Power equipment intelligent green recovery method and system based on material attribute characteristics

Abstract

The invention provides an electric power equipment intelligent green recovery method and system based on material attribute characteristics, and the method comprises the steps: carrying out the preprocessing of the attribute characteristics of to-be-recovered electric power equipment, so as to correct the distortion characteristics under the environment interference, and complement the characteristic parameters of the lack of novel materials; distributing a dynamic weight for each attribute index in the attribute index system through a dynamic weight distribution algorithm, and calculating a comprehensive score of the power equipment through a weighted summation calculation mode; determining an initial recommendation recovery scheme according to the comprehensive score; and calculating a green recovery index of the power equipment, evaluating whether the power equipment meets a green recovery requirement through the green recovery index, and adjusting the initial recommended recovery scheme according to an evaluation result. According to the method, accurate classification of high-difficulty recovery materials and adaptive recovery scheme recommendation are realized through multi-stage operation processing of attribute feature preprocessing, feature weighted fusion and green evaluation.

Description

Technical Field

[0001] This application relates to the field of power equipment recycling technology, and in particular to a smart green recycling method and system for power equipment based on material attribute characteristics. Background Technology

[0002] Currently, the power industry recycles retired power equipment to improve resource utilization, and with increasing environmental awareness, the adoption rate of green recycling technologies for power equipment is gradually increasing.

[0003] In related technologies, green recycling solutions for power equipment typically employ intelligent dismantling and automated sorting technologies combined with technologies such as the Internet of Things, artificial intelligence (AI), and blockchain to efficiently dismantle retired equipment, recycle materials, and reuse resources.

[0004] However, in practical applications, the recycling solutions in the aforementioned technologies are not suitable for complex recycling scenarios, mixed materials, and new materials, resulting in poor recycling efficiency and various problems such as environmental protection disadvantages. Summary of the Invention

[0005] This application aims to at least partially address one of the technical problems in the related art.

[0006] Therefore, the first objective of this application is to propose an intelligent green recycling method for power equipment based on the material attribute characteristics. This method achieves accurate classification of highly difficult-to-recycle materials and recommends suitable recycling schemes through multi-level computational processing of attribute characteristics, feature weighted fusion, and green assessment.

[0007] The second objective of this application is to propose an intelligent green recycling system for power equipment based on the material properties.

[0008] The third objective of this application is to propose an electronic device.

[0009] The fourth objective of this application is to provide a computer-readable storage medium.

[0010] To achieve the above objectives, the first aspect of this application is to propose an intelligent green recycling method for power equipment based on material attribute characteristics, comprising the following steps: The attribute characteristics of the power equipment to be recycled are preprocessed to correct the distorted characteristics under environmental interference and to supplement the missing characteristic parameters of new materials. A multi-dimensional attribute indicator system is constructed. A dynamic weight allocation algorithm is used to assign dynamic weights to each attribute indicator in the attribute indicator system. The comprehensive score of the power equipment is calculated based on each preprocessed attribute feature and its corresponding dynamic weight using a weighted summation method. The initial recommended recycling plan for the power equipment is determined based on the comprehensive score. Based on multiple attribute indicators related to green recycling in the attribute indicator system, the green recycling index of the power equipment is calculated. The green recycling index is used to assess whether the power equipment meets the green recycling requirements, and the initial recommended recycling plan is adjusted according to the assessment results.

[0011] Optionally, the preprocessing of the attribute characteristics of the power equipment to be recovered to correct distorted characteristics under environmental interference includes: determining the proportion of interference of environmental conditions on attribute characteristics by conducting actual measurements on multiple sets of samples; determining the correction weight by calculating the error rate of the corrected attribute characteristics, wherein the correction weight is used to balance the statistical error of the interference proportion; constructing an attribute correction formula based on the interference proportion and the statistical error, and correcting the collected attribute characteristics by the attribute correction formula; comparing the correction result with the true value of the attribute characteristics to verify whether the correction result meets the requirements.

[0012] Optionally, before determining the interference ratio of environmental conditions on attribute characteristics by conducting actual measurements on multiple sets of samples, the method further includes: determining the type of environmental conditions, and matching the corresponding interference ratio and statistical error for different types of environmental conditions.

[0013] Optionally, the multi-dimensional attribute index system includes four dimensions: material characteristics, physical properties, environmental attributes, and recycling efficiency. The attribute indexes corresponding to the material characteristics include spectral characteristic values ​​and density; the attribute indexes corresponding to the physical properties include normalized hardness, health, and service life; the attribute indexes corresponding to the environmental attributes include the proportion of harmful components and degradability; and the attribute indexes corresponding to the recycling efficiency include recycling energy consumption and resource recovery rate.

[0014] Optionally, the step of assigning dynamic weights to each attribute indicator in the attribute indicator system using a dynamic weight allocation algorithm includes: calculating the initial weights of each attribute indicator in the attribute indicator system based on the Analytic Hierarchy Process (AHP). Calculate the initial comprehensive score of multiple groups of samples based on the initial weights, classify and recover the multiple groups of samples based on the initial comprehensive scores, and calculate the recovery efficiency of the initial weights based on the recovery benefits of each group of samples; use the recovery efficiency to iteratively optimize the initial weights.

[0015] Optionally, the iterative optimization of the initial weights using the recovery efficiency includes: calculating the contribution deviation of each attribute indicator when the recovery efficiency is less than a preset efficiency threshold; calculating the weight adjustment coefficient of the corresponding attribute indicator based on the contribution deviation; determining the optimized weight of each attribute indicator based on the weight adjustment coefficient, and normalizing the optimized weights of each attribute indicator; and iteratively optimizing the weights using the optimized weights and updated multiple sets of samples until the updated recovery efficiency is greater than or equal to the efficiency threshold.

[0016] Optionally, calculating the green recycling index of the power equipment based on multiple attribute indicators related to green recycling in the attribute indicator system includes: positiveening the negative indicators among the multiple attribute indicators; calculating the product of the positively processed multiple attribute indicators to obtain the green recycling index; assessing whether the power equipment meets the green recycling requirements through the green recycling index and adjusting the initial recommended recycling plan according to the assessment results includes: comparing the green recycling index with a preset assessment threshold; executing the initial recommended recycling plan if the green recycling index is greater than or equal to the assessment threshold; and carrying out pollution control and safe disposal of the power equipment if the green recycling index is less than the assessment threshold.

[0017] Optionally, after calculating the product of multiple attribute indicators after positive processing to obtain the green recycling index, the method further includes: optimizing the green recycling index by setting a scenario-based adjustment coefficient for different recycling scenarios.

[0018] To achieve the above objectives, a second aspect of this application also proposes an intelligent green recycling system for power equipment based on material attribute characteristics, comprising the following modules: The preprocessing module is used to preprocess the attribute characteristics of the power equipment to be recycled, so as to correct the distorted characteristics under environmental interference and supplement the missing characteristic parameters of the new materials. The calculation module is used to construct a multi-dimensional attribute indicator system, assign dynamic weights to each attribute indicator in the attribute indicator system through a dynamic weight allocation algorithm, and calculate the comprehensive score of the power equipment based on each preprocessed attribute feature and corresponding dynamic weight in the attribute indicator system through a weighted summation calculation method. The recommendation module is used to determine the initial recommended recycling plan for the power equipment based on the comprehensive score; The evaluation module calculates the green recycling index of the power equipment based on multiple attribute indicators related to green recycling in the attribute indicator system, evaluates whether the power equipment meets the green recycling requirements through the green recycling index, and adjusts the initial recommended recycling plan according to the evaluation results.

[0019] To achieve the above objectives, a third aspect of this application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the intelligent green recycling method for power equipment based on material attribute characteristics as described in any one of the first aspects above.

[0020] To achieve the above objectives, the fourth aspect of this application also proposes a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the intelligent green recycling method for power equipment based on material attribute characteristics as described in any of the first aspects above.

[0021] The technical solution provided by the embodiments of this application brings at least the following beneficial effects: First, this application corrects feature distortion and completes the material feature parameters to provide reliable data for classification; then, it quantifies the recycling value through multi-dimensional dynamic weighting, adapts to different types of materials, avoids process mismatch, and improves classification accuracy; finally, it identifies high pollution risks by calculating the green index, ensuring that recycling meets environmental protection standards and avoiding environmental risks. Therefore, this application adopts a core recycling path from specific material scenarios to algorithmic steps, formula calculations, and classification results. Addressing the recycling difficulties in the power industry's green recycling, such as mixed materials, environmental interference, and new materials, it constructs a three-level algorithm system combining attribute feature preprocessing, feature weighted fusion, and green assessment. Through multi-dimensional parameter optimization, dynamic weight adjustment, and scenario-based calibration, it achieves accurate classification and recycling scheme recommendations for highly difficult-to-recycle materials. This application can reduce material detection time and manual processing workload, save coal energy consumption, reduce inventory and recycling costs, and achieve cost reduction and efficiency improvement. Furthermore, it is applicable to different polluted environments and new materials, improving the scenario adaptability of recycling.

[0022] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1A flowchart illustrating an intelligent green recycling method for power equipment based on material attribute characteristics, as proposed in this application embodiment; Figure 2 This is a flowchart of an attribute feature preprocessing method proposed in an embodiment of this application; Figure 3 This is a flowchart of a weight iterative optimization method proposed in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an intelligent green recycling system for power equipment based on material attribute characteristics, as proposed in an embodiment of this application. Detailed Implementation

[0024] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0025] It should be noted that the power equipment recycling solutions in the relevant embodiments can employ technologies such as intelligent dismantling and automated sorting, IoT and blockchain traceability management, material recycling and cascade utilization, and artificial intelligence and big data analysis. Specifically, regarding intelligent dismantling and automated sorting technologies, taking the intelligent dismantling of power grid equipment as an example, a green intelligent dismantling center can be established, using intelligent equipment such as robotic arms, magnetic separators, and eddy current separators to achieve automated dismantling of materials such as transformers and cables. Taking the fine separation of photovoltaic modules as an example, a high-pressure jet grinding method can be developed to dismantle retired photovoltaic panels into micron-sized particles such as glass shavings and silver dust.

[0026] For IoT and blockchain traceability management technologies, reverse logistics networks can be optimized to build a "regional warehouse combined with turnover warehouse" model. Through intelligent warehousing systems and dynamic logistics scheduling, the efficiency of allocating idle materials can be improved. Full lifecycle data management can also be implemented; for example, sensors can monitor the status of conductors in real time, automatically triggering a winding mechanism to retrieve broken wires when they break, reducing safety hazards and improving maintenance efficiency. For material recycling and cascade utilization technologies, high-value utilization of metals can be achieved. For artificial intelligence and big data analytics technologies, intelligent detection and classification can be performed, such as identifying meter appearance defects based on deep learning models. Predictive maintenance and resource scheduling can also be implemented; for example, building a UAV inspection system for power transmission lines, using cloud-fog-edge heterogeneous collaborative technology to achieve multi-drone, multi-task autonomous inspection.

[0027] However, the power equipment recycling solutions in the above-mentioned embodiments have the following problems: First, the intelligent dismantling and automated sorting technologies have poor adaptability in complex scenarios. When faced with irregularly shaped or complex equipment (such as transformer hybrid components and new types of meters), the robotic arm has poor adaptability and is prone to jamming; the sorting accuracy of mixed materials (such as fine copper wire and aluminum wire, photovoltaic film and glass slag) is insufficient, requiring secondary manual processing; new equipment iterates rapidly, and the AI ​​recognition model is lagging behind.

[0028] Second, the data collaboration between IoT and blockchain traceability management technologies is poor. In remote and harsh environments (such as mountainous areas and areas with strong electromagnetic fields), the data disconnection rate of IoT devices is high; cross-enterprise blockchain platform interfaces are incompatible, forming information silos; traceability only covers the logistics / disposal process and lacks core technical parameters of the equipment (such as battery cycle count), resulting in incomplete full lifecycle data.

[0029] Third, the purity and recycling efficiency of materials recycled and reused in the technology are low. For multi-system mixtures (such as ternary lithium and lithium iron phosphate batteries, rubber-plastic hybrid cables), the purity of the recycled materials is insufficient, making it difficult to achieve high value; the treatment cost of chemical recycling by-products (such as fluoride-containing wastewater and dust) is high; the detection efficiency before reuse is low (such as 30 minutes for single battery testing), and the consistency is not fully assessed.

[0030] Fourth, the environmental adaptability and linkage of AI and big data analytics technologies are poor. In harsh environments (such as heavy rain or oil pollution), the accuracy of AI recognition drops sharply (for example, the missed detection rate of drones rises to 25%); big data prediction models do not respond in real time to sudden working conditions (such as extreme high temperatures), and the judgment of retirement cycle is delayed; multi-scenario data (inspection, inventory, dismantling) are not linked, resulting in inefficient resource allocation.

[0031] To this end, this application proposes an intelligent green recycling method and system for power equipment based on the characteristics of material attributes. Through multi-dimensional parameter optimization, dynamic weight adjustment and scenario-based calibration, it can achieve accurate classification and recycling scheme recommendation for highly difficult-to-recycle materials.

[0032] The following description, with reference to the accompanying drawings, illustrates an intelligent green recycling method and system for power equipment based on material attribute characteristics, as proposed in the embodiments of this application.

[0033] Figure 1 This is a flowchart illustrating a smart green recycling method for power equipment based on material attribute characteristics, as proposed in an embodiment of this application. Figure 1 As shown, the method includes the following steps: Step S101: Preprocess the attribute characteristics of the power equipment to be recycled in order to correct the distorted characteristics under environmental interference and supplement the missing characteristic parameters of the new materials.

[0034] Specifically, this application first preprocesses the attribute characteristics of the target power equipment to be recycled to address environmental interference and complete material parameters. Attribute characteristic preprocessing is a fundamental step in the recycling algorithm system of this application, capable of resolving feature distortion issues caused by various harsh environments (such as those with oil pollution, dust, and strong electromagnetic fields) and parameter deficiencies in novel materials (such as novel photovoltaic films), providing high-quality, highly reliable foundational data for subsequent feature fusion and green assessment. This application can employ a combination of scenario-adaptive correction models and parameter completion algorithms for attribute characteristic preprocessing.

[0035] For example, to complete missing feature parameters, parameter completion algorithms can be used to complete feature parameters such as GBR parameters of devices like photovoltaic films. During preprocessing, defects in the battery U-Net model parameters can also be calibrated.

[0036] To more clearly illustrate the specific implementation process of correcting distorted features in the attribute feature preprocessing of this application, the following is an exemplary description of a specific feature correction method proposed in one embodiment of this application. Figure 2 This is a flowchart of an attribute feature preprocessing method proposed in an embodiment of this application, as shown below. Figure 2 As shown, the method includes the following steps: Step S201: Determine the proportion of interference of environmental conditions on attribute characteristics by conducting actual measurements on multiple groups of samples.

[0037] It should be noted that, in order to facilitate an intuitive understanding of the implementation process of the attribute feature correction method in this application embodiment, this embodiment uses the example of correcting the spectral characteristics of a rubber-plastic hybrid cable in an oily environment for illustration.

[0038] The analysis first examines the problem in this scenario. Decommissioned cables in the power industry often have contaminants such as insulating oil and dust adhering to them, causing severe distortion in the raw spectral values ​​(Sraw) collected by a near-infrared spectrometer. Specifically, the true spectral value of rubber (Strue,rubber) is approximately 0.75 (corresponding to the absorption coefficient at a specific wavelength, reflecting the molecular structure characteristics of the material), and the true spectral value of plastic (Strue,plastic) is approximately 0.55. Oil contamination causes an overall shift in the spectral values; the measured Sraw, rubber = 0.68 and Sraw, plastic = 0.60, a difference of only 0.08. Traditional spectral recognition algorithms cannot distinguish between these values, easily leading to sorting errors.

[0039] The core principle of the algorithm in this embodiment is based on the "interference quantification-reverse correction" logic. It determines the proportion (δ) of oil contamination affecting the spectral values ​​through actual measurement, establishing a mapping relationship of "original spectral value → interference elimination → true spectral value". The physical meaning of the interference coefficient δ is "the proportion of spectral value deviation caused by oil contamination to the true value", i.e., Sraw = Strue × (1-δ) (because oil contamination can obscure the material surface, leading to a weakening of the spectral absorption signal, thus resulting in a lower spectral value). The core of the correction is to deduce Strue from δ, while simultaneously introducing a correction weight (α) to balance the uncertainty of the interference (such as the influence of different oil contamination thicknesses).

[0040] In practice, the interference coefficient δ is calibrated first. 100 groups of the same type of oil-free cables (50 groups of rubber and 50 groups of plastic) are selected. In a laboratory environment, an oil-contaminated scenario in a power field is simulated (oil thickness 0.1-0.5mm, covering common insulating oil types). The true spectral value (Strue) when there is no oil contamination and the original spectral value (Sraw) after oil contamination are collected respectively. The δ value of each group is calculated using the formula δ=(Strue-Sraw) / Strue, and the average value is taken as the final δ (after actual calibration, δ=0.1, that is, the spectral value is 10% lower due to oil contamination).

[0041] Step S202: By calculating the error rate of the corrected attribute features, the correction weight is determined, wherein the correction weight is used to balance the statistical error of the interference ratio.

[0042] Specifically, determine the correction weight α. α is used to balance the statistical error of δ and can be optimized using cross-validation. For example, divide 100 calibration samples into 10 groups, and correct each group with α = 0.7, 0.8, 0.9, and 1.0 respectively. Calculate the error rate between the corrected spectral values ​​and the true values, and select the α with the lowest error rate (for example, the final α = 0.9, at which point the average error rate is <3%).

[0043] Step S203: Construct an attribute correction formula based on the interference ratio and statistical error, and correct the collected attribute features using the attribute correction formula.

[0044] Specifically, this step applies the spectral correction formula, which can be expressed as follows: Scorr = Sraw / [(1-δ)×α+(1-α)×0.95] Among them, (1-δ)×α is the main correction term, which reflects the core elimination logic of oil pollution interference; (1-α)×0.95 is the error compensation term, and 0.95 is the empirical coefficient (the upper limit of interference fluctuation based on historical data statistics), which avoids over-correction caused by a single parameter.

[0045] Step S204: Compare the correction result with the true value of the attribute feature to verify whether the correction result meets the requirements.

[0046] Specifically, the correction effect was verified by substituting actual measured data into calculations.

[0047] For rubber: Scorr, rubber = 0.68 / [(1-0.1)×0.9+(1-0.9)×0.95] = 0.68 / (0.81+0.095)≈0.68 / 0.905≈0.75, which is completely consistent with the true value of 0.75.

[0048] For plastic: Scorr, plastic = 0.60 / [(1-0.1)×0.9+ (1-0.9)×0.95]≈0.60 / 0.905≈0.66. Therefore, the error rate with the true value of 0.55 is <20%, and the difference with the rubber correction value is increased to 0.09, which improves the distinguishability by 12.5%.

[0049] In one embodiment of this application, before determining the interference ratio of environmental conditions on attribute characteristics by conducting actual measurements on multiple sets of samples, the method further includes: determining the type of environmental conditions, and matching the corresponding interference ratio and statistical error for different types of environmental conditions.

[0050] Specifically, this embodiment optimizes the above algorithm. Continuing with the example above, for different types of oil contamination (such as mineral insulating oil, vegetable oil, etc.), an "oil contamination type identification branch" can be added. By collecting the characteristic wavelengths of oil contamination with a spectrometer (for example, mineral oil has an absorption peak at 2930 cm⁻¹), the corresponding δ and α are automatically matched to further reduce the correction error (target error rate <5%).

[0051] Step S102: Construct a multi-dimensional attribute index system, assign dynamic weights to each attribute index in the attribute index system through a dynamic weight allocation algorithm, and calculate the comprehensive score of the power equipment based on each preprocessed attribute feature and its corresponding dynamic weight through a weighted summation calculation method.

[0052] Specifically, this application performs weighted fusion of preprocessed attribute features to calculate the comprehensive score F of the power equipment. Feature weighted fusion is the core component of this application's recycling algorithm system. By constructing a multi-dimensional attribute index system and combining it with dynamic weight allocation, the comprehensive score (F) of the materials is calculated, thereby quantifying their recycling value (such as high-value regeneration potential) and process adaptability (such as suitability for cascade utilization), providing a quantitative basis for classification and recommendation.

[0053] It should be noted that, in order to facilitate an intuitive understanding of the implementation process of the feature weighted fusion method in this step, a retired ternary lithium battery is used as an example for illustration.

[0054] In one embodiment of this application, the multi-dimensional attribute index system includes four dimensions: material characteristics, physical properties, environmental attributes, and recycling efficiency. The attribute indexes corresponding to material characteristics include spectral characteristic values ​​and density; the attribute indexes corresponding to physical properties include normalized hardness, health, and service life; the attribute indexes corresponding to environmental attributes include the proportion of harmful components and degradability; and the attribute indexes corresponding to recycling efficiency include recycling energy consumption and resource recovery rate.

[0055] Specifically, when constructing the attribute indicator system, based on the recycling needs of retired materials in the power industry, nine core attributes were selected from four dimensions: "material characteristics, physical properties, environmental attributes, and recycling efficiency," forming the indicator system, as shown in Table 1 below: Table 1. Multi-dimensional attribute index system of ternary lithium batteries

[0056] Furthermore, a dynamic weight allocation algorithm is executed to determine the dynamic weights. This application can use a fusion model of the Analytic Hierarchy Process (AHP) and recycling efficiency feedback to achieve dynamic weight allocation.

[0057] In some of the related embodiments, the traditional weight allocation (such as the simple AHP method) relies solely on expert experience and lacks feedback adjustment based on actual recycling efficiency, which can easily lead to a disconnect between the weights and actual needs. This application adopts a dynamic weight model of "AHP method + recycling efficiency feedback optimization", which can achieve scientific allocation and real-time updates of weights.

[0058] In one embodiment of this application, a dynamic weight allocation algorithm is used to assign dynamic weights to each attribute indicator in the attribute indicator system, including: calculating the initial weight of each attribute indicator in the attribute indicator system based on the analytic hierarchy process (AHP); calculating the initial comprehensive score of multiple groups of samples based on the initial weight; classifying and recovering multiple groups of samples based on the initial comprehensive score; calculating the recovery efficiency of the initial weight based on the recovery benefit of each group of samples; and iteratively optimizing the initial weight using the recovery efficiency.

[0059] Specifically, in this embodiment, the initial weights are first calculated based on the AHP method. During the calculation of the initial weights, a judgment matrix is ​​constructed first. For example, five experts in the field of electricity recycling (including two process engineers, two equipment experts, and one environmental expert) can be invited to compare the importance of the nine attributes pairwise using the "1-9 scale" (1 = equally important, 9 = extremely important), forming a judgment matrix A (9×9 order).

[0060] Next, consistency checks and weight calculations are performed. The largest eigenvalue λmax of the judgment matrix is ​​calculated, and consistency is checked using the consistency index CI=(λmax-n) / (n-1) (n=9) and the random consistency index RI (RI=1.45 from the table). (CR=CI / RI<0.1, indicating a pass). After passing, the initial weight vector W0=[w01,w02,...,w09] is calculated using the eigenvector method. Where w01=0.25 (Scorr), w02=0.03 (ρ), w03=0.03 (H), w04=0.15 (Ch), w05=0.04 (D), w06=0.20 (K), w07=0.10 (Tcomp), w08=0.05 (E), and w09=0.15 (R).

[0061] Understandably, the initial weights are based on expert experience, highlighting the importance of spectral purity, health, and recovery rate.

[0062] Then, initial weight feedback optimization based on recycling efficiency is performed.

[0063] Among them, recovery efficiency (η) is the core indicator for measuring the rationality of weights, defined as "actual recovery benefit / theoretical maximum benefit". The higher the η, the more the weight allocation conforms to actual needs. In this embodiment, the initial weights are iteratively optimized by establishing a correlation model of "weight-recovery efficiency".

[0064] As an example, in calculating the recycling efficiency η, 100 groups of retired ternary lithium battery samples are first selected, and a comprehensive score F0 is calculated according to the initial weights. Based on F0, the samples are divided into 3 categories (F0≥0.8 is high-value category, 0.5≤F0<0.8 is medium-value category, and F0<0.5 is low-value category). Corresponding recycling processes are adopted for each category (high-value category is used for secondary utilization, medium-value category is used for material recycling, and low-value category is used for safe disposal). The actual recycling revenue of each group is calculated (e.g., the sales revenue of batteries from secondary utilization and the sales revenue of metals from material recycling). Then, η is calculated using the following formula: η = (Σactual revenue) / (Σtheoretical maximum revenue), where the theoretical maximum revenue is the revenue of each sample using the optimal process.

[0065] Then, the calculated recovery efficiency is used to iteratively optimize the initial weights.

[0066] To more clearly illustrate the specific implementation process of iteratively optimizing the initial weights using recycling efficiency in this application, a specific iterative optimization method proposed in one embodiment of this application will be used as an example for illustration below. Figure 3 This is a flowchart of a weight iterative optimization method proposed in an embodiment of this application, as shown below. Figure 3 As shown, the method includes the following steps: Step S301: If the recovery efficiency is less than a preset efficiency threshold, calculate the contribution deviation of each attribute indicator.

[0067] Specifically, this step begins with calculating the weight adjustment coefficient γ. First, the recovery efficiency is compared with a preset efficiency threshold. If η < 0.9 (i.e., the efficiency threshold), it indicates that the initial weights need optimization. Then, the contribution deviation of each attribute indicator is calculated using the following formula: Contribution deviation Δi = (Actual contribution i - Theoretical contribution i) / Theoretical contribution i; Wherein, the actual contribution i = (sample mean of the attribute × initial weight wi) / F0 mean, and the theoretical contribution i = (optimal mean of the attribute × initial weight wi) / F0 optimal mean.

[0068] Step S302: Calculate the weight adjustment coefficient of the corresponding attribute indicator based on the contribution deviation.

[0069] Specifically, the weight adjustment coefficient γi is calculated based on Δi, which can be done using the following formula: γi = 1 + Δi (if Δi > 0, it means that the actual contribution of the attribute is higher than the theoretical contribution, and the weight needs to be increased; if Δi < 0, the weight needs to be decreased).

[0070] Step S303: Determine the optimized weight of each attribute indicator based on the weight adjustment coefficient, and normalize the optimized weight of each attribute indicator.

[0071] Specifically, the optimized dynamic weights W are determined. The optimized weights Wi = W0i × γi, and are simultaneously normalized, i.e., ΣWi = 1, to ensure that the sum of the weights is 1.

[0072] Step S304: Using the optimized weights and the updated multiple sets of samples, perform weight optimization iteratively until the updated recovery efficiency is greater than or equal to the efficiency threshold.

[0073] Specifically, iterative verification is performed: the optimized weight W is substituted into the new sample, F and η are recalculated, and steps S301 to S304 of this embodiment are repeated until η≥0.9. The weight at this time is the final dynamic weight after optimization. After actual optimization, the final weight vector W=[0.26(Scorr),0.02(ρ),0.02(H),0.14(Ch),0.05(D),0.21(K),0.09(Tcomp),0.06(E),0.15(R)].

[0074] Therefore, it can be seen that the weight of health status and energy consumption after adjustment has been slightly increased, which is more in line with the actual needs of recycling efficiency.

[0075] Furthermore, the comprehensive score of the power equipment is calculated using the comprehensive scoring formula and the optimized final dynamic weights. The comprehensive score F adopts a "weighted summation model," which can be expressed by the following formula: F = Σ(Wi × Xi), i = 1, 2, ..., 9, where Xi is the preprocessed value of the i-th attribute feature (interference eliminated and missing values ​​filled), and Wi is the dynamic weight of the i-th attribute feature. F ∈ [0, 1], and a higher F value indicates higher recycling value of the material and better process adaptability.

[0076] The following is a specific example illustrating the calculation of the comprehensive score F for retired ternary lithium batteries.

[0077] Given the preprocessed attribute vector X = [Scorr = 0.85, ρ = 2.6 g / cm³, H = 0.4, Ch = 0.3, D = 0.6, K = 0.6, Tcomp = 0.6625, E = 0.3, R = 0.9] of a retired ternary lithium battery, substituting it into the final weight W, we calculate: F = 0.26 × 0.85 + 0.02 × 2.6 + 0.02 × 0. 0.4 + 0.14 × 0.3 + 0.05 × 0.6 + 0.21 × 0.6 + 0.09 × 0.6625 + 0.06 × 0.3 + 0.15 × 0.9 = 0.221 + 0.052 + 0.008 + 0.042 + 0.03 + 0.126 + 0.059625 + 0.018 + 0.135 ≈ 0.691625 (F ≈ 0.69).

[0078] Step S103: Determine the initial recommended recycling plan for the power equipment based on the comprehensive score.

[0079] Specifically, based on the calculated range of the comprehensive score F, corresponding initial recycling classification rules are formulated. As one possible implementation, the initial recommended recycling scheme corresponding to the comprehensive score can be determined using Table 2 shown below: Table 2 Initial Recommended Recycling Plan

[0080] In step S102 of the above example, since F≈0.69, it can be seen from Table 2 above that the score belongs to the medium value level, the process adaptability is good, and the "material recycling" scheme is recommended, which is completely matched with the actual battery status (health level 0.6, which does not meet the requirements for cascade utilization).

[0081] Step S104: Based on multiple attribute indicators related to green recycling in the attribute indicator system, calculate the green recycling index of the power equipment, evaluate whether the power equipment meets the green recycling requirements through the green recycling index, and adjust the initial recommended recycling plan according to the evaluation results.

[0082] Specifically, this step calculates the green recycling index G to conduct a green assessment of the power equipment, and then determines whether the initial recommended recycling scheme determined in step S103 needs to be adjusted based on the assessment results.

[0083] Among them, the green recycling index G needs to reflect the green concept of "low harm, low energy consumption, high degradability and high recyclability". Therefore, this application can adopt the "product model" (multiplying each index after positive processing) to ensure that the shortcomings of any index will significantly reduce the G value, and avoid the problem that a single index is too excellent and covers up the defects of multiple indices.

[0084] In one embodiment of this application, the green recycling index of power equipment is calculated based on multiple attribute indicators related to green recycling in the attribute indicator system, including: positiveizing the negative indicators among the multiple attribute indicators; and calculating the product of the positively processed multiple attribute indicators to obtain the green recycling index.

[0085] Specifically, in the derivation of the green recycling index G formula in this embodiment, the indicators are first positiveized. For some indicators (such as Ch and E in the above indicator system) that are "negative indicators" (i.e., the higher the value, the worse), they need to be positiveized first and transformed into "positive indicators" (i.e., the higher the value, the better).

[0086] For example, the positive value for harmful components: Chpositive = 1 - Ch (i.e., the higher the Ch, the lower the positive Ch, which aligns with green principles). The positive value for energy consumption: Epositive = 1 - E (i.e., the higher the E, the lower the positive E, which aligns with green principles). Degradability and recyclability are "positive indicators," requiring no treatment; D and R can be used directly.

[0087] Next, the formula for the green index is derived. Based on the above positiveized indicators, the basic formula for the green recycling index G is: G = Chpositive × D × Epositive × R = (1-Ch) × D × (1-E) × R. In this formula, the weight of each indicator is 1. Since the four indicators are equally important to green recycling, if any indicator value is too low (such as Ch = 0.8, Chpositive = 0.2), the G value will decrease significantly, thereby triggering an environmental warning.

[0088] Furthermore, in this embodiment, the green recycling index is used to assess whether the power equipment meets the green recycling requirements, and the initial recommended recycling plan is adjusted according to the assessment results. This includes: comparing the green recycling index with a preset assessment threshold; if the green recycling index is greater than or equal to the assessment threshold, the initial recommended recycling plan is implemented; if the green recycling index is less than the assessment threshold, pollution control and safe disposal are carried out on the power equipment.

[0089] In one embodiment of this application, after calculating the product of multiple attribute indicators after positive processing to obtain the green recycling index, the method further includes: optimizing the green recycling index by adjusting the coefficient according to different recycling scenarios.

[0090] Specifically, this embodiment uses a scenario-based coefficient to adjust the basic formula of the green recycling index G in order to optimize G. The degree of influence of various indicators will change for different recycling scenarios (such as high-temperature environments or high-humidity environments), therefore a scenario-based adjustment coefficient (ε) needs to be introduced to optimize the basic formula.

[0091] For example, in humid environments, the impact of energy recovery (E) on environmental protection is greater (humidity leads to increased energy consumption and generates more pollutants), requiring an increase in the weight of positive E. In highly polluted areas (such as near chemical industrial parks), the impact of harmful components (Ch) is greater, requiring an increase in the weight of positive Ch. The optimized formula is: G=(positive Ch)^(ε1)×D^(ε2)×(positive E)^(ε3)×R^(ε4), where ε1+ε2+ε3+ε4=4 to ensure that the total weight is consistent with the basic formula, and ε1-ε4 are determined based on the scenario through expert scoring and calibration with actual data.

[0092] The following example illustrates the specific implementation process of step S104 using a concrete example of a green assessment application for fluorine-containing waste batteries and ordinary ternary lithium batteries.

[0093] In this example, the retired batteries of a power company include two types: fluorinated waste batteries (the cathode material contains fluoride, and the recycling process easily generates fluoride-containing wastewater) and ordinary ternary lithium batteries (fluoride-free). The comprehensive score F of both types of batteries is approximately 0.6 (medium value level), and a green assessment is needed to determine whether they are suitable for a recommended recycling program.

[0094] The preprocessed basic data is shown in Table 3 below: Table 3 Battery Basic Data Sheet

[0095] Based on the above data, we first use the basic formula to calculate G.

[0096] For a typical ternary lithium battery, Gtypical = (1-0.3)×0.6 + (1-0.3)×0.9 = 0.7×0.6×0.7×0.9 = 0.2646 (G≈0.26).

[0097] For fluorine-containing waste batteries, Gfluorine = (-0.6)0.2(1-0.6)×0.8 = 0.4×0.2×0.4×0.8 = 0.0256 (G≈0.03).

[0098] Furthermore, perform green evaluation threshold and decision-making. By statistically analyzing the historical data of green recycling in the power industry and combining relevant environmental protection standards, determine the green evaluation threshold G_threshold = 0.15. The decision-making rules are as follows: If G ≥ G_threshold: It meets the requirements of green recycling and can be implemented according to the scheme initially recommended by the comprehensive score (F); If G < G_threshold: It does not meet the requirements of green recycling, triggers an environmental protection risk warning, and the recycling scheme needs to be adjusted (such as adding pollution treatment processes) or changed to safe disposal.

[0099] Furthermore, based on the above judgment results, conduct evaluation results and scheme adjustment. For ordinary ternary lithium batteries: G ≈ 0.26 ≥ 0.15, which meets the green requirements. It is recommended to implement according to the "material regeneration" scheme corresponding to the F value without adjustment; for fluorine-containing waste batteries: G ≈ 0.03 < 0.15, triggering an environmental protection warning. The raw material regeneration scheme (which will produce fluorine-containing wastewater) needs to be adjusted to the "safe disposal + special pollution treatment" scheme (such as treating fluorine-containing wastewater by alkaline neutralization method and then disposing it) to avoid environmental pollution.

[0100] On this basis, it is also possible to perform optimization calculations for scenarios in highly polluted areas.

[0101] For example, if the evaluation scenario is a highly polluted area (such as near a chemical industrial park), it is necessary to increase the weight of Ch positive and set the scenario coefficients ε1 = 1.5, ε2 = 1.0, ε3 = 0.8, ε4 = 0.7 (total = 4). The optimized G calculation formula is as follows: For ordinary ternary lithium batteries: G'_ordinary = (0.7)^1.5(0.6)^1.0×(0.7)^0.8×(0.9)^0.7 ≈ 0.606×0.6×0.816× 0.919 ≈ 0.25 (still ≥ 0.15); For fluorine-containing waste batteries: G'_fluorine-containing = (0.4)^1.5(0.2)^1.0(0.4)^0.8(0.8)^0.7 ≈ 0.253×0.2×0.635×0.857 ≈ 0.028 (still < 0.15).

[0102] It can be seen from this that the optimized evaluation results are consistent with the basic formula, thus verifying the scenario adaptability of the recycling algorithm of this application.

[0103] In summary, the intelligent green recycling method for power equipment based on material attribute characteristics in this application first corrects feature distortion and completes material feature parameters to provide reliable data for classification. Then, it quantifies the recycling value through multi-dimensional dynamic weighting, adapting to different types of materials, avoiding process mismatches, and improving classification accuracy. Finally, it identifies high-pollution hazards by calculating a green index, ensuring that recycling meets environmental standards and mitigating environmental risks. Therefore, this method adopts a core recycling path from specific material scenarios to algorithmic steps, formula calculations, and classification results. Addressing the recycling difficulties in the power industry, such as mixed materials, environmental interference, and new materials, it constructs a three-level algorithm system combining attribute feature preprocessing, feature weighted fusion, and green assessment. Through multi-dimensional parameter optimization, dynamic weight adjustment, and scenario-based calibration, it achieves accurate classification and recycling scheme recommendations for highly difficult-to-recycle materials. This method can reduce material detection time and manual processing workload, save coal energy consumption, reduce inventory and recycling costs, and achieve cost reduction and efficiency improvement. Furthermore, it is applicable to different polluted environments and new materials, improving the scenario adaptability of recycling.

[0104] To achieve the above embodiments, this application also proposes an intelligent green recycling system for power equipment based on material attribute characteristics. Figure 4 This is a schematic diagram of the structure of an intelligent green recycling system for power equipment based on material attribute characteristics, as proposed in an embodiment of this application. Figure 4 As shown, the system includes: The preprocessing module 100 is used to preprocess the attribute characteristics of the power equipment to be recycled, so as to correct the distorted characteristics under environmental interference and supplement the missing characteristic parameters of the new materials.

[0105] The calculation module 200 is used to construct a multi-dimensional attribute index system. It assigns dynamic weights to each attribute index in the attribute index system through a dynamic weight allocation algorithm, and calculates the comprehensive score of the power equipment based on each preprocessed attribute feature and its corresponding dynamic weight through a weighted summation calculation method.

[0106] Recommendation module 300 is used to determine the initial recommended recycling plan for power equipment based on the comprehensive score.

[0107] The assessment module 400 calculates the green recycling index of power equipment based on multiple attribute indicators related to green recycling in the attribute indicator system. It assesses whether the power equipment meets the green recycling requirements through the green recycling index and adjusts the initial recommended recycling plan based on the assessment results.

[0108] It should be noted that the foregoing explanation of the embodiment of the intelligent green recycling method for power equipment based on material attribute characteristics also applies to the system of this embodiment, and will not be repeated here.

[0109] In summary, the intelligent green recycling system for power equipment based on material attribute characteristics in this application embodiment achieves accurate classification of highly difficult-to-recycle materials and recommends suitable recycling solutions through multi-level computational processing of attribute characteristics, weighted feature fusion, and green assessment.

[0110] To implement the above embodiments, this application also proposes an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the intelligent green recycling method for power equipment based on material attribute characteristics as described in any of the first aspect embodiments above.

[0111] To implement the above embodiments, this application also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the intelligent green recycling method for power equipment based on material attribute characteristics as described in any one of the first aspect embodiments above.

[0112] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0113] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0114] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0115] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0116] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0117] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0118] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0119] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A power equipment intelligent green recycling method based on material attribute characteristics, characterized in that, The method comprises the following steps: Preprocessing attribute features of the power equipment to be recycled to correct distorted features under environmental interference and complete missing feature parameters of new materials; A multi-dimensional attribute index system is constructed, a dynamic weight distribution algorithm is used to assign a dynamic weight to each attribute index in the attribute index system, and a weighted sum calculation method is used to calculate a comprehensive score of the power equipment according to each preprocessed attribute feature and the corresponding dynamic weight in the attribute index system; An initial recommended recycling scheme of the power equipment is determined according to the comprehensive score; A green recycling index of the power equipment is calculated based on multiple attribute indexes related to green recycling in the attribute index system, the green recycling index is used to evaluate whether the power equipment meets the green recycling requirements, and the initial recommended recycling scheme is adjusted according to the evaluation result.

2. The method of claim 1, wherein, The preprocessing of the attribute features of the power equipment to be recycled to correct the distorted features under environmental interference comprises: The interference proportion of environmental conditions on attribute features is determined by actual measurement of multiple groups of samples; The correction weight is determined by calculating the error rate of the corrected attribute features, wherein the correction weight is used to balance the statistical error of the interference proportion; An attribute correction formula is constructed based on the interference proportion and the statistical error, and the collected attribute features are corrected by the attribute correction formula; The correction result is compared with the true value of the attribute features to verify whether the correction result meets the requirements.

3. The method of claim 2, wherein, Before the interference proportion of environmental conditions on attribute features is determined by actual measurement of multiple groups of samples, the method further comprises: The type of the environmental conditions is determined, and the corresponding interference proportion and statistical error are matched for different types of the environmental conditions.

4. The method of claim 1, wherein, The multi-dimensional attribute index system comprises four dimensions of material features, physical properties, environmental protection attributes and recycling efficiency; The attribute indexes corresponding to the material features include spectral feature values and density, the attribute indexes corresponding to the physical properties include normalized hardness, health degree and service period, the attribute indexes corresponding to the environmental protection attributes include harmful component proportion and degradability, and the attribute indexes corresponding to the recycling efficiency include recycling energy consumption and resource recovery rate.

5. The method of claim 1, wherein, The dynamic weight distribution algorithm comprises: The initial weight of each attribute index in the attribute index system is calculated based on the analytic hierarchy process (AHP); The initial comprehensive scores of multiple groups of samples are calculated according to the initial weights, the multiple groups of samples are classified and recycled according to the initial comprehensive scores, and the recycling efficiency of the initial weights is calculated based on the recycling benefits of each group of samples; The initial weights are iteratively optimized using the recycling efficiency.

6. The method of claim 5, wherein, The iterative optimization of the initial weights using the recycling efficiency comprises: In the case that the recycling efficiency is less than a preset efficiency threshold, the contribution degree deviation of each attribute index is calculated; The weight adjustment coefficient of the corresponding attribute index is calculated according to the contribution degree deviation; Determine the optimized weight of each attribute index according to the weight adjustment coefficient, and normalize the optimized weight of each attribute index; Use the optimized weight and the updated multiple groups of samples to perform weight optimization in a loop until the updated recovery efficiency is greater than or equal to the efficiency threshold.

7. The method of claim 1, wherein, The green recovery index of the power equipment is calculated based on the multiple attribute indexes related to green recovery in the attribute index system, including: The negative indexes in the multiple attribute indexes are positively processed; The product of the multiple attribute indexes after positive processing is calculated to obtain the green recovery index; The green recovery index is used to evaluate whether the power equipment meets the green recovery requirement, and the initial recommended recovery scheme is adjusted according to the evaluation result, including: The green recovery index is compared with a preset evaluation threshold, and the initial recommended recovery scheme is executed if the green recovery index is greater than or equal to the evaluation threshold; In the case where the green recovery index is less than the evaluation threshold, the power equipment is subjected to pollution control and safe disposal.

8. The method of claim 7, wherein, After calculating the product of the multiple attribute indexes after positive processing to obtain the green recovery index, it further includes: The green recovery index is optimized by a scenario adjustment coefficient for different recovery scenarios.

9. An intelligent green recycling system for power equipment based on material attribute features, characterized in that, It includes the following modules: A preprocessing module for preprocessing the attribute characteristics of the power equipment to be recycled to correct the distorted features under environmental interference and complete the missing feature parameters of new materials; A calculation module for constructing a multi-dimensional attribute index system, assigning a dynamic weight to each attribute index in the attribute index system through a dynamic weight distribution algorithm, and calculating the comprehensive score of the power equipment according to each preprocessed attribute characteristic and the corresponding dynamic weight in the attribute index system through a weighted summation calculation method; A recommendation module for determining the initial recommended recovery scheme of the power equipment according to the comprehensive score; An evaluation module for calculating the green recovery index of the power equipment based on the multiple attribute indexes related to green recovery in the attribute index system, evaluating whether the power equipment meets the green recovery requirement through the green recovery index, and adjusting the initial recommended recovery scheme according to the evaluation result.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the intelligent green recovery method of power equipment based on material attribute characteristics as claimed in any one of claims 1-8.

Images