In the 20-year design life of MW-class wind turbines, the number of load alternations is up to 10, and the fatigue life assessment is essential for the long-term dynamic load of wind turbine components. The tower flange is connected to the high-strength bolts in Figure 1. The threaded joints are subject to high stress concentrations, so fatigue evaluation is particularly important. At present, the non-periodic stress fatigue life analysis method mainly uses the full-life SN curve method combined with the rain flow counting method and the Palmgren-Miner criterion. The key task is to determine the relationship between the flange load and the bolt stress and determine a match with it. Bolt fatigue rating SN curve. Petersen linear model [1], Faulhaber-Thomala curve model [2], Schmidt-Neuper segmentation model [3], FEA model and M. The Seidel curve model [4] can determine the relationship between flange loading and bolt stress. The Eurocode 3 specification [7] can be used to determine the fatigue level of high-strength bolts. Using the factors affecting the fatigue life of high-strength bolts, the fatigue life of high-strength bolts connected to the ring flange of the tower can be effectively evaluated to meet the GL specification [6]. Requirements for fatigue life assessment of high strength bolts. 1 Flange loading and bolt stress calculation As shown in the circular flange connection single-segment model in Figure 1, the tensile stress and bending moment stress of the bolt will fluctuate with the change of the external load. It is determined that the relationship between the flange load and the bolt stress is the bolt fatigue evaluation. The primary task can be determined as follows. 1.1 M.Seidel curve model [4] The GL specification [6] stipulates that the flange connection shall be assembled in accordance with DIN 18800 part 7 and the pre-tightening force FV shall be applied to the high-strength bolt. Under the influence of the eccentric clamping and eccentric loading of the annular flange, the working tension FS of the bolt is: The M.Seidel curve model simplifies the flange-loaded stress section center to a beam model that is fixed at one end, as shown in Figure 2. When the tower annular flange is subjected to the variable working load Z, a gap B* will appear on the upper and lower flange connection faces, and the position is changed. The position can be determined by the distance b* of the gap B* from the central axis of the bolt. When the gap B* appears on the right side of the bolt's center axis, b*>0, b*<0 appears on the left side of the bolt's center axis. For the simplified beam model of the center of the flange loaded stress section, the equilibrium conditions of applied force and bending moment can be obtained: (Note: If the bolt pre-tightening force Fv is large, if the Fs is large, then the M bending moment is small, the flange deformation angle is small, see formula (11) above) From equations (8) and (10), the working tension FS of the high-strength bolt under the working load Z can be determined. And the overturning moment MS can determine the tensile stress of the bolt σS=FS/AS (AS is the equivalent stress area of ​​the bolt) and the bending moment stress σw=MS/WS (WS is the bending modulus of the bolt). By changing the size of the working load Z, a nonlinear relationship between the flange loaded Z and the bolt stress can be obtained. 2.2 Schmidt-Neuper model method [3] The Schmidt-Neuper calculation model defines two critical states for a single flange joint to withstand the working load Z: state I is the gap at the outer edge of the upper and lower flange contact faces; state II is the upper and lower flange contact faces at the distance The bolt axis x is just separated. Reference [3] gives the relationship between the working load Z and the bolt preload force in the state I and state II as follows: ZI=[(a-0.5b)/(a+b)]·FV (13) ZII=[1/(λ*·q)]·FV (14) Where: λ*=(0.7a+b)/(0.7a); q———The ratio of the stiffness of the gasket to the flange in the flange joint. When the working load Z=0, Z=ZI, Z=ZII and Z are arbitrary values ​​Zarb and larger than ZII, the working tension of the high-strength bolt is FS, and the tensile stress σS is as shown in Table 1, so the working load can be obtained. A piecewise linear relationship between Z and bolt tensile stress. 2.3 Finite Element Analysis Method A single flange connector contact finite element model is created in the MSC. Marc/Mentat environment. As shown in Figure 3, the model uses hexahedral elements, including upper flanges, lower flanges, bolts, nuts, washers, and partial tower walls. The material of each component is low alloy high strength structural steel with an elastic modulus of 2.06×105 MPa, a Poisson's ratio of 0.3 and a density of 7.85×103 kg/m3. Contact relationship setting: The nut, gasket and bolt contact are set to be bonded, the upper and lower flange contacts are set to friction, the gasket is in contact with the flange to set friction, and the friction factor between the contact pairs is 0.15. In the flange connection contact analysis, the analysis process is divided into two conditions: pre-tightening and working. Under pre-tightening conditions, the pre-tightening force FV is applied to the high-strength bolt; under the working condition, the working load Z is applied. Set boundary constraints, constraints The degree of freedom of the node at the bottom of the lower flange wall. The boundary conditions and load settings for a single flange joint model are shown in Figure 4. Under working conditions, the working load Z is loaded by distributed loading. After the load is solved, the maximum stress point of the bolt is found, and the maximum stress of each load step bolt is extracted, and the working load Z is applied according to each step. The size and the corresponding bolt stress can be used to fit the relationship between the load Z of the flange and the stress of the high-strength bolt. 3 instance calculation Taking a ring-shaped flange connection structure of a MW-class horizontal axis wind turbine tower as an example, the structural geometric parameters are shown in Figure 5. Other parameters are as follows: high-strength bolt grade 10.9, size M36×226, pre-tightening force applied FV=510 kN[5], bolt hole dh=39 mm, gasket outer diameter da, w=66 mm, gasket inner diameter di, w=37 mm, gasket thickness tw=6 mm, bolt number n=112. Use M. separately The relationship between the load-carrying Z and the bolt stress obtained by the Seidel curve model, Schmidt/Neuper method and finite analysis method is shown in Fig. 6. M.Seidel curve model, can calculate the bending moment stress caused by eccentric loading, the relationship between bending moment stress variable, tensile stress variable and external load Z is shown in Fig. 7. 4 bolt fatigue life analysis The fatigue life analysis method of the bolt adopts the full-life S-N curve analysis method combining the rain flow cycle counting method and the Palmgren-Miner damage criterion. 4.1 Fatigue conditions and treatment Based on the GL specification [6], the fatigue wind load conditions and related parameters are defined in the wind turbine design software GHBLADED, and the load history of the fatigue working condition at the center of the annular flange connection section of the tower is obtained. Considering the load components Mxy and FZ which mainly affect the bolt stress change under fatigue conditions, the Mxy and FZ time load history under a single fatigue condition is shown in Fig. 8. Equation (15) [1] determines the relationship between the external load Z and Mxy and FZ, so the relationship between the external load Z and the bolt stress can be obtained, as shown in Figure 8. 4.2 Establishment of material S-N curve The fatigue strength of high-strength bolts is related to the diameter of the bolts and can be selected according to EC 3 [7] and VDI 2230. According to VDI2230, for the bolts of 8.8~12.9 with more than 2×106 cycles and the survival rate is 99%, the bolts are heat-treated after the thread is rolled. The fatigue strength of the bolts can be expressed by the following formula: σASV=0.85×(150/d+45) (16) Where: d———the nominal diameter of the bolt. According to the Eurocode3 fatigue rating of 71, for bolts larger than 30mm in diameter, the S-N curve reduction factor KS[6] is introduced to reduce the reference value σC of the fatigue strength at 2×106 of the S-N curve. The corrected fatigue strength is: σC, mod=kS·σC (17) kS=(30/d)0.25 (18) The S-N curve of the high-strength bolt was corrected using Eurocode 3 considering the material safety factor 1.15 [6] and the bolt diameter dimension. The S-N curve is set as shown in Figure 9. 4.3 Fatigue calculation criteria The Palmgren-Miner damage criterion means that under cyclic loading, fatigue damage can be linearly accumulated. Each stress is independent and independent of each other. When the accumulated damage reaches a certain value, the specimen or component will fatigue. damage. In the 20-year design life of a wind turbine, the conditions under which the bolt does not suffer from fatigue damage are: 5 Results and conclusions Based on the GL specification, the requirements for the high-strength bolts of the MW-class wind turbine tower ring flange are connected. The full-life S-N curve analysis method combined with the rain flow cycle counting method and the Palmgren-Miner linear damage criterion, high-strength bolt fatigue Damage assessment can be done at MSC. Fracture is done in the software. In the 20-year design life of the wind turbine, the fatigue life calculation result of the tower flange flange connection high-strength bolt is M. The Seidel curve model is 0.335 4, the Schmidt-Neuper model is 0.443 8, and the finite element analysis model is 0.321 6. Based on the analysis process and results, the following conclusions can be drawn: (1) M. The Seidel curve model considers the bending moment stress caused by the eccentric loading of the high-strength bolt, and the bending moment stress variation and the tensile stress variable of the bolt are in an order of magnitude, and the fatigue damage caused by the bending moment stress cannot be ignored, so this model It is reasonable, the relationship curve is consistent with the measured value [4], but the calculation process is more complicated. (2) Schmidt-Neuper piecewise linear model, the method is simple, but does not consider the bending moment stress generated by eccentric loading on high-strength bolts, and overestimates the bolt stress, so it has certain theoretical defects, which can be used to roughly evaluate bolt fatigue. life. (3) Finite element analysis method, simple method, relationship curve result and M. Seidel is roughly consistent, but under extreme loads, the bolt stress is relatively low, overestimating the fatigue capacity of the bolt. (4) The S-N curve selected by the Eurocode 3 specification is relatively conservative compared to the SDI curve determined by VDI2230, which is closer to the measured value. references [1] Petersen C. Stahlbau[M]. Auflage Braunschweig: Wiesbad-en: Vieweg, 1997. [2] Faulhaber A, ThomalaW. Erl觌uterungenzurRichtlinie VDI2230Blatt1 (1986): Dernichtlineare Berechnungsansatz [J]. VDI-ZBd. 129 Nr. 9 1987. [3] Schmidt H, Neuper M. Zum elastostatischen tragverhalten exzentrisch gezogener LSt觟βe mit vorgespannten schrauben [J]. Stahlbau, 66 (1997), S: 163-168. [4] Seidel M. Ermittlung der ermüdungsbeanspruchung vonschrauben exzen trisch belasteter flanschverbindungen [J]. Hannover, Diss, 2001. [5] VDI2230 Part 1, Systematic calculation of high duty bolted joints joints with one cylindrical bolt [S]. Verein Deutsche Ingenieure, Düsseldorf, Germany, 2003: 48-51. [6] Guideline for the certification of wind turbine[S]. Germanischer Lloyd Wind Energize GmbH. Germany, 2010. [7] Eurocode3, Design of steel structures, Part 1 -9 : Fatigue strength of steel structure [S]. PREN 1993-1-9, 2002. Heat Lamp Oval Marble Carving Station,Oval Marble Carving Station,4-Lamp Carving Station,Marble Culinary Station Shaoxing Biaoyi Hardware Products Co., Ltd. , https://www.byeob.com
