System and method for making intelligent recovery decisions using machine learning

Abstract

An apparatus may receive historical operational data of a mechanical system, such as a rotor blade of a wind turbine. The apparatus may determine one or more quality levels for each of one or more materials of a system (e.g., a rotor blade). One or more quality levels may be determined by processing historical operational data using a data model. The data model may be trained using machine learning based on one or both of historical operational data of similar systems (e.g., other rotor blades) and end of life (EOL) test data thereof. The device may determine a recovery recommendation based on the one or more quality levels. The recovery recommendations may include instructions related to recovering one or more materials. The device may deliver the recovery recommendation to another device or recipient.

Description

TECHNICAL FIELD

[0001] The present invention relates generally to methods of using machine learning to make intelligent recycling decisions when recycling composite materials, and more particularly to using machine learning to make intelligent recycling decisions when recycling materials of wind turbines, such as wind turbine blades. BACKGROUND

[0002] Recycling is the act or process of converting recyclable materials into reusable materials. Generally, the recycling process involves manipulating or destroying the recyclable materials (e.g., through a melting process, etc.) such that the resulting materials can be used to create new products.

[0003] A wind turbine is a device that converts the kinetic energy of wind into electrical energy. For example, a wind turbine can convert the kinetic energy of wind into electrical power by using air forces generated by the interaction of wind with rotor blades to turn a generator.

[0004] Efforts to recycle components or materials of wind turbines have largely been non-existent or ineffective. One reason is that many wind turbine components are made using composite materials, in which multiple materials are bonded together using an epoxy resin. Much of the value of recycling the composite materials is the actual removal or disposal of the “waste” materials, which would otherwise be placed in a landfill or need to be recycled in another (and even more expensive) way (e.g., in situations where landfills are not permitted). Without an efficient and / or effective way to separate these materials, many components and / or materials of wind turbines have been placed or buried in landfills without being recycled. However, new methods of recycling components and / or materials of wind turbines have been discovered that involve separating each material of a composite material by exposing the composite material to a swelling agent (e.g., formic acid, etc.) for a certain period of time. This causes the epoxy resin to break down, allowing each material bonded together using the epoxy resin to be separated.

[0005] While this breakthrough provides a way to begin recycling materials of wind turbines, there is currently no efficient and / or effective method of processing to recycle these materials. For example, recyclable materials used in rotor blades of wind turbines can be used to create a variety of different new products. However, the new products can have a material specification that requires the materials to have a certain quality level. For example, if the new product is in any way associated with a safety risk or hazard that is amplified when the materials of the new product have a lower quality level, the material specification of the new product can require the materials to have a higher quality level.

[0006] The residual strength of a material can be measured. The residual strength is the load or force that an object or material can bear or support (e.g., before reaching a breaking point, a failure point, etc.). Factors that contribute to the residual strength of a material include the toughness of the material, the presence (or absence) of a fracture, the size, shape, and / or orientation of the fracture, etc.

[0007] The quality of a recyclable material can vary for many different reasons. For example, during the entire life of a product, different events can occur to reduce the residual strength of the material of the product. In particular, the residual strength of a material can naturally degrade over time, and further degradation can be caused by events such as when the material is exposed to harsh weather conditions, subjected to blunt physical impacts, etc.

[0008] Further, different portions of the same material can have different residual strengths. For example, assume that a component of a product is made using a certain material, but only a portion of the component is damaged. In this case, a first portion of the material corresponding to the damaged component can have a low residual strength, while other portions of the material can have a higher residual strength.

[0009] Due to the variations in the residual strength of a material (and variations in different portions of the same material), some recyclable materials can be limited to new product uses that are typically reserved for lower quality grades of recyclable materials. For example, in the absence of an efficient and / or effective way to demonstrate that a recyclable material has a certain quality (e.g., has a certain residual strength), the material, once recycled, can be limited to uses that are reserved for lower quality grades of materials, even if the material or a portion of the material has a higher quality grade.

[0010] It is an object of the present invention to alleviate or overcome some or all of the above-described problems. SUMMARY

[0011] In one aspect of the present invention, a method of making a recycling recommendation by classifying a material of a rotor blade of a wind turbine is disclosed. The method includes receiving, by a computing device, historical operational data of a rotor blade of a wind turbine. The method further includes determining, by the computing device, one or more quality grades for each of one or more materials of the rotor blade. The one or more quality grades are determined by processing the historical operational data using a data model. The data model is trained using machine learning based on one or both of historical operational data of other rotor blades and end-of-life (EOL) test data of other rotor blades. The method further includes determining, by the computing device, a recycling recommendation based on the one or more quality grades. The recycling recommendation includes instructions related to recycling the one or more materials of the rotor blade. The method further includes delivering, by the device, the recycling recommendation to another device or a recipient.

[0012] In an embodiment of the application, determining one or more quality levels can include determining a quality level of the material based on an expected reusability score (score). For example, historical operational data can be provided as input to the data model to cause the data model to output an expected reusability score for the material.

[0013] In another embodiment of the application, the expected reusability score can be based on at least one of an expected residual strength of the material, an expected structural integrity of the material, or an expected end-of-life (EOL) fiber length of fibers associated with the particular material.

[0014] In another embodiment of the application, the EOL test data can include a reusability score for a material of the other rotor blade. The corresponding reusability score indicates a degree to which the material or a portion of the material is reusable.

[0015] In another embodiment of the application, the historical part data can be used to train the data model, the historical part data including data identifying boundaries of components of the other rotor blade and data identifying boundaries of materials of the corresponding components. The data model can be trained to associate each corresponding boundary and / or a subset of boundaries with a reusability score.

[0016] In another embodiment of the application, the historical operational data of the rotor blade can include at least one of installation data, service data, or weather data. The historical operational data of the other rotor blade can include daily operational data and at least one of installation data, service data, or weather data. In this way, certain types of daily operational data that can be used to train the data model can not be needed to accurately determine a quality level of a material of the rotor blade.

[0017] In another embodiment of the application, determining one or more quality levels can include determining a first quality level of a first material of a component of the rotor blade and determining a second quality level of a second material of the component of the rotor blade, where the first quality level is different than the second quality level. In another embodiment of the application, determining one or more quality levels can include determining a first quality level of a first portion of a material of the rotor blade and determining a second quality level of a second portion of the same material of the rotor blade, where the first quality level is different than the second quality level.

[0018] In another embodiment of the invention, determining the recycling recommendation can include determining a set of cutting instructions that instruct cutting the rotor blade in a manner that separates material or respective subsets of material having different quality grades. In yet another embodiment of the invention, determining the recycling recommendation can include determining a set of bin placement instructions that instruct sorting material or subsets of material into different recycling receptacles such that each respective recycling receptacle corresponds to a particular quality grade.

[0019] In another embodiment of the invention, determining the recycling recommendation can include determining a set of cutting instructions that instruct cutting the rotor blade such that the cutting separates a first portion of material and a second portion of material. In another embodiment of the invention, determining the recycling recommendation can include determining a set of bin placement instructions that instruct placing the first portion of material and the second portion of material into different recycling receptacles.

[0020] In another aspect of the invention, an apparatus that performs any of the above- described methods is disclosed. In another aspect of the invention, a computer readable medium that performs any of the above-described methods is disclosed.

[0021] The above summary presents a simplified overview of some embodiments of the application to provide a basic understanding of certain aspects discussed in the present document. This summary is not intended to provide an extensive overview of the application nor is it intended to identify any key or critical elements of the application or to delineate the scope of the application. The sole purpose of the summary is to present some concepts in a simplified form as a prelude to the detailed description presented below. BRIEF DESCRIPTION OF DRAWINGS

[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate various embodiments of the application and together with the general description of the application given above, and the detailed description of the embodiments given below, serve to explain the embodiments of the present application.

[0023] Figure 1 A wind turbine is shown in which the present application can be advantageously used.

[0024] Figure 2 A cross-sectional view of a rotor blade of a wind turbine is shown.

[0025] Figure 3A is a diagram of a recycling management platform that receives historical data to be used to train a data model.

[0026] Figure 3B is a diagram of a recycling management platform determining features to be used to train a data model.

[0027] Figure 3Cis a diagram of a recycling management platform training data model.

[0028] Figure 4A is a diagram of a recycling management platform obtaining historical operational data for a rotor blade based on a request from a user device.

[0029] Figure 4B is a diagram of a recycling management platform using a data model to determine one or more quality grades for a respective material of a rotor blade.

[0030] Figure 4C is a diagram of a recycling management platform determining a recycling recommendation based on the one or more quality grades and providing the recycling recommendation to a user device.

[0031] Figure 5 is a diagram of an example environment in which the systems and / or methods described herein can be implemented.

[0032] Figure 6 is a diagram of example components of one or more devices of Figure 5

[0033] It should be appreciated that the Figures are not necessarily drawn to scale and that certain features of the Figures can have been exaggerated or simplified in the interest of clarity and understanding. Specific design features of the operational sequences disclosed herein, including, for example, the particular sizes, orientations, locations, and shapes of various illustrated components, can be determined in part by the particular intended application and use environment. Certain features of the illustrated embodiments can have been enlarged or distorted relative to others to facilitate visualization and clear understanding. DETAILED DESCRIPTION

[0034] The following detailed description of example embodiments refers to the accompanying drawings, which are depicted by way of illustration and not by way of limitation. These embodiments are described in sufficient detail to enable those skilled in the art to practice the application, and the words herein are used to describe the best information available. Other embodiments can be implemented and structural, logical, and electrical changes can be made without departing from the scope of the present application as defined in the appended claims.

[0035] Figure 1 ​A wind turbine 100 in which the products of the present application can be advantageously used is shown. The wind turbine 100 comprises a tower 102, a nacelle 104 having a rotor 106 mounted thereon, and a yaw bearing 108. The yaw bearing 108 rotatably connects the tower 102 to the nacelle 104. The rotor 106 comprises a rotor hub to which three rotor blades (shown as rotor blades 110, 112, and 114, respectively) are attached in this wind turbine 100. The yaw bearing 108 is configured such that the nacelle 104, together with the rotor 106 mounted thereon, is rotatable relative to the tower 102. In this way, the nacelle 104 can be arranged such that the rotor 106 is oriented towards the wind. When oriented towards the wind, the wind turbine 100 is operable to generate more electrical power.

[0036] Although one rotor 106 is illustrated in Figure 1 , it will be appreciated that multiple rotors can be carried by the tower 102. For example, a dual rotor configuration can be seen in US Publication No. 2022 / 0025866, and a configuration with two pairs of rotors can be seen in WO-A1-2018 / 157897. Accordingly, aspects of the present application are applicable to multiple rotor wind turbines or other types of wind turbines, and should not be limited to those shown in Figure 1 .

[0037] Figure 2 is a cross-sectional view of a rotor blade 200 of a wind turbine. In some embodiments, the rotor blade 200 can correspond to rotor blade 110, rotor blade 112, and / or rotor blade 114. The rotor blade 200 has a plurality of spar caps 212, 214, 216, and 218. The rotor blade 200 has an outer shell 202 made of two half shells 204, 206. The shells 204, 206 are molded from glass fiber reinforced plastic (GRP). Portions of the outer shell 202 are a sandwich panel construction and include a blade core of lightweight form (e.g., polyurethane) sandwiched between inner and outer GRP layers or “skins”.

[0038] The rotor blade 200 includes a first pair of spar caps and a second pair of spar caps 212, 214, 216, and 218 disposed between the sandwich panel regions of the outer shell 208. One spar cap of each pair is integrated with the windward shell, and the other spar cap of each pair is integrated with the leeward shell. The respective pairs of spar caps 212, 214, 216, and 218 oppose one another and extend longitudinally along the length of the rotor blade 200. A first longitudinally extending shear web 220 bridges the first pair of spar caps 212, 214, and a second longitudinally extending shear web 222 bridges the second pair of spar caps 216, 218. The shear webs 220, 222, in combination with the spar caps 212, 214, 216, 218, form a pair of I-beam structures that efficiently transfer loads from the rotor blade 200 to the hub (not shown) of the wind turbine. The spar caps 212, 214, 216, 218, in particular, transfer tensile and compressive bending loads, while the shear webs 220, 222 transfer shear stresses in the rotor blade 200.

[0039] In some embodiments, the spar caps (e.g., spar caps 212, 214, 216, 218) can be constructed using a stack of pultruded strips that are bonded together. The pultruded strips are bonded together by an adhesive, such as a resin (e.g., an epoxy resin). The pultruded strips are also arranged in the stack with their respective bonded surfaces in opposing relationship to one another. The adhesive can be applied to the bonded surfaces of the pultruded strips directly, or via another technique, such as a resin infusion process. In the infusion process, liquid resin is supplied to the stack, and the resin infuses between the opposing bonded surfaces of the pultruded strips. The wind turbine blade 200 described above is exemplary, and wind turbine blades have a wide range of designs and constructions. Accordingly, it should be understood that aspects of the present disclosure are applicable to wind turbine arrangements other than those described above.

[0040] Figures 3A-3C is a diagram of an example method 300 for training a data model using machine learning to determine a quality grade of a material of a rotor blade of a wind turbine. The quality grade of the material can then be used to make a recycling recommendation.

[0041] While one or more embodiments described herein relate to training a data model to determine a quality grade of a material of a rotor blade, it should be understood that this is provided by way of example. In practice, the example method 300 can be used to train a data model to determine a quality grade of a material used within other components of a wind turbine. Additionally or alternatively, the example method 300 can be used to train a data model to determine a quality grade of a material of other equipment / machines, such as a material of a printed circuit board (PCB), a material of a vehicle (e.g., a boat, a train, a car, etc.), a composite material, etc.

[0042] As Figure 3AAs shown, the recycling management platform 302 can receive historical data to be used to train a data model. For example, as shown by reference number 304, the recycling management platform 302 can receive historical part data, such as data pertaining to a rotor blade belonging to a set of wind turbines, from one or more data storage devices. The historical part data can be received via a communication interface, such as via an application programming interface (API) or similar type of interface, over a network (e.g., the Internet, etc.). The historical part data for a rotor blade can include component data providing information about respective components of the rotor blade and / or material data providing information about respective materials used within the rotor blade.

[0043] The component data can include information related to one or more spar caps, one or more shear webs, a shell of the rotor blade, components corresponding to a leading edge of the rotor blade, components corresponding to a trailing edge of the rotor blade, etc. At a value level, the component data can include a component identifier identifying the component, model data indicating a model of the component, type data identifying a type of the component, a rotor blade identifier (e.g., identifying the rotor blade to which the component is attached), data describing one or more features of the component, etc.

[0044] Additionally or alternatively, the component data can include measurement data identifying (or usable to identify) a boundary of the component and / or a boundary of a zone or region within the component. For example, the measurement data can include data identifying a length of the component, a height of the component, a width of the component, an area and / or a surface area of the component, etc. In some embodiments, the recycling management platform 302 can process the measurement data to generate a three-dimensional (3D) part map defining a boundary of the component of the rotor blade using coordinates.

[0045] The material data can provide information related to respective materials used within the rotor blade, such as a core element (e.g., foam, wood, etc.), a fiber layer material (e.g., synthetic fiber, semi-synthetic fiber, regenerated fiber, carbon fiber, basalt fiber, glass fiber, metal fiber, etc.), a resin matrix (e.g., made using epoxy resin), a composite material (e.g., including structural units and fibers disposed in an epoxy resin matrix), a material used as part of a lightning protection mesh, a coating, a root insert, a sensor, etc. The material data for a material can also include a material identifier, type data identifying a type of the material, a component identifier and / or a rotor blade identifier (e.g., identifying a component and / or a rotor blade in which the material is used), data describing one or more features of the material, etc.

[0046] Additionally or alternatively, the material data can include measurement data that identifies (or can be used to identify) boundaries of respective materials within the rotor blade (and / or within a given component of the rotor blade). For example, the measurement data can include data that identifies a length of the material, a height of the material, a width of the material, an area of the material, and / or a surface area, etc. In some embodiments, the recycling management platform 302 can process the measurement data to add the material to the 3D part map (e.g., such as by having coordinates define material boundaries of the material as used within the rotor blade).

[0047] Additionally or alternatively, the material data can include manufacturing data that provides information about conditions under which the respective material was manufactured. For example, the manufacturing data for a material can include data related to a particular manufacturing action. To provide a particular example, when a material is cured, the manufacturing data that is collected can include temperature data that indicates a temperature at which the material was cured, time data that indicates a duration of time for which the material was cured, pressure data that indicates a pressure to which the material was exposed during the curing process, data that indicates a difference between a measured property of the material during the curing process and a baseline property (e.g., which can be a standard / norm for curing that type of material), etc. Similar types of data can be collected for other actions or processes associated with manufacturing. To the extent these actions or processes associated with manufacturing are applicable to a particular component of a rotor blade, similar types of data can be collected for the respective component of the rotor blade.

[0048] As shown by reference number 306, the recycling management platform 302 can receive historical operational data for the rotor blade. For example, the recycling management platform 302 can receive the historical operational data for the rotor blade from one or more data storage devices. The historical operational data can be received via a communication interface (such as via an API or similar type of interface) over a network (e.g., the Internet).

[0049] The historical operational data for the respective rotor blade can include installation data that provides information related to installation of the respective rotor blade, routine operation data that provides information related to one or more routine operations performed over the life of the respective rotor blade, service data that provides information related to services (e.g., inspections, maintenance, repairs, etc.) performed for the respective rotor blade, weather data that provides information related to weather forecasts made in a geographic location corresponding to the respective rotor blade, and / or weather data measured on-site at the operational site, etc.

[0050] Installation data for the rotor blade can include time data indicating a time and / or date of installation of the rotor blade, event data indicating one or more events that occurred during the installation process, and / or the like. To provide a particular example, if the rotor blade is dropped or damaged during the installation process, event data can be created that includes an event identifier, time data indicating a time and / or date of the event, data describing the event, and / or the like. The data describing the event can include, for example, data identifying one or more components or materials damaged or affected by the event, data identifying a zone or area within the rotor blade where the damage occurred, and / or the like.

[0051] Routine operation data for the rotor blade can include energy generation data, operational time data, measured sensor data, and / or the like. The energy generation data can include data identifying a measured amount of energy generated by the wind turbine (to which the rotor blade is attached) over a given time period and / or data identifying an amount or degree of contribution of the rotor blade to that energy generation. The operational time data can include data identifying a duration of operation of the wind turbine (and / or rotor blade), measured sensor data including data identifying one or more sensor measurements collected over a given time period (e.g., a period of operation of the rotor blade). For example, the measured sensor data can include rotational data identifying a number of rotations of the rotor blade over a given time period, velocity data identifying a velocity of the rotor blade over a given time period, vibration data identifying a measurable degree of vibration of the rotor blade over a given time period, thrust data identifying a thrust applied to the rotor blade over a given time period, and / or the like.

[0052] Service data for the rotor blade can include inspection data providing information related to an inspection of the rotor blade, maintenance / repair data providing information related to any maintenance and / or repairs performed on the rotor blade, and / or the like. The inspection data can include time data indicating a time and / or date of the inspection, type data indicating a type of the inspection, result data describing results of the inspection, and / or the like. The maintenance / repair data can include time data indicating a time and / or date that maintenance and / or repairs were performed on the rotor blade, type data indicating a type of the maintenance and / or repairs performed, part data indicating a component, a zone or area within a component, a material, and / or a portion of a material that was subject to the maintenance and / or repairs, result data describing results following the maintenance and / or repairs that were performed, and / or the like.

[0053] In some embodiments, the inspection data can include non-destructive testing (NDT) data. NDT is used to test various components and / or materials of a wind turbine, such as components and / or materials of a rotor blade. The NDT data can provide information related to one or more NDTs, such as strain measurements, acoustic emission, ultrasonic vibration, thermal imaging, machine vision, etc. For example, the NDT data can include type data indicating a type of NDT, time data indicating a time and / or date that the NDT was performed, part data indicating a component, material, and / or location within a component and / or material that was targeted for the NDT, description data describing the NDT that was performed, result data indicating one or more results of the test, etc.

[0054] The maintenance / repair data can include time data indicating a date and / or time that maintenance and / or repair of the rotor blade was performed, type data indicating a type of maintenance and / or repair that was performed, part data indicating a component, material, and / or location within a component and / or material that was targeted for the maintenance and / or repair, description data describing the maintenance and / or repair, etc. To provide a few examples, maintenance / repair data can be collected for leading edge corrosion repair (e.g., a coating, tape, or shroud can be used as an additional layer to protect the leading edge), injection repair and viscous flow (e.g., filling and sealing by injecting a resin or suitable material), patch / crimp repair (e.g., to remove a damaged area, leaving a straight, stepped, or tapered hole, and applying a patch to the hole), cure and adhesive development type repair, replacement of failed parts, etc. Additional types of repair can include repair from damage caused during transport, cosmetic damage, coating repair, and / or complex blade repair, such as blade split tip, structural repair, lightning strike repair, internal repair, laminate repair, tip shell replacement, etc.

[0055] The weather data can include data related to weather forecasts or weather data measured at a geographic location of the respective wind turbine. For example, the weather data can include weather data for a geographic location where the wind turbine is installed. The weather data can include data indicating a temperature, data indicating a wind speed, data indicating a wind shear, data indicating wind turbulence, data indicating an amount of rain or snow, data indicating whether lightning is present, etc. The weather data can be collected at periodic intervals. In some embodiments, the weather data can include data identifying one or more average values related to weather forecasts, such as an average wind speed, an average temperature, etc. over a time interval.

[0056] As shown by reference number 308, the recycling management platform 302 can receive end-of-life (EOL) test data for a rotor blade. For example, the recycling management platform 302 can receive the EOL test data from one or more data storage devices. The EOL test data can be received via a communication interface, such as via an API or similar type of interface, over a network (e.g., the Internet).

[0057] EOL test data can include information related to one or more performance tests performed when a respective rotor blade reaches the end of its life span. For example, the EOL test data can provide information related to a quality measurement value indicative of a quality of a respective component and / or material of the rotor blade. To provide a particular example, the EOL test data can include a reusability score of a respective component and / or material of the rotor blade. The reusability score can be based on, for example, a residual value of the material, a structural integrity value of the material, and / or a measured length of fibers of the material.

[0058] In some embodiments (not shown), the recycling management platform 302 can perform one or more pre-processing operations to standardize the historical data into a uniform data type, data format, etc. prior to training the data model using the historical data. For example, the historical data can be received in different file types and / or formats, and the recycling management platform 302 can apply appropriate standardization techniques to the different data types or data formats such that the historical data is converted into a uniform data type, data format, etc.

[0059] As shown by reference number 308, the recycling management platform 302 can receive historical data. The historical data can include historical part data, historical operation data, and / or EOL test data. The historical part data can include information related to one or more components of one or more wind turbines. For example, the historical part data can include information related to a component type, a material type, a measured value of the component, a measured value of the material, a zone or region within the rotor blade where the component is located, a zone or region within the rotor blade where the material is located, etc. The historical operation data can include information related to one or more wind turbines. For example, the historical operation data can include information related to installation data, daily operation data, service data, and / or weather data. The EOL test data can include information related to one or more performance tests performed when a respective rotor blade reaches the end of its life span. For example, the EOL test data can provide information related to a quality measurement value indicative of a quality of a respective component and / or material of the rotor blade. To provide a particular example, the EOL test data can include a reusability score of a respective component and / or material of the rotor blade. The reusability score can be based on, for example, a residual value of the material, a structural integrity value of the material, and / or a measured length of fibers of the material. Figure 3B As shown by reference number 310, the recycling management platform 302 can select features for training the data model. For example, the recycling management platform 302 can identify a feature set including features that can be used to train the data model, and can select a subset of the identified features.

[0060] A feature can be a measurable property or characteristic that can be used to train a data model using machine learning. A feature set can include features of historical data values, features of aggregated historical data values, features of combinations of historical data values, features of benchmarks of one or more of these values, features of relationships between two or more of these values, etc. Particular example features are provided below.

[0061] Features of the historical part data can include, for example, a feature identifying a component type, a feature identifying a material type, a feature identifying a measured value of the component, a feature identifying a measured value of the material, a feature identifying a zone or region of the rotor blade where the component is located, a feature identifying a zone or region within the rotor blade where the material is located, etc.

[0062] Similar features can be determined for each of the other types of historical data. For example, the feature set can include features related to historical operation data, which can include features related to installation data, daily operation data, service data, and / or weather data. Additionally, the feature set can include features related to EOL test data, which can include features related to quality measurement values taken at the end of the life span of a respective wind turbine.

[0063] As described above, the set of features can include features that aggregate historical data values, features of combinations of historical data values, features of benchmarks, and / or features of relationships between two or more of these values. For example, features that aggregate data values can include features related to totals, averages, means, modes, etc. Features of different combinations of historical data values can include features of all (or some) combinations of two or more different historical data values. Features of benchmarks can include benchmark values of different types of historical data. For example, a benchmark value of EOL test data can be a value indicative of a (preferred) quality of a particular material and / or component. To provide another example, a benchmark value of daily operations data can include a value indicative of a benchmark performance level of a rotor blade and / or components thereof.

[0064] Features of relationships include values indicative of a relationship or association between two or more other values (i.e., two or more historical data values, combinations of values, aggregations of values, etc.), values indicative of trends found within the data, etc. For example, assume that a subset of rotor blades each have a material that shares the same EOL quality rating. If historical operations data for each of these rotor blades includes installation data indicative of the same material being damaged during the installation process, the recycling management platform 302 can identify as a feature the existence of a relationship between this particular installation data value and the quality rating of the material. In other words, this relationship can be described as a particular installation data value indicative of the quality rating of the material, affecting the quality rating of the material in some way, etc. Those skilled in the art will appreciate that a variety of different relationships between historical data values and quality ratings can be identified.

[0065] In some embodiments, the recycling management platform 302 can identify the set of features by processing the historical data using one or more feature identification techniques. The one or more feature identification techniques can include text mining and latent semantic analysis (LSA) techniques, trend variable analysis techniques, interest diversity analysis techniques, neural network techniques, composite indicator analysis techniques, clustering analysis techniques, etc.

[0066] In some embodiments, the recycling management platform 302 can select a subset of the identified features by processing the set of features and / or the historical data using one or more feature selection techniques. The one or more feature selection techniques can include one or more filtering techniques, one or more wrapper techniques, one or more embedded techniques, etc. The one or more filtering techniques can be used to remove redundant features that are duplicates. These techniques can include chi-squared tests, correlation coefficients, variance thresholds, etc. The one or more wrapper techniques can involve iterative methods for feature selection (and / or reduction) and can include forward selection techniques, backward elimination techniques, bidirectional elimination techniques, exhaustive selection techniques, recursive elimination techniques, etc. The one or more embedded techniques can include regularization techniques, tree-based selection techniques, etc.

[0067] Additionally or alternatively, the recycling management platform 302 can receive features for training the data model. For example, a feature set can be determined by a domain expert and provided to (or made accessible to) the recycling management platform 302.

[0068] In this way, the recycling management platform 302 determines features for training the data model.

[0069] As shown by reference number 310, the recycling management platform 302 can determine features for training the data model. For example, the recycling management platform 302 can determine features based on historical data and / or based on a domain expert’s knowledge of the material’s expected reusability. In some embodiments, the recycling management platform 302 can determine features based on a domain expert’s knowledge of the material’s expected reusability. For example, the recycling management platform 302 can receive a domain expert’s knowledge of the material’s expected reusability and determine features based on the domain expert’s knowledge. In some embodiments, the recycling management platform 302 can determine features based on historical data. For example, the recycling management platform 302 can determine features based on historical data of a material’s expected reusability. Figure 3C As shown by reference number 312, the recycling management platform 302 can train the data model to determine a quality grade of the material of the rotor blade based on the material’s expected reusability. For example, the recycling management platform 302 can train the data model by using one or more machine learning techniques to analyze the historical data and / or the determined features. The one or more machine learning techniques can include classification driven training techniques, logistic regression based training techniques, Naive Bayes classifier techniques, support vector machine (SVM) techniques, neural networks, and / or the like.

[0070] In some embodiments, the recycling management platform 302 can train the data model using an iterative approach. For example, the recycling management platform 302 can use the data model to process historical training data of a particular rotor blade to cause the data model to determine an expected reusability score of the material of the rotor blade. The historical training data can exclude known outputs such that the data model effectively predicts a reusability score of the material.

[0071] Next, the expected reusability score can be used to determine a quality grade of the respective material. In some embodiments, as a final processing step, the data model can associate the respective reusability score with a corresponding quality grade. In some embodiments, this step can be performed externally to the data model. For example, the recycling management platform 302 can reference a data structure that associates expected reusability scores with particular quality grades of the material such that an expected reusability score output by the data model can be mapped to a corresponding quality grade.

[0072] Continuing the example, the quality grade of the material (e.g., which corresponds to the determined expected reusability score) can be compared to an actual quality grade of the respective material. For example, assume that the recycling management platform 302 determines an incorrect quality grade. In this case, the recycling management platform 302 can update one or more weight values and / or threshold values of the data model such that the data model is reconfigured to make more accurate quality determinations when making subsequent determinations of the expected reusability of the material of the rotor blade.

[0073] In this way, the recycling management platform 302 trains the data model to determine a quality grade of the material based on the material’s expected reusability.

[0074] As mentioned above, Figures 3A-3C This is provided as an example only. Other examples may differ from the reference. Figures 3A-3C The described example. For example, with Figures 3A-3C Compared to what is shown, there may be additional equipment and / or networks, fewer equipment and / or networks, different equipment and / or networks, or equipment and / or networks with different arrangements. Furthermore, Figures 3A-3C The two or more devices shown can be implemented within a single device, or Figures 3A-3C The single device shown can be implemented as multiple and / or distributed devices. Additionally or alternatively, a group of devices (e.g., one or more devices) of Example 100 can perform one or more functions described as being performed by another group of devices of Example Method 300.

[0075] Figures 4A-4C This is a diagram of an example method 400 for using machine learning to determine the quality grade of materials for wind turbine rotor blades and for making recycling recommendations based on the quality grade. While one or more embodiments described herein relate to determining the quality grade of rotor blade materials, it should be understood that this is provided as an example. In practice, example method 400 can be used to determine the quality grade of materials for other components of a wind turbine. Additionally or alternatively, example method 400 can be used to determine the quality grade of materials for other equipment / machines, such as materials for printed circuit boards (PCBs), materials for transportation vehicles (e.g., boats, trains, automobiles, etc.), composite materials, etc.

[0076] like Figure 4A As shown, example method 400 involves data communication between user equipment 402, recycling management platform 404, and data storage device 406. As indicated by reference numeral 408, a user, such as a recycling administrator, can input a request for recycling recommendations for the rotor blades of a wind turbine to be decommissioned. For example, the recycling administrator can interact with the user interface of user equipment 402 to input the request for recycling recommendations. In some embodiments, the user interface may be part of an application and / or API that allows users to submit recycling requests to recycling management platform 404. When the recycling administrator submits a request, request data can be provided to recycling management platform 404. The request data may include a wind turbine identifier for the wind turbine and / or rotor blade identifiers for the rotor blades.

[0077] As indicated by reference numeral 410 in the attached diagram, the recycling management platform 404 can obtain historical operational data of the rotor blades of a wind turbine to be decommissioned. For example, the recycling management platform 404 can, based on received request data, provide a request for historical operations of the rotor blades to the data storage device 406. The data storage device 406 can use data structures to store historical operational data associated with wind turbine identifiers and / or rotor blade identifiers. In this way, the data storage device 406 can use the received identifiers to identify the historical operational data and provide it to the recycling management platform 404.

[0078] Historical operating data for rotor blades can include installation data, daily operating data, service data, weather data, etc. Although combined... Figure 3A Each type of operational data is described in detail; however, it should be noted that not every type of data is necessary to make recycling recommendations. For example, certain types of daily operational data may have been specifically collected for training purposes. These types of daily operational data may require sensors to be fixed or within range of the rotor blades and can be specifically configured to collect additional daily operational data for training data models. However, the recycling management platform 404 can still make recycling recommendations without this additional daily operational data.

[0079] In some embodiments, the recycling management platform 404 may obtain other types of historical data from the data storage device 406. For example, the recycling management platform 404 may obtain historical parts data and / or end-of-life (EOL) test data of the rotor blades. Regarding historical parts data, the recycling management platform 404 may provide a wind turbine identifier to the data storage device 406. The data storage device 406 may store the wind turbine identifier in association with a model identifier that identifies the model of the rotor blade. The data storage device 406 may use the model identifier to obtain historical parts data for that specific model of the rotor blade and may provide the historical parts data to the recycling management platform 404.

[0080] like Figure 4B As shown by reference numeral 412 in the attached figure, the recycling management platform 404 can use a data model to determine one or more quality grades of the corresponding material for the rotor blades. For example, the recycling management platform 404 can provide historical operational data (and / or any other historical data already acquired) as input to the data model. Figures 3A-3C Describe the training of the data model. The data model can use machine learning to process historical operational data and can output one or more expected reusability scores. The expected reusability score of a material can be determined based on at least one of the material's expected residual strength, the material's expected structural integrity, or the expected end-of-life (EOL) fiber length of the fibers associated with the material.

[0081] In some embodiments, the recycling management platform 404 can use the data model to determine an expected reusability score for the material of the rotor blade. For example, by processing the historical operational data using the data model, the recycling management platform 404 can determine that the historical operational data includes data related to the material indicating a particular residual strength, data related to the material indicating a particular structural integrity, and / or data related to the material indicating a particular EOL fiber length. Specifically, the historical operational data can include installation data indicating that a portion of the rotor blade using the material was damaged during an installation process, daily operational data indicating that the rotor blade was used in a wind turbine having a high usage rate (e.g., resulting in additional wear and tear of the material), and service data indicating that certain repairs were made to the material. Each of the respective historical operational data values can indicate that the material has a low residual value and / or a low structural integrity (e.g., relative to an average residual value, an average structural integrity, etc.). Similarly, each of the respective historical operational data values can indicate that the fibers of the material have a low fiber length (e.g., relative to an average fiber length). This can result in the data model outputting a low expected reusability score for the material.

[0082] In some embodiments, the recycling management platform 404 can use the data model to determine expected reusability scores for multiple materials included in the composite material. For example, the composite material used for the rotor blade can include multiple materials, such as fibers, metals, resins, etc. By processing the historical operational data, the recycling management platform 404 can determine an expected reusability score for the fibers, an expected reusability score for the metals, and an expected reusability score for the resins. Each respective expected reusability score can be determined based on historical data values indicating a particular residual strength, a particular structural integrity, and / or a particular EOL fiber length. For example, if each material responds differently to the effects of aging, to impacts from external objects, to exposure to weather, etc., these expected reusability scores can be different.

[0083] In some embodiments, the recycling management platform 404 can use the data model to determine multiple expected reusability scores for the same material. For example, assume that the historical data used to train the data model includes historical part data for a rotor blade. The historical part data includes measurement data identifying boundaries of materials (e.g., within the rotor blade, within a component of the rotor blade, etc.). As such, the recycling management platform 404 can use the data model to determine that certain historical operational data values corresponding to a particular portion (e.g., zone, region, etc.) of the same material indicate a particular residual strength, a particular structural integrity, and / or a particular EOL fiber length.

[0084] In some embodiments, the recycling management platform 404 may determine the quality grade of a material or material portion based on a determined expected reusability score. For example, the recycling management platform 404 may access a data structure that associates a range of expected reusability scores with certain quality grades. Therefore, the recycling management platform 404 may refer to the data structure to identify the corresponding quality grade for each expected reusability score.

[0085] To provide an illustrative example, four quality levels can be used: Quality Level A, Quality Level B, Quality Level C, and Quality Level D. The data structure can correlate the quality level value for each corresponding quality level with a range of expected reusability scores. It should be understood that this is provided as an example, and in practice, different systems for classifying quality levels can be implemented.

[0086] In this way, the recycling management platform 404 determines one or more quality grades of the corresponding material for the rotor blades.

[0087] like Figure 4C As shown by reference numeral 414, the recycling management platform 404 can determine recycling recommendations based on quality levels. Recycling recommendations may include one or more sets of instructions for recycling the material from the rotor blades, such as a set of cutting instructions instructing how to cut the rotor blades, or a set of bin placement instructions instructing the sorting of material or a subset of material into specific recycling containers.

[0088] In some embodiments, the recycling management platform 404 can determine recycling recommendations by referring to a data structure. For example, the recycling management platform 404 can access a data structure that associates data identifying the quality grade of a material with different groups of instructions that can be used as part of a recycling recommendation. The recycling management platform 404 can refer to the data structure to determine which groups of instructions are stored associated with each identified quality grade. Multiple groups of instructions can be combined and used as an overall recycling recommendation.

[0089] In some embodiments, the recycling management platform 404 may determine a set of cutting instructions that instruct the rotor blades to be cut in a manner that separates materials of different quality grades or subsets thereof. For example, if different regions of the spar cap have different quality grades, the set of cutting instructions may instruct the spar cap to be cut in a manner that creates spar cap pieces that are effectively separated based on quality grades. In some embodiments, the set of cutting instructions may instruct where to cut each corresponding material.

[0090] In some embodiments, the recycling management platform 404 may determine a set of bin placement instructions that instruct the sorting of materials or subsets of materials into different recycling containers. This allows materials and / or subsets of materials to be sorted into different recycling containers, such that each recycling container is dedicated to a specific quality grade.

[0091] As shown by reference number 416, the recycling management platform 404 can deliver the recycling recommendation to the user device 402. In some embodiments, the recycling recommendation can be delivered to another recipient, such as an email account or a related account. As shown by reference number 418, the user device 402 can display the recycling recommendation. This can allow the recycling manager to use the recycling recommendation to optimize a process of recycling one or more materials of the recycling rotor blade.

[0092] In some embodiments, the recycling recommendation can be provided to a controller (e.g., a computing device having a processor) that is part of an automated or semi-automated recycling facility. The automated or semi-automated recycling facility can have one or more pieces of equipment that are configured to automatically perform certain recycling tasks. The controller can be configured to send instructions to the pieces of equipment to allow the equipment to autonomously perform the recycling tasks. In this case, the controller can use the recycling recommendation to generate instructions that allow or cause the pieces of equipment to perform the recycling tasks, such as to cut the material in a certain way, place the material in a certain recycling receptacle, etc.

[0093] In some embodiments, the recycling recommendation can be provided to a user device of an operator of the recycling equipment. For example, the recycling recommendation can be provided to and displayed on a user interface of the user device. The user interface can display instructions that the operator can use to perform the recycling tasks, such as the cutting instructions and / or the recycling receptacle instructions described herein.

[0094] As described above, Figures 4A-4C are provided by way of example only. Other examples can vary from those described Figures 4A-4C in the examples described herein. For example, there can be additional devices and / or networks, fewer devices and / or networks, different devices and / or networks, or differently arranged devices and / or networks than those shown in Figures 4A-4C the examples described herein. Additionally or alternatively, a single device described as several devices in the examples described herein can be implemented as a single device, and / or a Figures 4A-4C device described as a single device in the examples described herein can be implemented as several, and / or distributed, devices. Additionally or alternatively, a group of devices described as performing one or more functions in the example methods 400 can perform differing or no functions from those described, and / or a device described as performing an example function can perform differing or additional functions from those described. Figures 4A-4C

[0095] Figure 5 is a diagram of an example environment 500 in which systems and / or methods described herein can be implemented. As shown, environment 500 includes a device 502, which can be a user device 402, a recycling management platform 404, a recycling facility 506, and a network 508, which can include one or more networks. Figure 5 ​As shown, environment 500 can include user device 502, data storage device 504, recycling management platform 506 supported within cloud computing environment 508, and / or network 512. The devices of environment 500 can be interconnected via wired connections, wireless connections, or a combination of wired and wireless connections. User device 502 can correspond to user device 402. Data storage device 504 can correspond to data storage device 406. Recycling management platform 506 can correspond to recycling management platform 302 and / or recycling management platform 404.

[0096] User device 502 includes one or more devices capable of receiving, generating, storing, processing, and / or providing information associated with recycling recommendations. User device 502 can include a device such as a tablet computer (e.g., iPad, etc.), a mobile phone (e.g., a smart phone, a wireless phone, etc.), a laptop computer, a handheld computer, a server computer, a gaming device, a wearable communication device (e.g., a smart watch, a pair of smart glasses, etc.), or a similar type of device.

[0097] In some embodiments, user device 502 can provide a request for a recycling recommendation to recycling management platform 506. In some embodiments, user device 502 can receive a recycling recommendation from recycling management platform 506.

[0098] Data storage device 504 includes one or more devices capable of receiving, storing, processing, and / or providing historical data. Data storage device 504 can include a server device or a group of server devices. In some embodiments, data storage device 504 can support data structures that associate different types of historical data. In some embodiments, data storage device 504 can receive a request (e.g., a query) for historical data from recycling management platform 506. This can cause data storage device 504 to provide the historical data to recycling management platform 506.

[0099] Recycling management platform 506 includes one or more devices capable of receiving, storing, processing, and / or providing information associated with wind turbines and / or recycling. Recycling management platform 506 can include a server device (e.g., a host server, a web server, an application server, etc.), a data center device, or a similar device.

[0100] In some embodiments, recycling management platform 506 can receive a request for a recycling recommendation from user device 502. In some embodiments, recycling management platform 506 obtains historical data associated with one or more rotor blades of a wind turbine from data storage device 504. For example, recycling management platform 506 can provide a request for historical data to data storage device 504. This can cause data storage device 504 to provide the historical data to recycling management platform 506.

[0101] In some embodiments, the recycling management platform 506 stores or has access to a data model that has been trained using machine learning. In some embodiments, the recycling management platform 504 can train the data model using historical data for a set of wind turbines. In some embodiments, the recycling management platform 504 can receive a trained data model from another device. In some embodiments, the recycling management platform 504 can provide a recycling recommendation to the user device 502.

[0102] In some embodiments, as shown, the recycling management platform 506 can be hosted in a cloud computing environment 508. Notably, while the embodiments described herein describe the recycling management platform 506 as being hosted in a cloud computing environment 508, in some embodiments, the recycling management platform 506 can not be cloud-based (i.e., can be implemented outside of a cloud computing environment) or can be partially cloud-based.

[0103] The cloud computing environment 508 includes an environment that hosts the recycling management platform 506. The cloud computing environment 508 can provide computing, software, data access, storage, etc. services that do not require end users to know the physical location and configuration of the systems and / or devices that host the recycling management platform 506. As shown, the cloud computing environment 508 can include a set of computing resources 510 (collectively, “computing resources 510” and individually, “computing resource 510”).

[0104] The computing resources 510 include one or more personal computers, workstation computers, server devices, or another type of computation and / or communication device. In some embodiments, the computing resources 510 can host the recycling management platform 506. The cloud resources can include computation examples executing in the computing resources 510, storage devices provided in the computing resources 510, data transfer devices provided by the computing resources 510, etc. In some embodiments, the computing resources 510 can communicate with other computing resources 510 via wired connections, wireless connections, or a combination of wired and wireless connections.

[0105] As further shown in Figure 5 the computing resources 510 can include a set of cloud resources such as one or more applications (APPs) 510a, one or more virtual machines (VMs) 510b, virtualized storage (VS) 510c, one or more hypervisors (HYPs) 510d, etc.

[0106] Applications 510a can include one or more software applications that can be provided to and / or accessed by user device 502 and / or recycling management platform 506. Applications 510a can obviate the need to install and execute software applications on these devices. In some embodiments, one application 510a can send / receive information to / from one or more other applications 510a via virtual machines 510b. In some embodiments, an application 510a can be a recycling management application. The recycling management application can include one or more user interfaces that, when displayed on user device 502, allow a user to submit a request for a recycling recommendation.

[0107] Virtual machines 510b can include a software implementation of a machine (e.g., a computer) that executes programs like a physical machine. Virtual machines 510b can be system virtual machines or process virtual machines, depending on the extent and corresponding use of any real machine by virtual machines 510b. System virtual machines can provide a complete system platform that supports execution of a complete operating system (“OS”). Process virtual machines can execute a single program and can support a single process. In some embodiments, virtual machines 510b can execute on behalf of another device (e.g., user device 502) and can manage infrastructure of cloud computing environment 508, such as data management, synchronization, or long-duration data transfers.

[0108] Virtualized storage 510c can include one or more storage systems and / or one or more devices that use virtualization technology within a storage system or device of computing resources 510. In some embodiments, within the context of a storage system, types of virtualization can include block virtualization and file virtualization. Block virtualization can refer to abstraction (or separation) of logical storage from physical storage, such that a storage system can be accessed without regard to physical storage or heterogeneous structure. The separation can allow an administrator of the storage system flexibility in how the administrator manages storage for end users. File virtualization can eliminate dependencies between data accessed at a file level and locations of physical storage files. This can enable optimization of performance for storage usage, server consolidation, and / or non-disruptive file migration.

[0109] Hypervisors 510d can provide a hardware virtualization technology that allows multiple operating systems (e.g., “guest operating systems”) to execute concurrently on a host computer, such as computing resources 510. Hypervisors 510d can present a virtual operating platform to guest operating systems and can manage execution of the guest operating systems.

[0110] Network 512 may include one or more wired and / or wireless networks. For example, network 512 may include cellular networks (e.g., fifth-generation (5G) networks, fourth-generation (4G) networks, such as Long Term Evolution (LTE) networks, third-generation (3G) networks and / or Code Division Multiple Access (CDMA) networks), Public Land Mobile Networks (PLMNs), Local Area Networks (LANs), Wide Area Networks (WANs), Metropolitan Area Networks (MANs), Telephone Networks (e.g., Public Switched Telephone Networks (PSTN)), Private Networks, Self-organizing Networks, Intranets, the Internet, Fiber-based Networks, Cloud Computing Networks, etc., and / or combinations of these or other types of networks.

[0111] Figure 5 The number and arrangement of devices and networks shown are provided as an example. In reality, with... Figure 5 Compared to what is shown, there may be additional equipment and / or networks, fewer equipment and / or networks, different equipment and / or networks, or equipment and / or networks with different arrangements. Furthermore, Figure 5 The two or more devices shown can be implemented within a single device, or Figure 5 The single device shown can be implemented as multiple distributed devices. Additionally or alternatively, a group of devices in environment 500 (e.g., one or more devices) can perform one or more functions described as being performed by another group of devices in environment 500.

[0112] Figure 6 This is a diagram of example components of device 600. Device 600 may correspond to user device 502, data storage device 504, and / or recycling management platform 506. In some embodiments, user device 502, data storage device 504, and / or recycling management platform 506 may include one or more devices 600 and / or one or more components of device 600. Figure 6 As shown, device 600 may include bus 602, processor 604, memory 606, storage component 608, input component 610, output component 612 and / or communication interface 614.

[0113] Bus 602 includes a component that permits communication among the multiple components of device 600. Processor 604 is implemented in hardware, firmware, and / or a combination of hardware and software. Processor 604 includes a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), a microprocessor, a microcontroller, a digital signal processor (DSP), a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), and / or another processing component. In some embodiments, processor 604 includes one or more processors capable of being programmed to perform a function. Memory 606 includes a random access memory (RAM), a read only memory (ROM), and / or another type of dynamic or static storage device (e.g., a flash memory, a magnetic storage device, and / or an optical storage device) that stores information and / or instructions for use by processor 604.

[0114] Storage component 608 stores information and / or software related to the operation and use of device 600. For example, storage component 608 can include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, and / or a solid state disk), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, and / or another type of non-transitory computer-readable medium, along with a corresponding drive.

[0115] Input component 610 includes a component that permits device 600 to receive information, such as via user input (e.g., a touch screen display, a keyboard, a keypad, a mouse, a button, a switch, and / or a microphone). Additionally, or alternatively, input component 610 can include a sensor for sensing information (e.g., a global positioning system (GPS) component, an accelerometer, a gyroscope, and / or an actuator). Output component 612 includes a component that provides output information from device 600 (e.g., a display, a speaker, and / or one or more light-emitting diodes (LEDs)).

[0116] Communication interface 614 includes a transceiver-like component (e.g., a transceiver and / or a separate receiver and transmitter) that enables device 600 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of wired and wireless connections. Communication interface 614 can permit device 600 to receive information from another device and / or provide information to another device. For example, communication interface 614 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi interface, a cellular network interface, and / or the like.

[0117] Device 600 can perform one or more processes described herein. Device 600 can perform these processes based on software instructions stored in a non-transitory computer-readable medium (such as memory 606 and / or storage component 608) executed by processor 604. Computer-readable media are defined herein as non-transitory memory devices. Memory devices include memory space within a single physical storage device or memory space distributed across multiple physical storage devices.

[0118] Software instructions may be read into memory 606 and / or storage component 608 via communication interface 614 from another computer-readable medium or from another device. When executed, the software instructions stored in memory 606 and / or storage component 608 may cause processor 604 to perform one or more processes described herein. Additionally or alternatively, hard-wired circuitry may be used in place of or in combination with software instructions to perform one or more processes described herein. Therefore, the embodiments described herein are not limited to any particular combination of hardware circuitry and software.

[0119] Figure 6 The number and arrangement of components shown are provided as an example. In reality, with... Figure 6 Compared to those shown, device 600 may include additional components, fewer components, different components, or components arranged differently. Additionally or alternatively, a set of components of device 600 (e.g., one or more components) may perform one or more functions described as being performed by another set of components of device 600.

[0120] The foregoing disclosure provides illustrations and descriptions, but is not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. Modifications and variations can be made based on the foregoing disclosure, or modifications and variations can be obtained from the practice of the embodiments.

[0121] This document describes several embodiments in conjunction with thresholds. As used herein, depending on the context, satisfying a threshold can mean a value greater than the threshold, more than the threshold, higher than the threshold, greater than or equal to the threshold, less than the threshold, less than the threshold, lower than the threshold, less than or equal to the threshold, equal to the threshold, etc.

[0122] Certain user interfaces have been described herein and / or illustrated in the accompanying drawings. The user interfaces can include graphical user interfaces, non-graphical user interfaces, text-based user interfaces, etc. The user interfaces can provide information for display. In some embodiments, a user can interact with the information, such as by providing input via an input component of a device that provides the user interface for display. In some embodiments, the user interfaces can be configurable by the device and / or user (e.g., a user can change a size of the user interface, information provided via the user interface, a location of information provided via the user interface, etc.). Additionally or alternatively, the user interfaces can be preconfigured to a standard configuration, a particular configuration based on a type of device on which the user interface is displayed, and / or a set of configurations based on capabilities and / or specifications associated with the device on which the user interface is displayed.

[0123] It will be apparent to those skilled in the art that the systems and / or methods described herein can be implemented in different forms of hardware, firmware, and / or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and / or methods is not limiting of the embodiments. Thus, the operation and behavior of the systems and / or methods were described herein without reference to specific software code — it being understood that software and hardware can be used to implement the systems and / or methods based on the description herein.

[0124] No element, act, or instruction used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the articles “a” and “an” are intended to include one or more items, and can be used interchangeably with “one or more.” Furthermore, as used herein, the term “set” is intended to include one or more items (e.g., related items, unrelated items, a combination of related and unrelated items, etc.), and can be used interchangeably with “one or more.” Where only one item is intended, the phrase “only one” or similar language is used. Also, as used herein, the terms “has,” “have,” “having,” or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.

[0125] While all applications have been illustrated by various embodiments, and while these embodiments have been described in considerable detail, it is not the intention of the applicant to restrict or in any way limit the scope of the appended claims to such detail. Additional advantages and modifications will readily appear to those skilled in the art. Therefore, the application in its broader aspects is not limited to the specific details, representative apparatus, and methods, and illustrative examples shown and described. Accordingly, departures can be made from such details without departure from the spirit or scope of applicant’s general inventive concept.

Claims

1. A method of making a recycling recommendation by classifying materials of a rotor blade of a wind turbine, the method comprising: receiving, by a computing device, historical operational data of the rotor blade; determining, by the computing device, one or more quality grades for each of one or more materials of the rotor blade, wherein the one or more quality grades are determined by processing the historical operational data using a data model, and wherein the data model has been trained based on one or both of historical operational data and end-of-life (EOL) test data of other rotor blades using machine learning, determining, by the computing device, the recycling recommendation based on the one or more quality grades, wherein the recycling recommendation comprises instructions related to recycling the one or more materials of the rotor blade; and delivering, from the computing device, the recycling recommendation to another device or recipient.

2. The method of claim 1, wherein, determining a quality grade of the one or more quality grades for a material of the one or more materials comprises: providing the historical operational data as input to the data model to cause the data model to output an expected reusability score for the material, and determining the quality grade for the material based on the expected reusability score.

3. The method of claim 2, wherein, the expected reusability score is based on at least one of: an expected residual strength of the material, an expected structural integrity of the material, or an expected end-of-life (EOL) fiber length of fibers associated with a particular material.

4. The method of claim 2 or 3, wherein, the EOL test data comprises reusability scores for materials of the other rotor blades, and wherein a respective reusability score indicates a degree to which a material or a portion of that material is reusable.

5. The method of claim 4, wherein, training the data model using historical part data, the historical part data comprising data identifying boundaries of components of the other rotor blades and data identifying boundaries of materials of respective components, and wherein the data model is trained to associate each respective boundary and / or subset of boundaries with a reusability score.

6. The method according to any one of the preceding claims, wherein, the historical operational data of the rotor blade comprises at least one of installation data, service data, or weather data, and wherein the historical operational data of the other rotor blades used to train the data model comprises daily operational data and at least one of installation data, service data, or weather data.

7. The method according to any of the preceding claims, wherein, determining the one or more quality grades comprises: determining a first quality grade for a first material of a component of the rotor blade, and determining a second quality grade for a second material of the component of the rotor blade, wherein the first quality grade is different than the second quality grade.

8. The method of any of the preceding claims, wherein, determining the recycling recommendation comprises: determining a set of cutting instructions that instruct cutting the rotor blade such that the cutting separates materials or subsets of respective materials having different quality grades.

9. The method of any of the preceding claims, wherein, determining the recycling recommendation comprises: determining a set of bin placement instructions that instruct sorting materials or subsets of materials into different recycling bins such that each respective recycling bin corresponds to a particular quality grade.

10. The method of any of the preceding claims, wherein, determining the one or more quality levels includes: determining a first quality level of a first portion of a material of the rotor blade, and determining a second quality level of a second portion of the same material of the rotor blade, wherein the first quality level is different than the second quality level.

11. The method of claim 10, wherein, determining the recycling recommendation includes: determining a set of cutting instructions that instruct cutting the rotor blade such that the cutting separates the first portion of the material and the second portion of the material.

12. The method of claim 10 or 11, wherein, determining the recycling recommendation includes: determining a set of bin placement instructions that instruct placing the first portion of the material and the second portion of the material into different recycling receptacles.

13. An apparatus comprising: one or more memories; and one or more processors communicatively coupled to the one or more memories to: receive historical operational data of a rotor blade of a wind turbine; determine one or more quality levels of each of one or more materials of the rotor blade, wherein the one or more quality levels are determined by processing the historical operational data using a data model, and wherein the data model has been trained using machine learning based on one or both of historical operational data of other rotor blades and end-of-life (EOL) test data of the other rotor blades, determine a recycling recommendation based on the one or more quality levels, wherein the recycling recommendation includes instructions related to recycling the one or more materials of the rotor blade; and deliver the recycling recommendation to another apparatus or recipient.

14. A non-transitory computer-readable medium storing instructions, the instructions comprising: one or more instructions that, when executed by one or more processors, cause the one or more processors to: receive historical operational data of a rotor blade of a wind turbine; determine one or more quality levels of each of one or more materials of the rotor blade, wherein the one or more quality levels are determined by processing the historical operational data using a data model, and wherein the data model has been trained using machine learning based on one or both of historical operational data of other rotor blades and end-of-life (EOL) test data of the other rotor blades, determine a recycling recommendation based on the one or more quality levels, wherein the recycling recommendation includes instructions related to recycling the one or more materials of the rotor blade; and deliver the recycling recommendation to another apparatus or recipient.

Images