The present disclosure relates generally to structural analysis and design software, and more specifically to techniques for modeling monopiles (e.g., large diameter monopiles with small length-to-diameter ratios) in structural analysis and design software.
Wind, either onshore or offshore, has been recognized as an abundant source of clean, renewable energy. Offshore wind is of particular interest as it is available over large continuous areas, offers stable and high wind speeds with low turbulence and wind shear, and can better address issues of noise and visual impact than onshore wind. To be able to compete with other sources of energy, the life cycle cost of offshore wind projects needs to be minimized. On average, the support structures and foundations for an offshore wind project accounts for about 13% of the total life cycle cost. In order to minimize this portion of the cost, support structure and foundation designs should be safe and effective, but not overly conservative.
Monopiles are recognized as the most cost-effective type of foundation for fixed offshore wind turbines in water depths of up to 40 meters (m). Until recently, most monopiles have been of relatively small diameter (e.g., 5-6 m) with large length-to-diameter ratios. Currently, there is increasing interest in large diameter monopiles (e.g., 10 m+) with small length-to-diameter ratios (e.g., ≤6, for instance ≥2 but ≤6). Such large diameter monopiles may be better suited for deep water installations, where they may support larger wind turbines subject to higher wind, wave and current loads. However, existing techniques for modeling monopiles often lead to overly conservative designs for large diameter monopiles. Conventional structural analysis and design software often uses the “Winkler Model” that models a monopile as a beam supported by a series of nonlinear springs that represent soil response. In the Winkler Model, soil response consists of distributed lateral load along the pile only. That is, the lateral soil reaction force (p) at a particular point on the pile is a function of the lateral displacement (y) at the point solely. A nonlinear function referred to as a p-y curve is employed which relates p to y. Different p-y curves have been determined over the years for different soil types through field testing. The Winkler Model has proven well suited to the design of slender piles (e.g., <1 m) with large length-to-diameter ratios (e.g., >30), for example, those used in offshore jacket structures employed by the oil and gas industry. However, the Winkler Model typically leads to overly conservative designs for large diameter piles (e.g., 10 m+) with small length-to-diameter ratios (e.g., ≤6, for instance ≥2 but ≤6). Concurrent wind, wave and current loads may produce a large overturning moment, and the Winkler Model typically does not adequately capture this behavior.
Recently, attempts have been made to replace use of the Winkler Model in structural analysis and design software with a new model referred to as the Pile-Soil Analysis (PISA) Model. The PISA Model introduces three other soil reactions, namely: distributed moment along the length of the pile, base moment at the pile tip, and base shear force at the pile tip. Although the PISA Model has shown promise in the design of monopiles, it too suffers shortcomings. The PISA Model only considers lateral loading and does not consider axial or combined loading on a pile, which may play an important role due to gravity loading of the supported structure. Further, the PISA Model requires new soil reaction curves which generally need to be determined for different soil types through new field testing. Such new soil reaction curves have not yet been standardized.
Accordingly, there is a need for improved techniques for modeling monopiles (e.g., large diameter monopiles with small length-to-diameter ratios) in structural analysis and design software.
In example embodiments, a new model for modeling monopiles (e.g., large diameter monopiles) in structural analysis and design software is provided that, in addition to distributed lateral load along the monopile, considers distributed moment along the length of the pile, base moment at the pile tip, and base shear force at the pile tip. The new model may avoid overly conservative designs for large diameter piles (e.g., 10 m+) with small length-to-diameter ratios (e.g., ≤6, for instance ≥2 but ≤6), while using standardized reaction curves (i.e., p-y curves and t-z curves) and considering axial and combined loading.
In one example embodiment, structural analysis and design software accesses a p-y curve and a t-z curve for a type of soil about a monopile at each of one or more layers, the monopile including a plurality of nodes in the one or more layers. As understood in the field, each p-y curve is a nonlinear function that relates lateral soil reaction force (p) to lateral displacement (y), and each t-z curve is a nonlinear function that relates skin friction (t) to vertical displacement (z). The software determines a distributed moment along the monopile and based thereon a rotational stiffness of a given node, wherein the rotational stiffness is calculated as a function of both force per unit area from a respective t-z curve for the layer of the given node and slope of the respective t-z curve for the layer of the given node. The software determines a base moment at the monopile's tip and based thereon a base rotational stiffness, wherein the base rotational stiffness is calculated as a function of both force per unit area from a respective p-y curve for the layer of the pile tip and slope of the respective p-y curve for the layer of the pile tip. The software also determines a base shear force at the pile tip and based thereon base shear stiffness, wherein the base shear stiffness is calculated as a function of slope of a respective t-z curve for the layer of the pile tip and pile tip area. The software iteratively solves a system of non-linear equations that uses a structural stiffness matrix that includes the rotational stiffness, the base rotational stiffness and the base shear stiffness in addition to pile material structural stiffness to model the monopile, and eventually outputs a description of the modeled monopile.
It should be understood that a variety of additional features and alternative embodiments may be implemented other than those discussed in this Summary. This Summary is intended simply as a brief introduction to the reader and does not indicate or imply that the examples mentioned herein cover all aspects of the disclosure or are necessary or essential aspects of the disclosure.
The description below refers to the accompanying drawings of example embodiments, of which:
The example software architecture may be a portion of structural analysis and design software. In one embodiment, the structural analysis and design software may be the SACS™ software available from Bentley Systems, Inc. and the portion may be part of the Pile 3D™ module of such software. It should be remembered, however, that the new model for modeling monopiles may also be implemented using different structural analysis and design software and/or different portions thereof.
In one example embodiment, three sub modules of the Pile 3D™ module may perform the major operations needed to model a monopile. A pile input (pilein) module 110 may read input data regarding pile material and geometry from a repository (not shown). Likewise, a soil input (soilin) module 120 may read input data regarding soil properties for the soil about the pile form the repository. Behaviors of soil typically can be modeled based on standardized reaction curves, and the soil input (soilin) module 120 may be responsible for accessing those curves, when needed. When the new model for modeling monopiles is to be utilized, the soil input (soilin) module 120 may be responsible for fetching a p-y curve and t-z curve for the type of soil about the monopile at each of several layers in the soil. Soil type may differ along the pile with its characteristics varying by depth. A pile analysis (pilesn) module 130 may perform finite element (FE) analysis to model the monopile.
A number of further submodules make up modules 110-120 and be responsible for various subtasks. A FE pile (fepile) module 132 may organize the other submodules. A FE soil (fesoil) module 134 may be responsible for calculating soil stiffness components and corresponding soil resistance forces. The new model for modeling monopiles may be implemented mainly in the FE soil (fesoil) module 134.
A FE stiffness (festif) module 136 may assemble the soil stiffness components and corresponding soil resistance forces and as three springs attached to nodes along the pile and three springs attached to the pile tip and add them to a structural stiffness matrix that describes pile material structural stiffness and to a load vector that describes load on the pile. Rotational stiffness at a given element along the monopile, base rotational stiffness of the monopile and base shear stiffness of the monopile calculated as part of the new model may be added to an appropriate degree of freedom in the structural stiffness matrix by the FE stiffness (festif) module 136. The size of the structural stiffness matrix and load vector may depend on a number of elements the pile is divided into (e.g., 100 to 1000). In some implementations the number of elements may be user selectable. Give n elements of the pile, the load vector may be a on by 1 vector and the structural stiffness matrix may be a 6n by 6n matrix. Loading on top of the pile may be divided into a number of load steps m (e.g., 100 to 10000). In some implementations, the number of load step may be user selectable.
A FE solver (fesolve) 138 module may iteratively solve a system of non-linear equations, updating the structural stiffness matrix and load vector over each of the load steps, until convergence, in order to model the monopile. Depending on the size of the load steps, the number of iterations to achieve convergence may vary (e.g., up to 1000 iterations). Further, a FE stress (frstrs) module 139 may calculate internal stresses, stains and perform code checks to determine whether the monopile meets code requirements.
In one embodiment, the new model for modeling monopiles may be activated in the structural analysis and design software using options in input lines, configured in a user interface of the software.
The new model for modeling monopiles may be based on the foundations of the traditional Winkler Model. As discussed above, the Winkler Model assumes that lateral soil reaction force (p) at a particular point on the pile is a function of the lateral displacement (y) at the point solely, which is represented by a p-y curve. A similar assumption may be made that the soil reaction force due to skin friction (t) is a function of displacement (z) along with friction force on the surface, which is represented by a nonlinear function referred to as a t-z curve.
While the traditional Winkler Model assumes soil reaction includes only distributed lateral load along the pile, three other soil reactions may be desirable to include when designing large diameter piles (e.g., 10 m+) with small length-to-diameter ratios (e.g., ≤6, for instance ≥2 but ≤6), namely distributed moment, shear force at the pile tip, and moment at the pile tip.
The reactions shown in
Looking first to distributed lateral load and rotational stiffness kθ
First, surface area (δAj) of the element j 420 may be obtained. The x coordinate to element j can be expressed as:
dj=R Cos α (1)
Vertical length of element j for small rotation θ (tan θ≅0) can be obtained as:
Zj=θdj (2)
Therefore, the area of the element j can be obtained using equations (1) and (2) as:
δAj=R(δα)Zj=Rθdj(δα) (3)
Using a t-z curve, force per unit area on the element (Tj) can be obtained using the elemental area δAj. The force on element (δFj) may be written as:
δFj=Tj(δAj)=RθTjdj(δα) (4)
In equation (4), dj and Tj are functions of α. As such, moving around the pile perimeter, skin friction force on each element (δFj) is varying. Moment δMj produced by friction force on the element around the y-axis can be calculated using equations (1)-(4) as:
δMj=(δFj)dj=Tj(δAj)dj=RθTjdj(δα) (5)
Therefore, the total moment can be expressed as:
Integrating equation (6) numerically results in the total moment produced by skin friction around the y-axis for a corresponding node. Moreover, equations (5) and (6) show that moment produced by element friction is a function of element rotation (θ), and unit force Tj obtained from a t-z curve which is itself a function of element rotation, which may be written as:
Mj=Mj(Tj(Zj(θ)),θ) (7)
Therefore, rotational stiffness kθ
Substituting from equations (2) and (5) results in:
Therefore, rotational stiffness kθ
Parameters in equation (10) show that, when a t-z curve is used to calculate the distributed moment along the monopile, the rotational stiffness of a given node is a function of both the values of Tj and the slope of the t-z curve (dTj/dZj).
One may look next to base moment at the monopile's tip and base rotational stiffness.
In determining the force on an element, its surface area (δAj) may be determined. The x coordinate of element j can be expressed as:
dj=R Cos α (11)
Vertical length of element j due to pile tip rotation (θb) of element j can be obtained as:
Zj=θbdj (12)
Since perimeter of a circle is perpendicular to its radius, the width of the element (St) can be expressed as:
δt≃R Sin α(δα) (13)
The area of the element j on the pile tip surface, shown in
δAj=2R Sin α(δt)=2R2 Sin2α(δα) (14)
Using a p-y curve, force per unit area on the element (Pj) can be obtained using the vertical displacement Z. Therefore, total force on element (δFj) may be written as:
δFj=PjδAj=2R2 Sin2αPj(δαj) (15)
In equation (15), Pj is a function of a. In other words, rotation of the pile tip surface results in variation of force on each element δFj. Moment Mj produced by lateral movement of the pile tip with respect to the x-axis on the element around the y-axis may be calculated using equations (11)-(15) as follows:
δMb=(δFj)Zj=Pj(δAj)Zj=2R2 Sin2αPjθbdj(δα) (16)
Therefore, the base moment (Mb) can be obtained by integrating equation (16) with respect to α, such that:
Integrating equation (17) numerically results in the total base moment produced due to lateral soil resistance at the pile tip around the y-axis. Integration limits between 0 and π/2 will cover half of the pile tip area. Equations (16) and (17) show that moment produced by the element displacement is a function of element rotation (θb), and unit force (Pj) obtained from a p-y curve, which is itself a function of element rotation, such that:
Mb=Mb(Pj(Zj(θb)),θb) (18)
Therefore, rotational stiffness is given as the derivative of moment with respect to rotation, such that:
Substituting from equations (12) and (16) results in:
Therefore, base rotational stiffness KM
Parameters in equation (21) show that, when a p-y curve is used to calculate the base moment at the pile tip, the corresponding rotational stiffness is a function of both the values of Pj and the slope of the p-y curve (dPj/dZj).
One may finally look to base shear force at the pile tip and base shear stiffness KV
Fb=Tb(πR2) (22)
Equation (22) shows that shear force produced by the skin friction is a function of unit force Tb obtained from the t-z curve, which is itself a function of displacement, written as:
Fb=Fb(Tb(Xb)) (23)
Therefore, stiffness may be written as:
Substituting from equation (22) results in:
Equations (22) and (25) show that values obtained from a t-z curve and its slope (dTb/dXb) multiplied by the pile tip area (πR2) can be used as the base shear force and base shear stiffness, respectively.
The above-described techniques may have a number of advantages over techniques that use a conventional Winkler Model or PISA model. The new model may avoid overly conservative designs for large diameter piles (e.g., 10 m+) with small length-to-diameter ratios (e.g., ≤6, for instance ≥2 but ≤6). It may further use only standardized reaction curves (i.e., p-y curves and t-z curves). Likewise, it may consider both axial and combined loading. Such advantages may lead to better monopile designs (e.g., that are not overly conservative while properly consider all loading), and also may improve computation efficiency (e.g., by removing the need to perform additional field testing or and then calculating additional, non-standard reaction curves).
It should be understood that a wide variety of adaptations and modifications may be made to the techniques. While specific example software and hardware is discussed above, it should be understood that the techniques may be implemented using a variety of different types of software, hardware, and combination thereof. Such software may include executable instructions stored in a non-transitory computing device-readable medium, such as a volatile or persistent memory device, a hard-disk, or other data storage. Such hardware may include a variety of types of processors, memory chips, programmable logic circuits, application specific integrated circuits, and/or other types of hardware components that support execution of software. Combinations of software and hardware may be adapted to suit different environments and applications. Above all, it should be understood that the above descriptions are meant to be taken only by way of example.
Number | Name | Date | Kind |
---|---|---|---|
20160326708 | Sinnreich | Nov 2016 | A1 |
Number | Date | Country |
---|---|---|
112861409 | May 2021 | CN |
118110193 | May 2024 | CN |
Entry |
---|
Alderlieste, Etienne, et al., “Experimental Investigation into Pile Diameter Effects of Laterally Loaded Mono-Piles,” ASME, Proceedings of the ASME 2011 30th International Conference on Ocean, Offshore and Arctic Engineering, OMAE2011, OMAE2011-50068, Rotterdam, The Netherlands, Jun. 19-24, 2011, pp. 1-6. |
Allotey, Nii, et al., “Generalized Dynamic Winkler Model for Nonlinear Soil-Structure Interaction Analysis,” NRC Canada, NRC Research Press Web, Canada Geotechnical Journal, vol. 45, May 5, 2008, pp. 560-573. |
Brinkgreve, Ronald, et al., “Validation and Application of A New Software Tool Implementing the PISA Design Methodology,” MDPI, Journal of Marine Science and Engineering, vol. 8, No. 457, Jun. 21, 2020, pp. 1-17. |
Bouman, Thijs, “A Winkler Model for the Seismic Analysis of Monopile Foundations: An Exploratory Study on the Modeling of Soil-Structure Interaction During Earthquakes,” NTNU—Norwegian University of Science and Technology, Department of Marine Technology, Dec. 2018, pp. 1-106. |
Burd, Harvey J., et al., “Application of the PISA Design Model to Monopiles Embedded in Layered Soils,” ICE Publishing, Geotechnique, vol. 70, No. 11, Sep. 14, 2020, pp. 1067-1082. |
Burd, Harvey J., et al., “New Data Analysis Methods for Instrumented Medium-Scale Monopile Field Tests,” ICE Publishing, Geotechnique, vol. 70, No. 11, Nov. 15, 2019, pp. 961-969. |
Burd, H. J., et al., “PISA Design Model for Monopiles for Offshore Wind Turbines: Application to a Marine Sand,” ICE Publishing, Geotechnique, vol. 70, No. 11, Feb. 12, 2020, pp. 1048-1066. |
Byrne, Byron W., et al., “Monotonic Laterally Loaded Pile Testing in a Stiff Glacial Clay Till at Cowden,” ICE Publishing, Geotechnique, vol. 70, No. 11, Nov. 15, 2019, pp. 970-985. |
Byrne, Byron W., et al., “PISA Design Model for Monopiles for Offshore Wind Turbines: Application to a Stiff Glacial Clay Till,” ICE Publishing, Geotechnique, vol. 70, No. 11, Feb. 6, 2020, pp. 1030-1047. |
Gerolymos, Nikos, et al., “Winkler Model for Lateral Response of Rigid Caisson Foundations in Linear Soil,” Elsevier Ltd, Elsevier, Soil Dynamics and Earthquake Engineering, vol. 26, Jan. 24, 2006, pp. 347-361. |
Gupta, B. K., et al., “Applicability of Timoshenko, Euler-Bernoulli and Rigid Beam Theories in Analysis of Laterally Loaded Monopolies and Piles,” ICE Publishing, Geotechnique, vol. 68, No. 9, Jan. 3, 2018, pp. 772-785. |
Haiderali, A. E., et al., “Evaluation of the P-Y Method in the Design of Monopiles for Offshore Wind Turbines,” Offshore Technology Conference, OTC 24088, Houston, Texas, USA, May 6-9, 2013, pp. 1-21. |
Lam, Ignatius, P. O., “Diameter Effects on P-Y Curves,” Deep Foundations Institute, Jan. 1, 2013, pp. 1-15. |
Mcadam, Ross A., et al., “Monotonic Laterally Loaded Pile Testing in a Dense Marine Sand at Dunkirk,” ICE Publishing, Geotechnique, vol. 70, No. 11, Nov. 15, 2019, pp. 986-998. |
Ming-Yuan, Wang, et al., A Dynamic Winkler Model to Analyze Offshore Monopile in Clayey Foundation Under Cyclic Load, EJGE, vol. 21, Bund. 2, Jan. 2016, pp. 2029-2041. |
Terceros, M., et al., “Evaluation of P-Y Approaches for Piles in Soft Clay,” SUT—Society for Underwater Technology, Smarter Solutions for Future Offshore Developments, Pile Design and Installation, OSIG 2017, Sep. 12, 2017, pp. 724-731. |
Zdravkovic, Lidija, et al., “Finite-Element Modelling of Laterally Loaded Piles in a Stiff Glacial Clay Till at Cowden,” ICE Publishing, Geotechnique, vol. 70, No. 11, Nov. 15, 2019, pp. 999-1013. |
Zdravkovic, Lidija, et al., “Ground Characterisation for PISA Pile Testing and Analysis,” ICE Publishing, Geotechnique, vol. 70, No. 11, Nov. 15, 2019, pp. 945-960. |