset threshold using a preset coordinated control algorithm (e.g., proportional control or fuzzy control). For example, when the peak pulsating pressure exceeds the threshold by 10%, the control system proportionally generates a first coordinated adjustment value of -5% (i.e., reducing the opening of the first regulating valve by 5%) and a second coordinated adjustment value of -3% (i.e., reducing the opening of the second regulating valve by 3%), thereby reducing combustion oscillations by decreasing fuel supply.

[0058] Sub-step two involves controlling the first regulating valve to change the supply flow rate of the primary premixed fuel based on the aforementioned first coordinated adjustment amount. In practice, the aforementioned actuator superimposes the first coordinated adjustment amount onto the current opening of the first regulating valve, generating a coordinated adjustment command for the first regulating valve. This command is then sent to the actuator of the first regulating valve, which drives the valve core to actuate, changing the flow area of ​​the valve orifice, thereby adjusting the supply flow rate of the primary premixed fuel. For example, when the first coordinated adjustment amount is -5%, if the current opening of the first regulating valve is 34.1%, the target opening is 29.1%. The control system then drives the valve to close slightly, reducing the supply of primary premixed fuel to suppress combustion oscillations.

[0059] Sub-step three involves controlling the second regulating valve to change the supply flow rate of the secondary premixed fuel, based on the aforementioned second coordinated adjustment amount. In practice, the aforementioned actuator superimposes the second coordinated adjustment amount onto the current opening of the second regulating valve, generating a coordinated adjustment command for the second regulating valve. This command is then sent to the actuator of the second regulating valve, which drives the valve core to actuate, changing the flow area of ​​the valve orifice, thereby adjusting the supply flow rate of the secondary premixed fuel. For example, when the second coordinated adjustment amount is -3%, if the current opening of the second regulating valve is 23.7%, the target opening is 20.7%. The control system drives the valve to close slightly, reducing the supply of secondary premixed fuel and working in conjunction with the first regulating valve to suppress combustion oscillations.

[0060] Sub-step four: In response to the fact that all the collected post-adjustment pulsating pressure amplitudes within the first preset time window are lower than the aforementioned preset pulsating pressure threshold, a second regulating valve gradual opening command is generated. The first preset time window refers to a pre-set time length for determining whether the pulsating pressure remains stable, typically measured in seconds (s), such as 5 or 10 seconds. The post-adjustment pulsating pressure amplitude refers to the combustion chamber pulsating pressure amplitude re-collected after the coordinated adjustment of the first and second regulating valves. The second regulating valve gradual opening command is a control command used to control the second regulating valve to gradually increase its opening at a slow rate, aiming to smoothly restore fuel supply after combustion stabilizes and avoid triggering oscillations again. In practice, after executing the coordinated adjustment of sub-steps one through three, the aforementioned executing entity continuously monitors the combustion chamber pulsating pressure signal. When all collected pulsating pressure amplitudes within the first preset time window (e.g., 5 seconds) are lower than the preset pulsating pressure threshold, it indicates that combustion oscillations have been effectively suppressed and the combustion state has returned to stability. At this time, the control system generates the second regulating valve gradual opening command. If, after coordinated adjustment, the amplitude of each adjusted pulsating pressure collected within the first preset time window still fails to fall below the preset pulsating pressure threshold, it indicates that the coordinated adjustment has failed to effectively suppress combustion oscillations. In this case, the actuator can maintain the current opening of the first and second regulating valves unchanged and continue to monitor the pulsating pressure changes. For example, after coordinated adjustment, if the pulsating pressure amplitude drops from exceeding the threshold to a safe range and remains stable for 5 seconds, the control system generates a gradual opening command to prepare to restore the opening of the second regulating valve.

[0061] Step 5: In response to the fact that none of the peak values ​​in the obtained pulsating pressure peak sequence exceed the aforementioned preset pulsating pressure threshold, a gradual opening command for the second regulating valve is generated. In practice, during the initial stage of secondary premixed fuel introduction, if all peak values ​​in the pulsating pressure peak sequence do not exceed the preset pulsating pressure threshold, it indicates that the combustion state is stable after the secondary premixed fuel is introduced, and no obvious combustion oscillations occur. In this case, there is no need to perform oscillation suppression regulation, and the control system directly generates a gradual opening command for the second regulating valve. For example, after the secondary premixed fuel is introduced, if the control system detects that the pulsating pressure peak values ​​are all below the threshold, indicating stable combustion, it immediately generates a gradual opening command for the second regulating valve, preparing to gradually open the second regulating valve to the target opening degree.

[0062] Step six: Based on the generated gradual opening command for the second regulating valve, control the second regulating valve to gradually open until the corresponding target opening degree is reached. The target opening degree refers to the desired opening degree of the second regulating valve preset according to the current operating conditions (e.g., full-speed no-load or under load), usually expressed as a percentage (%). For example, the target opening degree under full-speed no-load conditions can be set to 23.7%. Gradual opening means controlling the second regulating valve to open slowly at a preset gradual rate (e.g., 5% per second), rather than a sudden jump, to avoid combustion oscillation caused by a sudden increase in fuel. In practice, the aforementioned actuator responds to the gradual opening command of the second regulating valve, generating a series of progressively increasing opening degree commands at a preset gradual rate, sending them to the actuator of the second regulating valve. The actuator drives the valve core to gradually open until the target opening degree is reached. For example, when the current opening of the second regulating valve is 20.7%, the target opening is 23.7%, and the gradual change rate is 1% per second, the control system outputs opening commands of 21.7%, 22.7%, and 23.7% sequentially within 3 seconds, driving the valve to open smoothly to the target position, ensuring a steady increase in fuel flow and maintaining stable combustion.

[0063] Optionally, the aforementioned implementing entity may perform the following steps: Step one involves executing the following steps in parallel after adjusting the secondary fuel shut-off valve and the second regulating valve. Adjusting the secondary fuel shut-off valve and the second regulating valve refers to the stage where the combustion chamber enters a relatively stable operating state after the secondary premixed fuel is introduced and related pulsation suppression control is completed. Parallel execution means that the engine speed signal, total fuel flow signal, and exhaust temperature signal are independently acquired and processed simultaneously. These three monitoring channels operate independently without interference, each generating its own corresponding stable sub-signal, which is ultimately confirmed through logic and judgment to establish a stable operating condition.

[0064] Sub-step one: Based on the acquired rotational speed signal, perform the following steps: The first sub-step generates the first speed judgment result. This first speed judgment result is a logical judgment value output after comparing the real-time acquired speed signal with the rated speed, used to characterize whether the current speed has reached the rated speed. The rated speed refers to the rated operating speed of the gas turbine under its design conditions, usually measured in revolutions per minute (rpm). For example, the rated speed of a 30MW pure hydrogen gas turbine can be set to 5163 rpm. In practice, the aforementioned execution entity compares the real-time acquired instantaneous speed value with the preset rated speed. If the instantaneous speed value is greater than or equal to the rated speed, the first speed judgment result is "rated speed reached"; if it is less than the rated speed, the first speed judgment result is "rated speed not reached". For example, when the speed signal is 5168 rpm, the first speed judgment result is that the rated speed has been reached.

[0065] The second sub-step, in response to the generated first speed judgment result indicating that the rated speed has been reached, involves collecting instantaneous speed values ​​within a second preset time window. The second preset time window refers to a pre-set time length for collecting speed data, typically in seconds (s), for example, 30 seconds. The instantaneous speed value refers to the speed value acquired at each sampling moment, usually in revolutions per minute (rpm). In practice, when the first speed judgment result indicates that the rated speed has been reached, the aforementioned execution entity starts a timer and continuously collects instantaneous speed values ​​within the second preset time window at a preset sampling frequency (e.g., 10Hz), forming a sequence containing multiple speed sampling values. For example, after the rated speed is reached, the control system collects a speed value every 0.1 seconds within 30 seconds, collecting a total of 300 instantaneous speed values.

[0066] The third sub-step involves processing the collected instantaneous speed values ​​to obtain the standard deviation of the speed within the second preset time window. Instantaneous speed value processing refers to statistically calculating the collected instantaneous speed value sequence to assess the degree of speed fluctuation. The standard deviation of the speed refers to the standard deviation of the instantaneous speed value sequence; it is a statistical indicator that measures the amplitude of speed fluctuation, and a smaller value indicates a more stable speed. In practice, the aforementioned execution entity first calculates the average of all instantaneous speed values ​​within the second preset time window, then calculates the sum of squares of the deviations of each instantaneous speed value from the average, divides it by the number of sampling points, and takes the square root to obtain the standard deviation. For example, if 300 instantaneous speed values ​​are collected within 30 seconds, with an average of 5165 rpm, the calculated standard deviation is 2.5 rpm, indicating that the speed fluctuates around 5165 rpm by approximately 2.5 rpm.

[0067] The fourth sub-step involves generating a first stable sub-signal in response to the obtained speed standard deviation being less than a preset speed fluctuation threshold. The preset speed fluctuation threshold is a pre-defined upper limit of the standard deviation used to determine whether the speed is stable, typically expressed in revolutions per minute (rpm), for example, 5 rpm. The first stable sub-signal is a stability confirmation signal generated by the speed channel, indicating that the speed has reached a stable state. In practice, the executing entity compares the calculated speed standard deviation with the preset speed fluctuation threshold. If the speed standard deviation is less than the speed fluctuation threshold, the first stable sub-signal is generated. For example, when the speed standard deviation is 2.5 rpm and the preset speed fluctuation threshold is 5 rpm, the condition is met, and the first stable sub-signal is generated.

[0068] Sub-step two: Based on the collected total fuel flow signal, perform the following steps: The first sub-step generates a first flow rate judgment result. This first flow rate judgment result is a logical judgment value output after comparing the real-time collected total fuel flow rate signal with the target flow rate value, used to characterize whether the current total fuel flow rate has reached the target flow rate value. The total fuel flow rate refers to the total flow rate of hydrogen entering all fuel pipelines (including the duty acceleration stage, first stage, and second stage), usually measured in kilograms per hour (kg / h). The target flow rate value is the expected total fuel flow rate preset according to the current operating conditions (e.g., full-speed no-load or under load). For example, under full-speed no-load conditions, the target flow rate value can be set to 1011 kg / h. In practice, the aforementioned execution entity compares the real-time collected instantaneous value of the total fuel flow rate with the preset target flow rate value. If the instantaneous value of the total fuel flow rate is greater than or equal to the target flow rate value, the first flow rate judgment result is "target flow rate value reached"; if it is less than the target flow rate value, the first flow rate judgment result is "target flow rate value not reached". For example, when the total fuel flow rate is 1015 kg / h, the first flow rate judgment result is "target flow rate value reached".

[0069] The second sub-step, in response to the generated first flow rate judgment result indicating that the target flow rate value has been reached, involves collecting instantaneous flow rate values ​​within a third preset time window. The third preset time window refers to a pre-set time length for collecting fuel flow rate data, typically measured in seconds (s), for example, 30 seconds. The instantaneous flow rate value refers to the fuel flow rate value acquired at each sampling moment, typically measured in kilograms per hour (kg / h). In practice, when the first flow rate judgment result indicates that the target flow rate value has been reached, the aforementioned execution entity starts a timer and continuously collects instantaneous flow rate values ​​within the third preset time window at a preset sampling frequency (e.g., 10Hz), forming a sequence containing multiple flow rate sampling values. For example, after the total fuel flow rate reaches the target value, the control system collects flow rate values ​​every 0.1 seconds for 30 seconds, collecting a total of 300 instantaneous flow rate values.

[0070] The third sub-step involves processing the collected instantaneous flow values ​​to obtain the standard deviation of the flow rate within a third preset time window. Instantaneous flow value processing refers to statistically calculating the collected instantaneous flow value sequence to assess the degree of fuel flow fluctuation. The standard deviation of the flow rate is the standard deviation of the instantaneous flow value sequence, a statistical indicator that measures the amplitude of flow fluctuation; a smaller value indicates a more stable fuel supply. In practice, the aforementioned execution entity first calculates the average of all instantaneous flow values ​​within the third preset time window, then calculates the sum of squares of the deviations of each instantaneous flow value from the average, divides it by the number of sampling points, and takes the square root to obtain the standard deviation. For example, if 300 instantaneous flow values ​​are collected within 30 seconds, with an average of 1015 kg / h, the calculated standard deviation is 3.2 kg / h, indicating that the flow rate fluctuates around 1015 kg / h by approximately 3.2 kg / h.

[0071] The fourth sub-step involves generating a second stable sub-signal in response to the obtained flow rate standard deviation being less than a preset flow rate fluctuation threshold. The preset flow rate fluctuation threshold is a pre-defined upper limit of the standard deviation used to determine whether the fuel flow rate is stable, typically expressed in kilograms per hour (kg / h), for example, 10 kg / h. The second stable sub-signal is a stability confirmation signal generated by the fuel flow channel, indicating that the fuel flow rate has reached a stable state. In practice, the executing entity compares the calculated flow rate standard deviation with the preset flow rate fluctuation threshold. If the flow rate standard deviation is less than the threshold, a second stable sub-signal is generated. For example, when the flow rate standard deviation is 3.2 kg / h and the preset flow rate fluctuation threshold is 10 kg / h, the condition is met, and a second stable sub-signal is generated.

[0072] Sub-step three: Based on the collected exhaust temperature signal, perform the following steps: The first sub-step involves acquiring the instantaneous exhaust temperature values ​​within a fourth preset time window. The fourth preset time window refers to a pre-defined time duration for collecting exhaust temperature data, typically measured in seconds (s), for example, 30 seconds. The exhaust temperature signal refers to the real-time exhaust temperature value collected by thermocouples installed in the exhaust diffuser section of the gas turbine, typically measured in degrees Celsius (°C), reflecting the flue gas temperature at the combustion chamber outlet. The instantaneous exhaust temperature value refers to the exhaust temperature value acquired at each sampling moment. In practice, the aforementioned actuator continuously collects instantaneous exhaust temperature values ​​within the fourth preset time window at a preset sampling frequency (e.g., 10Hz), forming a sequence containing multiple exhaust temperature sampling values. For example, the control system collects exhaust temperature values ​​every 0.1 seconds within 30 seconds, collecting a total of 300 instantaneous exhaust temperature values.

[0073] The second sub-step involves processing the acquired instantaneous exhaust temperature values ​​to obtain the standard deviation of the exhaust temperature within a fourth preset time window. This instantaneous exhaust temperature processing involves statistically calculating the collected sequence of instantaneous exhaust temperature values ​​to assess the degree of exhaust temperature fluctuation. The standard deviation of the exhaust temperature is the standard deviation of the instantaneous exhaust temperature value sequence; it is a statistical indicator that measures the amplitude of exhaust temperature fluctuations. A smaller value indicates a more uniform and stable combustion chamber outlet temperature. In practice, the aforementioned execution entity first calculates the average of all instantaneous exhaust temperature values ​​within the fourth preset time window, then calculates the sum of squares of the deviations of each instantaneous exhaust temperature value from the average, divides this sum by the number of sampling points, and takes the square root to obtain the standard deviation of the exhaust temperature. For example, if 300 instantaneous exhaust temperature values ​​are collected within 30 seconds, with an average of 300℃, the calculated standard deviation is 2.8℃, indicating that the exhaust temperature fluctuates around 300℃ by approximately 2.8℃.

[0074] The third sub-step involves generating a third stable sub-signal in response to the obtained exhaust temperature standard deviation being less than a preset exhaust temperature fluctuation threshold. The preset exhaust temperature fluctuation threshold is a pre-defined upper limit of the standard deviation used to determine whether the exhaust temperature is stable, typically expressed in degrees Celsius (°C), for example, 5°C. The third stable sub-signal is a stability confirmation signal generated by the exhaust temperature channel, used to indicate that the exhaust temperature has reached a stable state. In practice, the aforementioned executing entity compares the calculated exhaust temperature standard deviation with the preset exhaust temperature fluctuation threshold. If the exhaust temperature standard deviation is less than the exhaust temperature fluctuation threshold, a third stable sub-signal is generated. For example, when the exhaust temperature standard deviation is 2.8°C and the preset exhaust temperature fluctuation threshold is 5°C, the condition is met, and a third stable sub-signal is generated.

[0075] Step two: In response to the fact that the generation timestamps corresponding to the first, second, and third stable sub-signals are all within the same fifth preset time window, a stable operating condition confirmation signal is generated. Here, the generation timestamp refers to the system time stamp when each stable sub-signal is generated, usually in milliseconds (ms) or seconds (s), used to record the precise moment the signal is generated. The fifth preset time window is a pre-set time length used to determine whether the three stable sub-signals are generated within the same time period, usually in seconds (s), for example, it can be set to 10 seconds. The stable operating condition confirmation signal is the final confirmation signal output after the three parameters of combined speed, fuel flow, and exhaust temperature have all stabilized. It is used to characterize that the gas turbine has reached a stable operating condition and can serve as a trigger condition for subsequent control actions (such as shutting off fuel in the shift accelerator stage). In practice, the aforementioned executing entity records the generation timestamps of the first, second, and third stable sub-signals and determines whether these three timestamps all fall within the same fifth preset time window (i.e., the interval between any two timestamps does not exceed the duration of the fifth preset time window). If all three timestamps fall within the same fifth preset time window, it indicates that the engine speed, fuel flow rate, and exhaust temperature have all reached a stable state within the same time period, and a stable operating condition confirmation signal is generated. If a stable operating condition confirmation signal is not generated within an extended waiting time window (e.g., an additional 30 seconds), it indicates that the engine speed, fuel flow rate, or exhaust temperature has not yet reached a stable state. In this case, the aforementioned execution entity can perform at least one of the following operations: maintain the current operating condition without shutting off the shift fuel and issue a stable state abnormality alarm; if the stable state abnormality persists, automatically reduce the fuel supply to allow the unit to return to a lower load condition and re-stabilize. For example, if the first stable sub-signal is generated at t=10.2 seconds, the second stable sub-signal at t=10.5 seconds, and the third stable sub-signal at t=10.8 seconds, and the fifth preset time window is set to 5 seconds, then all three generation timestamps fall within the window from t=10.0 seconds to t=15.0 seconds, meeting the conditions, and a stable operating condition confirmation signal is generated. This signal indicates that the gas turbine has met the stable operating conditions and subsequent operations can be safely carried out.

[0076] Step 1042: In response to the identification of a stable operating condition confirmation signal after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is controlled to close.

[0077] In some embodiments, the aforementioned actuator can control the shut-off fuel shut-off valve of the shift acceleration stage to close in response to recognizing a stable operating condition confirmation signal after regulating the secondary fuel shut-off valve and the second regulating valve. The stable operating condition confirmation signal refers to a confirmation signal generated after all three parameters—integrated speed, fuel flow rate, and exhaust temperature—have reached a stable state. This signal characterizes that the gas turbine has met stable operating conditions and can serve as a trigger condition for shutting off the shift acceleration stage fuel. The shift acceleration stage fuel shut-off valve is a shut-off valve installed on the shift acceleration stage fuel pipeline, used to control the supply and disconnection of fuel to the shift acceleration stage. It is typically driven by a pneumatic or electric actuator, has rapid response characteristics, and remains open during ignition and acceleration phases to provide the combustion chamber with the fuel required for diffusion combustion. The valve closes when the combustion mode switches to fully premixed combustion. In practice, after generating a stable operating condition confirmation signal, the aforementioned actuator immediately generates a command to close the shift acceleration stage fuel shut-off valve. This command is sent to the actuator of the shift acceleration stage fuel shut-off valve, which drives the valve to close, cutting off the supply of shift acceleration stage fuel. This allows the combustion chamber to rely entirely on primary and secondary premixed fuels for combustion, thus achieving the transition from a mixed combustion mode to a fully premixed combustion mode. For example, after the gas turbine has stabilized under load (e.g., 0.7MW load), the control system generates a stable operating condition confirmation signal through multi-parameter fusion judgment, and then generates a shut-off command to close the shift acceleration stage fuel shut-off valve. At this time, combustion completely switches to premixed mode, the main nozzle flame monitoring temperature drops from approximately 310℃ to approximately 220℃, and nitrogen oxide emissions drop from approximately 18.35ppm (after conversion) to approximately 1.5ppm (after conversion), verifying that the fully premixed combustion mode has been successfully established after the shift acceleration stage fuel shut-off.

[0078] Step 1043: Perform flame temperature identification on the collected flame temperature drop signal to obtain the flame temperature identification result.

[0079] In some embodiments, the aforementioned executing entity can perform flame temperature identification on the collected flame temperature drop signal to obtain a flame temperature identification result. The flame temperature drop signal refers to the real-time sequence of flame temperature values ​​collected by a thermocouple installed at the main nozzle outlet after the shift acceleration stage fuel shut-off valve is closed. This signal reflects the dynamic process of the flame temperature gradually decreasing from its original level and stabilizing after the shift fuel shut-off. Flame temperature identification refers to the process of performing feature analysis on the flame temperature drop signal to determine whether the flame temperature has dropped to a stable value. The flame temperature identification result refers to the judgment result output after flame temperature identification, used to characterize whether the flame temperature has stabilized (i.e., dropped to a preset stable value and remained stable). This result serves as the direct basis for generating the full premixed combustion mode confirmation signal. In practice, after controlling the shift acceleration stage fuel shut-off valve to close, the aforementioned executing entity continuously collects the flame temperature signal at the main nozzle outlet of each combustion chamber, performs time-series processing, sliding window segmentation, statistical feature calculation, and multi-condition judgment on the signal, and finally outputs the flame temperature identification result. For example, after the fuel in the duty acceleration stage is shut off, the control system monitors that the flame temperature gradually decreases from 310°C to around 220°C. Through sliding window analysis, it is confirmed that the average temperature is lower than the preset threshold and the standard deviation of temperature fluctuation is less than the preset threshold. After the temperature remains stable for more than the preset time, an identification result characterizing the stability of the flame temperature is generated, indicating that the fully premixed combustion mode has been successfully established.

[0080] In some optional implementations of certain embodiments, the aforementioned execution entity can perform flame temperature identification on the collected flame temperature drop signal through the following steps to obtain the flame temperature identification result: Step one: Based on the collected flame temperature decrease signal, generate flame temperature decrease time-series data. This flame temperature decrease time-series data refers to a sequence of flame temperature sampling values ​​arranged chronologically. Each sampling point includes a timestamp and the corresponding temperature value. This sequence records the dynamic change of flame temperature over time after the standby acceleration stage fuel shut-off. In practice, the aforementioned actuator starts timing the instant the standby acceleration stage fuel shut-off valve closes, continuously collecting thermocouple signals from the main nozzle outlet of each combustion chamber at a preset sampling frequency (e.g., 10Hz or 20Hz). The collected analog quantities are then converted from analog to digital to generate continuous time-series data. For example, after the standby fuel shut-off, the control system reads the flame temperature value every 0.1 seconds, forming a temperature sequence including timestamps, and continues collecting data until the flame temperature stabilizes.

[0081] Step two involves performing sliding window segmentation on the aforementioned flame temperature decrease time-series data to obtain various decreasing time-series data subsequences. Sliding window segmentation refers to the process of sliding a fixed-length time window across the time-series data, dividing the continuous time-series data into multiple overlapping or non-overlapping subsequence segments. A decreasing time-series data subsequence refers to a subset of temperature sample values ​​within each time window obtained after sliding window segmentation; each subsequence includes multiple temperature sample points within that time window. In practice, the execution entity sets a fixed window length (e.g., 5 seconds) and a sliding step size (e.g., 1 second). Starting from the beginning of the flame temperature decrease time-series data, it sequentially extracts temperature sample points within each window according to the sliding step size, forming a series of decreasing time-series data subsequences. For example, for 30 seconds of decreasing time-series data, with a window length of 5 seconds and a sliding step size of 1 second, a total of 26 subsequences are obtained, such as seconds 0-5, seconds 1-6, seconds 2-7, etc.

[0082] Step 3: Process the flame temperature sampling points for each of the aforementioned falling time-series data subsequences to obtain the average flame temperature and standard deviation for each subsequence. Flame temperature sampling point processing involves statistically calculating the temperature sampling points within each subsequence to quantify the central tendency and dispersion of the flame temperature within that time window. The average flame temperature is the arithmetic mean of all temperature sampling points within the subsequence, reflecting the central level of the flame temperature within that time window. The standard deviation of the flame temperature is the standard deviation of all temperature sampling points within the subsequence, reflecting the fluctuation range of the flame temperature within that time window. In practice, for each falling time-sequence data subsequence, the execution entity first calculates the arithmetic mean of all temperature sampling points within the subsequence, then calculates the sum of squares of the deviations of each sampling point from the average, divides it by the number of sampling points, and takes the square root to obtain the standard deviation of the flame temperature. For example, for 50 temperature sampling points (sampling frequency 10Hz) within a 5-second window, the average is 225℃ and the standard deviation is 2.5℃, indicating that the flame temperature within that window is stable around 225℃, fluctuating by approximately 2.5℃.

[0083] Step four: In response to the existence of a first preset number of decreasing time-series data subsequences where the average flame temperature of each corresponding subsequence is lower than a preset flame temperature average threshold, a first stability judgment signal is generated. Here, the first preset number of subsequences refers to the minimum number of consecutive subsequences required to determine if the average flame temperature consistently meets the target; for example, it can be set to 3 or 5. The preset flame temperature average threshold is a preset upper limit for the average flame temperature, usually expressed in degrees Celsius (°C), for example, it can be set to 250°C or 220°C. This threshold is determined based on the stable value of the flame temperature under the fully premixed combustion mode. The first stability judgment signal is a stability confirmation signal generated by the flame temperature average channel, used to characterize that the average flame temperature has decreased and stabilized within the target range. In practice, the aforementioned execution entity traverses each decreasing time-series data subsequence in chronological order, comparing the average flame temperature of each subsequence with the preset flame temperature average threshold. If there are a first preset number of subsequences (e.g., 3 consecutive subsequences) where the average flame temperature is lower than the threshold, then the first stability judgment signal is generated. For example, when the average values ​​of three consecutive subsequences are 218℃, 220℃, and 222℃ respectively, all of which are lower than the threshold of 250℃, the first stable judgment signal is generated.

[0084] Step 5: In response to the existence of a second preset number of decreasing time-series data subsequences where the standard deviation of each flame temperature is lower than a preset flame temperature standard deviation threshold, a second stability judgment signal is generated. Here, the second preset number of subsequences refers to the minimum number of consecutive subsequences required to determine if the flame temperature standard deviation consistently meets the threshold; for example, it can be set to 3 or 5. The preset flame temperature standard deviation threshold is a preset upper limit for the flame temperature standard deviation, usually expressed in degrees Celsius (°C), for example, it can be set to 5°C or 3°C. This threshold is used to determine whether the fluctuation of the flame temperature has converged to a stable state. The second stability judgment signal is a stability confirmation signal generated by the flame temperature standard deviation channel, used to characterize that the fluctuation of the flame temperature has converged to a stable range. In practice, the above-mentioned execution entity traverses each decreasing time-series data subsequence in chronological order, comparing the flame temperature standard deviation of each subsequence with the preset flame temperature standard deviation threshold. If there are a second preset number of subsequences (e.g., 3 consecutive subsequences) where the flame temperature standard deviation is lower than the threshold, a second stability judgment signal is generated. For example, when the standard deviations of three consecutive subsequences are 2.1℃, 2.3℃, and 2.0℃, respectively, all of which are lower than the threshold of 5℃, a second stable judgment signal is generated.

[0085] Step six: In response to the generation of both the first and second stability judgment signals, a flame temperature identification result is generated, indicating that the flame temperature has stabilized. The flame temperature identification result refers to the final identification signal output after both the average flame temperature and the standard deviation of the flame temperature are satisfied. This signal indicates that the flame temperature has decreased to a preset stable value and remains stable. In practice, the aforementioned execution entity continuously monitors the generation status of the first and second stability judgment signals. When both signals have been generated, it indicates that the average flame temperature has decreased to the target range and the fluctuations have converged. At this point, a flame temperature identification result indicating that the flame temperature has stabilized is generated. For example, after both the first stability judgment signal (average value below 250℃ for three consecutive windows) and the second stability judgment signal (standard deviation below 5℃ for three consecutive windows) have been generated, the control system outputs the flame temperature stabilization identification result as the basis for subsequently generating the fully premixed combustion mode confirmation signal.

[0086] Step 1044: In response to the obtained flame temperature identification result indicating that the flame temperature is stable, a fully premixed combustion mode confirmation signal is generated.

[0087] In some embodiments, the aforementioned execution entity may generate a fully premixed combustion mode confirmation signal in response to the obtained flame temperature identification result indicating that the flame temperature is stable. The flame temperature identification result refers to the judgment result output after flame temperature identification in step 1043, used to indicate whether the flame temperature has dropped to a preset stable value and remained stable. This result includes two states: "flame temperature stable identified" or "flame temperature not identified as stable." The fully premixed combustion mode confirmation signal is the final confirmation signal generated by the control system when the flame temperature identification result indicates that the flame temperature is stable. It indicates that the gas turbine has successfully transitioned from a mixed combustion mode (coexistence of diffusion combustion and premixed combustion in the booster stage) to a fully premixed combustion mode (maintaining combustion solely relying on primary and secondary premixed fuels). In practice, after completing flame temperature identification in step 1043, the aforementioned execution entity judges the flame temperature identification result. If the result indicates that the flame temperature is stable, a fully premixed combustion mode confirmation signal is immediately generated. This confirmation signal can be recorded as a status flag by the control system to indicate that the combustion mode switch has been completed. It can also be presented to the operators on the display interface (such as the human-machine interface) to inform them that the unit has entered the fully premixed combustion mode. It can also serve as a prerequisite for the start of subsequent control logic (such as load regulation, emission monitoring, or shutdown procedures). If the flame temperature identification result indicates that the flame temperature has not been stabilized, it means that after the shift acceleration stage fuel shut-off valve is closed, the flame temperature has failed to successfully decrease and stabilize within the preset target range. At this time, the aforementioned execution entity can perform at least one of the following operations: continue to maintain the current combustion state (i.e., the shift acceleration stage fuel is shut off, and combustion is maintained only by the primary and secondary premixed fuels), and continuously monitor the flame temperature changes; if the flame temperature identification result still does not indicate that the flame temperature has been stabilized within the preset timeout window (e.g., 30 seconds), a fully premixed mode establishment failure signal is generated; in response to the above fully premixed mode establishment failure signal, the shift acceleration stage fuel shut-off valve is reopened to restore the mixed combustion mode; if multiple attempts still fail to establish stable fully premixed combustion, a safety shutdown procedure is executed, and fault information is recorded for subsequent analysis. For example, after the fuel shut-off valve of the shift acceleration stage is closed, the control system analyzes and confirms that the flame temperature has dropped from 310℃ to around 220℃ through sliding window analysis. Furthermore, the average temperature across multiple consecutive windows is below a preset threshold (e.g., 250℃), and the temperature standard deviation is below a preset threshold (e.g., 5℃). This flame temperature identification indicates that the flame temperature has stabilized. At this point, the control system generates a full premixed combustion mode confirmation signal, indicating that the combustion mode switching has been completed and the unit is operating stably in full premixed combustion mode. After this signal is generated, operators can observe the full premixed mode status indicator on the monitoring interface. Simultaneously, the control system records this status, providing a basis for subsequent load adjustments or shutdown procedures.

[0088] In addressing the stability issues of combustion mode switching and the reliability of fully premixed combustion in the aforementioned background technology using the aforementioned hydrogen gas turbine combustion mode switching and control methods, the following technical problem often arises in the application scenario: during the long-term stable operation phase after the pure hydrogen gas turbine successfully completes combustion mode switching and enters fully premixed combustion mode (such as full-speed no-load or load power generation conditions): In fully premixed combustion mode, hydrogen and air are premixed in the micro-premixing tube before being injected into the combustion chamber. Due to the extremely fast flame propagation speed of hydrogen (approximately 8-10 times that of natural gas) and high combustion reactivity, when the mixed gas flow velocity at the micro-premixing tube outlet drops below a critical value, the flame may propagate back into the micro-premixing tube, causing nozzle erosion or even damage to the combustion chamber structure. Existing technologies lack real-time online sensing methods for the mixed gas flow velocity at the micro-premixing tube outlet, making it impossible to accurately assess the risk of backfire. Furthermore, when the risk increases, there is a lack of proactive intervention mechanisms, relying solely on passive shutdown protection triggered after backfire occurs, resulting in high equipment damage risk and low operational reliability. To address the following requirements for this application scenario: In fully premixed combustion mode, it is necessary to quantify backfire risk in real time and establish a graded active protection mechanism. This mechanism should control backfire risk within a safe range from the initial stage of occurrence by adjusting the inlet guide vane opening or fuel flow rate, achieving proactive prevention and reliable suppression of backfire. Therefore, we have decided to adopt the following solution: Optionally, the aforementioned implementing entity may also perform the following steps: The first step involves real-time acquisition of the compressor outlet pressure signal, compressor outlet temperature signal, and total fuel flow signal in response to the generation of the aforementioned fully premixed combustion mode confirmation signal. The compressor outlet pressure signal refers to the pressure value acquired in real-time by an absolute pressure transmitter installed at the compressor outlet, typically in megapascals (MPa), reflecting the pressure level of the compressor supplying gas to the combustion chamber. The compressor outlet temperature signal refers to the temperature value acquired in real-time by a thermocouple installed at the compressor outlet, typically in degrees Celsius (°C), reflecting the temperature level of the compressor supplying gas to the combustion chamber. The total fuel flow signal refers to the total hydrogen flow rate acquired in real-time by a Coriolis mass flow meter installed on the fuel mains line, typically in kilograms per hour (kg / h). In practice, the aforementioned actuator continuously acquires these three signals at a preset sampling frequency (e.g., 10 Hz) to provide basic data for subsequent calculations. For example, under the full-speed no-load condition of the gas turbine, the control system reads the compressor outlet pressure, compressor outlet temperature and total fuel flow every 0.1 seconds, and records them as P_c=0.634MPa, T_c=258℃ and W_f=1064kg / h, respectively.

[0089] The second step involves generating air density information based on the collected compressor outlet pressure and temperature signals. Air density refers to the density of air at the compressor outlet, typically expressed in kilograms per cubic meter (kg / m³), and can be calculated using the ideal gas law: ρ_air = P_c / (R·T_c), where P_c is the compressor outlet pressure (converted to Pa), T_c is the compressor outlet temperature (converted to K), and R is the gas constant for air (287.06 J / (kg·K)). In practice, the executing entity substitutes the collected compressor outlet pressure (converted to Pa) and compressor outlet temperature (converted to K) into the above formula to calculate the air density information. For example, when the compressor outlet pressure is 0.634 MPa (i.e. 634000 Pa) and the compressor outlet temperature is 258℃ (i.e. 531 K), the air density is approximately 634000 / (287.06×531)≈4.16 kg / m³.

[0090] The third step involves generating fuel density information based on the collected total fuel flow signal. Fuel density refers to the density of the hydrogen fuel, typically expressed in kilograms per cubic meter (kg / m³), and can be calculated based on hydrogen properties and fuel temperature and pressure. In practice, the aforementioned entity calculates the fuel density information using a pre-calibrated density-pressure-temperature relationship table or the ideal gas law, based on the hydrogen fuel's temperature (collected by a fuel temperature sensor) and pressure (collected by a fuel pressure sensor). For example, when the fuel pressure is 1.8 MPa and the fuel temperature is 25°C, the fuel density, as found in the hydrogen property table, is approximately 1.2 kg / m³.

[0091] The fourth step involves generating the micro-premixed pipe outlet gas velocity based on the aforementioned air density and fuel density information. This gas velocity refers to the flow rate of the mixture ejected from the micro-premixed pipe outlet after the air and hydrogen have mixed within the pipe. It is typically measured in meters per second (m / s) and is a key parameter for assessing backfire risk. In practice, the executing entity first calculates the air velocity based on the air density and the geometric parameters of the micro-premixed pipe (such as the total flow area), then calculates the fuel velocity based on the fuel density and the pipe's geometric parameters. The two are then superimposed to obtain the gas velocity. The specific calculation formula is: V_mix = (W_a / (ρ_air·A_total)) + (W_f / (ρ_fuel·A_total)), where W_a is the total air flow rate (obtainable from the compressor characteristic curve), W_f is the total fuel flow rate, and A_total is the total flow area of ​​all micro-premixed pipes. For example, when the air density is 4.16 kg / m³, the air flow rate is 128 kg / s, and the total flow area is 0.3 m², the air velocity is approximately 102.6 m / s; when the fuel density is 1.2 kg / m³ and the fuel flow rate is 0.295 kg / s (i.e., 1064 kg / h), the fuel velocity is approximately 0.82 m / s; and the mixed gas velocity is approximately 103.4 m / s.

[0092] The fifth step involves real-time acquisition of combustion chamber pressure, temperature, and fuel equivalence ratio signals. The combustion chamber pressure signal refers to the internal pressure value of the combustion chamber, typically measured in megapascals (MPa), acquired in real-time by a pressure sensor installed on the combustion chamber casing. The combustion chamber temperature signal refers to the internal temperature value of the combustion chamber, typically measured in degrees Celsius (°C), acquired in real-time by a thermocouple installed at the combustion chamber outlet or main nozzle. The fuel equivalence ratio signal is the ratio of the actual fuel-to-air ratio to the stoichiometric ratio, reflecting the concentration of the combustion mixture and is a crucial factor affecting the risk of backfire. In practice, the aforementioned actuators continuously acquire these signals at a preset sampling frequency, providing input parameters for calculating the backfire risk factor. For example, under load conditions, the combustion chamber pressure is 0.634 MPa, the combustion chamber temperature is 260°C, and the fuel equivalence ratio is 0.5.

[0093] Step 6: Based on the micro-premixed pipe outlet gas flow rate, combustion chamber pressure signal, combustion chamber temperature signal, and fuel equivalence ratio signal, a backfire risk coefficient is generated. The backfire risk coefficient is a dimensionless value calculated from multiple combustion parameters, used to quantify the risk of backfire under the current operating conditions; a higher value indicates a higher risk. In practice, the aforementioned entity calculates the backfire risk coefficient in the following way: First, it determines whether the micro-premixed pipe outlet gas flow rate is within a safe range. According to the burner design characteristics, there is a critical safe value for the micro-premixed pipe outlet gas flow rate (e.g., 100 m / s). When the actual flow rate is higher than this critical value, the backfire risk is lower; when the actual flow rate is lower than this critical value, the lower the flow rate, the higher the backfire risk. Specifically, the ratio of the current flow rate to the critical safe value is calculated. The closer the flow rate is to or below the critical value, the larger the flow rate safety factor (e.g., gradually increasing from 0.5 to 2.0). Second, it determines whether the current fuel equivalence ratio is in the easy-backfire range. Based on the combustion characteristics of hydrogen, the flame propagation speed is fastest and the risk of backfire is highest when the fuel equivalence ratio is close to the stoichiometric ratio (i.e., the hydrogen and air ratio is exactly at full combustion). When the equivalence ratio is lean (less hydrogen) or rich (more hydrogen), the flame propagation speed slows down and the risk of backfire decreases. Specifically, the deviation of the current equivalence ratio from the stoichiometric ratio is calculated. The greater the deviation, the smaller the equivalence ratio correction factor (e.g., decreasing from 1.0 to 0.3), and the smaller the deviation, the larger the correction factor. Next, it is determined whether the combustion chamber pressure and temperature are in the high-risk range. According to the principles of combustion kinetics, the higher the combustion chamber pressure and temperature, the faster the chemical reaction rate, the faster the flame propagation speed, and the higher the risk of backfire. Specifically, the current combustion chamber pressure is compared with the design reference pressure, and the current combustion chamber temperature is compared with the design reference temperature. The higher the pressure or temperature, the larger the thermodynamic correction factor (e.g., gradually increasing from 0.5 to 2.0). Finally, the above three factors are multiplied to obtain the backfire risk coefficient. For example, when the gas velocity at the outlet of the micro-premixed pipe is 80 m / s (below the critical value of 100 m / s), the velocity safety factor is 1.5; when the fuel equivalence ratio is 0.9 (close to the stoichiometric ratio of 1.0), the equivalence ratio correction factor is 0.9; when the combustion chamber pressure is 0.7 MPa (above the reference pressure of 0.5 MPa) and the combustion chamber temperature is 800 K (above the reference temperature of 500 K), the thermal correction factor is 1.6. Multiplying these three factors together yields a flashback risk coefficient of 1.5 × 0.9 × 1.6 = 2.16. Multiplying this by a normalization factor of 0.3 (to ensure the risk coefficient falls within the range of 0 to 1) gives a final risk coefficient of approximately 0.65, indicating a medium-to-high level of flashback risk, requiring the activation of protective measures. When the gas velocity at the outlet of the micro-premixed pipe increases to 120 m / s, the velocity safety factor decreases to 0.7, and with other conditions remaining unchanged, the final risk coefficient decreases to approximately 0.30, indicating that the flashback risk is now within a safe range.

[0094] Step 7: In response to the generated backfire risk coefficient exceeding a first preset risk threshold, an inlet guide vane opening adjustment command is generated. The first preset risk threshold is a pre-set critical value for the risk coefficient used to trigger inlet guide vane adjustment, for example, it can be set to 0.6. The inlet guide vane opening adjustment command is a control signal used to control the compressor inlet guide vane actuator to change the opening, typically including a target opening value or the amount of opening change. In practice, the actuator compares the calculated backfire risk coefficient with the first preset risk threshold. When the backfire risk coefficient exceeds the threshold, an inlet guide vane opening adjustment command is generated. For example, when the backfire risk coefficient is 0.7, exceeding the threshold of 0.6, the control system generates an adjustment command to increase the inlet guide vane opening.

[0095] Step 8: Based on the aforementioned inlet guide vane opening adjustment command, control the inlet guide vane actuator to increase the airflow into the combustion chamber. The inlet guide vane refers to the adjustable blade installed at the compressor inlet. By changing the blade angle, the airflow into the compressor can be adjusted, thus affecting the air supply to the combustion chamber. The inlet guide vane actuator is the device that drives the inlet guide vane to change its angle, typically using electric or hydraulic drive. In practice, the aforementioned actuator sends the inlet guide vane opening adjustment command to the inlet guide vane actuator. The actuator drives the inlet guide vane to rotate in the direction of increasing the opening, increasing the airflow into the compressor, thereby increasing the gas velocity at the outlet of the micro-premixed pipe and reducing the risk of backfire. For example, when the inlet guide vane opening adjustment command is to increase by 5%, the actuator drives the blade angle to change, increasing the airflow by approximately 5%, and the gas velocity at the outlet of the micro-premixed pipe increases accordingly.

[0096] Step nine involves increasing the airflow into the combustion chamber in response to the control mechanism of the inlet guide vane, generating a post-adjustment backfire risk coefficient. This post-adjustment backfire risk coefficient is the backfire risk coefficient recalculated using the method from step six after increasing the airflow, and is used to evaluate the adjustment effect. In practice, after the inlet guide vane actuator completes its action, the actuator waits for a preset stabilization time (e.g., 5 seconds), then re-collects relevant parameters and recalculates the post-adjustment backfire risk coefficient. For example, after increasing the airflow, the outlet gas velocity of the micro-premixed pipe increases from 103.4 m / s to 115 m / s, and the recalculated backfire risk coefficient decreases from 0.7 to 0.3.

[0097] Step 10: In response to the generated adjusted backfire risk coefficient exceeding a second preset risk threshold, a fuel flow reduction command is generated. The second preset risk threshold is a pre-set risk coefficient threshold lower than a first preset risk threshold, for example, 0.5. It is used to determine whether increasing the airflow is sufficient to reduce the risk. If the adjusted risk coefficient still exceeds the second threshold, it indicates that simply increasing the airflow is insufficient to reduce the risk to a safe level, and further fuel flow reduction is required. The fuel flow reduction command is a control signal used to control the first or second regulating valve to reduce its opening. In practice, the aforementioned executing entity compares the adjusted backfire risk coefficient with the second preset risk threshold. If it still exceeds the second threshold, a fuel flow reduction command is generated. For example, if the adjusted backfire risk coefficient is 0.55, still exceeding the second threshold of 0.5, the control system generates a fuel flow reduction command.

[0098] Step 11: Based on the aforementioned fuel flow reduction command, control the first or second regulating valve to reduce the fuel supply flow. In practice, the aforementioned actuator selects to reduce primary or secondary fuel according to the current operating conditions, sending the fuel flow reduction command to the corresponding regulating valve actuator. The actuator drives the valve to reduce its opening, thereby reducing the fuel supply and lowering the combustion chamber heat load and backfire risk. For example, in fully premixed combustion mode, the control system selects to reduce secondary fuel, decreasing the opening of the second regulating valve from 26.8% to 24%, reducing the total fuel flow by approximately 10%, further lowering the backfire risk to a safe level.

[0099] Steps one through eleven of this disclosure are an inventive point of this disclosure, which solves the technical problem that "in the fully premixed combustion mode of pure hydrogen gas turbine, there is a lack of a graded active protection method that can quantify the risk of backfire in real time and actively adjust the inlet guide vanes and fuel supply, resulting in the inability to predict the risk of backfire in advance, the lack of active intervention means after the risk occurs, and the combustion oscillation caused by improper adjustment". Existing technologies for backfire protection have the following shortcomings: First, the gas velocity at the outlet of the micro-premixed pipe cannot be sensed online in real time, relying only on the theoretical velocity value during the design phase, which cannot track the impact of actual operating condition changes on backfire risk. Second, backfire risk assessment relies on a single parameter (such as velocity or equivalence ratio), lacking a multi-parameter comprehensive quantitative method based on physical principles. It fails to integrate factors such as compressor outlet pressure, temperature, combustion chamber pressure, temperature, and fuel equivalence ratio into a unified risk indicator, leading to inaccurate risk judgment. Third, when the risk of backfire increases, existing technologies lack active adjustment means, relying only on passive protection (such as triggering shutdown after backfire occurs), which cannot intervene in the early stages of risk. Fourth, even if adjustment means exist, there is a lack of a multi-level coordinated protection mechanism of "increasing gas first, then decreasing gas," and unilateral adjustment may lead to combustion oscillation or sudden load changes, affecting the stable operation of the unit. Solving the above problems would enable real-time quantification of backfire risk under fully premixed combustion conditions. Furthermore, by implementing graded active protection measures—from increasing the inlet guide vane opening to reducing fuel flow—the backfire risk could be kept within a safe range, achieving safe and stable fully premixed operation of the pure hydrogen gas turbine. To achieve this, this disclosure proposes the following steps: First, in response to the generation of the aforementioned fully premixed combustion mode confirmation signal, real-time acquisition of compressor outlet pressure, compressor outlet temperature, and total fuel flow signals is conducted. This establishes a real-time perception basis for the compressor's air supply status and fuel supply quantity, solving the problem of insufficient accurate input parameters for backfire risk assessment. Second, based on the acquired compressor outlet pressure and temperature signals, air density information is generated. This converts pressure and temperature into the key physical parameter of air density, providing a basis for accurately calculating the air velocity at the micro-premixed pipe outlet and solving the problem of missing density parameters in velocity calculation. Third, based on the acquired total fuel flow signal, fuel density information is generated. Therefore, by calculating the hydrogen density based on fuel temperature and pressure, a foundation is provided for accurately calculating the fuel flow rate at the micro-premixed pipe outlet, solving the problem of difficulty in obtaining real-time hydrogen density information as operating conditions change. The fourth step involves generating the micro-premixed pipe outlet mixed flow rate based on the aforementioned air and fuel density information. Thus, by calculating the micro-premixed pipe outlet mixed flow rate in real time, online sensing of key parameters for backfire risk is achieved for the first time, solving the problem that traditional methods cannot obtain flow rates in real time and can only rely on theoretical design values. The fifth step involves real-time acquisition of combustion chamber pressure signals, combustion chamber temperature signals, and fuel equivalence ratio signals.This establishes a real-time sensing foundation for the thermal state and mixture concentration inside the combustion chamber, addressing the limitation of relying solely on flow rate as a single parameter for risk assessment. The sixth step generates a backfire risk coefficient based on the aforementioned micro-premixed pipe outlet flow rate, combustion chamber pressure signal, combustion chamber temperature signal, and fuel equivalence ratio signal. This integrates and quantifies the multi-dimensional factors affecting backfire (flow rate, pressure, temperature, and equivalence ratio) into a unified evaluation index, resolving the issues of difficulty in quantifying backfire risk and ambiguous judgment criteria. The seventh step generates an inlet guide vane opening adjustment command in response to the generated backfire risk coefficient exceeding a first preset risk threshold. This triggers active adjustment at the first sign of increased backfire risk, addressing the lack of timely intervention after risk occurrence. The eighth step, based on the aforementioned inlet guide vane opening adjustment command, controls the inlet guide vane actuator to increase the airflow into the combustion chamber. This increases the airflow and improves the micro-premixed pipe outlet flow rate, reducing the backfire risk coefficient to a safe level and achieving active suppression of backfire risk. Step 9: In response to the control of the inlet guide vane actuator to increase the airflow into the combustion chamber, a backfire risk coefficient is generated after adjustment. This allows for real-time evaluation of the adjustment effect, providing a quantitative basis for determining whether further intervention is needed, and solving the problem of not being able to determine whether the risk has been eliminated after adjustment. Step 10: In response to the generated backfire risk coefficient exceeding a second preset risk threshold, a fuel flow reduction command is generated. Therefore, when increasing the airflow is still insufficient to reduce the risk to a safe level, a second-level protection is activated, solving the problem of insufficient single-stage adjustment capability and continued risk. Step 11: Based on the aforementioned fuel flow reduction command, the first or second regulating valve is controlled to reduce the fuel supply flow. This further reduces the combustion chamber heat load by reducing the fuel flow, forming a multi-level active protection mechanism of "increasing airflow first, then decreasing it," ensuring that the backfire risk is completely eliminated. In summary, steps one through eleven of this embodiment collaborate with each other, starting from the entire chain of "real-time flow rate calculation, multi-parameter risk modeling, and graded active control." By constructing a closed-loop, graded, and progressive active protection framework for backfire risk, it achieves real-time quantitative assessment and active suppression of backfire risk in the fully premixed combustion mode of a pure hydrogen gas turbine. This effectively solves the problems of traditional technologies, such as the inability to perceive backfire risk in real time, lack of active intervention, and limited adjustment methods. It ensures that backfire risk is always controlled within a safe range during fully premixed operation, providing key technical support for the safe and stable operation of pure hydrogen gas turbines.

[0100] The embodiments disclosed herein have the following beneficial effects: The hydrogen gas turbine combustion mode switching and control methods of some embodiments of this disclosure can achieve accurate identification and stable control of the combustion mode switching process, significantly improving the reliability and operational stability of the hydrogen gas turbine during the staged combustion transition phase. Specifically, traditional hydrogen gas turbine combustion mode switching technologies (such as existing schemes that rely on speed thresholds or exhaust temperature for judgment) may face problems such as the inability to accurately determine whether premixed combustion has been stably established after the first-stage premixed fuel is introduced, the lack of multi-dimensional verification when the shift acceleration stage fuel is shut off, and the lack of effective confirmation methods after the full premixed mode is established. If only a single parameter judgment or open-loop control is relied upon, there may be risks of combustion oscillation, flameout, or even backfire caused by improper switching timing, ultimately leading to the unit's inability to operate safely and stably. Based on this, the hydrogen gas turbine combustion mode switching and control methods of some embodiments of this disclosure firstly acquire the rotational speed signal of the hydrogen gas turbine rotor in real time, and then execute an ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold. This provides a clear rotational speed reference for subsequent staged fuel introduction, ensuring the safety and reliability of the ignition phase. Then, in response to the collected rotational speed signal reaching the second rotational speed threshold, the primary fuel shut-off valve is controlled to supply primary premixed fuel to the combustion chamber. A flame temperature change rate signal is generated based on the collected flame temperature signal, and fluctuation identification is performed on this signal to obtain the fluctuation identification result. Thus, by directly monitoring the flame temperature change in the combustion chamber, the dynamic response of the combustion state after the primary premixed fuel is introduced can be captured in real time, providing a direct physical basis for subsequent judgments. Next, in response to the obtained fluctuation identification result indicating a characteristic fluctuation of flame temperature that first rises and then falls, the first regulating valve is adjusted. Therefore, using the "rise then fall" characteristic fluctuation as a marker of stable premixed combustion, the first regulating valve is immediately closed-loop regulated upon identification of this characteristic fluctuation, ensuring rapid flame temperature stabilization and avoiding the risk of combustion oscillation or flameout due to inaccurate switching judgments. Then, in response to the collected rotational speed signal reaching the third rotational speed threshold, the secondary fuel shut-off valve and the second regulating valve are controlled. This enables on-demand introduction of secondary premixed fuel, providing a fuel base for full-speed no-load and loaded operation. Next, in response to the stable operating condition confirmation signal detected after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is closed. Thus, by confirming a stable operating condition through multi-parameter fusion judgment (speed, fuel flow, exhaust temperature) before shutting off the shift fuel, combustion is ensured to be in a steady state upon shift fuel withdrawal, achieving uninterrupted shift fuel withdrawal. Then, the collected flame temperature drop signal is analyzed for flame temperature identification, yielding the flame temperature identification result. Therefore, by directly monitoring the flame temperature change after shift fuel shutdown, real-time feedback is provided for establishing the full premixed mode.Finally, in response to the flame temperature identification result indicating that the flame temperature is stable, a confirmation signal for the fully premixed combustion mode is generated. Thus, by using a dual-condition judgment method—calculating the average temperature and standard deviation separately through sliding window segmentation—both the absolute temperature level meets the target and the convergence of temperature fluctuations is guaranteed, ultimately confirming that the fully premixed combustion mode has been stably established. Furthermore, because this method introduces a real-time identification and closed-loop adjustment mechanism based on directly measured parameters in the combustion chamber throughout the entire combustion mode switching process (first-stage input, second-stage input, shift shutdown, and full premix confirmation), it can effectively adapt to the characteristics of rapid flame propagation and high reactivity of hydrogen fuel, and possesses inherent suppression capabilities for transient disturbances during the switching process, thereby enhancing robustness and reliability under real-world complex operating conditions. Simultaneously, through the synergistic optimization of staged fuel input, characteristic fluctuation identification, multi-parameter stable operating condition judgment, and temperature drop confirmation, the entire switching process can be kept consistent with the final fully premixed operation target, ensuring combustion stability across the entire operating range. Thus, by combining characteristic fluctuation identification, closed-loop regulation, multi-parameter fusion judgment and temperature drop confirmation, the overall reliability, stability and safety of hydrogen gas turbine combustion mode switching process are improved to a certain extent, providing key technical support for hydrogen gas turbine to achieve low emission and high stability operation in a wide range of operating conditions.

[0101] Further reference Figure 4 , Figure 4 An internal test diagram of the fuel control valve in an embodiment of this disclosure is shown. The fuel control valve refers to the first or second regulating valve described in this embodiment, used to control the flow rate of hydrogen fuel entering the combustion chamber to achieve continuous regulation of the fuel flow rate. The design of this fuel control valve enables rapid and precise adjustment of the fuel supply flow rate based on flame temperature deviations or pulsating pressure changes during the switching of combustion modes in the hydrogen gas turbine, providing a reliable basis for combustion stability control.

[0102] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a hydrogen gas turbine combustion mode switching and control device, which are similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0103] like Figure 2As shown, the hydrogen gas turbine combustion mode switching and control device 200 in some embodiments includes: a data acquisition unit 201, an ignition unit 202, a first execution unit 203, and a second execution unit 204. The data acquisition unit 201 is configured to acquire the rotational speed signal of the hydrogen gas turbine rotor in real time; the ignition unit 202 is configured to perform an ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold; the first execution unit 203 is configured to perform the following steps in response to the acquired rotational speed signal reaching a second rotational speed threshold: controlling a first-stage fuel shut-off valve to introduce first-stage premixed fuel into the combustion chamber; generating a flame temperature change rate signal based on the acquired flame temperature signal, and performing fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result; and characterizing the identified flame temperature change rate signal based on the obtained fluctuation identification result. The speed exhibits a characteristic fluctuation of first rising and then falling, and the first regulating valve is adjusted accordingly. The second execution unit 204 is configured to perform the following steps in response to the collected speed signal reaching the third speed threshold: regulating the secondary fuel cut-off valve and the second regulating valve; in response to the identification of a stable operating condition confirmation signal after regulating the secondary fuel cut-off valve and the second regulating valve, controlling the duty acceleration stage fuel cut-off valve to close; identifying the flame temperature by performing flame temperature identification on the collected flame temperature drop signal, and obtaining the flame temperature identification result; in response to the obtained flame temperature identification result indicating that the flame temperature is stable, generating a fully premixed combustion mode confirmation signal.

[0104] It is understandable that the units described in the device 200 are related to the reference. Figure 1 The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 200 and the units contained therein, and will not be repeated here.

[0105] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0106] like Figure 3 As shown, the electronic device 300 may include a processing unit 301 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0107] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0108] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0109] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0110] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0111] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire the rotational speed signal of the hydrogen gas turbine rotor in real time; execute an ignition operation in response to the acquired rotational speed signal reaching a first rotational speed threshold; and execute the following steps in response to the acquired rotational speed signal reaching a second rotational speed threshold: controlling the first-stage fuel shut-off valve to introduce first-stage premixed fuel into the combustion chamber; generating a flame temperature change rate signal based on the acquired flame temperature signal, and performing fluctuation identification on the generated flame temperature change rate signal to obtain a fluctuation identification result; and responding to the obtained fluctuation... The identification result indicates that the flame temperature exhibits a characteristic fluctuation of first rising and then falling, and the first regulating valve is adjusted accordingly. In response to the collected speed signal reaching the third speed threshold, the following steps are executed: the secondary fuel shut-off valve and the second regulating valve are regulated; in response to the stable operating condition confirmation signal identified after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is closed; the collected flame temperature drop signal is used for flame temperature identification to obtain the flame temperature identification result; in response to the obtained flame temperature identification result indicating that the flame temperature is stable, a fully premixed combustion mode confirmation signal is generated.

[0112] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0114] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a data acquisition unit, an ignition unit, a first execution unit, and a second execution unit. The names of these units do not necessarily limit the specific unit itself; for example, a data acquisition unit may also be described as "a unit that acquires the rotational speed signal of a hydrogen gas turbine rotor in real time."

[0115] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0116] Some embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements any of the above-described hydrogen gas turbine combustion mode switching and control methods.

[0117] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for switching and controlling combustion modes in a hydrogen gas turbine, comprising: Real-time acquisition of the rotational speed signal of the hydrogen gas turbine rotor; When the collected speed signal reaches the first speed threshold, an ignition operation is performed; In response to the acquired speed signal reaching the second speed threshold, the following steps are executed: Control the primary fuel shut-off valve to supply primary premixed fuel to the combustion chamber; Based on the collected flame temperature signal, a flame temperature change rate signal is generated, and fluctuation identification is performed on the generated flame temperature change rate signal to obtain the fluctuation identification result. In response to the obtained fluctuation identification results, which indicate that the flame temperature first rises and then falls, the first regulating valve is adjusted. In response to the collected speed signal reaching the third speed threshold, the following steps are executed: The secondary fuel shut-off valve and the second regulating valve are regulated; In response to the identification of a stable operating condition confirmation signal after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is controlled to close. The collected flame temperature drop signal is used to identify the flame temperature, and the flame temperature identification result is obtained. In response to the obtained flame temperature identification results indicating that the flame temperature is stable, a fully premixed combustion mode confirmation signal is generated.

2. The method of claim 1, wherein, The process involves generating a flame temperature change rate signal based on the acquired flame temperature signal, and performing fluctuation identification on the generated flame temperature change rate signal to obtain fluctuation identification results, including: Based on the collected flame temperature signal, flame temperature time series data is generated; The flame temperature time series data is differentially processed to obtain the flame temperature change rate time series data; The time series data of the flame temperature change rate is subjected to extreme point identification processing to obtain positive and negative extreme points; Based on the obtained positive and negative extreme points, fluctuation identification results are generated.

3. The method of claim 1, wherein, The response to the obtained fluctuation identification result, which indicates that the flame temperature exhibits a characteristic fluctuation of first rising and then falling, and the adjustment of the first regulating valve includes: Based on the collected flame temperature signal and the preset target temperature, a temperature deviation signal is generated; The temperature deviation signal is proportionally amplified to obtain a proportional control quantity; The temperature deviation signal is processed by time integration to obtain the integral control quantity; The temperature deviation signal is subjected to time differentiation processing to obtain the differential control quantity; The obtained proportional control quantity, integral control quantity and derivative control quantity are superimposed to obtain the PID control quantity of the first regulating valve; Based on the PID control amount of the first control valve, an opening control command for the first control valve is generated. Based on the opening adjustment command, the first regulating valve is controlled to change the supply flow rate of the primary premixed fuel.

4. The method of claim 1, wherein, The regulation of the secondary fuel shut-off valve and the second regulating valve includes: Control the secondary fuel shut-off valve to supply secondary premixed fuel to the combustion chamber; Based on the collected combustion chamber pulsating pressure signal, pulsating pressure time series data of pulsating pressure amplitude changing with time is generated; Peak detection processing is performed on the generated pulsating pressure time series data to obtain the pulsating pressure peak sequence; In response to any pulsating pressure peak exceeding a preset pulsating pressure threshold in the obtained pulsating pressure peak sequence, the following steps are performed: Generate a first coordinated adjustment amount for the first regulating valve and a second coordinated adjustment amount for the second regulating valve; Based on the first coordinated adjustment amount, the first regulating valve is controlled to change the supply flow rate of the primary premixed fuel. Based on the second coordinated adjustment amount, the second regulating valve is controlled to change the supply flow rate of the secondary premixed fuel; In response to the fact that the amplitude of each adjusted pulsating pressure collected within the first preset time window is lower than the preset pulsating pressure threshold, a second regulating valve gradually opens command is generated. In response to the fact that none of the pulse pressure peaks in the obtained pulse pressure peak sequence exceed the preset pulse pressure threshold, a second regulating valve gradually opens command is generated. Based on the generated second regulating valve gradually opening command, the second regulating valve is controlled to gradually open until the corresponding target opening degree is reached.

5. The method of claim 1, wherein, The method further includes: In response to the adjustment of the secondary fuel shut-off valve and the second regulating valve, the following steps are performed in parallel: Based on the acquired rotational speed signal, perform the following steps: Generate the first rotational speed determination result; In response to the generated first speed judgment result indicating that the rated speed has been reached, the instantaneous speed values ​​within the second preset time window are collected; The instantaneous rotational speed values ​​collected are processed to obtain the standard deviation of rotational speed within the second preset time window; In response to the obtained speed standard deviation being less than a preset speed fluctuation threshold, a first stable sub-signal is generated; Based on the collected total fuel flow signal, perform the following steps: Generate the first flow judgment result; In response to the generated first flow judgment result indicating that the target flow value has been reached, the instantaneous flow values ​​of each flow within the third preset time window are collected; The instantaneous flow values ​​collected are processed to obtain the standard deviation of flow within the third preset time window; In response to the obtained standard deviation of the flow rate being less than the preset flow fluctuation threshold, a second stable sub-signal is generated; Based on the collected exhaust temperature signal, the following steps are performed: Obtain the instantaneous values ​​of each exhaust temperature within the fourth preset time window; The obtained instantaneous exhaust temperature values ​​are processed to obtain the standard deviation of exhaust temperature within the fourth preset time window. A third stable sub-signal is generated in response to the obtained exhaust temperature standard deviation being less than a preset exhaust temperature fluctuation threshold. In response to the fact that the generation timestamps corresponding to the first stable sub-signal, the second stable sub-signal, and the third stable sub-signal are all within the same fifth preset time window, a stable operating condition confirmation signal is generated.

6. The method of claim 1, wherein, The process of identifying the flame temperature from the collected flame temperature drop signal to obtain the flame temperature identification result includes: Based on the collected flame temperature drop signal, flame temperature drop time series data is generated; The flame temperature decrease time series data is processed by sliding window segmentation to obtain each decrease time series data subsequence; Each of the falling time series data subsequences is processed by flame temperature sampling points to obtain the average flame temperature and standard deviation of flame temperature for each falling time series data subsequence. In response to the existence of a first preset number of decreasing time-series data subsequences where the average flame temperature of each corresponding to a certain number of consecutive decreasing time-series data subsequences is lower than a preset average flame temperature threshold, a first stable judgment signal is generated. In response to the existence of a number of consecutive second-preset subsequences of decreasing time-sequence data where the standard deviation of each flame temperature is lower than a preset flame temperature standard deviation threshold, a second stability judgment signal is generated. In response to the generation of both the first and second stability judgment signals, a flame temperature identification result representing the identification of a stable flame temperature is generated.

7. A combustion mode switching and control device for a hydrogen gas turbine, comprising: The acquisition unit is configured to acquire the rotational speed signal of the hydrogen gas turbine rotor in real time; The ignition unit is configured to perform an ignition operation in response to the collected speed signal reaching a first speed threshold. The first execution unit is configured to perform the following steps in response to the acquired rotational speed signal reaching a second rotational speed threshold: Control the primary fuel shut-off valve to supply primary premixed fuel to the combustion chamber; Based on the collected flame temperature signal, a flame temperature change rate signal is generated, and fluctuation identification is performed on the generated flame temperature change rate signal to obtain the fluctuation identification result. In response to the obtained fluctuation identification results, which indicate that the flame temperature first rises and then falls, the first regulating valve is adjusted. The second execution unit is configured to perform the following steps in response to the acquired rotational speed signal reaching a third rotational speed threshold: The secondary fuel shut-off valve and the second regulating valve are regulated; In response to the identification of a stable operating condition confirmation signal after regulating the secondary fuel shut-off valve and the second regulating valve, the shift acceleration stage fuel shut-off valve is controlled to close. The collected flame temperature drop signal is used to identify the flame temperature, and the flame temperature identification result is obtained. In response to the obtained flame temperature identification results indicating that the flame temperature is stable, a fully premixed combustion mode confirmation signal is generated.

8. An electronic device, comprising: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 6.

9. A computer readable medium having stored thereon a computer program, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 6.

Images