Bearing performance degradation evaluation method for offshore wind power foundation structure

Abstract

The invention relates to the technical field of offshore wind power, and discloses a bearing performance degradation evaluation method for an offshore wind power foundation structure, which effectively solves the problem of neglecting corrosion heterogeneity in the prior art by acquiring initial three-dimensional geometric point cloud data and historical service data and constructing a first finite element model. The model better fits the actual structure state; the local damage evolution and cumulative effect are simulated by means of a second finite element model, meanwhile, material degradation evaluation is combined, the interaction of three key factors including corrosion thinning, local damage and material mechanical property degradation is comprehensively considered, and the defect that the multi-factor coupling influence is not fully solved in the prior art is overcome; the extreme environment load condition is introduced to construct the third finite element model, the performance degradation law of the structure in the extreme environment can be analyzed, and then the reliability and applicability of evaluation are improved by establishing a mathematical relation model and predicting the full life cycle bearing performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of offshore wind power, and particularly relates to a bearing performance degradation evaluation method of offshore wind power foundation structure. BACKGROUND

[0002] As a pillar technology in the field of clean energy, the structural safety and long-term service ability of offshore wind power directly relate to the stability of energy supply and the sustainability of economic benefits. Tower, jacket and foundation are the core bearing components of offshore wind power system, which are exposed to harsh marine environment for a long time. Corrosion inevitably becomes a key factor affecting their performance. Accurate evaluation of bearing performance degradation after corrosion is not only a necessary means to ensure the safe operation of the structure, but also an important way to extend the service life of wind power facilities and reduce maintenance costs. However, the current research and practice still face significant technical bottlenecks in this field.

[0003] Existing evaluation methods mostly rely on simplified assumptions or empirical formulas, such as bearing capacity estimation based on uniform corrosion thinning, or overall performance derived from local test data. These methods often ignore the non-uniformity of corrosion, the cumulative effect of local damage and the complex coupling effect of material mechanical performance degradation, resulting in a large deviation between the evaluation results and the actual service state. Especially in extreme marine environments such as typhoon or wind wave flow coupling, the true response characteristics of the structure are difficult to be accurately captured, and the reliability and applicability of the evaluation are questioned. SUMMARY

[0004] Therefore, the present application provides a bearing performance degradation evaluation method of offshore wind power foundation structure to solve the problem that the existing evaluation methods ignore the non-uniformity of corrosion, the cumulative effect of local damage and the complex coupling effect of material mechanical performance degradation, resulting in a large deviation between the evaluation results and the actual service state.

[0005] In the first aspect, the present application provides a bearing performance degradation evaluation method of offshore wind power foundation structure, which comprises:

[0006] An initial three-dimensional geometric point cloud data set and a historical service data set of an offshore wind power foundation structure are acquired; based on the initial three-dimensional geometric point cloud data set and the historical service data set, corrosion thinning characteristics of the offshore wind power foundation structure are analyzed to obtain a first geometric distribution characteristic data set and a first finite element model containing non-uniformity; based on the first geometric distribution characteristic data set, an evolution path and a cumulative effect of local damage of the offshore wind power foundation structure are simulated by using a second finite element model to obtain a target coupling distribution data set between local damage and stress concentration, the second finite element model being obtained by introducing an initial assumption condition of local damage in the first finite element model; based on the target coupling distribution data set and a preset experimental degradation data set of material mechanical properties, a weakening degree of material degradation on residual resistance is evaluated and a target material performance characterization data set after comprehensive degradation is determined by using the second finite element model; based on the target material performance characterization data set, a performance degradation law of the offshore wind power foundation structure is analyzed by using a third finite element model, and a mathematical relationship model between corrosion degree and bearing capacity is established, the third finite element model being obtained by introducing an extreme environmental load condition in the second finite element model; by using the mathematical relationship model and the third finite element model, a bearing performance degradation condition of the offshore wind power foundation structure is predicted to obtain a full life cycle bearing performance prediction result of the offshore wind power foundation structure.

[0007] The bearing performance degradation evaluation method of the offshore wind power foundation structure provided by the application can in-depth analyze corrosion thinning characteristics by acquiring an initial three-dimensional geometric point cloud data and historical service data, can construct a first finite element model containing non-uniformity, can effectively solve the problem of ignoring corrosion non-uniformity in the prior art, and can make the model more conform to the actual structure state. Further, the second finite element model is used to simulate local damage evolution and cumulative effect to obtain coupling distribution between local damage and stress concentration, and material degradation evaluation is combined to comprehensively consider the interaction of the three key factors of corrosion thinning, local damage and material mechanical property degradation, and the defects of not sufficiently solving the multi-factor coupling influence in the prior art are overcome. Further, the third finite element model is constructed by introducing an extreme environmental load condition, can analyze the performance degradation law of the structure under the extreme environment, and further can predict the full life cycle bearing performance by establishing a mathematical relationship model, changes the condition that the existing evaluation method is difficult to accurately capture the real response of the structure under the extreme environment, and improves the reliability and applicability of the evaluation. Further, the full life cycle bearing performance prediction result finally predicted can provide a scientific basis for safety evaluation and maintenance decision of the offshore wind power foundation structure, is helpful for prolonging the service life and reducing the operation and maintenance cost, and realizes effective evaluation of the long-term service ability of the structure.

[0008] In an optional embodiment, based on the initial three-dimensional geometric point cloud data set and the historical service data set, the corrosion thinning characteristics of the offshore wind power foundation structure are analyzed to obtain a first geometric distribution characteristic data set and a first finite element model containing non-uniformity, including:

[0009] Based on the initial three-dimensional geometric point cloud data set and the historical service data set, the geometric characteristics of the offshore wind power foundation structure are analyzed and a target digital model and a second geometric distribution characteristic data set of the offshore wind power foundation structure are determined; based on the non-uniformity parameter set, a fourth finite element model is constructed by using a finite element analysis method, and a stress distribution initial mapping relationship is determined according to the fourth finite element model; based on the non-uniformity parameter set, a fourth finite element model is constructed by using a finite element analysis method, and a stress distribution initial mapping relationship is determined according to the fourth finite element model, the stress distribution initial mapping is used to represent the mapping relationship from the corrosion geometric characteristics to the mechanical properties; based on the stress distribution initial mapping relationship, the influence degree of non-uniformity on the geometric characteristics is analyzed by using the fourth finite element model, and a first finite element model containing non-uniformity and a first geometric distribution characteristic data set are determined.

[0010] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the application uses a point cloud data-based surface roughness analysis method to extract non-uniformity parameters of corrosion thinning, which can more accurately describe the irregular distribution of corrosion compared with the simple estimation based on uniform corrosion thinning in the prior art. Further, the non-uniformity parameters are processed by a finite element analysis method to construct a fourth finite element model and determine a stress distribution initial mapping relationship, which realizes accurate mapping from corrosion geometric characteristics to mechanical properties and avoids the evaluation deviation caused by ignoring the influence of geometric characteristic changes on mechanical properties in the prior art. Further, the influence degree of non-uniformity is analyzed and a first finite element model is determined, so that the model can more truly reflect the geometric and mechanical characteristics of the structure after corrosion, and the accuracy of the first finite element model is improved.

[0011] In an optional embodiment, based on the initial three-dimensional geometric point cloud data set and the historical service data set, the geometric characteristics of the offshore wind power foundation structure are analyzed and a target digital model and a second geometric distribution characteristic data set of the offshore wind power foundation structure are determined, including:

[0012] Based on an initial three-dimensional geometric point cloud data set, a first digital model is established; based on a historical service data set, a first distribution range of corrosion thinning is determined by processing the first digital model using a Monte Carlo simulation method; according to the first distribution range, non-uniformity characteristics of the offshore wind power foundation structure are analyzed and a non-uniformity distribution data set is obtained; based on the non-uniformity distribution data set, the geometric characteristics of the offshore wind power foundation structure are analyzed and the first digital model is updated using a finite element analysis method, and a second digital model is obtained; the geometric change data set of the offshore wind power foundation structure is determined using the second digital model; when the geometric change data set does not meet the first preset requirement, a future distribution range of corrosion thinning is predicted using a machine learning algorithm, and a second distribution range is obtained; according to the second distribution range, the geometric characteristics of the second digital model are adjusted, and a target digital model and a second geometric distribution characteristic data set are obtained.

[0013] The bearing performance degradation evaluation method of the offshore wind power foundation structure provided by the application uses the Monte Carlo simulation method to process the historical service data and determine the distribution range of corrosion thinning, and simultaneously, uses the machine learning algorithm to predict the future distribution range, effectively solving the problem of inaccurate corrosion distribution prediction in the prior art, so that the model can dynamically adapt to the development of corrosion. Further, the digital model is updated and adjusted according to the corrosion distribution range, and a target digital model and a geometric distribution characteristic data set are obtained, ensuring that the model can accurately reflect the change of the geometric characteristics of the structure. Further, by judging whether the geometric change data set meets the preset requirement, the model is dynamically adjusted, enhancing the adaptability of the model to different corrosion degrees and structural changes, making the model more practical.

[0014] In an optional implementation, based on the first geometric distribution characteristic data set, the evolution path and cumulative effect of the local damage of the offshore wind power foundation structure are simulated using a second finite element model to obtain a target coupling distribution data set between the local damage and the stress concentration, including:

[0015] Based on the first geometric distribution characteristic data set, an initial assumption condition of the local damage is introduced into the first finite element model to generate a fifth finite element model and a third geometric distribution characteristic data set; based on the third geometric distribution characteristic data set, the fifth finite element model is iteratively solved to obtain an evolution path of the local damage; according to the evolution path, the cumulative effect of the local damage is analyzed and an initial coupling distribution data set between the local damage and the stress concentration is determined; according to the initial coupling distribution data set, a position change trend of a first key region of the stress concentration is determined; according to the position change trend, the boundary conditions of the fifth finite element model are updated and a second finite element model is determined; according to the second finite element model, the target coupling distribution data set is determined.

[0016] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the application introduces a local damage initial assumption condition in the first finite element model, and obtains a local damage evolution path through iterative solution, which can accurately simulate the development process of local damage, and overcomes the defect that the damage evolution is difficult to simulate in the prior art. Further, the cumulative effect of the local damage is analyzed according to the evolution path, the coupling distribution of the local damage and the stress concentration is determined, the stress concentration and failure mode transition caused by the accumulation of local damage are effectively captured, and the problem that the cumulative effect is not considered in the prior art is solved. Further, the model boundary conditions are updated according to the position change trend of the stress concentration key area, so that the second finite element model can more accurately reflect the mechanical properties of the structure in the damage evolution process, and the accuracy and reliability of the model are improved.

[0017] In an optional implementation, based on the target coupling distribution data set and the preset experimental degradation data set of the material mechanical properties, the second finite element model is used to evaluate the weakening degree of the material degradation on the residual resistance and determine the target material performance characterization data set after comprehensive degradation, including:

[0018] Based on the target coupling distribution data set and the preset experimental degradation data set, the initial mechanical property parameter set is determined; the time sequence in the preset experimental degradation data is processed by using a numerical mapping method to obtain a degradation parameter sequence; the degradation parameter sequence is input into the second finite element model to obtain the initial material performance characterization data set after simulation degradation; when the initial material performance characterization data set does not meet the third preset requirement, the residual resistance weakening degree is calculated and the resistance change trend is determined according to the material state data set; and the target material performance characterization data set after comprehensive degradation is determined according to the resistance change trend and the initial mechanical property parameter set.

[0019] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the application uses a numerical mapping method to integrate the material degradation time sequence parameters into the finite element model, which can quantitatively evaluate the weakening degree of the material degradation on the residual resistance, and changes the inaccurate evaluation of the influence of material mechanical property degradation in the prior art. Further, when the material performance after simulation degradation does not meet the requirement, the residual resistance weakening degree is calculated and adjusted in time, which ensures that the target material performance characterization data set can accurately reflect the performance state of the material after comprehensive degradation, and provides support for subsequent quantitative evaluation of the comprehensive influence of corrosion thinning, local damage and material degradation.

[0020] In an optional implementation, based on the target material performance characterization data set, the third finite element model is used to analyze the performance degradation law of the offshore wind power foundation structure, and a mathematical relationship model between the corrosion degree and the bearing capacity is established, including:

[0021] Based on the target material performance characterization dataset, the structural characteristics of the offshore wind power foundation structure are analyzed and the structural characteristic curve of the offshore wind power foundation structure is determined by using the third finite element model; according to the structural characteristic curve and the third finite element model, the residual bearing capacity of the offshore wind power foundation structure is quantitatively evaluated, and the quantitative evaluation result is obtained; according to the quantitative evaluation result, the performance degradation law of the offshore wind power foundation structure is analyzed, and the target performance degradation dataset is obtained; based on the target performance degradation dataset, a mathematical relationship model between the corrosion degree and the bearing capacity is established by using a polynomial regression method.

[0022] The bearing performance degradation evaluation method of the offshore wind power foundation structure provided by the application can analyze the structural characteristics by using the third finite element model and determine the structural characteristic curve, which can comprehensively reflect the static and dynamic response characteristics of the structure after extreme environment and comprehensive degradation, and solves the problem that it is difficult to comprehensively analyze the structural response in the prior art. Further, the residual bearing capacity is quantitatively evaluated according to the structural characteristic curve and the finite element model, and compared with the estimation based on the simplified assumption or empirical formula in the prior art, the evaluation result is more accurate and reliable. Further, the mathematical relationship model between the corrosion degree and the bearing capacity is established based on the performance degradation law and the polynomial regression method, which provides a scientific mathematical tool for the evaluation and prediction of the structural bearing performance, and is convenient for practical engineering application.

[0023] In an optional embodiment, based on the target material performance characterization dataset, the structural characteristics of the offshore wind power foundation structure are analyzed and the structural characteristic curve of the offshore wind power foundation structure is determined by using the third finite element model, comprising:

[0024] Based on the target material performance characterization dataset, an initial calculation framework is constructed in the third finite element model and an initial simulation basic dataset is generated; the initial simulation basic dataset is processed by using dynamic simulation technology and the change law of the static response and the dynamic response is calculated; the dynamic response distribution dataset is determined according to the change law of the static response and the dynamic response; the second key area is extracted from the dynamic response distribution dataset and the stress concentration distribution result is determined; the type and characteristics of the failure model of the offshore wind power foundation structure are analyzed according to the stress concentration distribution result and the failure mode evolution dataset is determined; the structural characteristic curve is determined according to the failure mode evolution dataset.

[0025] The method for evaluating the bearing performance degradation of the offshore wind power foundation structure provided by the application can process initial simulation foundation data sets and calculate the change rule of static and dynamic responses by using dynamic simulation technology, can truly simulate the dynamic response process of the structure under the action of extreme environmental load, and overcomes the defects of insufficient simulation of dynamic response in the prior art. Further, by analyzing the stress concentration distribution results and the failure mode evolution data sets, the failure mode type and characteristics of the structure can be accurately determined, which provides an important basis for safety evaluation and optimal design of the structure. Further, the structure characteristic curve is determined according to the failure mode evolution data sets, which comprehensively reflects the response characteristics of the structure at different loads and degradation stages, and provides intuitive data support for the performance evaluation and prediction of the structure.

[0026] In an optional embodiment, according to the structure characteristic curve and the third finite element model, the residual bearing capacity of the offshore wind power foundation structure is quantitatively evaluated to obtain a quantitative evaluation result, including:

[0027] According to the structure characteristic curve, the stress value and the displacement value of the key nodes are extracted in the third finite element model to generate an initial data set; according to the initial data set and the initial bearing performance, the change trend of the corrosion thinning on the bearing performance is determined; according to the change trend, the distribution characteristics of the local damage are determined; according to the distribution characteristics, the influence degree of the material degradation is determined; based on the change trend, the distribution characteristics and the influence degree, the weight distribution data set is obtained by processing through the support vector machine algorithm; and according to the weight distribution data set, the quantitative evaluation result of the residual bearing capacity of the offshore wind power foundation structure is determined.

[0028] The application provides a bearing performance degradation evaluation method of an offshore wind power foundation structure, which is characterized by the following steps: accurately positioning and extracting stress and displacement data of key nodes in the third finite element model from the structural characteristic curve to generate an initial data set, solving the problem of non-targeted data extraction in the prior art and ensuring that the analysis is based on the most critical stress position of the structure; comparing the initial data set with the initial bearing performance of the structure to quantify parameters such as stress increase and displacement change caused by corrosion thinning, and then determining the specific influence direction and degree of corrosion thinning on the bearing performance; further, based on the bearing performance change trend caused by corrosion thinning, the distribution law of local damage in the structure is analyzed in depth, the derivation from macro performance change to micro damage distribution is realized, and the problem that it is difficult to associate corrosion and damage distribution in the prior art is solved; further, the weakening degree of the material mechanical performance degradation on the residual resistance of the structure is evaluated in combination with the local damage distribution characteristics, the influence difference of material degradation at different damage positions is determined, and the status of general evaluation of the influence of material degradation in the prior art is changed; further, the change trend of corrosion thinning, the local damage distribution characteristics and the influence degree of material degradation are comprehensively analyzed by using a support vector machine algorithm, and the weight distribution of the influence of each factor on the bearing performance is determined, the subjectivity of artificial experience assignment is avoided, the weight distribution is more in line with the law of actual multi-factor coupling, and the scientificity and accuracy of the evaluation are improved; further, according to the weight distribution data set, the influences of corrosion thinning, local damage and material degradation on the bearing performance are weighted and calculated, and a quantitative evaluation result of the residual bearing capacity can be obtained, which provides an intuitive and quantitative basis for the safety evaluation and maintenance decision of the structure, and solves the problem of lack of accurate quantification of the evaluation result in the prior art.

[0029] In an optional embodiment, according to the quantitative evaluation result, the performance degradation law of the offshore wind power foundation structure is analyzed to obtain a target performance degradation data set, including:

[0030] Based on the quantitative evaluation result, a Monte Carlo simulation method is used to introduce random variables and generate a bearing capacity data set under different corrosion degrees and extreme environment combinations; the bearing capacity data set is statistically analyzed to determine the bearing capacity change trend; based on the bearing capacity change trend, a probability distribution model is used to determine the probability distribution interval of the bearing performance; the performance degradation parameters are calculated according to the probability distribution interval to determine the target law data set of the performance degradation; according to the target law data set, the performance degradation law data set is determined; and according to the performance degradation law data set, the target performance degradation data set is determined.

[0031] The application provides a bearing performance degradation evaluation method of an offshore wind power foundation structure. The method introduces random variables by using a Monte Carlo simulation method, simulates bearing capacity changes under different corrosion degrees and extreme environment combinations, fully considers random and uncertain factors in actual engineering, and makes the evaluation result more in line with actual conditions. Further, the method determines a probability distribution interval of bearing performance by statistical analysis and a probability distribution model, provides an evaluation on a probability level for long-term service prediction of the structure, and changes the limitation of a deterministic evaluation in the prior art. Further, the method establishes a target performance degradation data set based on a performance degradation parameter and a data fitting method, realizes dynamic description of a structure performance degradation law, and can more accurately predict bearing capacity changes of the structure at different service stages.

[0032] In an optional embodiment, the bearing performance degradation of the offshore wind power foundation structure is predicted by using a mathematical relationship model and a third finite element model, and full life cycle bearing performance prediction results of the offshore wind power foundation structure are obtained, including:

[0033] The prediction accuracy of the mathematical relationship model at different service stages is judged, and a bearing performance dynamic evaluation relationship based on corrosion evolution is determined. According to the bearing performance dynamic evaluation relationship based on corrosion evolution, the boundary conditions and load parameters of the third finite element model are adjusted, and an adjusted model parameter set is obtained. The model parameter set is simulated, and a structure behavior data set is generated. Extreme environment features are extracted from the structure behavior data set, and an influence coefficient of the extreme environment on the residual resistance is determined according to the extreme environment features. The influence coefficient is processed by using a regression analysis algorithm, and a target influence trend data set of the residual resistance of the offshore wind power foundation structure is determined. According to the target influence trend data set and a preset life cycle time sequence, a bearing capacity change curve is determined. The bearing capacity change curve is analyzed, and full life cycle bearing performance prediction results of the offshore wind power foundation structure are determined.

[0034] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the application can adapt to the performance changes of the structure in different service stages by judging the prediction accuracy of the mathematical relationship model in different service stages and establishing a dynamic evaluation relationship based on corrosion evolution, and realizes dynamic evaluation of the bearing performance of the structure. Further, adjusting the boundary conditions and load parameters of the third finite element model according to the dynamic evaluation relationship can make the model more fit the stress state and environmental conditions of the structure in the actual service process. Further, simulating the adjusted model parameter set can generate a data set reflecting the behavior of the structure in the future service scenario, overcoming the defect that the future service scenario simulation in the prior art is not comprehensive. Further, extracting extreme environmental features from the structure behavior data set can quantitatively analyze the influence degree of extreme environment on the residual resistance of the structure and determine the influence coefficient, solving the problem that it is difficult to accurately evaluate the influence of extreme environment on the residual resistance of the structure in the prior art. Further, using a regression analysis algorithm to process the influence coefficient can fit the trend of the residual resistance changing with the service time and environmental conditions and form a target influence trend data set. Further, combining the target influence trend data set and the preset life cycle time sequence to determine the bearing capacity change curve and deeply analyzing the bearing capacity change curve can determine the bearing performance state of each key node of the structure in the whole life cycle, and then obtain the final whole life cycle bearing performance prediction result, which provides a scientific basis for the whole life cycle management of the offshore wind power foundation structure, helps engineers to make a reasonable maintenance plan and determine the best maintenance time and scheme, thereby prolonging the service life of the structure, reducing the operation and maintenance cost, and improving the safety and economy of the structure. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the specific embodiments of the application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0036] Figure 1 is a flowchart of the offshore wind power foundation structure bearing performance degradation evaluation method according to an embodiment of the application;

[0037] Figure 2 is a flowchart of the bearing performance degradation evaluation method after corrosion of the wind power tower, guide pipe frame and foundation according to an embodiment of the application;

[0038] Figure 3 is a flowchart of step S13 according to an embodiment of the application;

[0039] Figure 4is a flowchart of step S16 according to the embodiment of the present application. DETAILED DESCRIPTION

[0040] To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0041] The core challenge of the current evaluation method is how to comprehensively depict the interaction of the three key technical factors of corrosion thinning, local damage and material mechanical property degradation and their comprehensive influence on the bearing capacity. Due to the irregular distribution of corrosion thinning, the geometric characteristics of the structure change complexly; the accumulation of local damage can cause stress concentration and failure mode transition; and the degradation of material mechanical property further weakens the resistance of the structure. These factors are not fully solved, which leads to the difficulty in accurately simulating the static and dynamic response characteristics of the structure after corrosion, and further leads to the difficulty in accurately predicting the residual bearing capacity.

[0042] Therefore, how to establish a refined finite element model, comprehensively consider the non-uniformity of corrosion thinning, the evolution law of local damage and the influence of material mechanical property degradation, and simulate the bearing capacity change law of offshore wind tower, jacket and foundation in extreme marine environment under different corrosion degrees, becomes a key problem to be solved urgently.

[0043] According to the embodiments of the present application, a bearing capacity degradation evaluation method for offshore wind power foundation structure is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described here can be executed in different order.

[0044] In the present embodiment, a bearing capacity degradation evaluation method for offshore wind power foundation structure is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers and the like, Figure 1 is a flowchart of the bearing capacity degradation evaluation method for offshore wind power foundation structure according to the embodiments of the present application, as shown in Figure 1 The flowchart includes the following steps:

[0045] Step S101, obtaining an initial three-dimensional geometric point cloud data set and a historical service data set of the offshore wind power foundation structure.

[0046] The offshore wind power foundation structure can include offshore wind tower, jacket and foundation.

[0047] Further, the initial three-dimensional geometric point cloud data set represents the initial three-dimensional geometric data of the acquired offshore wind tower, jacket, and foundation. In actual operation, due to the complex and dangerous offshore working environment, it is difficult and risky to directly perform three-dimensional point cloud scanning on the offshore jacket structure, and therefore the design drawings can be acquired. The design drawings record the precise dimensions, shapes, and assembly relationships of the offshore wind foundation structure in an ideal state in detail, covering the height, diameter, and wall thickness of the tower, the truss structure size and node connection form of the jacket, and the type (such as single pile foundation, suction bucket foundation, etc.) and specific geometric parameters of the foundation. Through professional three-dimensional modeling software, the two-dimensional information in the drawings is converted into an accurate three-dimensional model, and corresponding point cloud data is generated, which contains spatial coordinate information of each part of the structure and can truly restore the initial geometric shape of the structure. In addition, if conditions permit, the point cloud data generated from the design drawings can also be supplemented and corrected in combination with the three-dimensional scanning data in the land manufacturing stage, to ensure the accuracy and integrity of the initial three-dimensional geometric point cloud data set, and to provide a reliable basis for subsequent corrosion thinning feature analysis, finite element model construction, and bearing capacity evaluation based on data.

[0048] Further, the historical service data set represents various data related to corrosion of the acquired offshore wind foundation structure during service, which can include:

[0049] (1) Corrosion monitoring data: regularly detected corrosion depth, area, position distribution, corrosion rate (such as annual thinning amount), etc.;

[0050] (2) Environmental load data: extreme marine environmental parameters (such as typhoon wind speed, wave height, seawater salinity, temperature, humidity, etc.) experienced during service;

[0051] (3) Operation and maintenance records: structural maintenance measures (such as repair of corrosion-resistant coating, local replacement, etc.), damage history (such as crack, deformation record);

[0052] (4) Operation data: load conditions of wind power equipment (such as normal operation load, start-stop frequency, vibration frequency, etc.).

[0053] Further, by combining the initial three-dimensional geometric point cloud data set and the historical service data set, the distribution range and non-uniformity characteristics of corrosion thinning can be analyzed.

[0054] Step S102, based on the initial three-dimensional geometric point cloud data set and the historical service data set, the corrosion thinning characteristics of the offshore wind foundation structure are analyzed, and a first geometric distribution characteristic data set and a first finite element model containing non-uniformity are obtained.

[0055] The corrosion thinning feature represents the characteristic of the offshore wind power foundation structure (such as a tower, a jacket, and a foundation) in which the thickness of the structural material decreases due to corrosion when serving in the long-term marine environment, mainly including the distribution range of the corrosion thinning (such as the numerical range of the thickness thinning) and the non-uniformity feature (such as the difference in the local corrosion depth).

[0056] Further, the first geometric distribution feature data set represents a data set obtained by analyzing and processing the corrosion thinning feature, records the distribution of the geometric characteristics of the offshore wind power foundation structure after corrosion, and can include information such as the thickness change and shape change of each part of the structure, and can accurately describe the change in the geometric characteristics of the structure due to corrosion thinning.

[0057] Further, the first finite element model represents a refined grid model containing the non-uniformity feature of the corrosion thinning, and the influence of the non-uniformity of the corrosion thinning on the geometric characteristics of the structure is considered in the construction process, which can more realistically simulate the mechanical behavior of the structure after corrosion, and provides a reliable model basis for analyzing the influence of local damage evolution, stress concentration distribution, and material degradation on the bearing capacity of the structure.

[0058] Specifically, by combining the obtained initial three-dimensional geometric point cloud data set and the historical service data set, the corrosion thinning feature of the offshore wind power foundation structure can be analyzed in depth, and then through the analysis, the first geometric distribution feature data set which can accurately reflect the corrosion state of the structure and the first finite element model containing the non-uniformity feature can be obtained.

[0059] Further, the constructed first finite element model can capture the stress concentration caused by corrosion, solving the deviation problem of the uniform corrosion assumption in the prior art.

[0060] In step S103, based on the first geometric distribution feature data set, the second finite element model is used to simulate the evolution path and cumulative effect of the local damage of the offshore wind power foundation structure, and a target coupling distribution data set between the local damage and the stress concentration is obtained.

[0061] The second finite element model is obtained by introducing an initial assumption condition of local damage in the first finite element model. The initial assumption condition of local damage represents the initial state parameters of local damage artificially set in the first finite element model, including the damage starting position, the initial damage value, and the critical condition of damage evolution, etc.

[0062] Further, the evolution path of the local damage represents the development process of the local damage (such as micro-cracks and plastic deformation) of the offshore wind power foundation structure with the increase of time or load under the action of corrosion and external load, etc., which can include the starting, expansion direction, and degree change of the damage, etc.

[0063] Further, the cumulative effect represents the influence of the continuous accumulation of local damage on the overall mechanical properties of the offshore wind power foundation structure, including stress concentration aggravation, bearing capacity decline, and failure mode transformation, etc.

[0064] Further, the target coupling distribution dataset represents the corresponding relationship data between the degree of local damage and the position and intensity of stress concentration determined by simulating the evolution of local damage and the cumulative effect, which can reflect the spatial coupling characteristics of damage distribution and stress concentration area in the structure.

[0065] Specifically, the initial assumption condition of local damage is introduced in a finite element model to obtain a corresponding second finite element model.

[0066] Further, based on the first geometric distribution characteristic dataset, the occurrence position and intensity of stress concentration can be determined by simulating the evolution path and cumulative effect of local damage through iterative calculation using the second finite element model, and then the coupling distribution of local damage and stress concentration can be obtained and the corresponding target coupling distribution dataset can be acquired.

[0067] Further, by simulating the evolution path and cumulative effect of local damage of the offshore wind power foundation structure, the defect that the cumulative effect of damage is not considered in the prior art is solved, and then the weak area of the structure can be accurately located.

[0068] Step S104, based on the target coupling distribution dataset and the preset experimental degradation dataset of material mechanical properties, the second finite element model is used to evaluate the weakening degree of material degradation on the residual resistance and determine the target material performance characterization dataset after comprehensive degradation.

[0069] The preset experimental degradation dataset of material mechanical properties represents a collection of degradation data of material mechanical properties with time or environmental condition changes obtained through experiments, which is used to characterize the degradation law of material mechanical properties under the action of factors such as corrosion, high temperature, and fatigue, and can include the numerical values of mechanical properties parameters such as yield strength, tensile strength, elastic modulus, and elongation at different degradation stages, as well as the corresponding degradation time, environmental conditions (such as temperature, humidity, and salt fog concentration) and other information.

[0070] Further, the weakening degree of material degradation on the residual resistance represents the degree of decline in the ability of the structure to resist external loads caused by the degradation of material mechanical properties, which is usually represented by quantitative indicators (such as the percentage of resistance reduction and the increase of stress concentration coefficient), and can reflect the decay amplitude of the residual bearing capacity of the structure relative to the initial bearing capacity after the degradation of material performance, as well as the specific influence of material degradation on the resistance of key parts (such as stress concentration area) of the structure.

[0071] Further, the target material performance characterization dataset represents a quantitative description set of the material mechanical performance state after comprehensively considering the multi-factor coupling effects of corrosion thinning, local damage and material degradation, and can include various mechanical performance parameters of the material after corrosion, damage and degradation (such as the yield strength, elastic modulus, fatigue limit and the like after degradation), and the spatial distribution characteristics of these parameters in the structure (such as the material performance difference corresponding to different corrosion regions and damage degrees).

[0072] Specifically, on the basis of the target coupling distribution dataset and the preset experimental degradation dataset of the material mechanical performance, the time sequence parameters of material degradation can be integrated into the second finite element model through a numerical mapping method.

[0073] Further, the weakening degree of the remaining resistance caused by material degradation is determined, and the material performance characterization after comprehensive degradation is obtained, and the corresponding target material performance characterization dataset is obtained.

[0074] Further, through material degradation evaluation, the interaction of the three key factors of corrosion thinning, local damage and material mechanical performance degradation is comprehensively considered, and the defects of the prior art that do not sufficiently solve the multi-factor coupling effect are overcome.

[0075] Step S105, based on the target material performance characterization dataset, the performance degradation law of the offshore wind power foundation structure is analyzed by using a third finite element model, and a mathematical relationship model between the corrosion degree and the bearing capacity is established.

[0076] The third finite element model is obtained by introducing extreme environmental load conditions in the second finite element model. The extreme environmental load conditions represent the load combination of the offshore wind power foundation structure under rare or severe environmental action that may be encountered during service, and can include extreme wind load, extreme wave load, extreme temperature load, flow ice or ship impact load and the like.

[0077] Further, the performance degradation law of the offshore wind power foundation structure represents the evolution trend of the bearing performance of the offshore wind power foundation structure with the change of service time or corrosion degree under the multi-factor coupling effects of corrosion, damage and material degradation, and can include the bearing capacity attenuation trend, the response characteristic change law, the failure mode transition law, the probability distribution characteristics and the like.

[0078] Further, the mathematical relationship model represents a mathematical expression for quantitatively describing the mapping relationship between the corrosion degree and the structure bearing capacity through data fitting technology, and can be a polynomial, a linear or a nonlinear function. In this embodiment, the mathematical relationship model is shown in the following relationship formula (1):

[0079] Y=aX 2 +bX+c (1)

[0080] In the formula, Y represents the bearing capacity; X represents the corrosion degree; a, b, c represent fitting coefficients.

[0081] Specifically, extreme environmental load conditions are introduced in the second finite element model to obtain a corresponding third finite element model.

[0082] Further, on the basis of the target material performance characterization dataset, the third finite element model is used to analyze the performance degradation law of the offshore wind power foundation structure.

[0083] Further, on the basis of the analysis results, the relationship between the corrosion degree and the bearing capacity can be fitted by a fitting method and a corresponding mathematical relationship model is established.

[0084] Further, through the above process, the third finite element model is constructed by introducing extreme environmental load conditions, which can analyze the performance degradation law of the structure under extreme environments, and further predict the full life cycle bearing performance by establishing a mathematical relationship model, which changes the situation that the existing evaluation method is difficult to accurately capture the true response of the structure under extreme environments, and improves the reliability and applicability of the evaluation.

[0085] In step S106, the mathematical relationship model and the third finite element model are used to predict the bearing performance degradation of the offshore wind power foundation structure, and the full life cycle bearing performance prediction result of the offshore wind power foundation structure is obtained.

[0086] Specifically, on the basis of the mathematical relationship model and the third finite element model, by dynamically adjusting the model parameters and simulating future service scenarios, the full life cycle bearing capacity of the offshore wind power foundation structure can be predicted, and the final full life cycle bearing performance prediction result is obtained, which provides a scientific basis for the safety evaluation and maintenance decision of the offshore wind power foundation structure, helps to prolong the service life and reduce the operation and maintenance cost, and realizes the effective evaluation of the long-term service ability of the structure.

[0087] The load-bearing performance degradation assessment method for offshore wind power infrastructure provided in this embodiment, by acquiring initial three-dimensional geometric point cloud data and historical service data, can deeply analyze the corrosion thinning characteristics and construct a first finite element model that includes non-uniformity. This effectively solves the problem of ignoring corrosion non-uniformity in the existing technology, making the model more consistent with the actual structural state. Furthermore, with the help of a second finite element model, the evolution and cumulative effects of local damage are simulated to obtain the coupled distribution of local damage and stress concentration. At the same time, combined with material degradation assessment, it comprehensively considers the interaction between the three key factors of corrosion thinning, local damage, and degradation of material mechanical properties, overcoming the defect of the existing technology that does not fully address the influence of multiple factors. Furthermore, by introducing extreme environmental load conditions to construct a third finite element model, it is possible to analyze the performance degradation law of the structure under extreme environments. Then, by establishing a mathematical relationship model and predicting the load-bearing performance throughout the entire life cycle, it changes the situation where the existing assessment method is difficult to accurately capture the true response of the structure in extreme environments, and improves the reliability and applicability of the assessment. Furthermore, the final predicted results of the full life cycle load-bearing performance prediction can provide a scientific basis for the safety assessment and maintenance decision-making of offshore wind power infrastructure, help extend the service life, reduce operation and maintenance costs, and realize the effective assessment of the long-term service capability of the structure.

[0088] In some optional implementations, the above step S102 includes:

[0089] Step S1021 : Based on the initial three-dimensional geometric point cloud dataset and the historical service dataset, the geometric characteristics of the offshore wind power infrastructure are analyzed to determine a target digital model and a second geometric distribution characteristic dataset of the offshore wind power infrastructure.

[0090] Specifically, a high-precision digital model (the target digital model) is generated from the initial 3D geometric point cloud dataset. Simultaneously, the geometric characteristics of the offshore wind turbine foundation are analyzed in conjunction with the historical service dataset to determine the distribution range and non-uniformity of corrosion thinning. This results in a preliminary description of the changes in the structural geometric characteristics (the second geometric distribution characteristic dataset).

[0091] In some optional implementations, the above step S1021 includes:

[0092] Step a1: establishing a first digital model based on an initial three-dimensional geometric point cloud data set.

[0093] Specifically, the initial three-dimensional geometric point cloud data set can be used to establish a corresponding initial digital model.

[0094] Furthermore, the initial 3D geometric point cloud dataset is registered and filtered.

[0095] Further, on the basis of the processed initial three-dimensional geometric point cloud data set, the established initial digital model can be reconstructed by using a Poisson surface reconstruction algorithm to obtain a corresponding first digital model.

[0096] The Poisson surface reconstruction algorithm represents a calculation method for converting three-dimensional point cloud data into a continuous triangular mesh surface, and the core is to fit the implicit surface of the point cloud by solving the Poisson equation, thereby generating a high-precision three-dimensional digital model.

[0097] Step a2, based on the historical service data set, the first distribution range of corrosion thinning is determined by using a Monte Carlo simulation method to process the first digital model.

[0098] The Monte Carlo simulation method represents a numerical calculation method for solving complex problems through random sampling and statistical analysis, and the core idea is to simulate the behavior of the system by using a probability model, and to obtain an approximate solution through a large number of repeated experiments.

[0099] Specifically, according to the description in step S101, the historical service data set can include corrosion rate, service time, environmental parameters (salt spray concentration, temperature and humidity cycle), etc.

[0100] Further, on the basis of the historical service data set, the variables such as corrosion rate and environmental factors are randomly sampled, and the Monte Carlo simulation method is used for simulation, and then the distribution range of corrosion thinning can be predicted according to the simulation results.

[0101] For example, combined with the historical service data, the Monte Carlo simulation method is used to predict the corrosion thinning, assuming that the corrosion rate is 0.5 millimeters per year, and the simulation times are 10,000 times, and then the distribution range of corrosion thinning is 1 to 8 millimeters.

[0102] Step a3, according to the first distribution range, the non-uniformity characteristics of the offshore wind power foundation structure are analyzed and a non-uniformity distribution data set is obtained.

[0103] Specifically, according to the first distribution range of corrosion thinning, by calculating the standard deviation of the local corrosion depth, the maximum thinning amount and other parameters, the corrosion depth difference of each part of the offshore wind power foundation structure can be quantitatively analyzed and the corresponding non-uniformity distribution data set is obtained.

[0104] For example, when the distribution range of corrosion thinning is 1 to 8 millimeters, it can be further determined by analysis that the non-uniformity characteristics are that the local corrosion depth difference can reach 3 millimeters.

[0105] Further, the non-uniformity distribution data set is stored in the form of thickness deviation matrix of each region in the point cloud model, which can reflect the irregular distribution characteristics of corrosion.

[0106] Step a4, based on the non-uniform distribution dataset, the geometric characteristics of the offshore wind power foundation structure are analyzed by using the finite element analysis method, and the first digital model is updated to obtain a second digital model.

[0107] Specifically, the non-uniform distribution dataset can be imported into the finite element analysis software, and the structural geometric characteristics (such as wall thickness, cross-sectional shape) are analyzed.

[0108] Further, the grid parameters, node coordinates, etc. of the first digital model can be adjusted according to the influence of corrosion thinning on geometric size, and the corresponding second digital model is generated.

[0109] For example, when the tower cylinder wall thickness is thinned from 20 mm to 12 mm and the corrosion depth at the jacket node is set to 5 mm, the stress distribution changes, the maximum stress increases from 150 MPa to 165 MPa, and the safety factor decreases from 5 to 3, and then the first digital model is updated to obtain the second digital model.

[0110] Step a5, using the second digital model to determine the geometric change dataset of the offshore wind power foundation structure.

[0111] Specifically, the updated second digital model is used to analyze the change trend of the structure geometry, and the structure geometric change data is obtained.

[0112] For example, the thickness change amount, cross-sectional moment of inertia change value, overall deformation amount, etc. of each part of the offshore wind power foundation structure can be extracted from the second digital model, and combined with stress, displacement and other mechanical response data to generate a geometric change dataset. For example, after the foundation leg is thinned by 7 mm, the overall stiffness of the structure decreases by 20%, and the corresponding displacement increment is 15 mm.

[0113] Step a6, when the geometric change dataset does not meet the first preset requirement, a machine learning algorithm is used to predict the future distribution range of corrosion thinning to obtain a second distribution range.

[0114] Specifically, when the geometric change dataset exceeds the preset threshold (such as stress increase of more than 20%, safety factor less than 3.5), a machine learning algorithm such as support vector regression (SVR) can be used, inputting multiple sets (1000 sets) of historical service data as a training set to establish a corrosion thinning prediction model.

[0115] Further, based on the current corrosion trend and environmental parameters, the distribution range of future corrosion thinning, i.e. the second distribution range, can be predicted, and the prediction error can be controlled within 5%.

[0116] Step a7, adjusting the geometric characteristics of the second digital model according to the second distribution range to obtain a target digital model and a second geometric distribution characteristic dataset.

[0117] Specifically, the geometric parameters of the second digital model (such as the thickness of each region and the location of the corrosion pit) can be dynamically corrected according to the second distribution range.

[0118] Furthermore, the final target digital model is generated through the adjusted geometric parameters, ensuring that the target digital model can accurately reflect the current and future corrosion status.

[0119] Furthermore, by integrating the geometric parameters, corrosion distribution and mechanical response data in the target digital model, a complete second geometric distribution characteristic data set can be formed.

[0120] Through the above process, Monte Carlo simulation methods were used to process historical service data and determine the distribution range of corrosion thinning. Simultaneously, machine learning algorithms were combined to predict future distribution ranges. This effectively addressed the inaccurate corrosion distribution predictions in existing technologies and enabled the model to dynamically adapt to the development of corrosion. Furthermore, the digital model was updated and adjusted based on the corrosion distribution range to obtain the target digital model and geometric distribution characteristic dataset, ensuring that the model accurately reflects changes in the structural geometric characteristics. Furthermore, by determining whether the geometric change dataset meets preset requirements and dynamically adjusting the model, the model's adaptability to varying corrosion levels and structural changes was enhanced, making the model more practical.

[0121] Step S1022 : Based on the non-uniformity parameter set, a fourth finite element model is constructed by processing using a finite element analysis method, and an initial mapping relationship of stress distribution is determined according to the fourth finite element model.

[0122] Specifically, based on the second geometric distribution characteristic data set, a surface roughness analysis method based on point cloud data can be used, such as fitting the surface profile using the least squares method to calculate the standard deviation σ of the local thickness variation and the maximum thinning amount.

[0123] Furthermore, the point cloud data of the corroded area is sampled in the target digital model, and the thickness value of each sampling point is calculated. For example, the standard deviation σ of the local thickness variation is measured to be 1.5 mm, and the maximum thinning is 8 mm.

[0124] Furthermore, a set of non-uniformity parameters of corrosion thinning can be obtained through statistical analysis.

[0125] Furthermore, the accuracy of parameter extraction can be ensured by comparing it with on-site corrosion detection data (such as ultrasonic thickness gauge measurements) and the error can be controlled within 0.5 mm.

[0126] Step S1023 : Based on the non-uniformity parameter set, a fourth finite element model is constructed by processing using a finite element analysis method, and an initial mapping relationship of stress distribution is determined according to the fourth finite element model.

[0127] Among them, the initial stress distribution mapping is used to characterize the mapping relationship from corrosion geometric characteristics to mechanical properties.

[0128] Specifically, based on the non-uniform parameter set, the fourth finite element model was constructed using an adaptive meshing technique (eg, setting the mesh density to 1 mm in the corrosion area and 5 mm in the non-corrosion area).

[0129] Furthermore, typical operating loads (such as an internal pressure of 10 MPa and a design wind speed load) can be applied to the fourth finite element model, and an elastic-plastic material model (such as a bilinear isotropic hardening model) can be used for analysis.

[0130] Furthermore, a finite element solver was used to calculate the stress distribution cloud of the model, determining the stress difference between the corroded and non-corroded areas. For example, the maximum stress in the corroded area reached 350 MPa, exceeding the material's yield strength of 300 MPa.

[0131] Furthermore, by analyzing the correspondence between corrosion geometric parameters (such as thickness and pit shape) and stress distribution, an initial mapping relationship from corrosion geometric characteristics to mechanical properties can be established. For example, for every 1 mm reduction in wall thickness, the local stress increases by 10 MPa.

[0132] Step S1024 : Based on the initial stress distribution mapping relationship, the fourth finite element model is used to analyze the influence of the non-uniformity on the geometric characteristics and determine a first finite element model containing the non-uniformity and a first geometric distribution characteristic data set.

[0133] Specifically, based on the initial mapping relationship of stress distribution, the fourth finite element model was used to analyze the effect of non-uniform corrosion on geometric properties. For example, local corrosion pits caused the section inertia moment to decrease by 15%, the maximum displacement of the structure to increase from 10 mm to 15 mm, and the fatigue life to increase from 10 mm to 15 mm. 6 times down to 10 4 Second, the evolution trend of geometric characteristics and mechanical responses can be determined through analysis.

[0134] Furthermore, the second geometric distribution characteristic data set may be updated according to the evolution trend.

[0135] Furthermore, if the updated second geometric distribution characteristic dataset meets preset requirements (e.g., the stress prediction error is less than a threshold), the fourth finite element model can be directly used as the final first finite element model including the inhomogeneity. Simultaneously, the updated second geometric distribution characteristic dataset is directly used as the final first geometric distribution characteristic dataset.

[0136] Further, if the updated second geometric distribution characteristic data set does not meet the preset requirement (such as the stress prediction error being greater than a threshold value), a Kriging interpolation algorithm can be used to adjust the thickness distribution standard deviation, and the fourth finite element model grid and material parameters are further adjusted according to the adjustment result, and then the corresponding first finite element model and first geometric distribution characteristic data set are generated.

[0137] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the embodiment extracts the non-uniformity parameters of corrosion thinning by using the surface roughness analysis method based on point cloud data, which can more accurately describe the irregular distribution of corrosion compared with the estimation based on simple uniform corrosion thinning in the prior art. Further, the non-uniformity parameters are processed by the finite element analysis method, the fourth finite element model is constructed, and the initial mapping relationship of the stress distribution is determined, which realizes accurate mapping of the corrosion geometric characteristics to the mechanical properties and avoids the evaluation deviation caused by ignoring the influence of the change of geometric characteristics on the mechanical properties in the prior art. Further, the first finite element model is determined by analyzing the influence degree of non-uniformity, so that the model can more truly reflect the geometric and mechanical characteristics of the structure after corrosion, and the accuracy of the first finite element model is improved.

[0138] In some optional embodiments, the step S103 comprises:

[0139] Step S1031, based on the first geometric distribution characteristic data set, introducing an initial assumption condition of local damage into the first finite element model and generating a fifth finite element model and a third geometric distribution characteristic data set.

[0140] Specifically, based on the first geometric distribution characteristic data set, the initial assumption condition of local damage described in the step S103 is introduced into the first finite element model.

[0141] Further, the initial assumption condition is embedded into the nodes or elements of the first finite element model, the material properties or element state of the model are modified through the secondary development interface (such as APDL, Python script) of the finite element software (such as ANSYS, ABAQUS), and then the fifth finite element model containing the initial condition of local damage can be generated.

[0142] Further, by integrating the geometric characteristics (such as the grid coordinates of the damage area and the element size) in the fifth finite element model, the initial damage distribution parameters (such as the position of the damage element and the initial damage value), and the associated corrosion geometric parameters (such as the wall thickness after thinning), the corresponding third geometric distribution characteristic data set can be formed.

[0143] Step S1032, based on the third geometric distribution characteristic data set, iteratively solving the fifth finite element model to obtain the evolution path of local damage.

[0144] Specifically, a nonlinear finite element solving algorithm (e.g., Newton-Raphson algorithm) can be employed to iteratively calculate the fifth finite element model. The iteration step and convergence criterion can be determined according to actual requirements.

[0145] Further, in the iterative calculation, as the load is gradually applied (or time is advanced), the damage model can be used to determine whether the damage evolves according to the energy release rate:

[0146] (1) When the energy release rate of a certain element exceeds the critical value (1000 J / m 2 ), the damage value increases according to a predetermined rule (e.g., linear or exponential growth), wherein the critical value is the critical energy release rate set according to the damage model based on the energy method;

[0147] (2) The damage value gradually increases from the initial 0 until it reaches the critical value (e.g., 0.8-1.0), at which point the element stiffness degrades and the stress is redistributed.

[0148] Further, the damage value change of the key nodes or elements is recorded in each iteration, and the corresponding local damage evolution path is formed.

[0149] Step S1033, analyze the cumulative effect of local damage according to the evolution path and determine the initial coupling distribution data set of local damage and stress concentration.

[0150] Specifically, based on the damage evolution path, the influence of damage accumulation on the mechanical properties of the structure can be quantified:

[0151] (1) Stress concentration intensifies: Damage accumulation leads to material stiffness degradation, and stress concentrates in the damaged area. For example, when the damage value increases from 0.5 to 0.8, the stress in this area increases from 300 MPa to 500 MPa.

[0152] (2) Failure mode transition: From global elastic deformation to local plastic failure, such as when the damage accumulates to 0.9, the element appears a plastic hinge, and the structure failure mode changes from strength failure to fatigue failure.

[0153] Further, by correlating the damage evolution path with the stress distribution cloud map, the stress concentration location and intensity corresponding to each damage state can be determined:

[0154] For example, when the damage value reaches 0.8, the stress concentration location (e.g., model edge node A) can be determined by the Von Mises stress criterion, and the stress value is 500 MPa; when the damage value increases to 0.9, the stress concentration area expands to form a band-shaped distribution, and the stress value increases to 700 MPa.

[0155] Further, by integrating the damage value, stress concentration location, stress intensity and other parameters, the corresponding initial coupling distribution data set can be formed.

[0156] Step S1034, determining the position change trend of the first key region of stress concentration according to the initial coupling distribution data set.

[0157] Specifically, the region where the stress value exceeds the yield strength of the material can be selected from the initial coupling distribution data set as the first key region.

[0158] Further, by comparing the positions of the stress concentration regions at different damage stages, the evolution direction thereof can be determined.

[0159] (1) Spatial expansion: damage accumulation causes the stress concentration region to expand from a single point (node A) to a strip-shaped region (the line connecting node A to node B).

[0160] (2) Depth migration: in a three-dimensional structure, the stress concentration position migrates from the surface to the region where the material degradation is severe.

[0161] For example, when the damage value increases from 0.8 to 0.9, the stress concentration region expands 10 mm circumferentially along the tower drum and migrates 5 mm radially inward to the inner wall.

[0162] Further, the position change trend can be described by using parameters such as displacement vector, expansion angle, migration distance, etc.

[0163] Step S1035, updating the boundary conditions of the fifth finite element model according to the position change trend and determining a second finite element model.

[0164] The initial boundary conditions of the third finite element model are set according to the actual service state of the structure and are used to simulate the real stress environment, which can include:

[0165] (1) Foundation constraint: such as pile foundation fixation (completely constrained translational and rotational degrees of freedom), fixed or hinged connection of the jacket with the seabed;

[0166] (1) Connection constraint: flange connection of tower drum segments (simulating bolt constraint), rigid or flexible connection of jacket member nodes;

[0167] (3) Load boundary: environmental load under initial working conditions (such as hydrostatic pressure, dead weight, etc.).

[0168] Specifically, the boundary conditions of the fifth finite element model can be modified according to the position change trend of the stress concentration key region.

[0169] Further, by applying the updated boundary conditions, the mesh is re-divided or the element properties are corrected, and the second finite element model is generated.

[0170] Furthermore, the effectiveness of the second finite element model can be verified by comparing the stress distribution cloud maps of the new and old models, thereby ensuring that the model can more accurately reflect the evolution of stress concentration after the boundary conditions are updated.

[0171] Step S1036: determining a target coupling distribution data set according to the second finite element model.

[0172] Specifically, by performing iterative calculations on the second finite element model, the target coupling distribution data set can be further determined. The specific process can be referred to the description of steps S1032 to S1033 above, which will not be repeated here.

[0173] The load-bearing performance degradation assessment method for offshore wind power infrastructure provided in this embodiment introduces the initial assumption of local damage in the first finite element model, and obtains the local damage evolution path through iterative solution, which can accurately simulate the development process of local damage and overcome the defect of the existing technology that it is difficult to simulate the evolution of damage. Furthermore, the cumulative effect of local damage is analyzed according to the evolution path, and the coupling distribution of local damage and stress concentration is determined, which effectively captures the stress concentration and failure mode transition caused by the accumulation of local damage, and solves the problem of insufficient consideration of the cumulative effect in the existing technology. Furthermore, the model boundary conditions are updated according to the position change trend of the key area of ​​stress concentration, so that the second finite element model can more accurately reflect the mechanical properties of the structure during the damage evolution process, thereby improving the accuracy and reliability of the model.

[0174] In some optional implementations, the above step S104 includes:

[0175] Step S1041 : determining an initial mechanical property parameter set based on the target coupling distribution data set and the preset experimental degradation data set.

[0176] Specifically, as described in steps S1031 through S1035 above, the target coupled distribution dataset contains the spatial correspondence between local damage and stress concentration. For example, a jacket node experiences a stress concentration of 500 MPa when the damage value is 0.8, and this rises to 700 MPa when the damage accumulates to 0.9.

[0177] Furthermore, combined with the stress concentration area in the target coupling distribution data, the corresponding material experimental degradation parameters are matched, and a corresponding set of initial mechanical property parameters is formed, which may include initial mechanical property parameters of the material such as yield strength, tensile strength, elastic modulus, elongation, performance parameters at different time points or degradation stages, and the mapping relationship between stress concentration areas and material degradation parameters.

[0178] Step S1042 : Processing the time series in the preset experimental degradation data using a numerical mapping method to obtain a degradation parameter sequence.

[0179] Specifically, discrete points of material performance changing with time in the pre-set experimental degradation data are extracted.

[0180] Further, the discrete time series can be converted into a continuous function by using methods such as cubic spline interpolation, polynomial fitting, or exponential smoothing.

[0181] Further, by mapping the continuous function according to the nodes or elements of the second finite element model, a degradation parameter sequence corresponding to the model nodes can be generated.

[0182] Step S1043, inputting the degradation parameter sequence into the second finite element model to obtain the initial material performance characterization dataset after simulation degradation.

[0183] Specifically, by using the parametric modeling function of the finite element software (such as the APDL language of ANSYS), the degradation parameter sequence is input as the material attribute and assigned to the corresponding nodes or elements of the second finite element model.

[0184] Further, by applying boundary conditions and loads (such as internal pressure 10 MPa, temperature field 200℃) to the second finite element model, the stress distribution, deformation, and other responses of the material in the degradation state can be simulated and calculated.

[0185] Further, the material performance parameters (such as strength, modulus, and Poisson's ratio) of each node after degradation can be extracted from the simulation calculation results and form the corresponding initial material performance characterization dataset.

[0186] Step S1044, when the initial material performance characterization dataset does not meet the third pre-set requirement, calculating the remaining resistance weakening degree and determining the resistance change trend according to the material state dataset.

[0187] Specifically, by comparing the simulation results with the experimental data or design specifications, it can be checked whether the third pre-set requirement is met. For example, whether the stress prediction error is less than the corresponding threshold value, whether the performance parameters of the material after degradation deviate from the experimental values by less than the corresponding threshold value, and other third pre-set requirements.

[0188] Further, if the error exceeds the threshold value (such as stress error > 10%), the remaining resistance weakening degree caused by material degradation can be calculated by finite element analysis. The remaining resistance weakening degree caused by material degradation can include the overall resistance reduction ratio, the resistance weakening of key parts, and the like.

[0189] Further, according to the calculated remaining resistance weakening degree, the resistance change law at different degradation stages can be analyzed and the corresponding resistance change trend can be determined.

[0190] For example, the resistance change trend can be:

[0191] (1) Linear trend: the overall resistance decreases by 5% for every 100 hours of degradation time.

[0192] (2) Non-linear trend: the resistance decreases slowly at the beginning of degradation (10% reduction in 0-300 hours) and accelerates later (15% reduction in 300-600 hours).

[0193] In step S1045, the target material performance characterization dataset after comprehensive degradation is determined according to the resistance change trend and the initial mechanical performance parameter set.

[0194] Specifically, by combining the resistance change trend with the initial mechanical performance parameter set, the degradation parameter sequence can be corrected.

[0195] Further, according to the resistance change trend, the performance parameters of each node after degradation are recalculated.

[0196] Further, by integrating the corrected material performance parameters, stress concentration influence coefficients, damage coupling effects, etc., the target material performance characterization dataset after comprehensive degradation can be formed, which can include: complete performance parameter sequence of each node at different degradation stages; performance characterization considering corrosion, damage, and material degradation multi-factor coupling; and mapping relationship corresponding to each node of the second finite element model.

[0197] The bearing performance degradation evaluation method of the offshore wind power foundation structure provided in this embodiment uses a numerical mapping method to integrate material degradation time sequence parameters into a finite element model, which can quantitatively evaluate the weakening degree of residual resistance caused by material degradation, and changes the inaccurate evaluation of the influence of material mechanical performance degradation in the prior art. Further, when the simulated material performance after degradation does not meet the requirements, the weakening degree of residual resistance is calculated in time and adjusted, ensuring that the target material performance characterization dataset accurately reflects the performance state of the material after comprehensive degradation, and providing support for subsequent quantitative evaluation of the comprehensive influence of corrosion thinning, local damage, and material degradation.

[0198] In some optional embodiments, the above step S105 includes:

[0199] In step S1051, based on the target material performance characterization dataset, a third finite element model is used to analyze the structural characteristics of the offshore wind power foundation structure and determine the structural characteristic curve of the offshore wind power foundation structure.

[0200] Specifically, by introducing extreme environmental load conditions into the second finite element model to obtain the third finite element model through the target material performance characterization dataset after comprehensive degradation, and then using dynamic simulation technology to calculate the change law of static response and dynamic response, the evolution trend of stress concentration and failure mode under load is determined, and the response characteristic curve of the offshore wind power foundation structure is obtained.

[0201] In some optional embodiments, the step S1051 comprises:

[0202] Step b1, based on the target material performance characterization dataset, constructing an initial calculation framework in the third finite element model and generating an initial simulation basis dataset.

[0203] Specifically, the comprehensive degraded target material performance characterization dataset is imported into the third finite element model. At the same time, the initial calculation framework is constructed in the third finite element model.

[0204] Further, the third finite element model is loaded with extreme environmental load conditions, and combined with the input target material performance characterization dataset and the constructed initial calculation framework, an initial simulation basis dataset corresponding to the model grid information, material attribute matrix, load vector, etc. can be output.

[0205] Step b2, processing the initial simulation basis dataset by using dynamic simulation technology and calculating the change law of static response and dynamic response.

[0206] Specifically, by using dynamic simulation technology to process the simulation basis data, the change law of static response and dynamic response can be further calculated.

[0207] In some optional embodiments, based on the initial simulation basis dataset, an explicit dynamics solver (such as ANSYS Explicit) is used for transient analysis. Wherein, the time step can be set to 0.1 seconds to ensure that the structure response under dynamic load can be captured.

[0208] Further, the static and dynamic responses are calculated:

[0209] (1) Static response: calculate the displacement and stress distribution of the structure under constant load, for example, the maximum displacement of the tower top under 10 MPa internal pressure is 15 mm;

[0210] (2) Dynamic response: calculate the acceleration and velocity time history curve under impact load, for example, the acceleration peak of the foundation under 500g impact is 800g.

[0211] Further, the displacement, stress, strain and other parameters of each node at different time steps are obtained in real time, and the time-response curve is generated to reflect the change law of static response and dynamic response.

[0212] Step b3, determining the dynamic response distribution dataset according to the change law of static response and dynamic response.

[0213] Specifically, the time-history response data of each node can be integrated according to the spatial position to form a three-dimensional dynamic response distribution. For example, the displacement distribution cloud chart in the tower cylinder height direction, the maximum displacement appears at the top 15 mm; or the stress distribution at the foundation connection, the maximum stress is 350 MPa (close to the tensile strength of 380 MPa).

[0214] Further, the response distribution can be visualized by means of contour maps, heat maps, etc., and the areas exceeding the material yield strength (such as elements with stress > 300 MPa) are screened out, and then the corresponding key response data set, i.e. the dynamic response distribution data set, is formed.

[0215] Step b4, extracting the second key area from the dynamic response distribution data set and determining the stress concentration distribution result.

[0216] Specifically, the corresponding second key area can be extracted from the dynamic response distribution data set.

[0217] For example, based on the dynamic response distribution data set, the area with stress concentration coefficient K>2 is identified, such as the structure corner, the severely corroded thinning site, and then the second key area is determined.

[0218] Further, the maximum stress value, stress concentration coefficient and its change over time of the second key area can be calculated, and the corresponding stress concentration distribution result is formed.

[0219] Step b5, analyzing the type and characteristics of the failure model of the offshore wind power foundation structure according to the stress concentration distribution result and determining the failure mode evolution data set.

[0220] Specifically, for the stress concentration distribution result, by analyzing the type and characteristics of the failure mode and judging the change direction of the evolution trend, the evolution data of the failure mode can be obtained.

[0221] For example, the failure mode can be determined according to the stress concentration distribution result, for example:

[0222] (1) Static failure: stress exceeding tensile strength leading to fracture;

[0223] (2) Fatigue failure: stress amplitude exceeding fatigue limit causing crack propagation.

[0224] Further, the Miner linear cumulative damage rule can be used to calculate the fatigue damage accumulation under cyclic loading. For example, when the stress amplitude is 250 MPa, the damage factor reaches 1.0 (failure) after 10^4 cycles.

[0225] Further, by combining the damage factor, crack propagation length and other parameters under different load cycle numbers, the corresponding failure mode evolution data set is generated.

[0226] Step b6, determining the structural property curve according to the failure mode evolution dataset.

[0227] Specifically, the response property curve is fitted by the failure mode evolution dataset, and then the discrete points in the curve are processed by interpolation method to obtain the smooth property curve data. Further, the inflection points and extreme values in the property curve data are obtained to determine the overall response property of the structure, and the final structural curve distribution, i.e., the structural property curve, is obtained.

[0228] For example, the key parameters in the failure mode evolution dataset can include the number of load cycles, the damage factor, the crack length, the stress amplitude, the failure mode characteristics at different degradation stages (such as the critical conditions of static failure and fatigue failure), etc.

[0229] Further, the least square method, polynomial regression or nonlinear regression method can be used to construct the fitting model.

[0230] Further, the failure mode evolution dataset is input into the constructed fitting model, and the fitting error is minimized by iterative optimization of parameters, and then the corresponding response property curve is formed.

[0231] Further, the distribution of the curve data points after fitting is analyzed, and the interval of the discrete points is determined.

[0232] Further, the cubic spline interpolation, line interpolation and other methods can be used to generate new intermediate point data between the discrete points to form a continuous property curve data sequence, i.e., to obtain the smooth property curve data.

[0233] Further, the inflection points and extreme values are extracted from the property curve data sequence, and the inflection points, extreme points and smoothed curve data are integrated to generate a complete structural property curve distribution, i.e., the final structural property curve.

[0234] Step S1052, quantitatively evaluating the residual bearing capacity of the offshore wind power foundation structure according to the structural property curve and the third finite element model, to obtain a quantitative evaluation result.

[0235] Specifically, according to the structural property curve, the stress value and displacement value of the key nodes are taken from the third finite element model, and are compared and analyzed with the initial bearing capacity.

[0236] Further, according to the comparison and analysis result, the comprehensive influence of corrosion thinning, local damage and material degradation on the bearing capacity is judged, and the quantitative evaluation result of the residual bearing capacity can be obtained.

[0237] In some optional embodiments, the above step S1052 comprises:

[0238] Step c1, according to the structural characteristic curve, the stress value and displacement value of the key node in the third finite element model are extracted and an initial data set is generated.

[0239] Specifically, the node with the maximum stress concentration or displacement can be selected from the corresponding region of the structural characteristic curve and the corresponding stress value and displacement value are extracted. For example, the tower drum mid-span node, foundation connection node are selected and the stress value (such as 120 MPa) and displacement value (15 mm) are extracted.

[0240] Further, the response data of the key nodes are integrated and the corresponding initial data set is formed.

[0241] Step c2, according to the initial data set and the initial bearing performance, the change trend of the bearing performance caused by corrosion thinning is determined.

[0242] Wherein, the initial bearing performance can be obtained by the following ways:

[0243] (1) Design data: theoretical calculation results (such as ultimate bearing capacity, safety factor) in structural design;

[0244] (2) Prototype experiment: static / dynamic load experiment on new structure or standard specimen to measure stress-displacement curve;

[0245] (3) Simulation model: based on ideal geometric model and material parameters without corrosion and damage, the initial bearing capacity is obtained by finite element analysis.

[0246] Specifically, by comparing the initial data set with the bearing performance of the structure without corrosion, the performance degradation such as stress increment, displacement increment caused by corrosion is calculated.

[0247] Further, according to the performance degradation, the change trend of the bearing performance caused by corrosion thinning can be further determined. For example, corrosion thinning leads to stress concentration aggravation (increment 20%), displacement increase (increment 50%), which indicates that the bearing performance presents a degradation trend.

[0248] Step c3, according to the change trend, the distribution characteristics of local damage are determined.

[0249] Specifically, the local damage area can be located according to the change trend.

[0250] Further, the spatial distribution of damage value in the structure can be counted, that is, the distribution characteristics of local damage are determined.

[0251] For example, according to the node with the maximum stress increment (such as node A with stress 120 MPa), the local damage evolution data in S103 are matched to determine that the damage value corresponding to the node is 0.8.

[0252] Furthermore, the area with damage value > 0.5 is concentrated in the area with corrosion depth > 5 mm, forming a band-like distribution, that is, the distribution characteristic of local damage is a band-like distribution.

[0253] Step c4: determining the impact degree of material degradation based on the distribution characteristics.

[0254] Specifically, based on the distribution characteristics, the material degradation parameters of key nodes can be queried through the target material performance characterization dataset. For example, the elastic modulus of node A drops from 70 GPa to 60 GPa, a decrease of 14.3%.

[0255] Furthermore, the reduction in resistance caused by material degradation, i.e., the degree of impact of material degradation, can be calculated based on the material degradation parameters. For example, a decrease in elastic modulus causes a 10% decrease in the overall stiffness of the structure, corresponding to a 10% decrease in bearing capacity.

[0256] In step c5, based on the change trend, distribution characteristics and impact degree, the weight distribution data set is obtained through support vector machine algorithm processing.

[0257] Specifically, the change trend (stress increase of 20%), distribution characteristics (damage value of 0.8) and impact degree (modulus decrease of 14.3%) were input into the support vector machine as feature vectors.

[0258] Furthermore, historical data can be used to train the model and optimize the penalty parameter C and kernel function parameter γ to make the weight distribution consistent with the actual coupling effect.

[0259] Furthermore, the influence weight of each factor on the load-bearing performance is output and a corresponding weight distribution data set is formed.

[0260] Step c6: determining a quantitative assessment result of the residual bearing capacity of the offshore wind power infrastructure according to the weight distribution data set.

[0261] Specifically, the residual bearing capacity can be calculated based on the weight distribution data set: residual bearing capacity = initial bearing capacity × (1-corrosion weight × stress increase - damage weight × damage effect - material weight × degradation effect).

[0262] Furthermore, the remaining bearing capacity is expressed as a percentage or a specific value, which is a quantitative assessment result.

[0263] Step S1053 : Analyze the performance degradation law of the offshore wind power infrastructure according to the quantitative evaluation result to obtain a target performance degradation data set.

[0264] Specifically, based on the quantitative evaluation results, the Monte Carlo simulation method is used to introduce random variables, simulate the changing trend of load-bearing capacity under different corrosion degrees and extreme environmental combinations, determine the probability distribution range of load-bearing performance, and then obtain the performance degradation law suitable for long-term service prediction.

[0265] In some optional embodiments, the step S1053 comprises:

[0266] Step d1, based on the quantitative evaluation results, introducing random variables and generating a bearing capacity dataset under different corrosion levels and extreme environment combinations using the Monte Carlo simulation method.

[0267] Step d2, performing statistical analysis on the bearing capacity dataset and determining the bearing capacity trend.

[0268] Step d3, based on the bearing capacity trend, determining the probability distribution interval of the bearing performance using the probability distribution model.

[0269] Step d4, calculating the performance degradation parameters and determining the target law dataset of performance degradation according to the probability distribution interval.

[0270] Step d5, determining the performance degradation law dataset according to the target law dataset.

[0271] Step d6, determining the target performance degradation dataset according to the performance degradation law dataset.

[0272] Specifically, random variables such as corrosion level (1-5 mm / year uniform distribution), extreme environment parameters (temperature -20℃ to 50℃, humidity 30%-90%) can be set and different combinations can be simulated.

[0273] Further, for different combinations, Monte Carlo simulation can be used to generate multiple groups of samples, for example, the corresponding bearing capacity decline rate can be calculated according to the generated multiple groups of samples. For example, a sample with a corrosion level of 3 mm / year, a temperature of 30℃, and a humidity of 60% has a corresponding bearing capacity that has dropped to 85% of the initial value.

[0274] Further, by integrating the corrosion level, environmental parameters, bearing capacity values, and other data of each sample, a corresponding bearing capacity dataset is formed.

[0275] Further, by calculating the mean, standard deviation, and quantile of the bearing capacity dataset, and using linear regression, polynomial regression, and other methods, the change trend of bearing capacity with corrosion level can be fitted.

[0276] Further, normal distribution, Weibull distribution, and other methods can be used to fit the probability density of bearing capacity. For example, by Kolmogorov-Smirnov test, it is determined that the Weibull distribution is optimal, with shape parameter k=5 and scale parameter λ=15 years.

[0277] Further, the bearing capacity range under the 95% confidence interval can be selected as the final probability distribution interval.

[0278] Further, the performance degradation parameter is calculated by the probability distribution interval, and a preliminary law of performance degradation, i.e., a target law data set, is obtained according to the calculation result.

[0279] Further, if the target law data set exceeds a preset threshold, the simulation parameter is further adjusted in combination with long-term service data to obtain a corrected degradation law, i.e., a performance degradation law data set.

[0280] Further, according to the corrected performance degradation law data set, a prediction law suitable for long-term service, i.e., a target performance degradation data set, can be further determined.

[0281] In step S1054, a mathematical relationship model between the corrosion degree and the bearing capacity is established by using a polynomial regression method based on the target performance degradation data set.

[0282] Specifically, the law data can be processed by using a data fitting technology, and a mathematical model of the corrosion degree and the bearing capacity is established by using a polynomial regression, such as the above relationship (1).

[0283] The offshore wind power foundation structure bearing performance degradation evaluation method provided by the embodiment utilizes dynamic simulation technology to process initial simulation foundation data sets and calculate the change law of static and dynamic response, can truly simulate the dynamic response process of the structure under the action of extreme environmental load, and overcomes the defects of insufficient dynamic response simulation in the prior art. Further, by analyzing the stress concentration distribution results and failure mode evolution data sets, the failure mode type and characteristics of the structure can be accurately determined, providing an important basis for safety evaluation and optimal design of the structure. Further, according to the failure mode evolution data set, the structure characteristic curve is determined, which comprehensively reflects the response characteristics of the structure at different loads and degradation stages. Further, by accurately positioning and extracting the stress and displacement data of the key nodes in the third finite element model from the structure characteristic curve, the initial data set is generated, solving the problem of non-targeted data extraction in the prior art, and ensuring that the analysis is based on the most critical stress position of the structure. Further, by comparing the initial data set with the initial bearing performance of the structure, the stress increase, displacement change and other parameters caused by corrosion thinning can be quantified, and the specific influence direction and degree of corrosion thinning on the bearing performance are further clarified. Further, based on the bearing performance change trend caused by corrosion thinning, the distribution law of local damage in the structure is analyzed in depth, realizing the deduction from macro performance change to micro damage distribution, and solving the problem that it is difficult to associate corrosion and damage distribution in the prior art. Further, combined with the local damage distribution characteristics, the weakening degree of the material mechanical performance degradation on the residual resistance of the structure is evaluated, and the influence difference of material degradation at different damage positions is clarified, changing the status quo of general evaluation of the influence of material degradation in the prior art. Further, the support vector machine algorithm is used to comprehensively analyze the change trend of corrosion thinning, the local damage distribution characteristics and the influence degree of material degradation, and determine the weight distribution of each factor on the bearing performance, avoiding the subjectivity of human experience value assignment, making the weight distribution more consistent with the law of actual multi-factor coupling, and improving the scientificity and accuracy of the evaluation. Further, according to the weight distribution data set, the influence of corrosion thinning, local damage and material degradation on the bearing performance is weighted and calculated, and the quantitative evaluation result of the residual bearing capacity can be obtained, providing an intuitive and quantitative basis for safety evaluation and maintenance decision of the structure, solving the problem of lack of accurate quantification of the evaluation result in the prior art. Further, the Monte Carlo simulation method is used to introduce random variables and simulate the bearing capacity change under different corrosion degrees and extreme environment combinations, fully considering the randomness and uncertainty factors in actual engineering, making the evaluation result more consistent with the actual situation. Further, the probability distribution interval of the bearing performance is determined through statistical analysis and probability distribution model, providing a probability level evaluation for long-term service prediction of the structure, changing the limitations of the deterministic evaluation in the prior art. Further, the target performance degradation data set is established based on the performance degradation parameters and data fitting method, realizing the dynamic description of the structure performance degradation law, and more accurately predicting the bearing capacity change of the structure at different service stages.Further, a mathematical relationship model between the corrosion degree and the bearing capacity is established based on the performance degradation law and a polynomial regression method, thereby providing a scientific mathematical tool for the evaluation and prediction of the structural bearing performance, and facilitating practical engineering applications.

[0284] In some optional embodiments, the step S106 includes:

[0285] In step S1061, the prediction accuracy of the mathematical relationship model at different service stages is judged, and a bearing performance dynamic evaluation relationship based on corrosion evolution is determined.

[0286] Specifically, for the established mathematical relationship model, verification data at different service stages can be obtained, and the prediction accuracy is judged by calculating the deviation between the predicted value and the actual value. If the prediction accuracy is lower than the preset threshold, the mathematical model is adjusted by increasing the sample data to obtain an optimized relationship model.

[0287] Further, according to the change trend of the relationship model, the characteristic parameters of the evolution process are obtained, the dynamic evaluation formula is updated through the characteristic parameters, and the bearing performance dynamic evaluation relationship based on corrosion evolution is obtained, as shown in the following relationship (2):

[0288] Z = f (X, t) (2)

[0289] In the formula, Z represents the dynamic bearing capacity, X represents the corrosion degree, and t represents the service time.

[0290] In step S1062, the boundary conditions and load parameters of the third finite element model are adjusted according to the bearing performance dynamic evaluation relationship based on corrosion evolution, and the adjusted model parameter set is obtained.

[0291] Specifically, the third finite element model can be initialized by using the bearing performance dynamic evaluation relationship based on corrosion evolution, and the boundary conditions and load parameters of the third finite element model are adjusted.

[0292] Further, the model parameters of the adjusted third finite element model are obtained to form a model parameter set.

[0293] In step S1063, the model parameter set is simulated to generate a structure behavior data set.

[0294] Specifically, the real-time simulation technology can be used to simulate the adjusted model parameters, and the corresponding structure behavior data set is generated for the service scene.

[0295] In step S1064, extreme environment features are extracted from the structure behavior data set, and an influence coefficient of the extreme environment on the residual resistance is determined according to the extreme environment features.

[0296] Specifically, extreme environmental characteristics are extracted from the structural behavior data, and then the influence coefficient of the extreme environment on the residual resistance is determined according to the extracted extreme environmental characteristics.

[0297] Illustratively, the displacement, stress, strain and other response information of the key nodes are extracted from the structural behavior data, and the extreme environmental load parameters are extracted.

[0298] Further, the mapping relationship between the extreme environmental characteristic parameters and the structural response quantities is established, and the variation law of the structural response under different environmental conditions is analyzed.

[0299] Further, the specific influence of the extreme environmental load on the residual resistance of the structure can be determined through mechanical analysis or numerical calculation.

[0300] Further, regression analysis, damage mechanics model or other quantitative methods can be used to convert the change of environmental characteristic parameters and structural response into influence coefficients. For example, based on the Miner linear cumulative damage rule, combined with fatigue life analysis, the damage accumulation rate under different environmental load combinations is calculated, and then the influence coefficient of the environment on the residual resistance is determined.

[0301] Further, the reasonableness of the influence coefficient can also be verified by comparing the simulation results or experimental data under different working conditions, and corrected according to the actual situation.

[0302] Step S1065, using a regression analysis algorithm to process the influence coefficient and determine the target influence trend data set of the residual resistance of the offshore wind power foundation structure.

[0303] Specifically, the influence coefficient can be further determined by the regression analysis algorithm to determine the target influence trend of the residual resistance of the offshore wind power foundation structure, and then the corresponding target influence trend data set can be obtained.

[0304] Illustratively, the determined influence coefficient of the extreme environment on the residual resistance (such as the influence coefficient corresponding to different wind speed, wave load, temperature and other environmental parameters) and the long-term service related parameters such as service time and corrosion degree are obtained.

[0305] At the same time, the quantitative indicators of the residual resistance (such as the bearing capacity degradation rate and the stress concentration coefficient change) are determined.

[0306] Further, according to the data characteristics, a suitable regression model can be selected, which can be a linear regression model, a nonlinear regression model, a multivariate regression model, etc.

[0307] Further, the influence coefficient and the residual resistance data are standardized or normalized to eliminate the influence of dimensions. Further, the outliers are removed to ensure the reliability of the data.

[0308] Further, the regression model parameters are solved by using methods such as least squares method, maximum likelihood estimation, etc.

[0309] Further, based on the solved regression model parameters, the regression model is extrapolated to long-term service time (such as 10 years, 20 years), and combined with the time evolution law of the influence coefficient (such as the change of corrosion rate with time), the long-term change of the residual resistance is predicted.

[0310] Further, the degradation rate of the residual resistance can also be judged by the slope, curvature and other parameters of the regression model.

[0311] Further, the critical value of the residual resistance can also be calculated by inputting the extreme environmental combination into the regression model, and the long-term safety of the structure under extreme conditions is evaluated.

[0312] Further, the target influence trend data set of the residual resistance of the offshore wind power foundation structure can be determined through the above process.

[0313] Step S1066, according to the target influence trend data set and the preset life cycle time sequence, the bearing capacity change curve is determined.

[0314] Specifically, by corresponding the target influence trend data set to the nodes of the preset life cycle time sequence, the corresponding bearing capacity change curve can be generated.

[0315] Step S1067, the bearing capacity change curve is analyzed and the full life cycle bearing performance prediction result of the offshore wind power foundation structure is determined.

[0316] Specifically, the bearing capacity change curve is analyzed in segments, and the key change nodes in the full life cycle are determined.

[0317] Further, if the key change node exceeds the preset threshold, the boundary conditions are adjusted for re-iteration simulation, and the optimized bearing capacity prediction is obtained.

[0318] Further, by comparing the simulation results under different environmental conditions, the long-term influence trend of extreme environment on the residual resistance is determined.

[0319] Further, by using the life prediction model and combining the material degradation law and the corrosion rate, the bearing capacity change of the offshore wind power foundation structure in the full life cycle can be predicted, and the corresponding full life cycle bearing performance prediction result is formed.

[0320] The method for assessing the load-bearing performance degradation of offshore wind turbine foundation structures provided in this embodiment determines the prediction accuracy of a mathematical relationship model at different service stages and establishes a dynamic assessment relationship based on corrosion evolution. This method can adapt to changes in the performance of the structure during different service stages, thus achieving dynamic assessment of the structure's load-bearing performance. Furthermore, by adjusting the boundary conditions and load parameters of the third finite element model based on the dynamic assessment relationship, the model can be made more consistent with the stress state and environmental conditions of the structure during actual service. Furthermore, simulation of the adjusted model parameter set can generate a dataset reflecting the structure's behavior in future service scenarios, overcoming the shortcomings of existing technologies in comprehensively simulating future service scenarios. Furthermore, by extracting extreme environmental characteristics from the structural behavior dataset, the degree of impact of extreme environments on the structure's residual resistance can be quantitatively analyzed and the impact coefficient can be determined, addressing the difficulty in accurately assessing the impact of extreme environments on the structure's residual resistance in existing technologies. Furthermore, by processing the impact coefficient using a regression analysis algorithm, the trend of residual resistance changes with service time and environmental conditions can be fitted, forming a target impact trend dataset. Furthermore, by combining the target impact trend dataset and the preset life cycle time series to determine the bearing capacity change curve, and conducting in-depth analysis of the bearing capacity change curve, it is possible to determine the bearing performance status of the structure at each key node throughout its life cycle, and then obtain the final full life cycle bearing performance prediction results, which provides a scientific basis for the full life cycle management of offshore wind power infrastructure, helps engineers formulate reasonable maintenance plans, determine the best maintenance time and plan, thereby extending the service life of the structure, reducing operation and maintenance costs, and improving the safety and economy of the structure.

[0321] In one example, Figure 2 As shown, a method for evaluating the bearing performance degradation of wind turbine towers, jackets, and foundations after corrosion is provided, specifically including:

[0322] S11. Obtain initial 3D geometric data of offshore wind turbine towers, jackets, and foundations. Use 3D modeling software to convert the 2D information in the drawings into an accurate 3D model to generate a high-precision digital model. Combined with historical service data, determine the distribution range and non-uniformity of corrosion thinning, and obtain a preliminary description of the changes in the structural geometric characteristics.

[0323] The initial digital model is generated by obtaining three-dimensional geometric data of the offshore wind power tower, jacket and foundation, and a high-precision digital model is obtained. Combined with historical service data, the high-precision digital model is processed by a data analysis method to determine the distribution range of the corrosion thinning. For the distribution range of the corrosion thinning, non-uniformity characteristics are analyzed to obtain non-uniformity distribution data. Through the non-uniformity distribution data, the geometric characteristics in the digital model are updated to generate an updated digital model. The updated digital model is used to analyze the change trend of the structure geometry to obtain structure geometry change data. If the structure geometry change data exceeds a preset threshold, a machine learning algorithm is used to predict the future distribution range of the corrosion thinning to obtain a prediction result. According to the prediction result, the geometric characteristics of the digital model are adjusted to generate a final structure geometry characteristic description.

[0324] Specifically, by obtaining the initial three-dimensional geometric data of the offshore wind power tower, jacket and foundation, the point cloud data generated after registration and filtering processing is used to generate a high-precision digital model using a Poisson surface reconstruction algorithm, and the model error is controlled within 5 mm. Combined with historical service data, the Monte Carlo simulation method is used to predict corrosion thinning, assuming that the corrosion rate is 5 mm per year and the simulation number is 10,000 times, the corrosion thinning distribution range is 1 to 8 mm, and the non-uniformity characteristic is that the local corrosion depth difference can reach 3 mm. The preliminary change of the structure geometry characteristics is described by using a finite element analysis software, assuming that the tower wall thickness is thinned from the initial 20 mm to 12 mm, the corrosion depth of the jacket node is 5 mm, and the corrosion thinning of the foundation pile leg is 7 mm. The analysis result shows that the structure stress distribution changes, the maximum stress concentration area appears in the position with larger corrosion depth, the stress value increases from the initial 150 MPa to 165 MPa, and the structure safety factor decreases from 5 to 3. On this basis, a machine learning algorithm such as support vector regression is further used to model the non-uniformity of the corrosion thinning, the training data set contains 1,000 groups of historical service data, the prediction error is controlled within 5%, and the distribution probability density function of the corrosion thinning is obtained, which provides data support for subsequent structure health monitoring and life assessment. Through the above technical means, the comprehensive quantitative analysis of the geometry characteristics of the offshore wind power foundation structure is realized, which provides a scientific basis for structure maintenance and optimized design.

[0325] S12, for the preliminary change description, the non-uniformity parameters of the corrosion thinning are extracted from the digital model, a refined grid model containing non-uniformity characteristics is constructed by using a finite element analysis method, the specific influence degree of the corrosion thinning on the geometry characteristics is judged, and an updated geometry characteristic distribution is obtained.

[0326] The digital model is used to obtain the non-uniformity parameters of corrosion thinning and determine the preliminary data distribution. The finite element method is used to construct a mesh model that incorporates the non-uniformity characteristics and obtain an initial mapping of the geometric properties. The mesh model is used to analyze the impact of non-uniformity on the geometric properties and determine the trend of change. The geometric property distribution is updated based on the trend of change to obtain optimized characteristic parameters. If the characteristic parameters exceed the preset threshold, the mesh model is adjusted through data processing to determine the corrected distribution. The updated geometric properties are extracted from the corrected distribution to obtain the final mapping result. A machine learning algorithm is used to verify the final mapping result and determine the consistency of the distribution.

[0327] Specifically, when extracting the non-uniform parameters of corrosion thinning through a digital model, a surface roughness analysis method based on point cloud data can be used. For example, the least squares method is used to fit the surface profile, the standard deviation σ of the local thickness variation is calculated to be 15 mm, and the maximum thinning amount is determined to be 8 mm. In the finite element analysis, a refined mesh model is constructed based on these parameters, and adaptive meshing technology is used to set the mesh density to 1 mm in the corrosion area and 5 mm in the non-corrosion area to ensure a balance between calculation accuracy and efficiency. By applying boundary conditions and loads, such as an internal pressure of 10 MPa, an elastic-plastic material model is used for analysis to obtain a stress distribution cloud map. The results show that the maximum stress in the corrosion area reaches 350 MPa, exceeding the material yield strength of 300 MPa. Based on the analysis results, the geometric property distribution is updated, and the thickness data is spatially interpolated using the Kriging interpolation algorithm to generate a new geometric model. The standard deviation of the thickness distribution is reduced to 1 mm, and the maximum thinning amount is revised to 75 mm, providing a reliable basis for subsequent structural integrity assessment.

[0328] S13, such as Figure 3 As shown in the figure, based on the updated geometric property distribution, the initial assumption of local damage is introduced into the finite element model. The evolution path and cumulative effect of local damage are simulated through iterative calculation, the occurrence location and intensity of stress concentration are determined, and the coupled distribution of local damage and stress concentration is obtained.

[0329] The initial assumptions of local damage are loaded through a pre-established finite element model, and updated data on the geometric property distribution are obtained. The evolution path of the local damage is obtained through iterative calculation. The cumulative effect is analyzed based on the evolution path, and the location of stress concentration is determined by the cumulative effect. If the stress concentration exceeds the preset threshold, the strength value of the stress concentration is calculated through strength analysis to obtain strength distribution data. The finite element model is used to verify the strength distribution data to obtain the coupled distribution of local damage and stress concentration. The key areas of stress concentration are extracted through the coupled distribution, and the position change trend of the key areas is determined. The boundary conditions of the finite element model are updated according to the change trend, and the optimized coupled distribution is obtained through iterative calculation. The final stress concentration location and strength value are determined through the optimized coupled distribution.

[0330] Specifically, in the finite element model, first, according to the updated geometric characteristic distribution, it is assumed that the initial damage value of a certain local area is 0, and the damage model based on the energy method is adopted, and the critical energy release rate of damage evolution is set to 1000 J / m 2 . Through iterative calculation, the Newton-Raphson algorithm is used to solve the nonlinear equations, with a step size of 0.1 for each iteration until the convergence condition is met (the residual is less than 1e-5). During the iteration process, the local damage value gradually increases, and when it reaches 0.8, stress concentration begins to appear. Through VonMises stress analysis, it is determined that the stress concentration occurs at node A in the edge area of ​​the model, and its stress value is 500MPa, which is much higher than the average stress value of 200MPa of the surrounding nodes. Further analysis shows that as the damage accumulates to 0.9, the stress value of node A rises to 700MPa, forming a significant stress concentration zone. Finally, through the finite element analysis of coupled damage and stress distribution, the spatial distribution map of local damage and stress concentration is obtained, which provides an important basis for subsequent structural optimization design.

[0331] S14. Obtain coupling distribution data, combine it with the experimental degradation curve of the material mechanical properties, and use the numerical mapping method to integrate the time series parameters of material degradation into the finite element model to determine the degree of weakening of the residual resistance caused by material degradation, and obtain the material performance characterization after comprehensive degradation.

[0332] The coupling distribution data and experimental degradation data are acquired through the acquisition system to obtain the initial mechanical performance parameters. The time series in the experimental degradation data are processed using a numerical mapping method to obtain a degradation parameter sequence. The degradation parameter sequence is integrated into a pre-established finite element model to obtain the material state after simulated degradation. If the simulation results show that the material degradation exceeds the preset threshold, the degree of weakening of the residual resistance is calculated to obtain the resistance change trend. Based on the resistance change trend and combined with the mechanical performance parameters, the performance characterization data after comprehensive degradation is determined. The correlation between material degradation and the degree of weakening is analyzed through the performance characterization data to obtain the degradation impact distribution. A machine learning algorithm is used to cluster the degradation impact distribution to determine the key impact areas of material degradation.

[0333] Specifically, the team first obtained coupled material distribution data through experiments. For example, the time-varying curve of the tensile strength of an aluminum alloy exposed to high temperature showed that after 300 hours of continuous high temperature exposure, the tensile strength of the material dropped from an initial 450 MPa to 380 MPa. Next, combining the experimental degradation curves of the material's mechanical properties, a numerical mapping method was used to incorporate the time series parameters of the material degradation into the finite element model.

[0334] Specifically, experimental data is mapped onto the nodes of the finite element model using interpolation algorithms, such as cubic spline interpolation, to ensure that the degradation process of each node aligns with the experimental data. Then, the finite element analysis software is used to simulate the stress distribution and deformation of the material at different stages of degradation.

[0335] For example, after 100 hours of degradation, the model shows that the stress concentration factor at a critical location increases from 2 to 5, indicating a significant decrease in the resistance of the local area of the material. Finally, based on the results of the finite element analysis, the degree of weakening of the residual resistance of the material due to degradation is determined, and the comprehensive performance characterization of the degraded material is obtained.

[0336] For example, through calculation, it is found that the overall resistance of the material after 300 hours has decreased by 15%, and the stress concentration effect at the critical location has significantly increased. This process not only quantifies the impact of material degradation, but also provides reliable data support for subsequent structural safety evaluation.

[0337] S15, by comprehensively characterizing the performance of the degraded material, extreme environmental load conditions are applied in the finite element model, and dynamic simulation techniques are used to calculate the variation of static and dynamic responses, determine the evolution trend of stress concentration and failure mode under load, and obtain the response characteristic curve of the structure.

[0338] Through the performance data of the degraded material, an initial calculation framework is constructed in the finite element model, and load conditions under extreme environments are applied to obtain preliminary simulation basic data. Dynamic simulation techniques are used to process the simulation basic data, calculate the variation of static and dynamic responses, and obtain dynamic distribution data of the responses. Key areas are extracted from the dynamic distribution data of the responses, the location and intensity of stress concentration are determined, and the stress concentration distribution results are obtained. Based on the stress concentration distribution results, the type and characteristics of the failure mode are analyzed, the change direction of the evolution trend is determined, and the evolution data of the failure mode are obtained. The response characteristic curve is fitted by the evolution data, the discrete points in the curve are processed by interpolation method, and the smooth characteristic curve data are obtained. The inflection points and extreme values in the characteristic curve data are determined, the overall response characteristics of the structure are determined, and the final structural curve distribution is obtained. Key parameters are extracted from the structural curve distribution, the stability of the structure is judged by the preset threshold, and the evaluation result is obtained.

[0339] Specifically, through comprehensive characterization of material properties after degradation, the aluminum alloy was first subjected to tensile testing using a tensile testing machine, measuring a yield strength of 280 MPa, a tensile strength of 320 MPa, and an elongation of 15% at room temperature. The material was then subjected to aging treatment at 200°C for 1000 hours, with the yield strength dropping to 250 MPa, the tensile strength to 300 MPa, and the elongation to 12%. Based on these data, a constitutive relationship for the material was established in a finite element model, using the Johnson-Cook plasticity model with a strain hardening coefficient of 45, a strain rate sensitivity coefficient of 0.12, and a temperature softening coefficient of 0. In the finite element analysis, the structure was subjected to extreme environmental loading conditions, including a temperature field increase from 20°C to 200°C and a dynamic impact load with a peak acceleration of 500 g and a duration of 10 ms. Dynamic simulations were performed using an explicit dynamics solver to calculate the static and dynamic responses of the structure. By extracting the displacement, velocity and acceleration response curves of key nodes, it was found that the maximum displacement occurred in the middle of the structure, reaching 15mm, and the maximum acceleration was 800g. Further analysis of the stress distribution and the VonMises stress criterion were used to determine that the stress concentration area appeared at the corner of the structure, with a maximum stress value of 350MPa, which is close to the tensile strength of the material. Based on fatigue damage theory, the Miner linear cumulative damage law was used to calculate the fatigue life of the structure under cyclic loads. When the fatigue damage factor reaches 0, the structure fails. By drawing the stress-life curve and determining the response characteristic curve of the structure, it was found that when the stress amplitude exceeded 250MPa, the fatigue life dropped sharply, and the number of cycles dropped from 10^6 times to 10^4 times. Finally, the structural design was optimized according to the simulation results, the fillet radius at the corner was increased, and the maximum stress value was reduced to 280MPa, which significantly improved the fatigue life of the structure.

[0340] S16, such as Figure 4 As shown in the figure, according to the response characteristic curve, the stress and displacement values ​​of the key nodes are extracted from the finite element model. By comparing and analyzing with the initial bearing capacity, the comprehensive influence of corrosion thinning, local damage and material degradation on the bearing capacity is judged, and the quantitative evaluation results of the residual bearing capacity are obtained.

[0341] The response characteristic curve is obtained by the finite element model, and the stress value and displacement value of the key node are extracted to obtain preliminary data. According to the comparison between the preliminary data and the initial bearing performance, the change trend of the bearing performance caused by corrosion thinning is judged. The distribution characteristics of local damage are analyzed through the change trend to determine the influence degree of material degradation. The support vector machine algorithm is used for classification according to the comprehensive influence to obtain the weight distribution of corrosion thinning, local damage and material degradation. If a factor in the weight distribution exceeds the preset threshold, the response characteristics are adjusted through the finite element model to obtain updated stress values and displacement values. The remaining bearing capacity is calculated according to the updated stress values and displacement values to obtain the quantitative evaluation result. The final change state of the bearing performance is determined by analyzing the deviation between the quantitative evaluation result and the initial bearing performance.

[0342] Specifically, in the finite element model, the response characteristic curve can be generated by extracting the stress value and displacement value of the key node to evaluate the bearing performance of the structure.

[0343] For example, in a certain bridge model, the midspan node is selected as the key node, and the stress value of 120 MPa and the displacement value of 15 mm are extracted. Through comparison and analysis with the initial bearing performance, it is found that corrosion thinning causes the stress value to increase by 20 MPa, local damage causes the displacement value to increase by 5 mm, and material degradation further increases the stress value by 10 MPa. Through comprehensive analysis of these factors, the linear superposition algorithm is used to calculate the remaining bearing capacity as 75% of the initial bearing capacity. The specific algorithm is: remaining bearing capacity = initial bearing capacity × (1-stress increase ratio-displacement increase ratio-material degradation ratio) = 100% × (1-167-333-083) = 75%. This quantitative evaluation result provides an important basis for subsequent structure maintenance and reinforcement, ensuring the safety and durability of the bridge.

[0344] S17, for the quantitative evaluation result, a Monte Carlo simulation method is used to introduce random variables to simulate the change trend of the bearing capacity under different corrosion degrees and extreme environment combinations, determine the probability distribution interval of the bearing performance, and obtain the performance degradation law suitable for long-term service prediction.

[0345] The Monte Carlo simulation method is used to introduce random variables to generate bearing capacity data under different corrosion levels and extreme environment combinations. Statistical analysis tools are used to process the bearing capacity data to obtain the change trend of the bearing capacity. A probability distribution model is applied to the change trend to determine the probability distribution interval of the bearing performance. The performance degradation parameter is calculated through the probability distribution interval to obtain the preliminary law of performance degradation. If the performance degradation parameter exceeds the preset threshold, the simulation parameters are adjusted in combination with long-term service data to obtain the corrected degradation law. According to the corrected degradation law, a prediction law suitable for long-term service is determined. After obtaining the prediction law, the applicability of the prediction law under different environment combinations is verified through the Monte Carlo simulation method to obtain the final bearing performance prediction result.

[0346] Specifically, on the basis of quantitative evaluation results, the Monte Carlo simulation method is used to introduce random variables to simulate the change trend of bearing capacity under different corrosion levels and extreme environment combinations. First, the corrosion level is set to be uniformly distributed from 1 mm / year to 5 mm / year, and the extreme environment conditions include temperature fluctuation range of-20℃ to 50℃ and humidity range of 30% to 90%. By generating 10,000 random samples, the bearing capacity under each combination is calculated in combination with the material mechanical performance model.

[0347] For example, for a sample with a corrosion level of 3 mm / year, a temperature of 30℃, and a humidity of 60%, a linear regression model is used to predict that its bearing capacity will decrease to 85% of the initial value. By statistically analyzing the bearing capacity data of all samples, the probability distribution interval is determined, such as the bearing capacity under the 95% confidence interval being 75% to 90% of the initial value. Further, a time series analysis method is used to establish the change law of bearing performance with service time, and it is predicted that within 10 years of service, the bearing capacity will decrease to 70% to 80% of the initial value. Finally, by fitting the Weibull distribution function, the performance degradation law suitable for long-term service prediction is obtained, with a shape parameter of 5 and a scale parameter of 15 years, indicating that within 15 years, there is a high probability that the bearing capacity will decrease to less than 50% of the initial value.

[0348] S18, obtain performance degradation law data, establish a mathematical relationship model between corrosion level and bearing capacity through data fitting technology, judge the prediction accuracy of the model at different service stages, and obtain a bearing performance dynamic evaluation formula based on corrosion evolution.

[0349] Obtain performance degradation related law data, obtain complete data set by collecting multi-source sensor data. Adopt data fitting technology to process law data, and establish a mathematical model of corrosion level and bearing capacity through polynomial regression, which is expressed as Y=aX 2+ bX + c. For the mathematical model, obtain the verification data of different service stages, and judge the prediction accuracy by calculating the deviation between the predicted value and the actual value. If the prediction accuracy is lower than the preset threshold, adjust the mathematical model by increasing the sample data to obtain the optimized relationship model. According to the change trend of the relationship model, obtain the characteristic parameters of the evolution process, update the dynamic evaluation formula through the characteristic parameters, and the formula is Z = f(X, t). Through the dynamic evaluation formula, judge the decline law of the bearing capacity in the evolution process, and obtain the performance degradation evaluation result based on the corrosion evolution. Adopt the dynamic evaluation result, combine the service stage data, and determine the bearing performance threshold of each stage through linear interpolation method.

[0350] Specifically, in actual engineering, in order to obtain the performance degradation rule data, the temperature, humidity, salt mist concentration and other parameters of the corrosion environment can be monitored through sensors, and the relationship between the corrosion degree and time can be established combined with the corrosion weight loss data of the material sample.

[0351] For example, a steel structure is in service in a salt mist environment, and the corrosion weight loss data is collected every 30 days. After 180 days of monitoring, the weight loss amounts are 1 gram, 25 grams, 45 grams, 7 grams, 0 grams and 35 grams. Using the least square method to fit these data, a quadratic polynomial relationship between the corrosion degree and time can be obtained: C(t) = 0.03t 2 + 0.1t + 0.05, where C(t) is the corrosion degree and t is the time (day). Then, through the laboratory loading test, the bearing capacity data under different corrosion degrees is measured, for example, when the corrosion degree is 1, 25, 45, 7, 0 and 35, the bearing capacity is 100 kN, 95 kN, 88 kN, 80 kN, 70 kN and 60 kN respectively. Using multiple linear regression analysis, a linear relationship model between bearing capacity and corrosion degree is established: P(C) = -30C + 103, where P(C) is the bearing capacity. In order to verify the model accuracy, the data is divided into training set and test set, the training set is used to fit the model, and the test set is used to evaluate the prediction error. By calculating the root mean square error (RMSE) and the determination coefficient (R 2 ), it is found that the prediction accuracy of the model is higher in the early stage of service, the RMSE is 5 kN, and the R 2 is 98; but in the later stage of service, the prediction error gradually increases, the RMSE rises to 2 kN, and the R 2 drops to 92. In order to dynamically evaluate the bearing performance, combine the corrosion evolution model, and substitute P(C) into C(t) to obtain the bearing performance dynamic evaluation formula based on time: P(t) = -30(0.03t 2 + 0.1t + 0.05) + 103, so as to realize the real-time prediction and evaluation of the bearing capacity of the structure.

[0352] S19, adjusting boundary conditions and load parameters in the finite element model according to the dynamic evaluation formula, simulating the structural behavior in future service scenarios using real-time simulation technology, determining the long-term influence trend of extreme environments on residual resistance, and obtaining the prediction of the change of the bearing capacity of offshore wind power foundation structure in the whole life cycle.

[0353] The finite element model is initialized by the dynamic evaluation formula, the boundary conditions and load parameters are adjusted, and the adjusted model parameters are obtained. Real-time simulation technology is used to simulate the adjusted model parameters, and structural behavior data is generated for the service scenario. Extreme environment features are extracted from the structural behavior data to determine the influence coefficient of extreme environments on residual resistance. The influence coefficient is processed by regression analysis algorithm to determine the long-term influence trend of residual resistance. The long-term influence trend data is obtained, and the bearing capacity change curve is obtained by combining the life cycle time series. The bearing capacity change curve is analyzed in sections to determine the key change nodes in the whole life cycle. If the key change nodes exceed the preset threshold, the bearing capacity prediction is obtained by adjusting the boundary conditions and reiterating the simulation.

[0354] Specifically, in the finite element model, first, the ANSYS software is used to establish a fine mesh model of the offshore wind power foundation structure, and the mesh size is set to 5 meters to ensure the calculation accuracy. By adjusting the boundary conditions, the foundation constraint is set to be completely fixed to simulate the actual installation situation. In terms of load parameters, the extreme wind load of 25 meters per second and the impact load of 10 meters of wave height are input, and these parameters are based on the marine environment data of once in a hundred years. Real-time simulation technology is used to perform transient analysis with an explicit dynamic algorithm with a time step of 0.1 seconds to simulate the dynamic response of the structure in future service scenarios. During the analysis process, by monitoring the displacement and stress of the key nodes, it is found that the maximum displacement occurs at the top of the tower, reaching 15 meters, and the maximum stress occurs at the foundation connection, reaching 350 megapascals. Further, the fatigue damage accumulation theory is used to calculate the fatigue life of the structure in 20 years of service period based on the Miner linear cumulative damage rule. By comparing the simulation results under different environmental conditions, the long-term influence trend of extreme environments on residual resistance is determined. Finally, using the life prediction model, combined with the material degradation law and corrosion rate, the bearing capacity change of the offshore wind power foundation structure in the whole life cycle is predicted, and the results show that the structural bearing capacity decreases by about 12% at the 15th year and by about 18% at the 20th year. This prediction result provides an important basis for structure maintenance and life assessment.

[0355] The load bearing performance degradation evaluation method for the wind power tower, jacket and foundation after corrosion provided by the example is provided, initial geometric data of the structure is acquired, corrosion thinning characteristics are analyzed in combination with historical service data, and a refined finite element model containing non-uniformity is constructed. Local damage assumption is introduced in the model, damage evolution path is simulated, and stress concentration distribution is determined. Meanwhile, material degradation parameters are integrated into the model, and the influence of the material degradation parameters on the residual resistance is evaluated. By applying extreme environmental load, the structural response characteristics are calculated, and the comprehensive influence of corrosion, damage and material degradation on the load bearing performance is quantitatively evaluated. Finally, a mathematical relationship model of the corrosion degree and the load bearing capacity is established, the change of the load bearing capacity in the whole life cycle is predicted, a scientific basis is provided for the safety evaluation and maintenance decision of the offshore wind power foundation structure, the service life is prolonged, and the operation and maintenance cost is reduced.

[0356] Although the embodiments of the present application are described in conjunction with the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A method for evaluating the degradation of the bearing performance of an offshore wind power foundation structure, characterized in that: The method comprises: Obtain the initial 3D geometric point cloud dataset and historical service dataset of the offshore wind power foundation structure; Analyzing the corrosion and thinning characteristics of the offshore wind power foundation structure based on the initial three-dimensional geometric point cloud dataset and the historical service dataset to obtain a first geometric distribution characteristic dataset and a first finite element model including non-uniformity; Based on the first geometric distribution characteristic data set, simulating the evolution path and cumulative effect of local damage of the offshore wind turbine foundation structure using a second finite element model to obtain a target coupled distribution data set between local damage and stress concentration, wherein the second finite element model is obtained by introducing an initial assumption condition of local damage into the first finite element model; Based on the target coupled distribution dataset and a preset experimental degradation dataset of material mechanical properties, using the second finite element model to evaluate the degree of weakening of the residual resistance caused by material degradation and determine a target material property characterization dataset after comprehensive degradation; Based on the target material performance characterization data set, using a third finite element model to analyze the performance degradation law of the offshore wind power infrastructure and establish a mathematical relationship model between the degree of corrosion and the load-bearing capacity, the third finite element model being obtained by introducing extreme environmental load conditions into the second finite element model; The mathematical relationship model and the third finite element model are used to predict the degradation of the bearing performance of the offshore wind power foundation structure, and a full life cycle bearing performance prediction result of the offshore wind power foundation structure is obtained.

2. The method according to claim 1, characterized in that Based on the initial three-dimensional geometric point cloud dataset and the historical service dataset, the corrosion thinning characteristics of the offshore wind power foundation structure are analyzed to obtain a first geometric distribution characteristic dataset and a first finite element model including non-uniformity, including: Analyzing the geometric characteristics of the offshore wind power infrastructure structure based on the initial three-dimensional geometric point cloud dataset and the historical service dataset and determining a target digital model and a second geometric distribution characteristic dataset of the offshore wind power infrastructure structure; Based on the second geometric distribution characteristic data set, a surface roughness analysis method based on point cloud data is used to extract a non-uniform parameter set of corrosion thinning in the target digital model; Based on the non-uniform parameter set, a fourth finite element model is constructed after processing using a finite element analysis method, and an initial stress distribution mapping relationship is determined according to the fourth finite element model, wherein the initial stress distribution mapping is used to characterize a mapping relationship from corrosion geometric characteristics to mechanical properties; Based on the initial stress distribution mapping relationship, the fourth finite element model is used to analyze the influence of non-uniformity on geometric characteristics and determine the first finite element model and the first geometric distribution characteristic data set containing non-uniformity.

3. The method according to claim 2, characterized in that Analyzing the geometric characteristics of the offshore wind power infrastructure structure based on the initial three-dimensional geometric point cloud dataset and the historical service dataset and determining a target digital model and a second geometric distribution characteristic dataset of the offshore wind power infrastructure structure, including: Establishing a first digital model based on the initial three-dimensional geometric point cloud data set; Based on the historical service data set, the first digital model is processed using a Monte Carlo simulation method to determine a first distribution range of corrosion thinning; Analyzing the non-uniformity characteristics of the offshore wind power infrastructure according to the first distribution range and obtaining a non-uniformity distribution data set; Based on the non-uniform distribution data set, using a finite element analysis method to analyze the geometric characteristics of the offshore wind power foundation structure and update the first digital model to obtain a second digital model; determining a geometric variation dataset of the offshore wind power infrastructure using the second digital model; When the geometric change data set does not meet the first preset requirement, a machine learning algorithm is used to predict the future distribution range of corrosion thinning to obtain a second distribution range; The geometric characteristics of the second digital model are adjusted according to the second distribution range to obtain the target digital model and the second geometric distribution characteristic data set.

4. The method according to claim 1, wherein Based on the first geometric distribution characteristic data set, a second finite element model is used to simulate the evolution path and cumulative effect of local damage of the offshore wind power foundation structure to obtain a target coupled distribution data set between local damage and stress concentration, including: Based on the first geometric distribution characteristic data set, introducing an initial assumption of local damage into the first finite element model and generating a fifth finite element model and a third geometric distribution characteristic data set; Iteratively solving the fifth finite element model based on the third geometric distribution characteristic data set to obtain an evolution path of local damage; Analyzing the cumulative effect of local damage according to the evolution path and determining an initial coupled distribution data set of local damage and stress concentration; determining a position change trend of a first critical area of ​​stress concentration according to the initial coupled distribution data set; updating the boundary conditions of the fifth finite element model according to the position change trend and determining the second finite element model; The target coupling distribution dataset is determined according to the second finite element model.

5. The method according to claim 1, wherein Based on the target coupled distribution dataset and the preset experimental degradation dataset of the material mechanical properties, the second finite element model is used to evaluate the degree of weakening of the residual resistance caused by material degradation and determine the target material performance characterization dataset after comprehensive degradation, including: determining an initial mechanical property parameter set based on the target coupling distribution data set and the preset experimental degradation data set; Processing the time series in the preset experimental degradation data using a numerical mapping method to obtain a degradation parameter sequence; Inputting the degradation parameter sequence into the second finite element model to obtain an initial material property characterization data set after simulated degradation; When the initial material property characterization data set does not meet the third preset requirement, calculating the residual resistance weakening degree and determining the resistance change trend according to the material state data set; According to the resistance variation trend and the initial mechanical property parameter set, a performance characterization data set of the target material after comprehensive degradation is determined.

6. The method according to claim 1, characterized in that Based on the target material performance characterization data set, a third finite element model is used to analyze the performance degradation law of the offshore wind power foundation structure, and a mathematical relationship model between the degree of corrosion and the load-bearing capacity is established, including: Based on the target material performance characterization data set, using the third finite element model to analyze the structural characteristics of the offshore wind power foundation structure and determine a structural characteristic curve of the offshore wind power foundation structure; performing a quantitative evaluation of the residual bearing capacity of the offshore wind power foundation structure according to the structural characteristic curve and the third finite element model to obtain a quantitative evaluation result; Analyzing the performance degradation law of the offshore wind power infrastructure according to the quantitative assessment results to obtain a target performance degradation data set; Based on the target performance degradation data set, a polynomial regression method is used to establish the mathematical relationship model between the degree of corrosion and the load-bearing capacity.

7. The method according to claim 6, characterized in that Analyzing the structural characteristics of the offshore wind power foundation structure using the third finite element model based on the target material performance characterization data set and determining a structural characteristic curve of the offshore wind power foundation structure includes: Based on the target material performance characterization data set, constructing an initial calculation framework in the third finite element model and generating an initial simulation basic data set; Using dynamic simulation technology to process the initial simulation basic data set and calculate the changing rules of static response and dynamic response; Determine a dynamic response distribution data set according to the variation rules of the static response and the dynamic response; Extracting a second key area from the dynamic response distribution data set and determining a stress concentration distribution result; Analyzing the type and characteristics of the failure model of the offshore wind power infrastructure according to the stress concentration distribution result and determining a failure mode evolution data set; The structural characteristic curve is determined according to the failure mode evolution data set.

8. The method according to claim 6, characterized in that A quantitative assessment is performed on the residual bearing capacity of the offshore wind power foundation structure according to the structural characteristic curve and the third finite element model to obtain a quantitative assessment result, including: extracting stress values ​​and displacement values ​​of key nodes in the third finite element model according to the structural characteristic curve and generating an initial data set; Determining a change trend of the load-bearing performance due to corrosion thinning based on the initial data set and the initial load-bearing performance; determining the distribution characteristics of local damage according to the change trend; determining the degree of impact of material degradation based on the distribution characteristics; Based on the change trend, the distribution characteristics and the impact degree, a weight distribution data set is obtained through support vector machine algorithm processing; The quantitative assessment result of the residual bearing capacity of the offshore wind power infrastructure is determined according to the weight distribution data set.

9. The method according to claim 6, characterized in that Based on the quantitative assessment results, the performance degradation law of the offshore wind power infrastructure is analyzed to obtain a target performance degradation data set, including: Based on the quantitative evaluation results, a Monte Carlo simulation method is used to introduce random variables and generate a load-bearing capacity data set under different combinations of corrosion degrees and extreme environments; Performing statistical analysis on the carrying capacity data set and determining a carrying capacity change trend; Based on the load-bearing capacity variation trend, determining a probability distribution interval of the load-bearing performance using a probability distribution model; Calculating the performance degradation parameter according to the probability distribution interval and determining a target law data set of performance degradation; Determining a performance degradation law data set based on the target law data set; The target performance degradation data set is determined according to the performance degradation law data set.

10. The method according to claim 1, characterized in that The mathematical relationship model and the third finite element model are used to predict the degradation of the bearing performance of the offshore wind power foundation structure, and a full life cycle bearing performance prediction result of the offshore wind power foundation structure is obtained, including: Determining the prediction accuracy of the mathematical relationship model at different service stages and determining a dynamic evaluation relationship for load-bearing performance based on corrosion evolution; According to the dynamic evaluation relationship of load-bearing performance based on corrosion evolution, the boundary conditions and load parameters of the third finite element model are adjusted to obtain an adjusted model parameter set; simulating the model parameter set and generating a structural behavior data set; extracting extreme environmental characteristics from the structural behavior data set, and determining an influence coefficient of the extreme environment on the residual resistance based on the extreme environmental characteristics; Processing the influence coefficient using a regression analysis algorithm and determining a target influence trend data set of the residual resistance of the offshore wind power foundation structure; Determine a carrying capacity change curve based on the target impact trend dataset and a preset life cycle time series; The load-bearing capacity variation curve is analyzed to determine the full life cycle load-bearing performance prediction result of the offshore wind power infrastructure.

Images