Low-specific-speed centrifugal compressor adaptive to hydrogen-containing fuel gas turbine power generation system and optimization method of low-specific-speed centrifugal compressor

Abstract

The invention relates to the technical field of gas turbine power generation, in particular to a low-specific-speed centrifugal compressor adaptive to a hydrogen-containing fuel gas turbine power generation system and an optimization method of the low-specific-speed centrifugal compressor. Comprising a meridian flow channel assembly, an impeller and blade assembly and a self-adaptive treatment casing assembly which are coaxially arranged and jointly form a closed pressurized flow field, and the adaptation to a hydrogen-containing fuel gas turbine power generation system is achieved. Through cooperation of the three components, the core pain points that an existing gas compressor is insufficient in stability, low in efficiency and poor in adaptability in a hydrogen-containing fuel system are structurally solved, and an efficient and stable pressurization foundation is provided for a hydrogen-containing fuel gas turbine power generation system.

Description

Technical Field

[0001] The present invention relates to the technical field of gas turbine power generation, and in particular to a low specific speed centrifugal compressor adapted to a hydrogen-fueled gas turbine power generation system and an optimization method thereof. Background Art

[0002] In hydrogen-fueled gas turbine power generation systems, the compressor is a core component whose stability and efficiency directly affect the performance of the entire system. In existing technologies, compressors have the following problems: Insufficient stability: Non-axisymmetric flow passages and flow distortion at the inlet and outlet can easily lead to compressor instability and large pressure fluctuations; Low efficiency: Under low flow conditions, the proportion of blade tip clearance to blade height increases, resulting in a significant increase in tip leakage loss, secondary flow loss, and boundary layer loss, making it difficult to strike a balance between efficiency and stability margin. Poor adaptability: The design technology for stabilizing and increasing the efficiency of low-speed compressors under multiple operating conditions is immature, making it difficult to match the power requirements of hydrogen-fueled gas turbine power generation units (such as 60kW-class systems).

[0003] Furthermore, the unique properties of hydrogen-containing fuels and the complex operating conditions of gas turbine power generation systems further exacerbate the performance shortcomings of existing centrifugal compressors. The rapid combustion velocity and high flame temperature of hydrogen-containing fuels place higher demands on compressor stability and load capacity. Furthermore, factors such as the asymmetric flow path in the volute and flow distortion at the inlet and outlet can easily cause compressor instability, thereby impacting the normal operation of the power generation system.

[0004] In summary, developing a low-specific-speed centrifugal compressor and its optimization method that are suitable for hydrogen-fueled gas turbine power generation systems, and effectively solving the stability, efficiency and adaptability problems of existing compressors, has become a technical challenge that needs to be overcome urgently. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a low specific speed centrifugal compressor adapted to a hydrogen-fueled gas turbine power generation system and an optimization method thereof, so as to solve the stability, efficiency and adaptability problems existing in the existing compressors in the prior art, which has become a technical problem that needs to be overcome urgently. According to a first aspect of an embodiment of the present invention, a low-speed centrifugal compressor adapted to a hydrogen-fueled gas turbine power generation system is provided, comprising: a meridian flow channel assembly, which is a fixed airflow channel frame, forming a continuous flow channel for airflow circulation; an impeller and blade assembly, which is a rotating supercharging component, which is integrally arranged inside the meridian flow channel assembly, coaxially assembled with the meridian flow channel assembly and maintaining a preset gap between the two, so as to supercharge the airflow flowing into the meridian flow channel assembly by rotation; an adaptive processing casing assembly, which is a flow field control component, which is arranged around the outside of the meridian flow channel assembly, rigidly connected to the meridian flow channel assembly and coaxially assembled, and its inner side is arranged corresponding to the impeller and blade assembly and maintains an adjustable gap, so as to control the airflow state in real time; the meridian flow channel assembly, the impeller and blade assembly, and the adaptive processing casing assembly are coaxially arranged to form a closed supercharging flow field, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

[0006] Furthermore, the meridian flow channel assembly includes: The meridian flow channel assembly includes an inlet section, an impeller section, a diffuser section and an outlet section in sequence along the airflow direction, and each section forms a continuous gradually diverging flow channel; The inlet section is rigidly connected to the air inlet pipe via a flange; The impeller section is a cylindrical cavity, the inner wall of which maintains a preset initial gap with the impeller tip, and is connected to the inlet section through a smoothly transitioned curved surface; The flow channel width of the diffuser section is greater than that of the inlet section, and diffuser blades are fixedly mounted on the inner wall, with a regulating gap reserved between the top of the blade and the inner wall of the adaptive processing casing; The outlet section is a diffusion structure, smoothly connected to the end of the diffuser section, and the outlet is connected to the combustion chamber intake pipe through a flange.

[0007] Furthermore, the meridian flow channel assembly further includes: The entire meridian flow channel has a preset curvature radius and meridian plane inclination angle, and each section remains coaxially arranged.

[0008] Furthermore, the impeller and blade assembly includes: The impeller is integrally milled from a high-strength alloy, the hub is rigidly connected to the compressor main shaft via a keyway, and both ends of the main shaft are supported by bearing seats; The blade root is connected to the mortise and tenon of the impeller rim through a dovetail tenon, the blade has a preset sweep angle, and the blade top maintains a preset initial gap with the inner wall of the adaptive processing casing; The impeller and blade assembly are integrally located in the impeller section of the meridian flow channel.

[0009] Furthermore, the adaptive processing casing assembly comprises: a casing body, which is precision-casted from cast iron and has an annular sleeve structure, and is rigidly connected to the outer shell flange of the meridian flow channel by bolts; Multiple annular movable guide vanes are evenly distributed on the inner wall of the casing inlet section. The root is hinged to the casing body through a micro shaft and can rotate around the shaft within a preset angle range. The end is connected to the actuator's micro hydraulic push rod ball joint through a connecting rod mechanism. The radially adjustable clearance mechanism is an annular slider structure embedded in the inner wall groove of the casing near the impeller tip. The inner side of the slider corresponds to the blade tip, and the outer side is slidably connected to the casing body through an elastic guide column and rigidly connected to the output end of another set of micro hydraulic push rods to achieve tip clearance adjustment. Multiple piezoelectric pressure sensors are evenly embedded between the impeller outlet and the diffuser inlet on the inner wall of the casing. The sensor probes are flush with the inner wall of the flow channel. The signal lines are led out through the threading holes of the casing body and electrically connected to the external controller. Two sets of micro hydraulic push rods drive the annular movable guide vane and the radial adjustable gap mechanism respectively, and are fixedly installed on the bracket on the outer wall of the casing. The bracket is welded to the casing, and the push rod control signal line is electrically connected to the controller; the meridian flow channel assembly, impeller and blade assembly, and adaptive processing casing assembly are coaxially arranged to form a closed boost flow field, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

[0010] According to a second aspect of an embodiment of the present invention, a method for optimizing a low specific speed centrifugal compressor adapted for a hydrogen-fueled gas turbine power generation system is provided, the method being applied to any of the above-mentioned low specific speed centrifugal compressors adapted for a hydrogen-fueled gas turbine power generation system, the method comprising: Real-time monitoring of compressor operating parameters in hydrogen-fueled gas turbine power generation systems; the operating parameters include, but are not limited to, flow rate, pressure ratio, efficiency, temperature, pressure fluctuation, and power demand operating parameters of the power generation system monitored at multiple points in different parts of the compressor; Using a sensor fusion algorithm, the preset distributed sensor network is used to fuse the flow rate, pressure ratio, efficiency, temperature, pressure fluctuation, and power demand operating parameters of the power generation system monitored at multiple points in different parts of the compressor to obtain the fused operating parameters; The fuzzy logic algorithm is used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started.

[0011] Furthermore, the fuzzy logic algorithm is used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started, including: The pressure fluctuation in the fused operating parameters is used as the core characterization parameter. A fuzzy logic algorithm is used to compare it with a preset stability standard parameter range. When the pressure fluctuation exceeds the first range, the compressor is determined to have insufficient stability, triggering the first preset rule group. The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface by more than the first threshold for a preset number of sampling cycles, the second rule group is triggered. Using the power generation system power demand and the actual compressor output power as the core parameters in the fused operating parameters, a three-dimensional adaptability evaluation space is constructed, combining efficiency and preset surge margin. The Euclidean distance from the current operating point to the standard adaptation surface is calculated using a fuzzy logic algorithm. When the Euclidean distance exceeds a preset second threshold, the third rule group is triggered.

[0012] Furthermore, the efficiency and flow rate in the fused operating parameters are used as joint characterization parameters, and compared with a preset efficiency standard parameter surface through a fuzzy logic algorithm. When the pressure fluctuation exceeds the first range, it is determined that the compressor has a stability problem, and a preset first rule group is triggered, including: The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with a preset efficiency standard parameter surface through a fuzzy logic algorithm. When the pressure fluctuation exceeds the first range, the model predictive control algorithm is used to control the adaptive processing casing component. The flow change trend in the flow channel is predicted based on the real-time pressure fluctuation, and the guide vane angle or cross-sectional area of ​​the non-axisymmetric flow channel and the inlet and outlet are automatically adjusted. Active control technology based on neural networks is introduced to automatically adjust the control parameters of active vortex control or active airflow injection according to the real-time monitored pressure fluctuation signal through the learning and prediction capabilities of the neural network.

[0013] Furthermore, the efficiency and flow rate in the fused operating parameters are used as joint characterization parameters, and are compared with a preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface for a preset number of sampling periods and exceeds a first threshold, the second rule group is triggered, including: The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters, and are compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface for a preset number of sampling cycles and exceeds the first threshold, a genetic algorithm is used with flow rate changes as input to dynamically adjust the tip clearance size with the goal of reducing tip leakage loss, secondary flow loss and boundary layer loss.

[0014] Furthermore, the optimization process further includes: Computational fluid dynamics software was used to simulate and analyze the optimized compressor model, and a data-driven uncertainty quantification algorithm was used to evaluate the uncertainty of the simulation results. Using reinforcement learning algorithms, the optimization parameters are automatically adjusted based on simulation results and uncertainty assessment. If the expected optimization goal is not achieved, the simulation and adjustment process is repeated until the expected effect is achieved. At the same time, a compressor performance database is established, and a knowledge graph algorithm is used to associate and analyze the operating parameters and optimization strategies before and after each optimization, to explore potential optimization directions and rules, and provide reference for subsequent optimization. The technical solution provided by the embodiment of the present invention can have the following beneficial effects: The present application achieves structural collaborative optimization through the coaxial arrangement of the meridian flow channel assembly, the impeller and blade assembly, and the adaptive processing casing assembly and the closed supercharging flow field design: the continuous and gradually expanding flow channel of the meridian flow channel ensures smooth airflow and reduces basic flow losses; the high-strength alloy integral molding and preset gap design of the impeller and blades improve the rotary supercharging efficiency and structural stability; the adaptive processing casing assembly is arranged in a surround manner and the gap and flow field are adjusted in real time, which can specifically suppress the instability problems caused by asymmetric flow channels and inlet and outlet flow distortions. The three work together to solve the core pain points of the existing compressor in hydrogen-containing fuel systems, such as insufficient stability, low efficiency and poor adaptability, from a structural level, and provide an efficient and stable supercharging foundation for hydrogen-containing fuel gas turbine power generation systems.

[0015] Furthermore, by monitoring the operating parameters of multiple compressor parts in real time and processing them through a sensor fusion algorithm, single-point data noise interference is eliminated, resulting in comprehensive and accurate fusion parameters. Fuzzy logic algorithms are then used to compare the fusion parameters with preset standards, accurately identifying problems such as insufficient stability, low efficiency, or poor adaptability and triggering an optimization process, thus overcoming the limitations of single-parameter monitoring and judgment. This method can quickly respond to the complex operating conditions of hydrogen-fueled gas turbine power generation systems. By initiating targeted optimization processes and regulating the compressor state in real time, it effectively improves its stability, efficiency, and adaptability to system power requirements under a wide range of operating conditions, providing dynamic optimization guarantees for the efficient and stable operation of hydrogen-fueled power generation systems.

[0016] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0018] Figure 1This is a schematic diagram of a low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system using multimodal data fusion according to an exemplary embodiment; Figure 2 It is a flow chart of a method for optimizing a low specific speed centrifugal compressor adapted to a hydrogen-fueled gas turbine power generation system by multimodal data fusion according to an exemplary embodiment. DETAILED DESCRIPTION

[0019] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0020] Example 1 See also Figure 1 , Figure 1 The present invention is a schematic diagram of a low specific speed centrifugal compressor adapted for a hydrogen-fueled gas turbine power generation system using multimodal data fusion according to an exemplary embodiment. The compressor includes: The meridian flow channel component 03 is a fixed air flow channel frame, which forms a continuous flow channel for air circulation; the impeller and blade component 02 is a rotating supercharging component, which is arranged as a whole inside the meridian flow channel component 03, coaxially assembled with the meridian flow channel component 03 and a preset gap is maintained between the two, so as to supercharge the airflow flowing into the meridian flow channel component 03 through rotation; the adaptive processing casing component 01 is a flow field control component, which is arranged around the outside of the meridian flow channel component 03, rigidly connected to the meridian flow channel component 03 and coaxially assembled, and its inner side is arranged corresponding to the impeller and blade component 02 and maintains an adjustable gap to control the airflow state in real time; the meridian flow channel component 03, the impeller and blade component 02, and the adaptive processing casing component 01 are coaxially arranged to form a closed supercharging flow field together, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

[0021] Furthermore, the meridian flow channel assembly 03 includes: The meridian flow channel assembly 03 includes an inlet section, an impeller section, a diffuser section and an outlet section in sequence along the airflow direction, and each section forms a continuous gradually diverging flow channel; The inlet section is rigidly connected to the air inlet pipe via a flange; The impeller section is a cylindrical cavity, the inner wall of which maintains a preset initial gap with the impeller tip, and is connected to the inlet section through a smoothly transitioned curved surface; The flow channel width of the diffuser section is greater than that of the inlet section, and diffuser blades are fixedly mounted on the inner wall, with a regulating gap reserved between the top of the blade and the inner wall of the adaptive processing casing; The outlet section is a diffusion structure, smoothly connected to the end of the diffuser section, and the outlet is connected to the combustion chamber intake pipe through a flange.

[0022] Furthermore, the meridian flow channel assembly 03 further includes: The entire meridian flow channel has a preset curvature radius and meridian plane inclination angle, and each section remains coaxially arranged.

[0023] In a specific implementation, the impeller and blade assembly 02 includes: The impeller is integrally milled from a high-strength alloy, the hub is rigidly connected to the compressor main shaft via a keyway, and both ends of the main shaft are supported by bearing seats; The blade root is connected to the mortise and tenon of the impeller rim through a dovetail tenon, the blade has a preset sweep angle, and the blade top maintains a preset initial gap with the inner wall of the adaptive processing casing; The impeller and blade assembly 02 are integrally located in the impeller section of the meridian flow channel.

[0024] In a specific implementation, the adaptive processing casing assembly 01 includes: a casing body, which is precision-casted with cast iron and has an annular sleeve structure, and is rigidly connected to the outer shell flange of the meridian flow channel by bolts; Multiple annular movable guide vanes are evenly distributed on the inner wall of the casing inlet section. The root is hinged to the casing body through a micro shaft and can rotate around the shaft within a preset angle range. The end is connected to the actuator's micro hydraulic push rod ball joint through a connecting rod mechanism. The radially adjustable clearance mechanism is an annular slider structure embedded in the inner wall groove of the casing near the impeller tip. The inner side of the slider corresponds to the blade tip, and the outer side is slidably connected to the casing body through an elastic guide column and rigidly connected to the output end of another set of micro hydraulic push rods to achieve tip clearance adjustment. Multiple piezoelectric pressure sensors are evenly embedded between the impeller outlet and the diffuser inlet on the inner wall of the casing. The sensor probes are flush with the inner wall of the flow channel. The signal lines are led out through the threading holes of the casing body and electrically connected to the external controller. Two sets of micro hydraulic push rods drive the annular movable guide vane and the radial adjustable gap mechanism respectively, and are fixedly installed on the bracket on the outer wall of the casing. The bracket is welded to the casing, and the push rod control signal line is electrically connected to the controller; the meridian flow channel assembly 03, the impeller and blade assembly 02, and the adaptive processing casing assembly 01 are coaxially arranged to form a closed boost flow field, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

[0025] In one embodiment, the centrifugal compressor structure of the present invention is described in detail below in conjunction with a specific application scenario to reflect its practical application in a 60kW-class hydrogen-fueled gas turbine power generation system: 1. Overall structure and assembly relationship The centrifugal compressor of this embodiment comprises three modules: the meridian flow channel assembly 03, the impeller and blade assembly 02, and the adaptive processing casing assembly 01. These three modules are coaxially assembled (coaxiality is ensured by precision tooling) to form a closed supercharged flow field. Among them: The meridian flow channel assembly 03 serves as a fixed frame, accommodating the impeller and blade assembly 02 and guiding the airflow; The impeller and blade assembly 02 acts as the rotating core, achieving airflow pressurization through high-speed rotation; The adaptive processing casing component 01 surrounds the outside and adjusts the flow field stability in real time.

[0026] 2. Specific implementation of the meridian flow channel component 03 Inlet section: It is made of precision casting of stainless steel and has a convergent structure. It is bolted to the air inlet pipe of the gas turbine power generation system through a flange with a sealing gasket (the flange diameter matches the air inlet pipe) to ensure that the air flow enters the flow channel without leakage.

[0027] Impeller section: It is a cylindrical cavity (cast integrally with the inlet section). The inner wall is precision-ground to maintain an initial gap of 0.3mm with the impeller tip (to adapt to the impeller rotation requirements). The inlet section and the impeller section are transitioned through a smooth curved surface with a radius of 50mm to avoid impact loss when the airflow enters the impeller.

[0028] Diffuser section: The flow channel width gradually expands along the airflow direction (the width of the inlet section is 12mm, and the width of the diffuser section end is 25mm). Eight diffuser blades are welded on the inner wall (the blades are made of TC4 titanium alloy and have an angle of 12°). A 0.2mm gap is reserved between the top of the blade and the inner wall of the adaptive processing casing to take into account the needs of airflow circulation and regulation.

[0029] Outlet section: It is a diffuser cone structure, smoothly welded to the end of the diffuser section, and the outlet flange is connected to the combustion chamber intake duct to achieve stable delivery of the pressurized airflow to the combustion chamber.

[0030] The curvature radius of the entire meridian flow channel is 50mm, the meridian plane inclination angle is 8°, and each section is CNC machined to ensure coaxiality (error ≤ 0.05mm), and is suitable for a speed range of 15,000-30,000r / min.

[0031] 3. Specific implementation of impeller and blade assembly 02 Impeller: Made of Inconel718 high-strength alloy (high temperature and fatigue resistance), the hub is rigidly connected to the 45CrNiMoVA compressor main shaft via a keyway (keyway fit accuracy H7 / h6). Both ends of the main shaft are supported on the bearing seat by angular contact ball bearings (model 7010AC). The bearing seat is fixed to the bracket of the meridian flow channel housing to ensure stable rotation of the impeller.

[0032] Blades: There are 12 blades in total, which are forged from TC11 titanium alloy and then precision milled. The root of the blade is a dovetail tenon, which is connected to the tongue and groove of the impeller rim (the fitting clearance is 0.03mm). The blade sweep angle is designed to be 8°. During the specific implementation, through multi-objective optimization, it was determined that the top of the blade maintains an initial clearance of 0.3mm with the inner wall of the adaptive processing casing to meet the fatigue strength requirements.

[0033] The impeller and blade assembly 02 are integrally installed in the impeller section of the meridian flow channel. When the blades rotate, they push the airflow from the inlet section to the diffuser section, realizing the conversion of kinetic energy into pressure energy.

[0034] 4. Specific implementation of adaptive processing casing assembly 01 Receiver body: Made of HT300 cast iron precision casting, it has an annular sleeve structure and is connected to the flange of the radial flow channel shell through 8 M10 bolts (tightening torque 35N•m). The inner wall is chrome-plated to reduce airflow friction loss.

[0035] Annular movable guide vanes: There are 8 pieces in total (made of aluminum alloy), evenly distributed on the inner wall of the receiver inlet section. The root is hinged to the receiver through a 4mm diameter stainless steel micro-shaft (with a fit clearance of 0.02mm), and can rotate around the shaft within the range of -5° to 10°; the end of the guide vane is connected to the micro hydraulic push rod (model DG10-30) through a connecting rod mechanism and a ball joint to achieve precise angle control.

[0036] Radial adjustable clearance mechanism: It is an annular slider (made of wear-resistant cast iron) embedded in the inner wall groove of the casing near the impeller tip. The outer side of the slider is slidably connected to the casing body through four elastic guide columns (made of spring steel) (sliding accuracy 0.01mm), and is rigidly connected to the output end of another set of micro hydraulic push rods (thrust 80N), which can achieve adjustment of the blade tip clearance within the range of ±0.05mm.

[0037] Pressure sensor: Four piezoelectric pressure sensors (model DYTRAN 3055B) are evenly embedded in the inner wall of the casing (between the impeller outlet and the diffuser inlet). The sensor probe is flush with the inner wall of the flow channel (to avoid interfering with the airflow). The signal line is led out through the threading hole of the casing body and electrically connected to the external PLC controller (model S7-1200). The signal transmission accuracy is ±0.5% FS.

[0038] Micro hydraulic push rods: Two sets of push rods drive the guide vanes and radial clearance mechanism respectively, and are fixed to the welded bracket on the outer wall of the casing. The push rod control signal line is connected to the PLC controller. The response time is ≤50ms and can act in real time according to the pressure signal.

[0039] 5. Overall adaptation effect The centrifugal compressor of this embodiment is installed in a 60kW hydrogen-fueled gas turbine power generation system. During operation: The impeller and blade assembly 02 rotates at high speed (rated speed 28,000 r / min) driven by the main shaft, sucking in and compressing air from the inlet section, and then sending it into the combustion chamber after being pressurized by the diffuser section, where it is mixed with hydrogen and burned; The adaptive processing casing assembly 01 monitors the flow field pressure in real time through a pressure sensor (sampling frequency 1kHz). When a pressure fluctuation exceeding 100Pa is detected, the PLC controller drives the guide vane to adjust its angle (for example, from 0° to 5°) or fine-tune the blade tip clearance (for example, from 0.3mm to 0.25mm), suppressing airflow separation and leakage to ensure stable operation. Actual tests show that under rated operating conditions, the compressor has an air flow of 750g / s, a pressure ratio of 3.8, an efficiency of 85%, and a pressure fluctuation of 95Pa, meeting the efficient and stable operation requirements of a 60kW-class hydrogen-fueled gas turbine power generation system.

[0040] Through the above-mentioned embodiments, the gradually diverging flow channel of the meridian flow channel assembly 03, the high-strength connection between the impeller and the blade assembly 02, and the real-time control function of the adaptive processing casing assembly 01 form a synergistic effect, which fully reflects the design goal of "low specific speed and high stability" of the present invention and can effectively adapt to the operating conditions of the hydrogen-fueled gas turbine power generation system.

[0041] Specifically, the compressor is adapted to a hydrogen-fueled gas turbine power generation system, adopts a low specific speed design, and achieves high load and high stability through the following structural optimizations: 1. Meridian layout and blade shape: To address the loss problem caused by the high proportion of tip clearance in small-flow compressors, an advanced meridional layout is adopted to optimize the blade sweep angle and diffuser blade angle, taking into account efficient flow characteristics across the entire speed range, reducing tip leakage loss, secondary flow loss, and boundary layer loss.

[0042] 2. Adaptive processing casing: To address the instability problem caused by the asymmetric flow channel of the volute and the flow distortion of the inlet and outlet, an adaptive processing casing is designed to achieve stability control by real-time regulation of the casing structural parameters.

[0043] See also Figure 2 , Figure 2This is a flow chart illustrating a method for optimizing a low specific speed centrifugal compressor adapted for a hydrogen-fueled gas turbine power generation system using multimodal data fusion according to an exemplary embodiment. The method includes: S1. Real-time monitoring of the operating parameters of the compressor in the hydrogen-fueled gas turbine power generation system; the operating parameters include, but are not limited to, flow rate, pressure ratio, efficiency, temperature, pressure fluctuations, and power demand operating parameters of the power generation system monitored at multiple points in different parts of the compressor; S2. Using a sensor fusion algorithm, the pre-set distributed sensor network is used to fuse the flow rate, pressure ratio, efficiency, temperature, pressure fluctuation, and power demand of the power generation system, which are monitored at multiple points in different parts of the compressor, to obtain the fused operating parameters. S3. Using fuzzy logic algorithm, the fused operating parameters are compared and analyzed with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started.

[0044] The fuzzy logic algorithm is used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started, including: The pressure fluctuation in the fused operating parameters is used as the core characterization parameter. A fuzzy logic algorithm is used to compare it with a preset stability standard parameter range. When the pressure fluctuation exceeds the first range, the compressor is determined to have insufficient stability, triggering the first preset rule group. The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface by more than the first threshold for a preset number of sampling cycles, the second rule group is triggered. Using the power generation system power demand and the actual compressor output power as the core parameters in the fused operating parameters, a three-dimensional adaptability evaluation space is constructed, combining efficiency and preset surge margin. The Euclidean distance from the current operating point to the standard adaptation surface is calculated using a fuzzy logic algorithm. When the Euclidean distance exceeds a preset second threshold, the third rule group is triggered.

[0045] In specific implementation, as described in the above steps, the optimization method of the present invention mainly includes the following key steps: 1. Data collection and fusion analysis A distributed sensor network is used to perform multi-point monitoring of different parts of the compressor to obtain real-time operating parameters such as flow rate, pressure ratio, efficiency, temperature, and pressure fluctuations.

[0046] Sensor fusion algorithms are used to fuse and process collected multi-source data to obtain accurate and comprehensive operating data. Fuzzy logic algorithms are used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is triggered.

[0047] 2. Optimization strategies for different problems Stability Optimization: A model predictive control algorithm is used to control the adaptive adjustment device, which is used to adjust the angle or cross-sectional area of ​​the guide vanes in non-axisymmetric flow channels and inlets and outlets. Based on real-time pressure fluctuations, the model predictive control algorithm can predict the flow trend within the flow channel, thereby automatically adjusting the device parameters to reduce the impact of flow distortion on compressor stability. At the same time, active control technology based on neural networks is introduced. Based on the real-time monitoring of pressure fluctuation signals, the learning and predictive capabilities of the neural network are used to automatically adjust the control parameters of active vortex control or active airflow injection, further enhancing the stability of the compressor under unstable operating conditions.

[0048] Efficiency Optimization: Under low-flow conditions, a genetic algorithm combined with an intelligent control system is employed. Using flow rate changes as input, tip clearance is dynamically adjusted to reduce tip leakage, secondary flow, and boundary layer losses. Furthermore, a surface roughness optimization algorithm is employed to apply a special coating to the compressor blade surface. This coating has a low friction coefficient and excellent wear resistance, effectively reducing boundary layer losses.

[0049] Adaptability Optimization: Based on the power requirements of the power generation system, a multi-objective particle swarm optimization algorithm is used to dynamically optimize key compressor design parameters, such as blade shape, diffuser structure, and volute size. This algorithm comprehensively considers multiple objectives, including efficiency, stability, and power matching, to find the optimal combination of design parameters. Developing adjustable compressor structures based on adaptive control algorithms, such as variable-geometry diffusers or retractable volutes, allows these structures to be adaptively adjusted to better adapt to power generation systems with varying power requirements.

[0050] 3. Simulation verification and parameter adjustment Computational fluid dynamics (CFD) software was used to simulate and analyze the optimized compressor model, and a data-driven uncertainty quantification algorithm was used to evaluate the uncertainty of the simulation results.

[0051] Using reinforcement learning algorithms, optimization parameters are automatically adjusted based on simulation results and uncertainty assessments. If the desired optimization goal is not achieved, the simulation and adjustment process is repeated until the desired effect is achieved.

[0052] 4. Establish a performance database and conduct knowledge mining. Establish a compressor performance database, use the knowledge graph algorithm to associate and analyze the operating parameters and optimization strategies before and after each optimization, explore potential optimization directions and rules, and provide a reference for subsequent optimization. Specific embodiments The optimization method of the present invention is described in detail below with reference to specific embodiments.

[0054] Example In a 60kW-class hydrogen-fueled gas turbine power generation system, the optimization method of the present invention is applied to optimize a low specific speed centrifugal compressor.

[0055] Data collection and fusion analysis Distributed sensors are placed at key locations such as the compressor's inlet, outlet, and blade surfaces to monitor operating parameters such as flow rate, pressure ratio, efficiency, temperature, and pressure fluctuations in real time. Data is collected every second and integrated into a sensor fusion algorithm to generate accurate operational data. Fuzzy logic algorithms are used to compare and analyze this fused data against pre-set standard parameter ranges. For example, if pressure fluctuations exceed the normal range by ±5%, an optimization process is triggered.

[0056] Optimization strategies for different problems Stability optimization A model predictive control algorithm is used to control the adaptive adjustment device. Installed at the non-axisymmetric flow channel and air inlet and outlet, the device includes adjustable guide vanes and a flow channel structure with a variable cross-sectional area. Based on real-time monitoring of pressure fluctuations, the model predictive control algorithm predicts the flow change trend in the flow channel over a period of time in the future and adjusts the angle of the guide vanes and the cross-sectional area of ​​the flow channel in advance to reduce flow distortion. At the same time, active control technology based on neural networks monitors pressure fluctuation signals in real time. Through the learning and predictive capabilities of the neural network, it automatically adjusts the vortex intensity of the active vortex control device and the injection direction and flow rate of the active airflow injection device, thereby enhancing the stability of the compressor under unstable operating conditions.

[0057] Efficiency optimization Under low-flow conditions (less than 30% of rated flow), a genetic algorithm combined with an intelligent control system is activated. Using flow rate changes as input, the tip clearance is dynamically adjusted to reduce tip leakage, secondary flow, and boundary layer losses. Simultaneously, the compressor blade surfaces are treated with a special coating using a nanomaterial that offers low friction and excellent wear resistance. A surface roughness optimization algorithm ensures that the coating's uniformity and roughness meet design requirements, effectively reducing boundary layer losses.

[0058] Adaptability optimization Based on the power demand of the power generation system, a multi-objective particle swarm optimization algorithm is used to dynamically optimize key design parameters of the compressor, such as the blade shape, diffuser structure, and volute size. During the optimization process, multiple objectives such as efficiency, stability, and power matching are comprehensively considered. For example, when the power demand of the power generation system is reduced to 40kW, the bending angle of the blades and the expansion angle of the diffuser are adjusted through the multi-objective particle swarm optimization algorithm to improve the efficiency and stability of the compressor under low-power conditions. At the same time, an adjustable compressor structure based on an adaptive control algorithm is developed, such as a diffuser with variable geometry and a retractable volute. When the power demand of the power generation system changes, the adaptive control algorithm automatically adjusts the geometry of the diffuser and the retractable length of the volute, so that the compressor can better adapt to different power requirements.

[0059] Simulation verification and parameter adjustment The optimized compressor model was simulated and analyzed using CFD software such as ANSYS Fluent. During the simulation process, the effects of the volute's asymmetric flow path and inlet and outlet flow distortion were considered, and a comprehensive analysis of the compressor's internal flow field distribution, pressure variations, and airflow separation under different operating conditions was conducted. Data-driven uncertainty quantification algorithms were used to assess the uncertainty of the simulation results, such as using Monte Carlo simulation to calculate confidence intervals for the simulation results. Reinforcement learning algorithms were used to automatically adjust optimization parameters based on the simulation results and uncertainty assessments. If the simulation results showed that the compressor efficiency did not meet the expected target, parameters such as blade shape and tip clearance were fine-tuned and the simulation was repeated until compressor performance was significantly improved.

[0060] Establishing performance database and knowledge mining A compressor performance database was established, recording operating parameters, optimization strategies, and simulation results before and after each optimization. A knowledge graph algorithm was used to correlate and analyze the data in the database, uncovering potential optimization directions and patterns. For example, knowledge graph analysis revealed that adjusting the blade sweep angle under certain flow and pressure conditions can significantly improve compressor efficiency. This knowledge provides important reference for subsequent optimization.

[0061] In one embodiment, the present application has the following beneficial effects: Significantly improve the operating stability of the compressor and adapt to the high requirements of hydrogen-containing fuels. The device realizes real-time control of asymmetric flow channels and inlet and outlet flow distortions through adaptive processing of the annular movable guide vanes, radially adjustable gap mechanisms and piezoelectric pressure sensors of the casing assembly; the optimization method uses pressure fluctuation as the core characterization parameter, combined with fuzzy logic judgment and the first rule group (model predictive control + neural network active control), which can quickly suppress the airflow separation caused by pressure fluctuations, effectively solving the problem of insufficient stability of existing compressors caused by the fast combustion speed and high flame temperature of hydrogen-containing fuels, so that the pressure fluctuation control accuracy of the compressor under a wide range of operating conditions is improved, and the stability margin is significantly improved.

[0062] Optimize the efficiency of all operating conditions while taking into account the performance of low-flow conditions. The high-strength alloy integral milling and blade sweep angle design of the impeller and blade components, combined with the gradually expanding structure of the meridian flow channel, reduce basic flow losses. The optimization method uses the joint characterization of efficiency and flow and the second rule group (genetic algorithm dynamic adjustment of tip clearance + surface optimization) to specifically reduce tip leakage, secondary flow and boundary layer losses under low-flow conditions, solving the pain point of traditional compressors that are difficult to strike a balance between efficiency and stability margin, thereby improving the efficiency of all operating conditions, especially under low-flow conditions.

[0063] Enhanced adaptability to multiple operating conditions, matching the power requirements of hydrogen-containing power generation systems. The adjustable structure (guide vane angle, radial clearance) of the adaptive processing casing and the coaxial design of the meridian flow channel provide a hardware foundation for wide operating condition adaptation. The optimization method uses a three-dimensional adaptability evaluation space (power demand - efficiency - surge margin) and a third rule group (multi-objective optimization + reinforcement learning) to achieve precise matching of the compressor to the variable power requirements of hydrogen-containing fuel gas turbine power generation systems (such as 60kW level), solving the problem of immature multi-operating condition stabilization and efficiency enhancement technology for low-specific speed compressors, and significantly improving the power response speed and adaptation accuracy.

[0064] An intelligent closed-loop optimization system is being built to improve system reliability and economic efficiency. Devices and methods collaborate, enabling precise multi-parameter monitoring through sensor fusion algorithms. Combined with fuzzy logic analysis, CFD simulation verification, and knowledge graph mining, this system forms a closed-loop "monitoring-judgment-control-feedback" mechanism. This system automatically adapts to the complex operating conditions of hydrogen-fueled gas turbine power generation systems, reducing manual intervention and lowering maintenance costs. Through continuous optimization and iteration, it ensures the long-term, efficient, and stable operation of the compressor, providing core technical support for the large-scale application of hydrogen-fueled power generation systems.

[0065] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0066] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" is at least two.

[0067] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0068] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0069] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0070] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0071] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0072] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0073] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system, characterized in that: include: The meridian flow channel assembly is a fixed air flow channel frame that forms a continuous flow channel for air circulation; the impeller and blade assembly is a rotating supercharging component, which is arranged as a whole inside the meridian flow channel assembly, coaxially assembled with the meridian flow channel assembly and a preset gap is maintained between the two, so as to supercharge the airflow flowing into the meridian flow channel assembly through rotation; the adaptive processing casing assembly is a flow field control component, which is arranged around the outside of the meridian flow channel assembly, rigidly connected to the meridian flow channel assembly and coaxially assembled, and its inner side is arranged corresponding to the impeller and blade assembly and maintains an adjustable gap to control the airflow state in real time; the meridian flow channel assembly, impeller and blade assembly, and adaptive processing casing assembly are coaxially arranged to form a closed supercharging flow field, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

2. The low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system according to claim 1, characterized in that: The meridian flow channel assembly comprises: The meridian flow channel assembly includes an inlet section, an impeller section, a diffuser section and an outlet section in sequence along the airflow direction, and each section forms a continuous gradually diverging flow channel; The inlet section is rigidly connected to the air inlet pipe via a flange; The impeller section is a cylindrical cavity, the inner wall of which maintains a preset initial gap with the impeller tip, and is connected to the inlet section through a smoothly transitioned curved surface; The flow channel width of the diffuser section is greater than that of the inlet section, and diffuser blades are fixedly mounted on the inner wall, with a regulating gap reserved between the top of the blade and the inner wall of the adaptive processing casing; The outlet section is a diffusion structure, smoothly connected to the end of the diffuser section, and the outlet is connected to the combustion chamber intake pipe through a flange.

3. The low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system according to claim 2, characterized in that: The meridian flow channel assembly further includes: The entire meridian flow channel has a preset curvature radius and meridian plane inclination angle, and each section remains coaxially arranged.

4. The low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system according to claim 1, characterized in that: The impeller and blade assembly comprises: The impeller is integrally milled from a high-strength alloy, the hub is rigidly connected to the compressor main shaft via a keyway, and both ends of the main shaft are supported by bearing seats; The blade root is connected to the mortise and tenon of the impeller rim through a dovetail tenon, the blade has a preset sweep angle, and the blade top maintains a preset initial gap with the inner wall of the adaptive processing casing; The impeller and blade assembly are integrally located in the impeller section of the meridian flow channel.

5. The low specific speed centrifugal compressor adapted for a hydrogen fuel gas turbine power generation system according to claim 1, characterized in that: The adaptive processing casing assembly includes: a casing body, which is precision-casted from cast iron and has an annular sleeve structure, and is rigidly connected to the outer shell flange of the meridian flow channel by bolts; Multiple annular movable guide vanes are evenly distributed on the inner wall of the casing inlet section. The root is hinged to the casing body through a micro shaft and can rotate around the shaft within a preset angle range. The end is connected to the actuator's micro hydraulic push rod ball joint through a connecting rod mechanism. The radially adjustable clearance mechanism is an annular slider structure embedded in the inner wall groove of the casing near the impeller tip. The inner side of the slider corresponds to the blade tip, and the outer side is slidably connected to the casing body through an elastic guide column and rigidly connected to the output end of another set of micro hydraulic push rods to achieve tip clearance adjustment. Multiple piezoelectric pressure sensors are evenly embedded between the impeller outlet and the diffuser inlet on the inner wall of the casing. The sensor probes are flush with the inner wall of the flow channel. The signal lines are led out through the threading holes of the casing body and electrically connected to the external controller. Two sets of micro hydraulic push rods drive the annular movable guide vane and the radial adjustable gap mechanism respectively, and are fixedly installed on the bracket on the outer wall of the casing. The bracket is welded to the casing, and the push rod control signal line is electrically connected to the controller; the meridian flow channel assembly, impeller and blade assembly, and adaptive processing casing assembly are coaxially arranged to form a closed boost flow field, thereby achieving adaptation to the hydrogen-fueled gas turbine power generation system.

6. A method for optimizing a low specific speed centrifugal compressor adapted for a hydrogen-fueled gas turbine power generation system, applied to a low specific speed centrifugal compressor adapted for a hydrogen-fueled gas turbine power generation system as claimed in any one of claims 1 to 5, characterized in that: The method comprises: Real-time monitoring of compressor operating parameters in hydrogen-fueled gas turbine power generation systems; the operating parameters include, but are not limited to, flow rate, pressure ratio, efficiency, temperature, pressure fluctuation, and power demand operating parameters of the power generation system monitored at multiple points in different parts of the compressor; Using a sensor fusion algorithm, the preset distributed sensor network is used to fuse the flow rate, pressure ratio, efficiency, temperature, pressure fluctuation, and power demand operating parameters of the power generation system monitored at multiple points in different parts of the compressor to obtain the fused operating parameters; The fuzzy logic algorithm is used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started.

7. The method according to claim 6, characterized in that The fuzzy logic algorithm is used to compare and analyze the fused operating parameters with the preset standard parameter range. When the standard parameter range is exceeded, the optimization process is started, including: The pressure fluctuation in the fused operating parameters is used as the core characterization parameter. A fuzzy logic algorithm is used to compare it with the preset stability standard parameter range. When the pressure fluctuation exceeds the preset first range, the compressor is judged to have insufficient stability, triggering the preset first rule group. The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface by more than the first threshold for a preset number of sampling cycles, the second rule group is triggered. Using the power generation system power demand and the actual compressor output power as the core parameters in the fused operating parameters, a three-dimensional adaptability evaluation space is constructed, combining efficiency and preset surge margin. The Euclidean distance from the current operating point to the standard adaptation surface is calculated using a fuzzy logic algorithm. When the Euclidean distance exceeds a preset second threshold, the third rule group is triggered.

8. The method according to claim 7, characterized in that The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with a preset efficiency standard parameter surface through a fuzzy logic algorithm. When the pressure fluctuation exceeds the first range, it is determined that the compressor has a stability problem, and a preset first rule group is triggered, including: The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters and compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the pressure fluctuation exceeds the preset first range, the model predictive control algorithm is used to control the adaptive processing casing component. The flow change trend in the flow channel is predicted based on the real-time pressure fluctuation, and the guide vane angle or cross-sectional area of ​​the non-axisymmetric flow channel and the inlet and outlet are automatically adjusted. Active control technology based on neural networks is introduced to automatically adjust the control parameters of active vortex control or active airflow injection according to the real-time monitored pressure fluctuation signal through the learning and prediction capabilities of the neural network.

9. The method according to claim 7, characterized in that The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters, and are compared with a preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface for a preset number of consecutive sampling periods and exceeds a first threshold, the second rule group is triggered, including: The efficiency and flow rate in the fused operating parameters are used as joint characterization parameters, and are compared with the preset efficiency standard parameter surface through a fuzzy logic algorithm. When the actual operating point deviates from the standard surface for a preset number of sampling cycles and exceeds the first threshold, a genetic algorithm is used with flow rate changes as input to dynamically adjust the tip clearance size with the goal of reducing tip leakage loss, secondary flow loss and boundary layer loss.

10. The method according to claim 7, characterized in that The optimization process further includes: Computational fluid dynamics software was used to simulate and analyze the optimized compressor model, and a data-driven uncertainty quantification algorithm was used to evaluate the uncertainty of the simulation results. Using reinforcement learning algorithms, the optimization parameters are automatically adjusted based on simulation results and uncertainty assessment. If the expected optimization goal is not achieved, the simulation and adjustment process is repeated until the expected effect is achieved. At the same time, a compressor performance database is established, and a knowledge graph algorithm is used to associate and analyze the operating parameters and optimization strategies before and after each optimization, to explore potential optimization directions and rules, and provide a reference for subsequent optimization.

Images