A design optimization method for long blade variable condition self-adaption of deep peak shaving steam turbine
By employing an adaptive design approach, this study addresses the issues of efficiency degradation and flow deterioration in long turbine blades under deep peak-shaving conditions. A load-sensing-dynamic matching design optimization method is used to proactively adjust blade geometric parameters, solving the technical challenge of low efficiency in traditional designs and improving the unit's operating efficiency and design reusability.
Patent Information
- Application Number
- CN202610729519.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-26
- Publication Date
- 2026-08-25
AI Technical Summary
Under deep peak shaving conditions, the efficiency of long turbine blades decreases and the flow deteriorates. Traditional design methods cannot actively adapt to load changes, resulting in a serious deviation between the design point and the operating point, flow separation and shock wave structure mismatch, difficulty in reusing design knowledge, long design cycle and high cost.
An adaptive design method based on load sensing and dynamic matching is adopted. By dividing the load condition zone, a dynamic response model of angle of attack and blade profile is established. The blade is segmented along the direction into an angle of attack adaptive segment, a shock wave adaptive segment, and a transition segment. Multi-condition coupled iterative optimization is carried out, and a load and performance response profile library is constructed to realize the active adjustment of blade geometric parameters.
Maintaining excellent aerodynamic performance within a load range of 15%-100%, shortening the design cycle, reducing costs, and achieving efficient operation across the entire operating range.
Smart Images
Figure CN122634684A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a design optimization method for steam turbine blades, specifically to a design optimization method for long blades of deep peak-shaving steam turbines that adapts to varying operating conditions. This invention belongs to the field of steam turbine blade aerodynamic design technology. Background Technology
[0002] In recent years, with the continuous increase in the proportion of new energy installed capacity, the transformation of thermal power units from traditional base load operation to peak-shaving operation has become an inevitable trend. Affected by factors such as the slowdown in electricity consumption demand and the rapid growth of non-fossil energy power generation, the power generation of thermal power units has begun to show a negative growth trend. At the same time, the proportion of residential and commercial electricity consumption has been rising year by year, resulting in a surge in the peak-valley difference of electricity load. This forces large-capacity thermal power units that were originally designed according to base load to participate in deep peak shaving, and they are usually operating at a lower load (below 40% of rated load).
[0003] Reduced unit load means decreased efficiency and increased coal consumption for power generation, failing to fully utilize the high efficiency advantages of large-capacity units. Among these factors, blades are the most sensitive to unit load, especially the long blades in the last stage, which are the first to enter a "small volumetric flow" state under low load conditions, resulting in the most severe flow deterioration. Traditional long blades optimized for rated operating conditions face the following technical challenges under deep peak-shaving conditions:
[0004] The design point deviates significantly from the operating point: Traditional long blade design methods typically optimize based on a certain rated operating condition (such as 100% THA or 75% THA), with fixed blade geometry parameters. When the unit operates at a low load condition below 40% for an extended period, the airflow angle of attack deviates significantly, resulting in flow separation at a large negative angle of attack in the blade root region and shock wave structure mismatch in the blade tip region, leading to a sharp decline in flow efficiency.
[0005] Passive adaptation rather than active matching: The "wide load airfoil" design concept in existing technologies still falls under the category of "static design". That is, through compromise design, the blade can maintain relatively acceptable performance over a wide load range. However, in essence, the blade geometry passively bears the changes in angle of attack and flow losses caused by variable operating conditions, and lacks the ability to actively adjust aerodynamic characteristics according to load changes.
[0006] One-dimensional and three-dimensional design are separated: In the traditional design process, one-dimensional thermal calculation and three-dimensional blade geometry design are relatively independent. Once the one-dimensional parameters are determined, it is difficult to make reverse adjustments based on the three-dimensional flow field diagnosis results, resulting in poor matching between thermal design and geometric design.
[0007] Difficulty in reusing design knowledge: The design of long blades for units of different capacities (300MW, 600MW, 1000MW) often requires starting from scratch, making it difficult to effectively accumulate and reuse design experience, resulting in long design cycles and high costs.
[0008] To address the aforementioned issues, this invention proposes an adaptive design optimization method for long blades in deep peak-shaving steam turbines under varying operating conditions. By establishing an adaptive design mechanism of "load perception-dynamic matching," the long blades can actively adjust their aerodynamic characteristics under different loads, achieving efficient operation across the entire operating range. Summary of the Invention
[0009] The purpose of this invention is to solve the technical problems of reduced efficiency and deteriorated flow of long turbine blades under deep peak shaving conditions in the prior art, and to provide a design optimization method for adaptive design of long turbine blades under varying operating conditions for deep peak shaving.
[0010] To solve the above problems, the present invention provides the following technical solution:
[0011] An adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions is provided, the method being implemented as follows:
[0012] Step 1: Load Condition Zoning and Characteristic Parameter Extraction:
[0013] The operating load range of the deep peak shaving unit is divided into multiple characteristic operating condition zones, and key aerodynamic characteristic parameters of each operating condition zone are extracted. The key aerodynamic characteristic parameters include load, angle of attack i, and Mach number Ma.
[0014] Step 2: Construct a dynamic response model for angle of attack and blade shape:
[0015] Establish a dynamic mapping relationship between the change in angle of attack and the adjustment of blade geometric parameters, wherein the blade geometric parameters include at least the geometric inlet angle β1p, the geometric outlet angle β2p, and the curvature distribution of the arc in the blade profile;
[0016] Step 3: Blade Segment Adaptive Design
[0017] The long blade is divided into multiple functional segments along the blade height direction. Based on the flow characteristics of each segment under different load conditions, different variable operating condition response mechanisms are assigned to each segment. The multiple functional segments are the angle of attack adaptive segment, the shock wave adaptive segment, and the transition connection segment.
[0018] Step 4: Multi-condition coupled iterative optimization:
[0019] The adaptive blade 3D models of the angle-of-attack adaptive section, shock wave adaptive section and transition section obtained in step 3 were constructed using UG software. Numerical simulation calculations under different load conditions were performed using CFD software. The dynamic response parameters of each section were optimized through multiple rounds of iteration until the preset efficiency target and flow stability index were met in all characteristic operating conditions.
[0020] Furthermore, the 15%~100% rated load range in step one is divided into three characteristic operating condition zones:
[0021] Low load zone: 15%~40% of rated load, characterized by small volumetric flow rate, large negative angle of attack, and high risk of flow separation at the root;
[0022] Medium load zone: 40%~70% of rated load, characterized by design flow rate, relatively stable flow, and high efficiency;
[0023] High load zone: 70%~100% of rated load, characterized by large volumetric flow rate, transonic flow at the top, and significant shock wave loss.
[0024] Furthermore, in step two, the mathematical expression of the angle of attack and blade dynamic response model is as follows:
[0025]
[0026] Where f(Δi) is the adjustment amount of the curvature of the blade's mid-curve, and Δi is the deviation of the angle of attack relative to the design point. , , These are the empirical coefficients obtained through CFD calculations and fitting experimental data.
[0027] Furthermore, in step three, the blade root section is designed as an angle-of-attack adaptive section to address the large negative angle-of-attack characteristics under low-load conditions. Its geometric intake angle β1p is dynamically adjusted according to a predetermined law as the load changes.
[0028] Blade tip section: Designed as a shock wave adaptive section to address the transonic flow characteristics under high load conditions, its blade profile curvature automatically optimizes the shock wave structure as the Mach number Ma changes.
[0029] Mid-leaf segment: Serves as a transitional segment, ensuring a smooth connection between the leaf root segment and the leaf tip segment.
[0030] Furthermore, in step three, the adaptive angle of attack mechanism of the blade root section is as follows: when the unit load is reduced to below 40%, the geometric inlet angle β1p of the blade root section automatically decreases according to the preset gradient as the load decreases, so that the airflow angle of attack is kept in the optimal range of -5° to +5°, thus delaying the flow separation under low load conditions.
[0031] Furthermore, the geometric intake angle β1p of the blade root section is dynamically adjusted according to the following rules as the load changes:
[0032] β1p(load) = β1p_design + α·(40% - load percentage)
[0033] Wherein, β1p_design is the geometric inlet angle at the design point under 75% load, with a value of 32.5°; α is the adjustment coefficient, with a value of 0.25° / %. When the load is reduced to 30%, β1p automatically decreases by 2.5°, improving the actual airflow angle of attack from -18.5° to -6.2°, effectively delaying flow separation.
[0034] Furthermore, in step three, the shock wave adaptive mechanism of the blade tip section is as follows: when the unit load increases to more than 70%, the curvature of the arc in the blade tip section airfoil is automatically adjusted as the Mach number Ma increases, so that the shock wave position moves backward and the shock wave intensity decreases, thereby reducing shock wave loss under transonic flow conditions.
[0035] Furthermore, it also includes step five: constructing a load and performance response line library, normalizing the optimal airfoil parameters obtained under different load conditions and storing them in the database to form a reusable adaptive airfoil template library.
[0036] Furthermore, the load and performance response type line library supports a rapid selection function. By inputting the peak shaving depth range of the target unit, the system automatically matches the optimal adaptive blade combination and generates a three-dimensional blade geometry model that can be directly used for engineering.
[0037] Furthermore, the long blades have a segmented adaptive function along the blade height direction, and can maintain excellent aerodynamic performance through active response of geometric parameters within a load range of 15% to 100%.
[0038] Compared with the prior art, the present invention has the following technical advantages:
[0039] (1) Innovative design concept: Breaking through the traditional "static optimization" design concept, the first "variable working condition adaptive" design method was created, which transformed the long blade from "passively bearing variable working conditions" to "actively matching variable working conditions", maintaining excellent aerodynamic performance within the 15%-100% load range.
[0040] (2) Segmented design innovation: In view of the large differences in flow characteristics along the blade height of long blades, a segmented adaptive design strategy is proposed for the blade root segment, blade middle segment and blade tip segment. The blade root segment focuses on the angle of attack adaptability, and the blade tip segment focuses on shock wave control to achieve precise optimization.
[0041] (3) Innovation of dynamic response mechanism: The “angle of attack-blade type” dynamic response model is established, which enables the blade geometric parameters to be automatically adjusted with load changes, effectively expanding the high-efficiency operation range and solving the technical problem of “high efficiency at the design point and low efficiency under changing working conditions” in traditional design.
[0042] (4) Design reuse innovation: Construct a "load-performance" response line library to realize the effective accumulation of design knowledge and cross-unit reuse, which greatly shortens the design cycle and reduces R&D costs. Attached Figure Description
[0043] Figure 1 This is a flowchart of the present invention;
[0044] Figure 2 This is a schematic diagram of load condition zoning;
[0045] Figure 3 This is a schematic diagram of the segmented adaptive design of medium and long blades;
[0046] Figure 4 This is a schematic diagram of the dynamic response model of angle of attack and blade shape;
[0047] Figure 5 This is a schematic diagram of the three-dimensional structure of a long blade;
[0048] Figure 6 This is a curve comparing the efficiency of the present invention and existing technologies under different loads. Detailed Implementation
[0049] Specific implementation method one: Combining Figures 1 to 6 This embodiment describes a design optimization method for adaptive design of long blades in deep peak-shaving steam turbines under varying operating conditions. The method is implemented as follows:
[0050] Step 1: Load Condition Zoning and Characteristic Parameter Extraction:
[0051] The operating load range of the deep peak shaving unit is divided into multiple characteristic operating condition zones, and key aerodynamic characteristic parameters of each operating condition zone are extracted. The key aerodynamic characteristic parameters include angle of attack i and Mach number Ma.
[0052] Step 2: Construct a dynamic response model for angle of attack and blade shape:
[0053] A dynamic mapping relationship is established between the change in angle of attack and the adjustment of blade geometric parameters. The blade geometric parameters include at least the geometric inlet angle β1p, the geometric outlet angle β2p, and the curvature distribution of the airfoil. The dynamic response principle of angle of attack and airfoil is to establish a functional relationship between the deviation of angle of attack and the adjustment of the curvature of the airfoil, so that the blade geometric parameters can be adjusted in a "responsive" manner according to the change in angle of attack caused by load changes, thereby achieving active maintenance of aerodynamic performance.
[0054] Step 3: Blade Segment Adaptive Design
[0055] The long blade is divided into multiple functional segments along the blade height direction. Based on the flow characteristics of each segment under different load conditions, different variable operating condition response mechanisms are assigned to each segment. The multiple functional segments are the angle of attack adaptive segment, the shock wave adaptive segment, and the transition connection segment.
[0056] Step 4: Multi-condition coupled iterative optimization:
[0057] The adaptive blade 3D models of the angle-of-attack adaptive section, shock wave adaptive section and transition section obtained in step 3 were constructed using UG software. Numerical simulation calculations under different load conditions were performed using CFD software. The dynamic response parameters of each section were optimized through multiple rounds of iteration until the preset efficiency target and flow stability index were met in all characteristic operating conditions.
[0058] Combination Figure 1 and Figure 2 As shown, the 15%~100% rated load range in step one is divided into three characteristic operating condition zones:
[0059] Low-load region: 15%~40% of rated load, characterized by small volumetric flow rate, large negative angle of attack, and high risk of flow separation at the root. The principle of designing from the low-pressure final stage is that the final stage is the first to enter the "small volumetric flow rate" condition at low loads, and the flow is most likely to deteriorate. Prioritize ensuring the adaptability of the final stage, and then match it stage by stage forward, so that the entire flow path can maintain matching at low loads and avoid step-by-step mismatch.
[0060] Medium load zone: 40%~70% of rated load, characterized by design flow rate, relatively stable flow, and high efficiency;
[0061] High load zone: 70%~100% of rated load, characterized by large volumetric flow rate, transonic flow at the top, and significant shock wave loss.
[0062] Combination Figure 1 and Figure 4 As shown, in step two, the mathematical expression of the angle of attack and blade dynamic response model is as follows:
[0063]
[0064] Where f(Δi) is the adjustment amount of the curvature of the blade's mid-curve, and Δi is the deviation of the angle of attack relative to the design point. , , These are the empirical coefficients obtained through CFD calculations and fitting experimental data.
[0065] Combination Figure 1 and Figure 3 As shown, in step three, the blade root section is designed as an angle-of-attack adaptive section to address the large negative angle-of-attack characteristics under low-load conditions. Its geometric intake angle β1p is dynamically adjusted according to a predetermined law as the load changes.
[0066] Blade tip section: Designed as a shock wave adaptive section to address the transonic flow characteristics under high load conditions, its blade profile curvature automatically optimizes the shock wave structure as the Mach number Ma changes.
[0067] Mid-leaf segment: Serves as a transitional segment, ensuring a smooth connection between the leaf root segment and the leaf tip segment.
[0068] Combination Figure 1 and Figure 2 As shown, in step three, the adaptive angle of attack mechanism of the blade root section is as follows: when the unit load is reduced to below 40%, the geometric inlet angle β1p of the blade root section automatically decreases according to the preset gradient as the load decreases, so that the airflow angle of attack is kept in the optimal range of -5° to +5°, thus delaying the flow separation under low load conditions.
[0069] Combination Figure 1 and Figure 2 As shown, the geometric intake angle β1p of the blade root section dynamically adjusts with load changes according to the following rules:
[0070] β1p(load) = β1p_design + α·(40% - load percentage)
[0071] Wherein, β1p_design is the geometric inlet angle at the design point under 75% load, with a value of 32.5°; α is the adjustment coefficient, with a value of 0.25° / %. When the load is reduced to 30%, β1p automatically decreases by 2.5°, improving the actual airflow angle of attack from -18.5° to -6.2°, effectively delaying flow separation.
[0072] Combination Figure 1 and Figure 2 As shown, in step three, the shock wave adaptive mechanism of the blade tip section is as follows: when the unit load increases to more than 70%, the curvature of the arc in the blade tip section airfoil is automatically adjusted as the Mach number Ma increases, so that the shock wave position moves backward and the shock wave intensity decreases, thereby reducing shock wave loss under transonic flow conditions.
[0073] Combination Figure 1 As shown, it also includes step five: constructing a load and performance response line library, normalizing the optimal airfoil parameters obtained under different load conditions and storing them in the database to form a reusable adaptive airfoil template library.
[0074] Combination Figure 1 As shown, the load and performance response type line library supports a rapid selection function. By inputting the peak shaving depth range of the target unit, the system automatically matches the optimal adaptive blade combination and generates a three-dimensional blade geometry model that can be directly used for engineering.
[0075] Combination Figure 1 and Figure 2 As shown, the long blades have a segmented adaptive function along the blade height direction, and can maintain excellent aerodynamic performance through active response of geometric parameters within the load range of 15% to 100%.
[0076] Example 1
[0077] This embodiment takes the last stage long blade (blade height 1200mm) of a 600MW thermal power unit as an example to illustrate the specific implementation process of the present invention.
[0078] Step 1: Load Condition Zoning and Characteristic Parameter Extraction
[0079] Combination Figure 2 As shown, the 15%~100% rated load range is divided into three characteristic operating condition zones:
[0080] Low load zone: 15%~40% of rated load. This zone is characterized by: small volumetric flow rate, large negative angle of attack (up to -15° to -20°) in the blade root region, high risk of flow separation at the root, and low Mach number (Ma<0.8) in the blade tip region.
[0081] Medium load zone: 40%~70% of rated load. This zone is characterized by: angle of attack close to the design value (-5° to +5°), relatively stable flow, and good matching of each stage.
[0082] High-load zone: 70%~100% of rated load. This zone is characterized by: large volumetric flow rate, transonic flow at the blade tip (Ma=0.9~1.3), significant shock wave losses, and relatively full flow at the blade root.
[0083] Key aerodynamic characteristic parameters such as load, angle of attack, and Mach number (Ma) were extracted for each operating condition using CFD calculations to establish a database linking load, angle of attack, and Mach number (Ma). The results show that at 30% load, the blade root angle of attack reaches -18.5°, and the blade tip angle of attack is -3.2°; at 100% load, the blade root angle of attack is +2.3°, the blade tip angle of attack is +8.7°, and the blade tip Mach number reaches 1.25.
[0084] Step 2: Construct a dynamic response model based on the angle of attack and leaf shape.
[0085] Establish a dynamic mapping relationship between changes in angle of attack and adjustments to blade geometry parameters. Combined with... Figure 3 As shown, in this embodiment, the curvature of the arc in the leaf shape is used as the core adjustment parameter to construct a quadratic function model:
[0086]
[0087] Where f(Δi) is the adjustment amount of the curvature of the arc in the leaf shape (unit: m). -1 Δi represents the deviation of the angle of attack from the design point, in degrees. Empirical coefficients were determined through extensive CFD calculations and fitting of planar blade cascade experimental data. =0.15, =0.08, =0.002.
[0088] The model shows that when the angle of attack deviates negatively (low load condition), f(Δi) is negative, that is, the curvature of the mid-curve is reduced (making the blade profile more gentle) to improve the flow under large negative angle of attack; when the angle of attack deviates positively (high load condition), f(Δi) is positive, that is, the curvature of the mid-curve is increased to adapt to the flow under large positive angle of attack.
[0089] Step 3: Blade Segment Adaptive Design
[0090] Combination Figure 3 As shown, the long blade is divided into three functional segments along the leaf height direction:
[0091] (1) Leaf root segment design (0%~30% of leaf height)
[0092] The blade root section is designed primarily for high negative angle-of-attack characteristics under low-load conditions. In this embodiment, the geometric intake angle β1p of the blade root section is dynamically adjusted according to the following rules as the load changes:
[0093] β1p(load) = β1p_design + α·(40% - load percentage)
[0094] Wherein, β1p_design is the geometric inlet angle at the design point (75% load), with a value of 32.5°; α is the adjustment coefficient, with a value of 0.25° / %. When the load decreases to 30%, β1p automatically decreases by 2.5°, improving the actual airflow angle of attack from -18.5° to -6.2°, effectively delaying flow separation.
[0095] The blade root section features a large leading-edge radius design (leading-edge radius / chord length = 0.12) to improve tolerance to large angle-of-attack variations. Simultaneously, the blade root section geometry exit angle β2p is optimized to ensure that the outlet airflow maintains a favorable flow direction even under low load conditions.
[0096] (2) Design of the blade tip section (70%~100% of the blade height)
[0097] The blade tip section is designed primarily for transonic flow characteristics under high load conditions. In this embodiment, the curvature of the arc line in the blade tip section airfoil is automatically adjusted according to the dynamic response model in step two as the Mach number Ma changes.
[0098] Specific implementation: In the blade 3D modeling parametric system, the mid-curvature of each section of the blade tip is defined as a function of the Mach number. When CFD calculations detect that the blade tip Mach number exceeds 0.9, the system automatically increases the mid-curvature, shifting the shock wave position backward and reducing its intensity. At 100% load (Ma=1.25), the mid-curvature of the blade tip increases by 15% compared to the design point, resulting in a 23% reduction in shock wave loss.
[0099] The blade tip section features a thin trailing edge design (trailing edge thickness / chord length = 0.02) to reduce wake loss in transonic flow. Simultaneously, the blade tip clearance profile is optimized with a "contraction-expansion" channel design to reduce leakage loss.
[0100] (3) Leaf mid-section design (30%~70% of leaf height)
[0101] The middle segment of the leaf serves as a transitional section, ensuring a smooth connection between the leaf root and leaf tip segments. Cubic spline interpolation is employed to ensure continuous variation of geometric parameters from the leaf root to the leaf tip, avoiding additional losses caused by abrupt geometric changes.
[0102] Step 4: Multi-condition coupled iterative optimization
[0103] The segmented adaptive blade 3D model obtained in step 3 is verified by CFD under different load conditions to establish a closed-loop iterative process of design, calculation, diagnosis and adjustment.
[0104] First iteration: CFD calculations were performed based on the preliminary design, with a computational grid of approximately 3 million cells, using SST.
[0105] The k-ω turbulence model shows that at 30% load, the separation zone at the blade root decreases, but the shock wave loss at the blade tip remains high; the efficiency meets the target at 100% load, but the efficiency is lower than expected at 50% load.
[0106] Second iteration: Adjusting the two parameters collaboratively based on the diagnostic results.
[0107] One-dimensional thermodynamic parameter adjustment: Optimize enthalpy drop distribution, reduce the final stage enthalpy drop by 5%, and reduce sensitivity to variable operating conditions;
[0108] Three-dimensional geometric parameter adjustment: Correcting the dynamic response coefficient of the blade tip section, The value was adjusted from 0.002 to 0.0025 to enhance shock wave control capabilities;
[0109] Third iteration: CFD calculations were performed again to verify that all indicators met the preset targets. Combined with... Figure 5 and Figure 6 As shown, the final optimization results indicate that within the load range of 30% to 100%, the stage efficiency fluctuation is less than 4 percentage points, which is 6 to 12 percentage points higher than that of the traditional design.
[0110] Step 5: Construct a load and performance-responsive line library
[0111] The optimal airfoil parameters obtained in this embodiment are normalized and stored in the response line library. The normalization method is as follows:
[0112] Coordinate normalization: Taking the blade root section as the reference, the geometric parameters of each section (chord length b, maximum thickness Dmax, installation angle βy, etc.) are expressed as multiples of the reference value to form a dimensionless description.
[0113] Aerodynamic normalization: The optimal airfoil under different loads is classified according to aerodynamic parameters (inlet angle β1, outlet angle β2, Mach number Ma) to form an aerodynamic template.
[0114] This type of blade library supports rapid selection: when the peak shaving depth range of the target unit is entered (such as 300MW unit, 30%~100% load peak shaving), the system automatically matches the optimal blade combination of this embodiment and generates a three-dimensional blade geometry model that can be directly engineered, realizing the cross-unit reuse of design knowledge.
[0115] Example 2
[0116] This embodiment uses a long blade (blade height 800mm) of the second and last stage of a 1000MW ultra-supercritical unit as an example to verify the applicability of the method of the present invention.
[0117] The difference from Example 1 is that:
[0118] (1) Load zoning adjustment: Based on the actual operating data of the 1000MW unit, the low load zone is adjusted to 25%~40% of the rated load (250MW-400MW), the medium load zone to 40%~80% of the rated load (400MW-800MW), and the high load zone to 80%~100% of the rated load (800MW-1000MW).
[0119] (2) Dynamic response model correction: Due to the larger blade load and more significant transonic flow in the 1000MW unit, the dynamic response model of the blade tip section is adjusted as follows:
[0120] f(Δi) = 0.12 + 0.10·Δi + 0.003·Δi²
[0121] That is, it enhances the shock wave control capability. (Increased from 0.002 to 0.003).
[0122] (3) Optimization results: The blades designed using the method of this invention have an efficiency that is 8.5 percentage points higher than that of the traditional design at 25% load, and the efficiency remains unchanged at 100% load, with an average efficiency increase of 5.2 percentage points across all operating conditions.
[0123] Other implementation methods
[0124] The method of this invention is also applicable to the following types of blades:
[0125] (1) Different capacity units: long blades of the last stage and the second last stage of steam turbines with different power levels such as 300MW, 600MW, and 1000MW.
[0126] (2) Different working fluids: In addition to steam turbines, it can also be applied to the design of long blades for other turbomachinery such as gas turbines and air turbines.
[0127] (3) Different peak shaving depths: The load zoning and dynamic response model parameters can be adjusted according to the actual peak shaving needs of the unit (such as 15%~100%, 30%~90%, etc.).
[0128] The empirical coefficients for the dynamic response model of angle of attack and airfoil in this specification , , The determination method is as follows: Using orthogonal experimental design, typical angle-of-attack deviations (-20°, -15°, -10°, -5°, 0°, +5°, +10°, +15°, +20°) are selected, and CFD calculations and planar blade cascade experiments are performed to verify the optimal mid-curvature adjustment amount under different angles of attack. Then, the coefficients are obtained by least-squares fitting. Those skilled in the art can use similar methods to determine coefficient values applicable to different application scenarios based on specific blade characteristics and design requirements.
[0129] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art can make various improvements and modifications without departing from the spirit and principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A design optimization method for adaptive design of long blades in deep peak-shaving steam turbines under varying operating conditions, characterized in that: The method is implemented in the following manner: Step 1: Load Condition Zoning and Characteristic Parameter Extraction: The operating load range of the deep peak shaving unit is divided into multiple characteristic operating condition zones, and key aerodynamic characteristic parameters of each operating condition zone are extracted. The key aerodynamic characteristic parameters include load, angle of attack i, and Mach number Ma. Step 2: Construct a dynamic response model for angle of attack and blade shape: Establish a dynamic mapping relationship between the change in angle of attack and the adjustment of blade geometric parameters, wherein the blade geometric parameters include at least the geometric inlet angle β1p, the geometric outlet angle β2p, and the curvature distribution of the arc in the blade profile; Step 3: Blade Segment Adaptive Design The long blade is divided into multiple functional segments along the blade height direction. Based on the flow characteristics of each segment under different load conditions, different variable operating condition response mechanisms are assigned to each segment. The multiple functional segments are the angle of attack adaptive segment, the shock wave adaptive segment, and the transition connection segment. Step 4: Multi-condition coupled iterative optimization: The adaptive blade 3D models of the angle-of-attack adaptive section, shock wave adaptive section and transition section obtained in step 3 were constructed using UG software. Numerical simulation calculations under different load conditions were performed using CFD software. The dynamic response parameters of each section were optimized through multiple rounds of iteration until the preset efficiency target and flow stability index were met in all characteristic operating conditions.
2. The design optimization method for adaptive design of long blades in deep peak-shaving steam turbines under varying operating conditions, as described in claim 1, is characterized in that: The 15%~100% rated load range in step one is divided into three characteristic operating condition zones: Low load zone: 15%~40% of rated load, characterized by small volumetric flow rate, large negative angle of attack, and high risk of flow separation at the root; Medium load zone: 40%~70% of rated load, characterized by design flow rate, relatively stable flow, and high efficiency; High load zone: 70%~100% of rated load, characterized by large volumetric flow rate, transonic flow at the top, and significant shock wave loss.
3. The design optimization method for adaptive design of long blades in deep peak-shaving steam turbines under varying operating conditions, as described in claim 1, is characterized in that: In step two, the mathematical expression of the angle of attack and blade dynamic response model is as follows: Where f(Δi) is the adjustment amount of the curvature of the blade's mid-curve, and Δi is the deviation of the angle of attack relative to the design point. , , These are the empirical coefficients obtained through CFD calculations and fitting experimental data.
4. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions according to claim 1, characterized in that: In step three, the blade root section is designed as an angle-of-attack adaptive section to address the large negative angle-of-attack characteristics under low-load conditions. Its geometric intake angle β1p is dynamically adjusted according to a predetermined law as the load changes. Blade tip section: Designed as a shock wave adaptive section to address the transonic flow characteristics under high load conditions, its blade profile curvature automatically optimizes the shock wave structure as the Mach number Ma changes. Mid-leaf segment: Serves as a transitional segment, ensuring a smooth connection between the leaf root segment and the leaf tip segment.
5. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions according to claim 4, characterized in that: In step three, the adaptive angle of attack mechanism of the blade root section is as follows: when the unit load is reduced to below 40%, the geometric inlet angle β1p of the blade root section automatically decreases according to the preset gradient as the load decreases, so that the airflow angle of attack is kept in the optimal range of -5° to +5°, thus delaying the flow separation under low load conditions.
6. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions, as described in claim 5, is characterized in that: The geometric intake angle β1p at the blade root section dynamically adjusts according to the following rules as the load changes: β1p(load) = β1p_design + α·(40% - load percentage) Wherein, β1p_design is the geometric inlet angle at the design point under 75% load, with a value of 32.5°; α is the adjustment coefficient, with a value of 0.25° / %. When the load is reduced to 30%, β1p automatically decreases by 2.5°, improving the actual airflow angle of attack from -18.5° to -6.2°, effectively delaying flow separation.
7. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions according to claim 4, characterized in that: In step three, the shock wave adaptive mechanism of the blade tip section is as follows: when the unit load increases to more than 70%, the curvature of the arc in the blade tip section airfoil is automatically adjusted as the Mach number Ma increases, so that the shock wave position moves backward and the shock wave intensity decreases, thereby reducing shock wave loss under transonic flow conditions.
8. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions according to claim 1, characterized in that: It also includes step five: building a load and performance response line library, normalizing the optimal airfoil parameters obtained under different load conditions and storing them in the database to form a reusable adaptive airfoil template library.
9. The adaptive design optimization method for long blades of deep peak-shaving steam turbines under varying operating conditions according to claim 6, characterized in that: The load and performance response type line library supports a rapid selection function. By inputting the peak shaving depth range of the target unit, the system automatically matches the optimal adaptive blade combination and generates a three-dimensional blade geometry model that can be directly used for engineering.
10. The design optimization method for adaptive design of long blades in deep peak-shaving steam turbines under varying operating conditions according to claim 1, characterized in that: The long blades have a segmented adaptive function along the blade height direction, and can maintain excellent aerodynamic performance through active response of geometric parameters within the load range of 15% to 100%.