Digital twin-driven retired wind turbine generator predictive maintenance decision-making system

Abstract

The invention relates to the technical field of wind turbine generator maintenance decision-making, and discloses a digital twin-driven retired wind turbine generator predictive maintenance decision-making system. The system comprises an operation data acquisition module, a component data analysis module, a digital twin model construction module, an adaptation compatibility evaluation module and a maintenance decision generation module. Formulating core maintenance parameters and a priority sequence thereof based on the operation and component data; constructing a digital twin model prototype by fusing core and alternative component materials, and analyzing signal detection characteristics of the digital twin model prototype; calculating a life prediction equivalent value according to durability test data of the model, and evaluating the adaptation compatibility between the life prediction equivalent value and a maintenance system; and comprehensively screening an optimal maintenance material and generating a decision. According to the method, joint simulation and compatibility quantitative evaluation are carried out on various material replacement schemes in a virtual environment, and the signal sensing capability and the system matching degree can be optimized before actual maintenance, so that the feasibility and decision reliability of life extension operation and maintenance of the retired unit are improved.

Description

Technical Field

[0001] This invention relates to the field of wind turbine maintenance decision-making technology, specifically a digital twin-driven predictive maintenance decision-making system for decommissioned wind turbines. Background Technology

[0002] At the data acquisition level, existing technologies mostly focus on collecting routine parameters during wind turbine operation, lacking a systematic and comprehensive mechanism for acquiring operating condition parameters and dynamic performance parameters during the decommissioning phase. Most maintenance plans rely only on some basic operating data and fail to develop targeted core maintenance parameters based on the characteristics of aging and performance degradation of decommissioned turbine components. This results in a lack of clear goal orientation in subsequent maintenance work, often falling into the dilemma of "comprehensive overhaul but failing to grasp the key points".

[0003] In component analysis, traditional methods typically focus only on surface damage or single performance indicators, failing to delve into the intrinsic data and structural design data of components. The lack of systematic analysis of the physical performance characteristics corresponding to the intrinsic data makes it impossible to accurately determine the importance and maintenance urgency of different components in decommissioned units, thus hindering the formation of a scientific parameter prioritization sequence. This often results in maintenance resources being allocated to non-critical components, leading to a waste of human and material resources, and may also overlook the potential failure risks of core components.

[0004] While the application of digital twin technology in industrial equipment maintenance is gradually increasing, its application in the field of decommissioned wind turbines remains immature. Existing digital twin models are mostly built based on material and structural data of the turbine in its new state, failing to fully consider the aging and deterioration of core components in decommissioned turbines, and rarely incorporating alternative component materials for comparative simulation. After model construction, only basic functions are simply verified; there is a lack of systematic collection and analysis of sensing accuracy data and durability test data, making it impossible to accurately grasp the model's signal detection characteristics. Consequently, the model fails to realistically reflect the actual operating state of the decommissioned turbine and cannot provide reliable simulation support for maintenance decisions.

[0005] In terms of compatibility evaluation, current technologies often overlook the compatibility between digital twin models and maintenance systems. In most cases, only the model's own performance indicators are considered, without combining durability test data to calculate the model's lifespan prediction equivalent value, or deeply analyzing the model's compatibility with actual maintenance systems. This results in the constructed digital twin model potentially being incompatible with existing maintenance processes and equipment, unable to effectively integrate into the maintenance decision-making system, and failing to play its due role.

[0006] During the maintenance decision-making phase, existing solutions largely rely on human experience and incorporate relatively singular parameters. Maintenance strategies are often formulated solely based on the degree of component damage or operating time, failing to comprehensively consider key factors such as the signal detection characteristics, performance stability, and compatibility of the digital twin model. Furthermore, in material selection, there is a lack of comprehensive comparison and screening of materials for core components and alternative components, making it difficult to determine the most suitable maintenance materials for decommissioned units. This results in maintenance decisions lacking scientific rigor and rationality, potentially failing to completely resolve existing unit problems and increasing subsequent maintenance costs and safety hazards due to inappropriate material selection. Summary of the Invention

[0007] The purpose of this invention is to provide a digital twin-driven predictive maintenance decision system for decommissioned wind turbines to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides a digital twin-driven predictive maintenance decision-making system for decommissioned wind turbines, the system comprising:

[0009] The operation data acquisition module is used to acquire the operating condition parameters and dynamic performance parameters of decommissioned wind turbines and to formulate the core maintenance parameters of wind turbines.

[0010] The component data analysis module is used to schedule the intrinsic data of wind turbine components and structural design data, analyze the physical performance characteristics corresponding to the intrinsic data of the components, and determine the parameter priority sequence of the core maintenance parameters.

[0011] The digital twin model construction module is used to query the core component materials and alternative component materials of the wind turbine, combine the core component materials and alternative component materials to simulate and build the wind turbine to obtain a digital twin model prototype, collect the sensing accuracy data and durability test data of the digital twin model prototype, and analyze the signal detection characteristics of the digital twin model prototype based on the sensing accuracy data.

[0012] The compatibility evaluation module is used to calculate the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, and analyze the compatibility of the digital twin model prototype in the maintenance system.

[0013] The maintenance decision generation module is used to combine the signal detection characteristics, the performance stability, and the compatibility to select the best maintenance material from the core component materials and the alternative component materials, and generate a maintenance decision for the wind turbine based on the core maintenance parameters, the parameter priority sequence, and the best maintenance material.

[0014] Preferably, the operation data acquisition module determines the working environment elements of the wind turbine based on the operating condition parameters and the dynamic performance parameters, analyzes the application scenario characteristics of the wind turbine based on the working environment elements, collects the existing fault information of the wind turbine, and formulates the core maintenance parameters of the wind turbine by combining the application scenario characteristics and the existing fault information.

[0015] The operational data acquisition module combines the application scenario characteristics and existing fault information to formulate the core maintenance parameters of the wind turbine, including:

[0016] Feature extraction is performed on the application scenario features to obtain scenario feature factors;

[0017] The existing fault information is classified to obtain a set of fault categories;

[0018] The scene feature factors are encoded to obtain a scene feature encoding vector;

[0019] The fault category set is encoded to obtain a fault category encoding vector;

[0020] Cluster analysis is performed on the scene feature encoding vectors to obtain scene feature cluster centers;

[0021] Cluster analysis is performed on the fault category encoding vectors to obtain the fault category cluster centers;

[0022] The correlation between the cluster centers of the scene features and the cluster centers of the fault categories is analyzed to obtain the correlation mapping matrix;

[0023] Based on the correlation mapping matrix, the key scenario characteristics and key fault categories of the wind turbine are determined;

[0024] Based on the key scenario characteristics and key fault categories, the design constraints of the wind turbine are analyzed.

[0025] Based on the aforementioned design constraints, the core maintenance parameters for the wind turbine are determined.

[0026] Preferably, the component data analysis module analyzes the physical performance characterization corresponding to the intrinsic data of the component, determines the signal propagation topology of the wind turbine based on the structural design data, evaluates the synergistic coupling effect between the signal propagation topology and the physical performance characterization, and determines the parameter priority sequence of the core maintenance parameters based on the synergistic coupling effect.

[0027] The component data analysis module analyzes the physical performance characterization corresponding to the intrinsic data of the component, including:

[0028] The intrinsic data of the component are standardized to obtain standardized component data;

[0029] Extract the physical attributes of the components corresponding to the standardized component data, and filter the physical attributes to obtain key physical attributes;

[0030] Calculate the physical performance index corresponding to the key physical attributes, and generate the physical performance characterization corresponding to the intrinsic data of the component based on the physical performance index.

[0031] Preferably, the component data analysis module determines the signal propagation topology of the wind turbine based on the structural design data, including:

[0032] The structural design data is preprocessed to obtain the target structural design data;

[0033] Extract the set of structural parameters of the wind turbine from the target structural design data;

[0034] The structural parameter set is subjected to material property association processing to obtain a property association parameter set;

[0035] Based on the attribute association parameter set, construct the signal propagation numerical model corresponding to the wind turbine;

[0036] The signal propagation numerical model is simulated and processed to obtain dynamic data of signal propagation.

[0037] The signal propagation dynamic data is subjected to topological abstraction processing to generate a preliminary signal propagation topology;

[0038] The topology data of the wind turbine is acquired, and the spatial topology representation of the preliminary signal propagation topology is optimized based on the topology data to obtain the optimized signal propagation topology.

[0039] Preferably, the component data analysis module evaluates the synergistic coupling effect between the signal propagation topology and the physical performance characterization, including:

[0040] Extract the signal propagation features corresponding to the signal propagation topology, and perform dimensionality reduction processing on the signal propagation features to obtain dimensionality-reduced signal propagation features;

[0041] Calculate the feature similarity index between the reduced-dimensional signal propagation features, and calculate the representation similarity index between the physical performance representations;

[0042] Calculate the correlation factor between the signal propagation topology and the physical performance characterization;

[0043] By combining the correlation factor, the feature similarity index, and the representation similarity index, the degree of synergistic coupling between the signal propagation topology and the physical performance representation is calculated.

[0044] Based on the aforementioned cooperative coupling degree, the cooperative coupling effect between the signal propagation topology and the physical performance characterization is evaluated.

[0045] Preferably, the calculation of the correlation factor between the signal propagation topology and the physical performance characterization includes:

[0046] The signal propagation topology and the physical performance representation are vectorized respectively to obtain the propagation topology vector and the performance representation vector.

[0047] Calculate the vector cosine between the propagation topology vector and the performance characterization vector;

[0048] Calculate the vector mutual information between the propagation topology vector and the performance representation vector;

[0049] By combining the vector cosine value and the vector mutual information, the correlation factor between the signal propagation topology and the physical performance characterization is calculated using a weight adjustment coefficient.

[0050] Preferably, the digital twin model construction module analyzes the signal detection characteristics of the digital twin model prototype based on the sensing accuracy data, including:

[0051] The sensing accuracy data is cleaned to obtain cleaned sensing accuracy data.

[0052] Analyze and extract the time-domain and frequency-domain features of the cleaning sensing accuracy data;

[0053] Based on the precision time-domain features and the precision frequency-domain features, a signal characteristic descriptor corresponding to the digital twin model prototype is generated;

[0054] Based on the signal characteristic descriptor, the signal detection characteristics of the digital twin model prototype are analyzed.

[0055] Preferably, the compatibility evaluation module calculates the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, evaluates the performance stability of the digital twin model prototype based on the lifetime prediction equivalent value, determines the installation location requirements and working condition constraints of the wind turbine in the maintenance system, collects the compatibility parameters of the digital twin model prototype, and analyzes the compatibility of the digital twin model prototype in the maintenance system by combining the installation location requirements, the working condition constraints and the compatibility parameters.

[0056] The compatibility evaluation module calculates the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, including:

[0057] Call the abnormal behavior response tag group to obtain the time data of the wind turbine, including the fault start time, fault development time and fault recovery time, calculate the time interval, and generate fault key time interval data;

[0058] Based on the data from the critical time intervals of the fault, analyze the fault density distribution information;

[0059] Based on the fault density distribution information, calculate the fault density offset feature value, identify the location interval of the fault density abnormal segment on the time axis, and generate the fault density offset time period.

[0060] Based on the fault density offset time period, the correlation between fault development and time segment is evaluated, continuous segments with offset fault development behavior are screened, and fault behavior delay segments are generated.

[0061] Based on the fault behavior delay segment, the lifetime prediction equivalent value of the digital twin model prototype is calculated.

[0062] Preferably, the compatibility evaluation module analyzes the compatibility of the digital twin model prototype in the maintenance system by combining the installation location requirements, the working condition constraints, and the compatibility parameters, including:

[0063] Based on the fault behavior delay segment, the maintenance time data of the wind turbine is extracted to obtain maintenance lag information;

[0064] Determine whether the maintenance lag value exceeds the maintenance completion baseline threshold, filter the component numbers that have not been maintained, extract the waiting time of the corresponding components, sort them according to the waiting time of the incomplete components, and generate a component maintenance waiting priority sequence;

[0065] Call the sorting information in the component maintenance waiting priority sequence, configure additional maintenance time windows for the components in sequence, adjust the maintenance time length within the total maintenance time range, record the component number and the corresponding adjusted maintenance time, and generate a maintenance time configuration table;

[0066] Based on the maintenance time configuration table, combined with the installation location requirements, the working condition constraints, and the adaptability parameters, the compatibility of the digital twin model prototype in the maintenance system is analyzed.

[0067] Preferably, the maintenance decision generation module generates maintenance decisions for the wind turbine, including:

[0068] Call the maintenance time configuration table, detect real-time wind turbine operation data, and mark component maintenance time points;

[0069] Determine whether the maintenance time point meets the maintenance conditions; if so, generate a maintenance start command.

[0070] Based on the maintenance initiation command, a maintenance decision for the wind turbine is generated.

[0071] Compared with the prior art, the beneficial effects of the present invention are:

[0072] The operation data acquisition module focuses on acquiring the operating condition parameters and dynamic performance parameters of decommissioned wind turbines. It can accurately capture the performance changes of turbine components due to aging and wear during the decommissioning stage. Based on this, the core maintenance parameters are developed, which closely match the actual maintenance needs of decommissioned turbines. This avoids the problem of maintenance parameters being out of touch with actual needs due to one-sided data acquisition in traditional technologies, and makes the maintenance work clear and targeted from the beginning.

[0073] The component data analysis module, by scheduling component intrinsic data and structural design data, deeply analyzes the physical performance characteristics corresponding to the component intrinsic data. This allows for a clear understanding of the role, performance status, and potential failure risks of different components in the overall operation of the decommissioned unit, thereby determining a scientifically sound priority sequence for core maintenance parameters. This process enables the rational allocation of maintenance resources according to the importance and urgency of components, avoiding resource waste or neglect of critical components in traditional maintenance, and improving the efficiency and effectiveness of maintenance work.

[0074] The digital twin model building module fully considers the actual conditions of decommissioned wind turbines during the model construction process, incorporating both core and alternative component materials into the simulation scope. This results in a digital twin prototype that more realistically reflects the structural characteristics and material conditions of the decommissioned turbine. Furthermore, by collecting sensing accuracy data and durability test data from the prototype model, and analyzing signal detection characteristics based on the sensing accuracy data, staff can accurately understand the model's perception capability and accuracy regarding the turbine's operating status in advance. This ensures the model provides a reliable simulation basis for subsequent maintenance decisions, solving the problem of significant discrepancies between simulation results and reality caused by traditional digital twin models that did not consider the characteristics of decommissioned turbines.

[0075] The compatibility evaluation module calculates the lifetime prediction equivalent value of the digital twin model prototype based on durability test data. Simultaneously, it analyzes the model's compatibility with the maintenance system, enabling early assessment of whether the model matches existing maintenance processes and equipment, as well as its stability and applicability during long-term use. This step effectively avoids situations where the model cannot be used due to incompatibility with the maintenance system after completion, ensuring that the digital twin model can be smoothly integrated into the maintenance decision-making system and fully leverage its role in condition simulation, fault prediction, and other areas.

[0076] The maintenance decision generation module comprehensively considers factors such as signal detection characteristics, performance stability, and compatibility, conducting a thorough comparison and screening of core component materials and alternative component materials to select the best maintenance materials for the decommissioned wind turbines. Based on this, the maintenance decisions generated by combining core maintenance parameters and parameter priority sequences balance the targeted nature of maintenance, the applicability of materials, and the reliability of the model. This overcomes the limitations of traditional maintenance decisions, which rely on human experience and rely on single parameters, making maintenance decisions more scientific and rational. It can effectively solve the problems existing in decommissioned wind turbines, reduce maintenance costs, and ensure the safe and stable operation of the units in the future. Attached Figure Description

[0077] Figure 1 This is a timing diagram of the predictive maintenance decision system for decommissioned wind turbines driven by digital twins as described in this invention.

[0078] Figure 2 A flowchart illustrating the working principle of the core maintenance parameters of the wind turbine was developed for the operational data acquisition module.

[0079] Figure 3 A flowchart illustrating the working principle of determining the priority sequence of core maintenance parameters for the component data analysis module. Detailed Implementation

[0080] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0081] Please see Figure 1 This invention provides a digital twin-driven predictive maintenance decision-making system for decommissioned wind turbines. The system includes: an operation data acquisition module, a component data analysis module, a digital twin model construction module, an adaptation and compatibility evaluation module, and a maintenance decision generation module.

[0082] The operation data acquisition module collects unit operating condition parameters and dynamic performance parameters, and formulates core maintenance parameters accordingly. The component data analysis module schedules component intrinsic data and structural design data, analyzes physical performance characteristics, and determines the priority sequence of core maintenance parameters. The digital twin model construction module queries core component materials and alternative component materials, generates a digital twin model prototype through simulation, and analyzes its signal detection characteristics based on sensing accuracy data. The compatibility evaluation module calculates the life prediction equivalent value based on durability test data and evaluates the model's compatibility with the maintenance system. The maintenance decision generation module comprehensively considers signal detection characteristics, performance stability, and compatibility to select the best maintenance materials, and generates the final maintenance decision by combining the core maintenance parameters and their priority sequence.

[0083] Example 1: See Figure 2 The module collects operating parameters and dynamic performance parameters of decommissioned wind turbines. Operating parameters include real-time monitoring data such as rotor speed, gearbox oil temperature, generator winding temperature, and nacelle vibration amplitude. Dynamic performance parameters cover performance indicators such as power output curves, energy conversion efficiency, and load response characteristics. This data is continuously collected and transmitted through a sensor network installed in various parts of the turbine. Based on the acquired operating parameters and dynamic performance parameters, the module further analyzes the wind turbine's operating environment factors. These factors include external factors such as atmospheric temperature change cycles, relative humidity distribution, wind speed fluctuation frequency, and salt spray corrosion concentration. These factors are obtained through data fusion analysis of environmental monitoring stations and the turbine's built-in sensors.

[0084] Based on a clear understanding of the working environment factors, the module conducts an in-depth analysis of the application scenario characteristics of wind turbines. These characteristics include specific climate patterns resulting from geographical location, such as the high salinity and high humidity environment of coastal areas, or the low temperature and strong wind conditions of plateau regions. It also considers the material aging and performance degradation trends caused by years of operation. These characteristics are extracted through long-term data accumulation and pattern recognition technology. The module simultaneously collects existing fault information of the wind turbines. This information comes from fault diagnosis reports in historical maintenance records, abnormal alarm logs from the real-time monitoring system, and lists of potential defects discovered during regular inspections. Common fault types include blade surface corrosion, increased bearing clearance, decreased insulation resistance, and control system signal drift.

[0085] Based on the characteristics of the application scenario and existing fault information, the module initiates the process of formulating core maintenance parameters. This process first extracts features from the application scenario to obtain scenario feature factors. Principal component analysis is used to reduce the dimensionality of multidimensional environmental data to identify key influencing factors such as corrosion acceleration factors, fatigue load cycle counts, and thermal stress fluctuation coefficients. Existing fault information is then classified to obtain a set of fault categories. Based on the fault mechanism and impact degree, hierarchical clustering algorithms are used to divide faults into major categories such as electrical system faults, mechanical structure faults, and control system faults. Each major category is further subdivided into specific fault modes such as cable joint oxidation, gear tooth surface pitting, and sensor accuracy failure.

[0086] Scene feature factors are encoded to obtain scene feature encoding vectors. One-hot encoding is used to convert discrete feature factors into binary vectors, while normalization is combined to scale continuous feature factors to a standard numerical range for unified data representation. Fault category sets are encoded to obtain fault category encoding vectors. The encoding process incorporates a fault severity weight coefficient, calculated based on downtime and repair costs caused by the fault, ensuring the encoded vector reflects not only the fault type but also its criticality. Cluster analysis is performed on the scene feature encoding vectors to obtain scene feature cluster centers. The K-means algorithm aggregates feature vectors under similar environmental conditions into clusters, with the center point of each cluster representing a typical environmental scenario, such as a high-temperature, high-humidity cluster or a low-temperature, high-load cluster. Fault category encoding vectors are clustered to obtain fault category cluster centers. The clustering process iteratively calculates based on fault frequency and repair urgency, ultimately forming clusters centered on high-frequency and high-risk faults.

[0087] The correlation mapping matrix is ​​obtained by analyzing the relationship between the cluster centers of scene features and the cluster centers of fault categories. Canonical correlation analysis is used to calculate the statistical correlation between environmental scenes and fault types, and the matrix element values ​​reflect the probability intensity of specific faults occurring under specific environmental conditions. Based on the correlation mapping matrix, key scene features and key fault categories of wind turbines are determined. Key scene features are concentrated on environmental factors significantly related to high-frequency faults, such as the combination of salt spray concentration and vibration intensity. Key fault categories are selected from fault modes most affected by environmental factors, such as bearing corrosion and insulation breakdown. Design constraints for wind turbines are analyzed based on key scene features and key fault categories. These constraints include limiting factors such as material selection meeting salt spray corrosion resistance requirements, structural design needing to withstand specific vibration spectrum loads, and electrical systems requiring humidity protection capabilities. Core maintenance parameters for wind turbines are formulated based on these design constraints. These core maintenance parameters are ultimately determined to be quantifiable monitoring indicators such as vibration acceleration threshold, lower limit of insulation resistance, lubricating oil acidity index, and blade surface roughness level. These parameters will serve as the direct basis for predictive maintenance decisions.

[0088] Example 2: See Figure 3 The module processes intrinsic component data, which originates from the original attribute records of each component of the wind turbine, including multi-dimensional information such as material composition certificates, mechanical performance test reports, and service history parameters. Standardization is achieved using a combination of the Z-score algorithm and range standardization. The Z-score algorithm transforms the data into a distribution with a mean of zero and a standard deviation of one, while range standardization linearly maps the data to the range of zero to one. This dual processing ensures the comparability of parameters with different dimensions and magnitudes; for example, it unifies the GPa unit of Young's modulus and the ppm / ℃ unit of the coefficient of thermal expansion to dimensionless values. When extracting the component physical properties corresponding to the standardized component data, the module calls a physical property database for matching and mapping. These physical properties cover intrinsic characteristics such as material density, yield strength, fatigue limit, thermal conductivity, and resistivity, which are determined through both materials science theoretical definitions and experimental measurement data.

[0089] Key physical attributes are obtained by screening the physical properties of components. The screening process is based on the correlation analysis between attributes and typical failure modes of wind turbines. Pearson correlation coefficient is used to calculate the correlation strength between each attribute and historical failure records, such as the correlation between vibration failure and material damping coefficient, and the correlation between overheating failure and thermal conductivity. Attributes with correlation coefficients exceeding a threshold are retained as key physical attributes. When calculating the physical performance indicators corresponding to the key physical attributes, the module defines the calculation rules according to the operating characteristics of the wind turbine. For example, the fatigue life index is calculated by stress-life curve and load spectrum, and the thermal conductivity coefficient takes into account the thermal conductivity of the material and the geometry of the component. These indicators are generated by the fusion calculation of physical models and empirical formulas. Based on the physical performance indicators, physical performance characterizations corresponding to the intrinsic data of the components are generated. The physical performance characterizations are constructed in the form of multi-dimensional vectors, with each dimension representing the value of a performance indicator. The similarity between vectors is measured by Euclidean distance for subsequent collaborative coupling analysis.

[0090] The signal propagation topology of the wind turbine is determined based on the structural design data, which includes structured information such as CAD drawings, finite element mesh models, and assembly relationship tables. Pre-processing is performed on the structural design data to obtain the target structural design data. This pre-processing includes median filtering to eliminate measurement errors for data denoising and K-nearest neighbor algorithm to fill in incomplete fields for missing values, ensuring data continuity and consistency. The structural parameter set of the wind turbine is extracted from the target structural design data. This set covers geometric dimensions such as blade length, shaft diameter, number of gear teeth, and bolt distribution coordinates, as well as the type and constraints of connection nodes. These parameters are automatically identified from the 3D model using feature extraction algorithms.

[0091] Material property association processing is performed on the structural parameter set to obtain an attribute-associated parameter set. This association process binds the geometric parameters of each component to its material properties by querying a material database; for example, it associates the module of a gear with the elastic modulus of alloy steel, and the dimensions of a bearing with the hardness of a ceramic ball, forming a parameter set containing both geometric and material attributes. Based on this attribute-associated parameter set, a numerical model of signal propagation for the wind turbine is constructed. The numerical model uses the finite element method to establish multiphysics coupling equations, simulating the propagation path of mechanical vibration waves in the structure, the diffusion process of heat flow between components, and the transmission characteristics of electromagnetic signals in the circuit. Adaptive mesh refinement technology is used for model mesh generation to ensure computational accuracy in key areas.

[0092] The signal propagation numerical model is simulated to obtain dynamic signal propagation data. The simulation includes transient analysis simulating signal response under impact loads, harmonic response analysis calculating steady-state propagation under periodic loads, and random vibration analysis evaluating signal characteristics under uncertain loads. Output data includes displacement contour maps, stress distribution maps, and temperature gradient maps for each node. Topological abstraction is performed on the dynamic signal propagation data to generate a preliminary signal propagation topology. Topological abstraction uses graph theory algorithms to abstract components as nodes and signal transmission paths as edges. Node weights represent component importance, and edge weights represent signal transmission efficiency, forming a weighted directed graph structure. The topological structure data of the wind turbine is acquired, including the hierarchical structure in the assembly relationship diagram and the causal chains in the signal flow diagram. This data is extracted from the control system logic diagram and sensor layout diagram. Based on the topological structure data, the preliminary signal propagation topology is optimized using spatial topological representation to obtain an optimized signal propagation topology. The optimization process uses the Delaunay triangulation algorithm to reconstruct node connections to ensure network connectivity and robustness. Virtual nodes are introduced to handle signal jumps across components, ultimately generating a topological network model that accurately reflects the signal transmission patterns within the wind turbine.

[0093] Example 3: Key features are extracted from the constructed signal propagation topology. These features include indicators describing the topology, such as node degree distribution, path propagation efficiency, signal attenuation coefficient, and network clustering coefficient. These features are calculated using graph theory algorithms. Dimensionality reduction is performed on the signal propagation features to obtain dimensionality-reduced features. Principal component analysis is used to retain the principal components with the highest contribution rates, thus reducing data dimensionality while preserving most of the original information. The dimensionality-reduced feature vectors are easier to calculate for subsequent similarity. The feature similarity index between dimensionality-reduced signal propagation features is calculated. This index measures the directional consistency between different feature vectors using an improved cosine similarity algorithm. The algorithm introduces weighting factors to adjust the contribution of different feature dimensions, giving key features higher discriminative weights. The representation similarity index between physical performance representations is calculated. Physical performance representations are multi-dimensional vectors, and their similarity is quantified using the reciprocal of the standardized Euclidean distance. Normalization of different physical dimensions is considered during distance calculation to ensure fairness in distance measurement. The correlation factor between signal propagation topology and physical performance representation is calculated. This step first vectorizes the signal propagation topology and physical performance representation. The vectorization process uses graph embedding technology to map the topology into a low-dimensional vector. At the same time, the physical performance representation is converted into a fixed-dimensional vector expression through eigenvalue decomposition. Finally, the propagation topology vector and performance representation vector are obtained.

[0094] The cosine value between the propagation topology vector and the performance representation vector is calculated. The cosine value is calculated by dividing the dot product by the magnitude product, reflecting the similarity of the two vectors in direction. The mutual information between the propagation topology vector and the performance representation vector is also calculated. This mutual information is based on the KL divergence between the joint probability distribution and the marginal probability distributions of the two vectors, measuring the statistical dependence between them. Combining the vector cosine value and the mutual information, a correlation factor is calculated using a weighted adjustment coefficient. This weighted adjustment coefficient is dynamically allocated based on feature importance, which is determined using a random forest feature importance evaluation algorithm. The formula for calculating the correlation factor is:

[0095]

[0096] in: Represents correlation factors. Represents the cosine value of a vector. Represents vector mutual information. and The weighting coefficients for the cosine value and mutual information are respectively, satisfying The constraints are as follows. The weight coefficients are optimized and determined using a grid search cross-validation method to ensure that the correlation factors can reflect directional similarity and statistical dependence in a balanced way.

[0097] The synergistic coupling degree between signal propagation topology and physical performance representation is calculated by combining correlation factors, feature similarity index, and characterization similarity index. The calculation process uses a weighted arithmetic mean formula to integrate the three indicators, and the weight allocation is dynamically adjusted based on the correlation strength between each indicator and system performance. The synergistic coupling degree is used to evaluate the synergistic coupling effect between signal propagation topology and physical performance representation. The synergistic coupling degree value range is mapped to the range of zero to one; the closer the value is to one, the stronger the synergy between the topology and physical performance. This system provides an objective basis for subsequent maintenance decisions through quantitative evaluation.

[0098] Example 4: When processing sensing accuracy data, the digital twin model construction module first collects raw signals from the sensor network installed at key parts of the wind turbine. These sensors include vibration accelerometers, temperature sensors, strain gauges, and current transformers. The collected data includes timestamps, measured values, sensor IDs, and status codes. The sensing accuracy data is then cleaned to obtain cleaned sensing accuracy data. The cleaning process uses a sliding window anomaly detection algorithm to identify and remove outliers that significantly exceed the physical range, such as data points where the rotational speed exceeds the rated value by 300% or the temperature jumps by more than 100°C instantaneously. Simultaneously, a Kalman filter algorithm is applied to smooth the data to eliminate random noise interference.

[0099] The precision time-domain and precision frequency-domain features corresponding to the cleaning induction precision data are analyzed and extracted. Time-domain feature calculations include statistics such as mean, variance, peak-to-peak value, kurtosis index, and waveform factor. These features reflect the amplitude distribution and fluctuation characteristics of the signal over time. Frequency-domain feature extraction uses Fast Fourier Transform to convert the signal to the frequency domain, calculating indices such as power spectral density, centroid frequency, frequency variance, and harmonic distortion. These features reveal the energy distribution pattern of the signal over frequency. Based on the precision time-domain and precision frequency-domain features, signal characteristic descriptors corresponding to the digital twin model prototype are generated. Descriptor construction employs feature fusion technology to combine time-domain and frequency-domain features into a high-dimensional feature vector, and normalization is performed on each dimension to eliminate dimensional differences, ultimately forming a digital description that comprehensively characterizes the signal properties. The signal detection characteristics of the digital twin model prototype are analyzed based on signal characteristic descriptors. The analysis process uses a pattern recognition algorithm to match and compare the current signal characteristics with a standard signal template. The evaluation indicators include sensitivity resolution, which reflects the system's ability to detect weak signals; linearity error, which measures the degree of deviation between input and output; dynamic range, which characterizes the maximum span of detectable signal amplitude; and repeatability error, which assesses the consistency of multiple measurement results. These characteristics together constitute the signal detection capability map of the digital twin model.

[0100] The compatibility evaluation module calculates the life prediction equivalent value of the digital twin model prototype based on durability test data. It calls the abnormal behavior response tag group to define the severity level and impact range of the fault. The tag group includes fields such as fault code, affected components, and urgency level. Time data of the wind turbine is acquired, including fault start time, fault development time, and fault recovery time. These timestamps are extracted from the historical event logs of the unit monitoring system. Time intervals are calculated to generate critical fault time interval data, reflecting the entire lifecycle of the fault from occurrence to repair. Based on the critical fault time interval data, fault density distribution information is analyzed. The kernel density estimation method is used to construct the probability density function of fault occurrence on the time axis. The density function curve shows the concentration of faults within a specific time period. Fault density offset feature values ​​are calculated based on the fault density distribution information. These offset feature values ​​are obtained by comparing the difference between the actual density distribution and the baseline distribution. The location intervals of abnormal fault density segments on the time axis are identified, generating fault density offset time periods. These time periods correspond to the dangerous periods of abnormal fault concentration.

[0101] The correlation between fault development and time intervals is assessed based on the fault density offset time period. Grey relational analysis is used to calculate the correlation coefficients between different time intervals and fault development. Continuous segments exhibiting offsets in fault development behavior are selected to generate fault behavior delay segments, which characterize the time lag between fault occurrence and actual performance. The lifetime prediction equivalent value of the digital twin model prototype is calculated based on the fault behavior delay segments. The calculation process uses a Weibull distribution model to fit the fault interval time data, and the probability distribution of remaining lifetime is estimated through shape and scale parameters. The median of the distribution is used as the prediction baseline value for the equivalent value. Table 1 shows the fault time interval data.

[0102] Table 1: Failure Time Intervals and Density Distribution Data for Wind Turbine Units

[0103] Fault Number Fault start time Fault development time Fault recovery time Time interval (hours) density value F2023001 2023-01-1208:30 2023-01-1214:45 2023-01-1309:20 24.83 0.087 F2023002 2023-02-0516:20 2023-02-0608:10 2023-02-0711:40 43.33 0.124 F2023003 2023-03-1810:15 2023-03-1909:30 2023-03-2014:00 51.75 0.156 F2023004 2023-04-2213:40 2023-04-2307:55 2023-04-2408:30 42.83 0.118 F2023005 2023-05-3009:00 2023-05-3110:20 2023-06-0116:45 55.75 0.192 F2023006 2023-07-1214:50 2023-07-1308:30 2023-07-1410:15 43.42 0.126 F2023007 2023-08-0811:30 2023-08-0909:45 2023-08-1012:20 48.83 0.142 F2023008 2023-09-1908:20 2023-09-2010:10 2023-09-2115:30 55.17 0.185 F2023009 2023-10-2515:40 2023-10-2608:50 2023-10-2709:30 41.83 0.115 F2023010 2023-11-1410:10 2023-11-1509:30 2023-11-1614:45 52.58 0.162

[0104] By fusing multi-source data and conducting time-series analysis, a complete evaluation chain from signal detection to lifetime prediction is established, providing a data foundation for assessing the suitability of digital twin models in maintenance systems. Fault density values ​​are calculated using a Gaussian kernel function for density estimation, with the bandwidth parameter adaptively determined based on the standard deviation of the time interval data. The density value reflects the degree of fault clustering along the time axis; a higher density value indicates that the fault occurrence frequency is significantly higher than the average level during that period. The calculation of lifetime prediction equivalent values ​​comprehensively considers multiple factors such as fault occurrence frequency, duration, and repair effectiveness, forming a quantitative assessment of the durability performance of the digital twin model.

[0105] Example 5: When processing the system adaptability of the digital twin model prototype, the compatibility evaluation module first extracts the maintenance time data of the wind turbine based on the fault behavior delay segment. The maintenance time data comes from the unit maintenance log database, which includes the planned start time, actual start time, completion time, and delay reason records of historical maintenance work orders. Maintenance lag information is calculated from these timestamp data. The maintenance lag information is specifically represented as a sequence of differences between the planned maintenance time and the actual execution time, as well as the distribution of average delay time for different component categories. It is then determined whether the maintenance lag value exceeds the maintenance completion benchmark threshold. The maintenance completion benchmark threshold is set according to the wind farm operation and maintenance specifications, which typically requires that the delay of non-critical components not exceed 48 hours and the delay of critical components not exceed 24 hours. The system automatically compares the actual delay value of each component with the threshold, filters out the component numbers for which maintenance has not been completed, and extracts the waiting time for the corresponding component. The waiting time is calculated by subtracting the planned completion time from the current system time.

[0106] The system generates a component maintenance waiting priority sequence by sorting components according to their waiting time. The sorting algorithm employs a multi-factor weighted evaluation, considering not only the absolute value of the waiting time but also a component criticality level coefficient and a failure risk index. The criticality level coefficient is based on the impact of component failure on the entire machine, while the failure risk index is calculated using historical failure frequency and severity. This results in a priority list sorted in descending order of maintenance urgency. The system then calls upon the sorting information from the component maintenance waiting priority sequence to allocate additional maintenance time windows to the components sequentially. The configuration process uses a dynamic programming algorithm to find the optimal allocation scheme within the total maintenance time, ensuring that high-priority components receive sufficient time resources while minimizing the impact on low-priority components. The maintenance time length is adjusted within the total maintenance time, with adjustment strategies including extending the duration of a single maintenance session, increasing maintenance frequency, or scheduling emergency maintenance windows. The system records the component number and the corresponding adjusted maintenance time to generate a maintenance time configuration table. This table includes fields such as component ID, original planned time, newly allocated time window, reason for adjustment, and expected completion time.

[0107] Based on the maintenance time configuration table, and considering installation location requirements, operating condition constraints, and adaptability parameters, this study analyzes the compatibility of the digital twin model prototype within the maintenance system. Installation location requirements include physical installation space size limitations, interface type matching requirements, and safety distance regulations for adjacent components; these requirements are extracted from the unit design drawings and installation specifications. Operating condition constraints involve operational boundary conditions such as ambient temperature range, maximum permissible vibration amplitude, and electromagnetic compatibility indicators; these constraints are obtained from the unit technical manual. Adaptability parameters include digital interface parameters such as the communication protocol version between the digital twin model and the physical system, the degree of data sampling rate matching, and the signal accuracy error range. The compatibility analysis employs a multi-objective optimization algorithm to calculate a weighted adaptability index. The algorithm uses installation matching degree, operating condition compliance, and parameter consistency as three optimization objectives, and obtains the optimal compatibility score through Pareto front solving. The score is mapped to a range of 0-100; a higher score indicates better integration and adaptability of the digital twin model within the maintenance system.

[0108] After the maintenance decision generation module starts, it calls the maintenance time configuration table and monitors real-time wind turbine operating data. This data is collected in real-time through the SCADA system and includes key monitoring parameters such as turbine power output, bearing temperature, gearbox oil pressure, and vibration spectrum. The system continuously compares the current data with the deviation from the normal operating range, identifying component maintenance time points. The time point determination is based not only on the planned schedule in the maintenance time configuration table but also on real-time equipment status assessment results. When monitored parameters exceed warning thresholds, the maintenance time point is automatically moved forward. The system then determines whether the maintenance time point meets the maintenance conditions, which include factors such as the equipment being in a shutdown state, ambient wind speed below safe operating limits, availability of on-site maintenance resources, and personnel qualifications. The system uses an IoT platform to acquire real-time wind farm meteorological data, maintenance team location information, and tool and equipment status for multi-dimensional condition compliance verification.

[0109] If the maintenance conditions are met, a maintenance initiation command is generated. The command generation process combines a rule engine and machine learning. The rule engine generates a basic command framework based on predefined safety procedures and operating processes, while the machine learning model optimizes the command details based on historical successful maintenance cases and real-time environmental variables. The final maintenance decision includes specific maintenance operation steps, a list of required tools, safety precautions, and expected completion time. The maintenance decision is automatically issued to the mobile terminals of on-site maintenance personnel in the form of work orders and is simultaneously updated to the digital twin platform of the wind farm management center, enabling full-process visual monitoring and dynamic adjustment of the maintenance process.

[0110] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A digital twin-driven predictive maintenance decision-making system for decommissioned wind turbines, characterized in that, The system includes: The operation data acquisition module is used to acquire the operating condition parameters and dynamic performance parameters of decommissioned wind turbines and to formulate the core maintenance parameters of wind turbines. The component data analysis module is used to schedule the intrinsic data of wind turbine components and structural design data, analyze the physical performance characteristics corresponding to the intrinsic data of the components, and determine the parameter priority sequence of the core maintenance parameters. The digital twin model construction module is used to query the core component materials and alternative component materials of the wind turbine, combine the core component materials and alternative component materials to simulate and build the wind turbine to obtain a digital twin model prototype, collect the sensing accuracy data and durability test data of the digital twin model prototype, and analyze the signal detection characteristics of the digital twin model prototype based on the sensing accuracy data. The compatibility evaluation module is used to calculate the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, and analyze the compatibility of the digital twin model prototype in the maintenance system. The maintenance decision generation module is used to combine the signal detection characteristics, the performance stability, and the compatibility to select the best maintenance material from the core component materials and the alternative component materials, and generate a maintenance decision for the wind turbine based on the core maintenance parameters, the parameter priority sequence, and the best maintenance material.

2. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 1, characterized in that, The operation data acquisition module determines the working environment elements of the wind turbine based on the operating condition parameters and the dynamic performance parameters. Based on the working environment elements, it analyzes the application scenario characteristics of the wind turbine, collects the existing fault information of the wind turbine, and formulates the core maintenance parameters of the wind turbine by combining the application scenario characteristics and the existing fault information. The operational data acquisition module combines the application scenario characteristics and existing fault information to formulate the core maintenance parameters of the wind turbine, including: Feature extraction is performed on the application scenario features to obtain scenario feature factors; The existing fault information is classified to obtain a set of fault categories; The scene feature factors are encoded to obtain a scene feature encoding vector; The fault category set is encoded to obtain a fault category encoding vector; Cluster analysis is performed on the scene feature encoding vectors to obtain scene feature cluster centers; Cluster analysis is performed on the fault category encoding vectors to obtain the fault category cluster centers; The correlation between the cluster centers of the scene features and the cluster centers of the fault categories is analyzed to obtain the correlation mapping matrix; Based on the correlation mapping matrix, the key scenario characteristics and key fault categories of the wind turbine are determined; Based on the key scenario characteristics and key fault categories, the design constraints of the wind turbine are analyzed. Based on the aforementioned design constraints, the core maintenance parameters for the wind turbine are determined.

3. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 1, characterized in that, The component data analysis module analyzes the physical performance characterization corresponding to the intrinsic data of the component, determines the signal propagation topology of the wind turbine based on the structural design data, evaluates the synergistic coupling effect between the signal propagation topology and the physical performance characterization, and determines the parameter priority sequence of the core maintenance parameters based on the synergistic coupling effect. The component data analysis module analyzes the physical performance characterization corresponding to the intrinsic data of the component, including: The intrinsic data of the component are standardized to obtain standardized component data; Extract the physical attributes of the components corresponding to the standardized component data, and filter the physical attributes to obtain key physical attributes; Calculate the physical performance index corresponding to the key physical attributes, and generate the physical performance characterization corresponding to the intrinsic data of the component based on the physical performance index.

4. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 3, characterized in that, The component data analysis module determines the signal propagation topology of the wind turbine based on the structural design data, including: The structural design data is preprocessed to obtain the target structural design data; Extract the set of structural parameters of the wind turbine from the target structural design data; The structural parameter set is subjected to material property association processing to obtain a property association parameter set; Based on the attribute association parameter set, construct the signal propagation numerical model corresponding to the wind turbine; The signal propagation numerical model is simulated and processed to obtain dynamic data of signal propagation. The signal propagation dynamic data is subjected to topological abstraction processing to generate a preliminary signal propagation topology; The topology data of the wind turbine is acquired, and the spatial topology representation of the preliminary signal propagation topology is optimized based on the topology data to obtain the optimized signal propagation topology.

5. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 4, characterized in that, The component data analysis module evaluates the synergistic coupling effect between the signal propagation topology and the physical performance characterization, including: Extract the signal propagation features corresponding to the signal propagation topology, and perform dimensionality reduction processing on the signal propagation features to obtain dimensionality-reduced signal propagation features; Calculate the feature similarity index between the reduced-dimensional signal propagation features, and calculate the representation similarity index between the physical performance representations; Calculate the correlation factor between the signal propagation topology and the physical performance characterization; By combining the correlation factor, the feature similarity index, and the representation similarity index, the degree of synergistic coupling between the signal propagation topology and the physical performance representation is calculated. Based on the aforementioned cooperative coupling degree, the cooperative coupling effect between the signal propagation topology and the physical performance characterization is evaluated.

6. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 5, characterized in that, The calculation of the correlation factor between the signal propagation topology and the physical performance characterization includes: The signal propagation topology and the physical performance representation are vectorized respectively to obtain the propagation topology vector and the performance representation vector. Calculate the vector cosine between the propagation topology vector and the performance characterization vector; Calculate the vector mutual information between the propagation topology vector and the performance representation vector; By combining the vector cosine value and the vector mutual information, the correlation factor between the signal propagation topology and the physical performance characterization is calculated using a weight adjustment coefficient.

7. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 1, characterized in that, The digital twin model construction module analyzes the signal detection characteristics of the digital twin model prototype based on the sensing accuracy data, including: The sensing accuracy data is cleaned to obtain cleaned sensing accuracy data. Analyze and extract the time-domain and frequency-domain features of the cleaning sensing accuracy data; Based on the precision time-domain features and the precision frequency-domain features, a signal characteristic descriptor corresponding to the digital twin model prototype is generated; Based on the signal characteristic descriptor, the signal detection characteristics of the digital twin model prototype are analyzed.

8. The predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 1, characterized in that, The compatibility evaluation module calculates the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, evaluates the performance stability of the digital twin model prototype based on the lifetime prediction equivalent value, determines the installation location requirements and working condition constraints of the wind turbine in the maintenance system, collects the compatibility parameters of the digital twin model prototype, and analyzes the compatibility of the digital twin model prototype in the maintenance system by combining the installation location requirements, the working condition constraints and the compatibility parameters. The compatibility evaluation module calculates the lifetime prediction equivalent value of the digital twin model prototype based on the durability test data, including: Call the abnormal behavior response tag group to obtain the time data of the wind turbine, including the fault start time, fault development time and fault recovery time, calculate the time interval, and generate fault key time interval data; Based on the data from the critical time intervals of the fault, analyze the fault density distribution information; Based on the fault density distribution information, calculate the fault density offset feature value, identify the location interval of the fault density abnormal segment on the time axis, and generate the fault density offset time period. Based on the fault density offset time period, the correlation between fault development and time segment is evaluated, continuous segments with offset fault development behavior are screened, and fault behavior delay segments are generated. Based on the fault behavior delay segment, the lifetime prediction equivalent value of the digital twin model prototype is calculated.

9. A predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 8, characterized in that, The compatibility evaluation module analyzes the compatibility of the digital twin prototype in the maintenance system by combining the installation location requirements, the working condition constraints, and the compatibility parameters, including: Based on the fault behavior delay segment, the maintenance time data of the wind turbine is extracted to obtain maintenance lag information; Determine whether the maintenance lag value exceeds the maintenance completion baseline threshold, filter the component numbers that have not been maintained, extract the waiting time of the corresponding components, sort them according to the waiting time of the incomplete components, and generate a component maintenance waiting priority sequence; Call the sorting information in the component maintenance waiting priority sequence, configure additional maintenance time windows for the components in sequence, adjust the maintenance time length within the total maintenance time range, record the component number and the corresponding adjusted maintenance time, and generate a maintenance time configuration table; Based on the maintenance time configuration table, combined with the installation location requirements, the working condition constraints, and the adaptability parameters, the compatibility of the digital twin model prototype in the maintenance system is analyzed.

10. A predictive maintenance decision-making system for decommissioned wind turbines driven by digital twins as described in claim 1, characterized in that, The maintenance decision generation module generates maintenance decisions for the wind turbine, including: Call the maintenance time configuration table, detect real-time wind turbine operation data, and mark component maintenance time points; Determine whether the maintenance time point meets the maintenance conditions; if so, generate a maintenance start command. Based on the maintenance initiation command, a maintenance decision for the wind turbine is generated.

Images