High-entropy alloy electrode production management method and system based on quality tracing

Abstract

The invention discloses a high-entropy alloy electrode production management method and system based on quality tracing, and belongs to the technical field of production management, and the method specifically comprises the steps: endowing an identifier for an electrode production batch, and collecting a technological parameter curve; dividing application scenes and storing data in a classified manner by associating batch identifiers with terminal product models; obtaining the statistical failure life of the terminal product and extracting related process curves to form an analysis sample set; quantitative fluctuation characteristics of the process curve are extracted, and correlation analysis is carried out on the quantitative fluctuation characteristics and corresponding batches of life; defining a process characteristic value interval in positive correlation with the long life in a specific scene and generating a corresponding production control parameter set; and matching an application scene according to the order product model, and calling a corresponding parameter set to execute directional production. According to the method, closed-loop tracing and correlation analysis between process data and terminal service performance are constructed, so that the performance suitability and the production control accuracy of the high-entropy alloy electrode in subdivision application are improved.

Description

Technical Field

[0001] This invention relates to the field of production management technology, specifically to a method and system for the production management of high-entropy alloy electrodes based on quality traceability. Background Technology

[0002] High-entropy alloy electrodes, as an emerging electrode material, exhibit significant potential in specific capacity, cycle stability, and high-temperature resistance due to the high-entropy effect, lattice distortion, and hysteresis diffusion effect brought about by their multi-component design. They are widely used in high-performance batteries, water electrolysis for hydrogen production, and other new energy fields. Their superior performance highly depends on the complex multi-principal component design and precise fabrication process. However, the multi-component characteristics also bring quality control challenges such as component segregation and complex phase transition behavior. Therefore, the stability of the entire process from melting and heat treatment to forming directly determines the performance consistency of the final product. Thus, establishing an effective quality traceability system is crucial for ensuring the reliability of high-entropy alloy electrodes.

[0003] Existing technical solutions typically revolve around ensuring the compliance of production parameters and the integrity of the data chain. Specifically, they involve using sensors to collect key process parameters such as melting temperature, holding time, and rolling rate, and then binding these parameters to specific production batches or units to create a production history archive. When product quality issues arise, this archive allows for reverse tracing to pinpoint the production stage or raw material batch where deviations may have occurred, thereby isolating and correcting the problem. This provides fundamental data support for production process standardization and problem attribution.

[0004] However, the aforementioned existing technical solutions still have significant limitations. Their core focus is on ensuring production parameters meet preset general process windows and establishing a one-way traceability chain, but they fail to deeply consider the differentiated service conditions and personalized performance requirements of high-entropy alloy electrodes in different end-use scenarios. For example, electrodes used in fast-charging batteries for electric vehicles need to withstand extremely high instantaneous current densities, while electrodes used in grid energy storage emphasize long-cycle capacity retention. Their core failure modes and performance optimization directions are fundamentally different. Existing traceability methods can only determine whether a product is "produced to standard," and cannot correlate the massive amounts of actual process parameter data, which naturally fluctuate, with the long-term service performance of the electrodes under specific harsh scenarios. Therefore, based on this traceability system, it is difficult to reverse-engineer the most critical process characteristics for specific application scenarios and optimize the production process accordingly. This leads to rigid production process parameters, making it difficult to dynamically adapt to diverse high-end application needs and restricting the product's performance competitiveness in niche markets. Summary of the Invention

[0005] The purpose of this invention is to provide a production management method and system for high-entropy alloy electrodes based on quality traceability, and to solve the following technical problems:

[0006] Existing technologies do not correlate actual process fluctuation data during production with the long-term service performance of high-entropy alloy electrodes in specific end-use scenarios, resulting in a lack of scenario-specific optimization of production processes and difficulty in dynamically adapting to diverse downstream performance requirements.

[0007] The objective of this invention can be achieved through the following technical solutions: A production management method for high-entropy alloy electrodes based on quality traceability includes the following steps: S1: Assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier; S2: Obtain the model of the terminal product assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify the application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios; S3: Obtain the statistical failure lifetime of the electrode batch corresponding to the terminal product model, and extract the process parameter curves of the relevant batches from the corresponding scenario based on the terminal product model to form an analysis sample set; S4: Extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch; S5: Based on the correlation analysis, define the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as the optimized process characteristics. Generate the corresponding production control parameter set based on the optimized process characteristics. S6: Determine the application scenario based on the terminal product model specified in the production order, and execute production by calling the production control parameter set corresponding to the application scenario.

[0008] As a further aspect of the present invention: in step S1, the specific process for forming the process parameter curve indexed by the batch identifier is as follows: By deploying data acquisition interfaces in each production unit, the original operating condition signals of the production equipment during the corresponding batch production period are acquired in real time; the original operating condition signals are converted from analog to digital and timestamped to generate discrete process parameter data points with time series. The data points are resampled and smoothed at equal intervals according to a preset sampling period to form a continuous process parameter data sequence. The data sequence is then bound to the batch identifier assigned to the corresponding production batch to obtain a process parameter curve with the batch identifier as the primary key index.

[0009] As a further aspect of the present invention: the specific process of calculating the failure lifetime in S3 is as follows: Receive a list provided by a downstream manufacturer containing the serial numbers of terminal products and their corresponding models; read the production start date and final failure date corresponding to the serial number of the terminal product; calculate the difference between the final failure date and the production start date to obtain the failure life of the terminal product; Based on the assembly association between the terminal product serial number and the electrode batch identifier, the failure life of the terminal product is classified into the corresponding electrode batch identifier; the failure life of all terminal products belonging to the same electrode batch identifier is calculated by arithmetic mean to obtain the statistical failure life of the electrode batch identifier; and an association record is established between the electrode batch identifier, the terminal product model and the statistical failure life.

[0010] As a further aspect of the present invention: in step S4, the specific process for obtaining the quantified fluctuation characteristics is as follows: Read each process parameter curve in the analysis sample set; divide the process parameter curve into time windows with a preset duration; within each time window, extract the corresponding process parameter data segment, and calculate the mean, maximum, and minimum values ​​of each process parameter data segment; combine the time sequence identifier of the time window, the mean, the maximum, and the minimum values ​​into a quadruple data structure; sort all the quadruple data structures corresponding to a process parameter curve according to the time sequence identifier to form the quantitative fluctuation characteristics of the process parameter curve.

[0011] As a further aspect of the present invention: the specific process of performing correlation analysis between the quantified fluctuation characteristics and the statistical failure lifetime of the corresponding batch in step S4 is as follows: Using the quantitative fluctuation characteristics identified by each electrode batch as input variables and the statistical failure lifetime identified by that batch as output variables, a set of data pairs for correlation analysis is constructed; the values ​​of all quantitative fluctuation characteristics in the set of data pairs are standardized. Using a linear regression model, a correlation analysis is performed on the standardized quantitative fluctuation characteristics and the statistical failure lifetime to obtain the regression coefficients and p-values ​​corresponding to the mean, maximum, and minimum values ​​at each time series position in the quantitative fluctuation characteristics. The mean, maximum, or minimum values ​​and their time series positions with p-values ​​lower than a preset correlation threshold are selected, and the time series positions are grouped into continuous time series intervals according to the continuity of the time axis. The regression coefficients corresponding to the mean, maximum, or minimum values ​​selected within the time series intervals are recorded together with the time series intervals as a correlation list.

[0012] As a further aspect of the present invention: in step S5, the specific process for obtaining the optimized process features is as follows: Based on the aforementioned list of relationships, filter the time series intervals where the regression coefficient of each electrode batch identifier is positive, and the corresponding mean, maximum, and minimum values. For each time point within each time series interval, electrode batch identifiers with positively correlated regression coefficients at that time point are selected from the analysis sample set, and the batch identifier with the largest statistical failure lifetime value is selected from these batch identifiers. Extract the mean, maximum and minimum values ​​of the process parameter curves of the selected electrode batch at the corresponding time points, and use them as the reference mean, reference maximum and reference minimum values ​​at that time point, respectively; form a reference mean sequence based on the reference mean of all time points, a reference maximum sequence based on the reference maximum of all time points, and a reference minimum sequence based on the reference minimum of all time points. The mean optimization interval boundary value for each time point in the time series interval is determined according to the reference mean sequence, the upper limit benchmark value for maximum optimization for each time point in the time axis is determined according to the reference maximum sequence, and the lower limit benchmark value for minimum optimization for each time point in the time axis is determined according to the reference minimum sequence. By summing the boundary values ​​of the mean optimization interval, the upper limit benchmark value of the maximum optimization, and the lower limit benchmark value of the minimum optimization, the optimized process characteristics are obtained.

[0013] As a further aspect of the present invention: the specific process of setting the production control parameter set in S5 is as follows; Read the mean optimization interval boundary value, the maximum optimization upper limit benchmark value, and the minimum optimization lower limit benchmark value recorded in the optimized process features; and calibrate the mean optimization interval boundary value as the target process parameter at the corresponding time point. The upper limit benchmark value for the maximum value optimization is directly set as the upper boundary of the operating range of the process parameters at the corresponding time point; the lower limit benchmark value for the minimum value optimization is directly set as the lower boundary of the operating range of the process parameters at the corresponding time point. The target process parameters, the upper and lower boundaries of the process parameter operating range are combined to form a closed-loop control parameter set for the corresponding time point; the closed-loop control parameter sets for each time point on the time axis are arranged and encapsulated according to the time sequence to form the production control parameter set bound to the application scenario.

[0014] A high-entropy alloy electrode production management system based on quality traceability, used in the aforementioned high-entropy alloy electrode production management method based on quality traceability, includes: The data acquisition module is used to assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier. The scenario classification module is used to obtain the terminal product model assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios. The sample acquisition module is used to obtain the statistical failure lifetime of electrode batches corresponding to the terminal product model, and extract the process parameter curves of relevant batches from the corresponding scenarios based on the terminal product model to form an analysis sample set. The data analysis module is used to extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and to perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch. The parameter generation module is used to define, based on the correlation analysis, the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as an optimized process characteristic, and generate a corresponding set of production control parameters based on the optimized process characteristic. The production control module is used to determine the application scenario based on the terminal product model specified in the production order, and to execute production by calling the production control parameter set corresponding to that application scenario.

[0015] The beneficial effects of this invention are: 1) This invention achieves a paradigm shift in quality traceability from process monitoring to performance attribution by establishing a full lifecycle data association based on application scenario grouping. First, this invention systematically associates electrode production batches with end-product models, classifying and storing process data according to specific application scenarios. Simultaneously, by integrating the actual failure lifespan data of end products, a complete data link of "process parameters - scenario tags - service results" is constructed. Based on this link, statistical correlation analysis can be performed on massive amounts of historical process data and corresponding long-term service performance within the same application scenario dimension, thereby identifying key process characteristics and fluctuation patterns that determine product lifespan in specific scenarios. This enables enterprises to transform production records, originally used only to determine pass / fail status, into data assets for evaluating and predicting the actual durability of products in target markets, providing a precise, evidence-based direction for process optimization.

[0016] 2) This invention achieves the explicit and standardized inheritance of implicit process knowledge by analyzing the mapping relationship between parameter fluctuations and service performance within a qualified process window. It is understandable that the final performance of high-entropy alloy electrodes (such as fatigue life and conductivity stability) is not solely determined by the average value of process parameters, but is more significantly influenced by the dynamic fluctuation characteristics of these parameters near their target values. These fluctuations are directly related to the micro-dynamic conditions during melting, solidification, and heat treatment. For example, instantaneous temperature fluctuations affect the diffusion rate and phase transformation path of elements, thereby altering key microstructures that determine service performance, such as grain size and phase composition. This invention transforms continuous process parameter curves into quantifiable and analyzable fluctuation characteristic sequences and uses statistical models to establish a correlation model between these characteristics and batch failure life. Essentially, it performs a data-driven reverse analysis of the causal chain of "process fluctuation - microstructure - macro performance." Through this correlation model, specific fluctuation value ranges significantly correlated with excellent service performance can be identified from numerous qualified production samples. Furthermore, by selectively integrating the beneficial fluctuations of multiple long-life batches at various key process stages, a "process feature template" representing the optimal microstructure for this application scenario is generated, thereby determining the optimal process production parameters for different application scenarios.

[0017] 3) This invention achieves the autonomous evolution of the production system from static execution to dynamic adaptation by constructing an automatic mapping and closed-loop update mechanism of "scenario requirements - optimized process - control parameters". First, this invention reverse-engineers the "process feature template" into a closed-loop control parameter set that drives specific production equipment, and strongly binds it to the application scenario. When a new production order specifies a product model, the corresponding control parameter set can be automatically invoked to execute production, thereby achieving scenario-oriented targeted manufacturing. More importantly, by continuously collecting service feedback data of products produced based on new parameters in the market, the associated model and process feature template can be periodically updated, thereby iteratively optimizing the control parameter set. This forms a complete data closed loop from "service feedback → data analysis → process optimization → production execution → new round of feedback", driving the manufacturing system to adapt to different high-end demands and continuously evolve towards higher scenario adaptability. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a schematic diagram of a high-entropy alloy electrode production management method based on quality traceability according to the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of a high-entropy alloy electrode production management system based on quality traceability according to the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 As shown, this invention is a production management method for high-entropy alloy electrodes based on quality traceability, comprising the following steps: S1: Assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier; First, at the start of production for each batch of high-entropy alloy electrodes, the production management system automatically generates a globally unique batch identifier (e.g., "HEA-LOT-2023-08-001"). During production, a high-precision sensor network deployed on key processes (such as vacuum melting furnaces, homogenization heat treatment furnaces, and rolling production lines) continuously collects process parameters in real time. Taking vacuum melting as an example, temperature sensors record the melt temperature several times per second, pressure sensors record the vacuum level inside the furnace, and mass spectrometers record the partial pressures of key gases. These heterogeneous time-series data from different devices are received, synchronized in time, and standardized in format through a unified IoT gateway. Subsequently, the system aligns and merges multi-dimensional data (temperature, pressure, etc.) within the same batch and the same time window to form a complete "process parameter curve" with time as the horizontal axis and containing multiple parameter dimensions. Finally, this curve is strongly bound to the unique identifier of the batch and stored in a time-series database, forming a traceable original data archive.

[0023] S2: Obtain the model of the terminal product assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify the application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios; When a batch of electrodes is delivered to downstream customers (such as battery manufacturers), the customers scan or record this batch identification code on their assembly lines. The customer's Manufacturing Execution System (MES) associates this electrode batch identification code with the unique serial number of the end product being assembled (such as "Model NEX-P100 fast-charging power battery pack") and synchronizes this association back to the electrode manufacturer's system through a secure interface. The electrode manufacturer's system automatically determines the application scenario to which the end product model belongs based on a predefined rule base (such as "all battery packs with model codes containing 'P' and rated charging rate ≥ 3C" are classified as "ultra-fast charging application scenarios"). After determination, the system moves or logically associates all process parameter curves under this batch identification code from the original database to a dedicated data partition for the corresponding application scenario for storage. For example, all electrode batch curves for battery packs used in "ultra-fast charging scenarios" are centrally stored in the same analysis library, thus laying the foundation for subsequent scenario-based data analysis.

[0024] S3: Obtain the statistical failure lifetime of the electrode batch corresponding to the terminal product model, and extract the process parameter curves of the relevant batches from the corresponding scenario based on the terminal product model to form an analysis sample set; After a certain market usage period, electrode manufacturers obtain quality feedback data from downstream customers or end-user after-sales channels. This data is usually provided in the form of a list, including the model and serial number of the end product that experienced performance degradation or failure, the production batch identifier of the electrode it was assembled with (derived through disassembly analysis or data statistics), and the average cycle life (i.e., statistical failure life) exhibited by that batch of electrodes in actual vehicles or tests, such as "batch HEA-LOT-2023-08-001 electrode, after an average of 1200 cycles in the NEX-P100 battery pack, its capacity decayed to 80%". Upon receiving this list, the first step is to determine the application scenario (e.g., "ultra-fast charging scenario") based on the end product model. Subsequently, from the dedicated data partition for that scenario, the complete process parameter curves corresponding to all electrode batch identifiers listed in the list are precisely retrieved and extracted. Finally, the process parameter curves of each batch are paired with their corresponding statistical failure life values ​​to form a structured analysis sample set. Each sample in this sample set clearly defines the correspondence between "the production process of a specific fluctuation pattern" and "its actual durability results in a specific scenario".

[0025] S4: Extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch; Each process parameter curve (e.g., melting temperature curve) in the analysis sample set is read and analyzed. First, the curve is preprocessed. Then, a sliding time window algorithm is used to divide the continuous curve into a series of overlapping segments. Within each time window segment, the mean (representing the average level), maximum, and minimum values ​​(representing the fluctuation amplitude) of the process parameter values ​​within that segment are calculated. These three statistics are then bound to the center time point of the window, forming a feature vector describing the fluctuation state at that moment. After traversing the entire curve, a quantitative fluctuation feature sequence composed of time series feature vectors is obtained. Subsequently, multivariate statistical analysis methods (such as multivariate linear regression) are used, with the fluctuation feature sequences of all samples as the independent variable matrix and the corresponding statistical failure lifetime as the dependent variable vector, for modeling and analysis. This analysis aims to calculate the regression coefficients and their statistical significance (p-value) between the mean, maximum, minimum values ​​and lifetime at each time point in the fluctuation characteristic sequence, thereby identifying which fluctuation characteristics of the parameters (such as "the average temperature near the 300th second" or "the maximum vacuum value near the 450th second") have a stable and significant positive or negative correlation with the long lifetime of the electrode during specific process periods.

[0026] S5: Based on the correlation analysis, define the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as the optimized process characteristics. Generate the corresponding production control parameter set based on the optimized process characteristics. Based on the results of S4 correlation analysis (i.e., regression coefficients and p-values), highly significant positively correlated feature points are selected. For each such feature point (e.g., "average temperature at the 300th second of the melting stage"), all electrode batches exhibiting values ​​favorable for long lifespan at that feature point are identified from the analysis sample set. By statistically analyzing the distribution of actual parameter values ​​of these "top-performing" batches at that feature point, an optimal numerical range (e.g., "1550°C-1565°C") is determined. By summarizing all the selected key feature points scattered across different process periods and their corresponding optimal numerical ranges, an optimized process feature template for this application scenario is formed. This template describes "the production curve should be guided to which numerical range at which key control points to achieve a long lifespan in this scenario." Next, process engineers convert this feature template into an executable file. For example, the requirement of "average temperature range of 1550-1565°C at the 300th second" is converted into specific parameters for the PID control loop setpoints, heating rate limits, and overshoot tolerance range of the melting furnace during the corresponding time period. All these equipment control parameters, adjusted to reproduce the optimized features, are packaged into a set of production control parameters that are tied to the application scenario.

[0027] S6: Determine the application scenario based on the terminal product model specified in the production order, and execute production by calling the production control parameter set corresponding to the application scenario.

[0028] When a new production order is received, it clearly specifies the end product model for which the produced electrodes will be used (e.g., "for the NEX-P200 battery pack"). The production planning system automatically parses this model and matches it to its application scenario according to predefined rules (e.g., matching it to "ultra-fast charging scenario"). Subsequently, a production instruction is issued to the Manufacturing Execution System (MES), which automatically retrieves the production control parameter set bound to that scenario from the parameter library. The MES downloads this parameter set to the production line controller in the corresponding workshop (e.g., the PLC for the melting furnace, the DCS for the heat treatment furnace). The production equipment will operate according to this optimized parameter set, rather than standard generic parameters, thereby producing high-entropy alloy electrodes that are pre-adapted to the performance requirements of the target application scenario in their technological genes.

[0029] It is worth noting that within the acceptable process window, different batches of high-entropy alloy electrodes will inevitably exhibit differentiated fluctuations in process parameters. The root cause lies in the subtle differences in the physical properties of the raw materials, the time-varying characteristics of the dynamic response of the production equipment, and the nonlinear coupling effects between multiple process parameters. These fluctuations are not meaningless noise, but rather key input signals that directly interfere with melt flow, solidification nucleation, atomic diffusion, and phase transformation kinetics. Different fluctuation patterns (such as the smoothness of the temperature profile and the residence stability in critical regions) will "write" different microstructural features such as grain size, compositional segregation, and second-phase distribution, thus fundamentally determining the differences in the final electrochemical performance and service life of the electrode. Based on this principle, this invention constructs a data closed loop of "process parameter curve - application scenario - statistical failure life" to systematically analyze the statistical correlation between different fluctuation patterns and terminal service performance within the qualified window, identify "beneficial fluctuation characteristics" that positively contribute to the long life of specific scenarios, and finally transform the traditionally passively constrained process window into an intelligent control target that can actively guide and reproduce optimized fluctuation patterns by generating a scenario-based production control parameter set. This achieves a paradigm shift from producing "statistically qualified products" to manufacturing "scenario-adapted optimal products".

[0030] In a preferred embodiment of the present invention, the specific process of forming the process parameter curve indexed by the batch identifier in step S1 is as follows: By deploying data acquisition interfaces in each production unit, the original operating condition signals of the production equipment during the corresponding batch production period are acquired in real time; the original operating condition signals are converted from analog to digital and timestamped to generate discrete process parameter data points with time series. The data points are resampled and smoothed at equal intervals according to a preset sampling period to form a continuous process parameter data sequence. The data sequence is then bound to the batch identifier assigned to the corresponding production batch to obtain a process parameter curve with the batch identifier as the primary key index.

[0031] In another preferred embodiment of the present invention, the specific process of calculating the failure lifetime in step S3 is as follows: Receive a list provided by a downstream manufacturer containing the serial numbers of terminal products and their corresponding models; read the production start date and final failure date corresponding to the serial number of the terminal product; calculate the difference between the final failure date and the production start date to obtain the failure life of the terminal product; Based on the assembly association between the terminal product serial number and the electrode batch identifier, the failure life of the terminal product is classified into the corresponding electrode batch identifier; the failure life of all terminal products belonging to the same electrode batch identifier is calculated by arithmetic mean to obtain the statistical failure life of the electrode batch identifier; and an association record is established between the electrode batch identifier, the terminal product model and the statistical failure life.

[0032] First, the system receives an electronic list provided periodically or event-triggered by downstream manufacturers (such as an electric vehicle company). The core content of this list is the unique serial number and corresponding product model of the confirmed failed or retired end products (such as a batch of battery packs). After reading the serial number, the system automatically queries its source—the downstream manufacturer's product lifecycle management database—to obtain two key timestamps for each serial number: one is the "production activation date" after the product is first put into use after rolling off the production line, marking the start of its lifecycle; the other is the "final failure date" when the product is taken out of service due to performance degradation to a threshold or malfunction, marking the end of its lifecycle. The system calculates the difference between these two dates, based on the principle that the time interval directly corresponds to the actual duration from product deployment to loss of function, thus obtaining the specific failure lifecycle data for that end product. Subsequently, the system needs to link these scattered product lifespan data with upstream electrode production batches. This is based on an "assembly association table" recorded and provided by downstream manufacturers during the assembly process. This table clearly records the batch identifier of the specific high-entropy alloy electrode used in the production and assembly of each end-product serial number. By matching this table, the system can accurately categorize the failure lifespan value of each end-product under the production batch identifier that provided it with the electrode raw materials. Next, for the failure lifespans of all end-products belonging to the same electrode batch identifier, the system performs an arithmetic mean calculation. The principle is to integrate the performance of the batch electrode across multiple end-products by calculating the average value, thereby obtaining a single, stable statistical characteristic value that can represent the overall lifespan level of the batch, namely, the "statistical failure lifespan." Finally, the system establishes and saves the calculated statistical failure lifespan result, along with the batch identifier and the end-product model information to which it was applied, as a linked record. This provides a complete index foundation for subsequent data analysis by product and scenario.

[0033] This process transforms fragmented, case-by-case product failure information from the downstream market into structured, batch-based performance characterization data for upstream production analysis. By averaging the lifespan of multiple end products, data noise caused by extreme abnormal usage conditions or accidental failures of individual products can be effectively smoothed out, thereby extracting statistical characteristics that more accurately reflect the quality and durability level of the electrode batch itself. This step serves as a bridge connecting "actual market performance" and "internal process parameters," transforming abstract market reputation for "durability" or "premature aging" into a numerical indicator that can be quantitatively correlated with specific production process curves. Only after this transformation is completed can subsequent steps analyzing the impact of process fluctuations on lifespan have a solid realistic foundation, making it possible for the entire method to ultimately achieve the goal of "optimizing production processes based on service performance." Otherwise, optimization will lose direction and lack empirical support.

[0034] In another preferred embodiment of the present invention, the specific process of obtaining the quantified fluctuation characteristics in step S4 is as follows: Read each process parameter curve in the analysis sample set; divide the process parameter curve into time windows with a preset duration; within each time window, extract the corresponding process parameter data segment, and calculate the mean, maximum, and minimum values ​​of each process parameter data segment; combine the time sequence identifier of the time window, the mean, the maximum, and the minimum values ​​into a quadruple data structure; sort all the quadruple data structures corresponding to a process parameter curve according to the time sequence identifier to form the quantitative fluctuation characteristics of the process parameter curve.

[0035] First, the system reads and analyzes the sample set. This means retrieving complete process parameter records for all relevant electrode batches in a specific application scenario from the categorized and stored data. For example, it retrieves continuous temperature change data for a batch of electrodes used in ultra-fast charging batteries during the vacuum melting process. Next, instead of directly processing the entire long curve, the system divides this continuous time curve into a series of consecutive or partially overlapping time windows at fixed intervals (e.g., every 30 seconds). The principle behind this is to discretize the continuous process into comparable standardized segments, allowing subsequent analysis to pinpoint specific process stages rather than the overall process. Within each defined time window, the system extracts the corresponding process parameter data segment, such as a 30-second set of temperature readings. It then calculates the arithmetic mean of all values ​​within this segment to represent the overall temperature level for that period. Simultaneously, it identifies the maximum (maximum) and minimum (minimum) values ​​within the segment to capture extreme temperature fluctuations. These three values ​​are calculated because the average reflects the steady state, while the maximum... The combination of the minimum and maximum values ​​clearly defines the range of fluctuations during that period, collectively describing the core statistical characteristics of that window. After the calculation is completed, the system combines the position identifier of the time window in the overall production time sequence (i.e., the time sequence identifier, such as representing the 300th to 330th second after the start of production), the calculated mean, maximum, and minimum values, into a structured data unit (i.e., a quadruple data structure). The principle of this combination is to bind "when it occurs" with "how the fluctuation occurs," ensuring that each feature point has a clear time attribute. Finally, the system sorts and concatenates all such quadruple data units corresponding to the same process parameter curve, generated in chronological order. The principle is to restore the original occurrence sequence of these feature points on the real production time axis, thereby forming a trajectory composed of discrete feature points that can characterize the fluctuation pattern of the original continuous curve. This trajectory is the quantitative fluctuation characteristic of the process parameter curve, which transforms a continuous curve that is difficult to calculate directly into a series of discrete points sorted by time that characterize local statistical properties.

[0036] By dividing the data into fixed windows and extracting statistics, the amount of data is significantly reduced, overcoming the "curse of dimensionality." This approach effectively extracts and retains core information characterizing process stability and fluctuation amplitude. Furthermore, by binding time-series identifiers, the temporal location information of fluctuation characteristics within the production process is fully preserved, which is crucial for identifying critical process windows. This approach directly serves the ultimate goal of the solution because it creates a common language—transforming "physical fluctuations in the process" into "computable numerical characteristics." Only by completing this transformation can the system use statistical tools to discover the hidden and complex correlation patterns between these fluctuation numerical characteristics and another key figure (statistical failure lifetime). Ultimately, this allows the system to extract key process knowledge that determines the long-term performance of products from massive amounts of production data and provides clear, quantifiable targets for proactively controlling process fluctuations.

[0037] In another preferred embodiment of the present invention, the specific process of performing correlation analysis between the quantified fluctuation characteristics and the statistical failure lifetime of the corresponding batch in step S4 is as follows: Using the quantitative fluctuation characteristics identified by each electrode batch as input variables and the statistical failure lifetime identified by that batch as output variables, a set of data pairs for correlation analysis is constructed; the values ​​of all quantitative fluctuation characteristics in the set of data pairs are standardized. Using a linear regression model, a correlation analysis is performed on the standardized quantitative fluctuation characteristics and the statistical failure lifetime to obtain the regression coefficients and p-values ​​corresponding to the mean, maximum, and minimum values ​​at each time position in the feature sequence. The mean, maximum, or minimum values ​​and their time positions with p-values ​​lower than a preset correlation threshold are selected, and the time positions are grouped into continuous time intervals according to the continuity of the time axis. The regression coefficients corresponding to the mean, maximum, or minimum values ​​selected within the time intervals are recorded together with the time intervals as a correlation list.

[0038] First, each electrode batch is treated as an independent analysis sample. The complete characteristic sequence (i.e., all quaternion data arranged in chronological order) transformed from the process parameter curves of that batch is used as a set of input variables. Simultaneously, the single statistical failure lifetime value calculated from downstream feedback for that batch is used as the corresponding output variable. Pairing these two variables establishes a correspondence between "cause" (process characteristics) and "effect" (lifetime performance), thus forming a data set suitable for mathematical model analysis. Next, the system standardizes the values ​​of all characteristic sequences in this set. Specifically, it calculates the overall mean and standard deviation of all values ​​at each time point (e.g., the mean of the "300-second window" for all batches). Then, each raw value is subtracted from the mean and divided by the standard deviation. This eliminates the incomparability caused by differences in units and absolute values ​​between different process parameters (such as temperature and vacuum), and transforms all characteristic data to a zero-based standard. The standard scale with a mean and standard deviation of 1 ensures that subsequent analyses can fairly measure the relative importance of each feature's impact on lifespan, without erroneously amplifying the influence of a particular parameter simply because its value is large. The system then analyzes this standardized dataset using a linear regression model. This model aims to find a set of weighted coefficients (i.e., regression coefficients) that best predict the output statistical failure lifespan from the linearly weighted combination of the input feature sequences. By solving this model, the system calculates a regression coefficient and a corresponding p-value for each specific statistic at each specific time point in the feature sequence (e.g., the mean at the 300th second, the maximum at the 300th second). The sign and magnitude of the regression coefficient indicate whether the feature is positively or negatively correlated with lifespan and the strength of the correlation. The p-value assesses whether the correlation is statistically significant, rather than accidental. Based on the p-value, the system sets a threshold for determining significance (e.g., 0).05), and automatically filters out feature points with p-values ​​below this threshold. This means that the correlation between these feature points and lifetime is highly unlikely to be random error and is a noteworthy "signal." Then, the system checks the continuity of these filtered significant feature points scattered across the time axis according to their positions on the original production time axis, merging adjacent or close points together to form several continuous time intervals (e.g., "second 280 to second 320"). The principle behind this is that the impact of the production process is usually a physicochemical process that lasts for a period of time, rather than an instantaneous effect. Merging intervals is more in line with engineering reality and facilitates the subsequent formulation of control strategies. Finally, the system creates a list of correlations. This list not only records each identified significant time interval but also clearly lists which type of statistic (mean, maximum, or minimum) is significant within that interval, along with its corresponding regression coefficient. This comprehensively records "when, what process characteristic, and what fluctuation trend produced a statistically significant and directional correlation with lifetime."

[0039] In another preferred embodiment of the present invention, the specific process of obtaining the optimized process features in step S5 is as follows: Based on the aforementioned list of relationships, filter the time series intervals where the regression coefficient of each electrode batch identifier is positive, and the corresponding mean, maximum, and minimum values. For each time point within each time series interval, electrode batch identifiers with positively correlated regression coefficients at that time point are selected from the analysis sample set, and the batch identifier with the largest statistical failure lifetime value is selected from these batch identifiers. Extract the mean, maximum and minimum values ​​of the process parameter curves of the selected electrode batch at the corresponding time points, and use them as the reference mean, reference maximum and reference minimum values ​​at that time point, respectively; form a reference mean sequence based on the reference mean of all time points, a reference maximum sequence based on the reference maximum of all time points, and a reference minimum sequence based on the reference minimum of all time points. The mean optimization interval boundary value for each time point in the time series interval is determined according to the reference mean sequence, the upper limit benchmark value for maximum optimization for each time point in the time axis is determined according to the reference maximum sequence, and the lower limit benchmark value for minimum optimization for each time point in the time axis is determined according to the reference minimum sequence. By summing the boundary values ​​of the mean optimization interval, the upper limit benchmark value of the maximum optimization, and the lower limit benchmark value of the minimum optimization, the optimized process characteristics are obtained.

[0040] First, based on the correlation list generated in the previous step, all entries with positive regression coefficients are selected. The principle is that a positive regression coefficient indicates a positive correlation between the process characteristic at that point (such as the average temperature at a certain time point) and the statistical failure life; that is, the larger the characteristic value, the longer the life tends to be. After selection, a set of process characteristic points that contribute positively to lifespan are obtained, scattered across different electrode batches and different time points. Next, in order to construct a complete, continuous, and optimal process characteristic template, an optimal reference value needs to be determined at each key time point. The specific approach is as follows: for each specific time point (e.g., the 305th second) within each positively correlated time series interval identified in the list, the sample set is retrieved and analyzed to find all electrode batch identifiers with positively correlated regression coefficients at that time point. This means that the process performance of these batches at that point has been confirmed to be related to long lifespan. Then, from these candidate batches, the batch identifier with the largest actual statistical failure life value is selected. The principle is to select the "longest-lived" batch as the benchmark, which can be considered as the "best" among all the "good students" in terms of process parameter performance at this point, and is the most representative practice. Subsequently, the system extracts the mean, maximum, and minimum values ​​calculated at that specific time point from the original process parameter curve corresponding to the selected optimal batch identifier. These three values ​​are recorded as the reference mean, reference maximum, and reference minimum values ​​at that time point, respectively. The principle behind this is to simultaneously capture the typical level (mean) and fluctuation range (maximum and minimum values) of the optimal batch at that moment, providing complete information for setting subsequent control targets. The system repeats the above "filtering-selection-extraction" operation at all covered time points and connects the reference mean values ​​obtained at each time point in chronological order to form a "reference mean sequence" describing the ideal average level. Similarly, all reference maximum and reference minimum values ​​are connected separately to form a "reference maximum sequence" and a "reference minimum sequence." Next, specific optimization objectives are formulated based on these reference sequences: For the mean, the system uses the reference mean sequence as a benchmark, setting a reasonable fluctuation range above and below it as the "mean optimization interval boundary value." The principle is to allow production to fluctuate within a small range around the optimal mean to maintain controllability. For the maxima and minima, the system directly uses the values ​​of the reference maxima and minima sequences as the "maximum optimization upper limit benchmark value" and the "minimum optimization lower limit benchmark value." The principle is to explicitly require that instantaneous fluctuations in the production process should not exceed these upper and lower limit boundaries defined by optimal batch practices. Finally, the three types of information—the mean optimization interval boundary value, the maximum optimization upper limit benchmark value, and the minimum optimization lower limit benchmark value—are summarized and encapsulated across all time points on the entire production timeline. The resulting complete dataset is the "optimized process characteristic," which defines an ideal process parameter fluctuation channel with a reasonable tolerance range.

[0041] By integrating and refining the fragmented "advantages" statistically discovered across different high-performing batches, a complete and executable "gold standard" process path is constructed. This avoids the risks of randomness or incomplete data that may arise from relying solely on a single optimal batch. Instead, it adopts a "broad-based" approach, selecting parameter values ​​at each time point that are verified to be related to long lifespan and perform best in long-life batches, thus synthesizing a theoretically more robust and reliable optimization target. This approach plays a crucial role in achieving the ultimate goal of the solution: it connects the objective laws discovered through correlation analysis with the generation of the control parameter set, serving as a key design step in transforming "data insights" into "production instructions." By generating such an optimized process characteristic, a specific and quantifiable catch-up target is provided for the production line, transforming "optimizing the process for a specific scenario" from an abstract concept into an engineering problem with clear numerical constraints. This makes subsequent precise control and production reproducibility possible, truly closing the intelligent loop from service performance feedback to production process optimization.

[0042] In another preferred embodiment of the present invention, the specific process of setting the production control parameter set in step S5 is as follows: Read the mean optimization interval boundary value, the maximum optimization upper limit benchmark value, and the minimum optimization lower limit benchmark value recorded in the optimized process features; and calibrate the mean optimization interval boundary value as the target process parameter at the corresponding time point. The upper limit benchmark value for the maximum value optimization is directly set as the upper boundary of the operating range of the process parameters at the corresponding time point; the lower limit benchmark value for the minimum value optimization is directly set as the lower boundary of the operating range of the process parameters at the corresponding time point. The target process parameters, the upper and lower boundaries of the process parameter operating range are combined to form a closed-loop control parameter set for the corresponding time point; the closed-loop control parameter sets for each time point on the time axis are arranged and encapsulated according to the time sequence to form the production control parameter set bound to the application scenario.

[0043] See Figure 2 The present invention also includes a high-entropy alloy electrode production management system based on quality traceability, used to implement the above-described high-entropy alloy electrode production management method based on quality traceability, comprising: The data acquisition module is used to assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier. The scenario classification module is used to obtain the terminal product model assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios. The sample acquisition module is used to obtain the statistical failure lifetime of electrode batches corresponding to the terminal product model, and extract the process parameter curves of relevant batches from the corresponding scenarios based on the terminal product model to form an analysis sample set. The data analysis module is used to extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and to perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch. The parameter generation module is used to define, based on the correlation analysis, the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as an optimized process characteristic, and generate a corresponding set of production control parameters based on the optimized process characteristic. The production control module is used to determine the application scenario based on the terminal product model specified in the production order, and to execute production by calling the production control parameter set corresponding to that application scenario.

[0044] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A production management method for high-entropy alloy electrodes based on quality traceability, characterized in that, Includes the following steps: S1: Assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier; S2: Obtain the model of the terminal product assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify the application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios; S3: Obtain the statistical failure lifetime of the electrode batch corresponding to the terminal product model, and extract the process parameter curves of the relevant batches from the corresponding scenario based on the terminal product model to form an analysis sample set; S4: Extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch; S5: Based on the correlation analysis, define the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as the optimized process characteristics. Generate the corresponding production control parameter set based on the optimized process characteristics. S6: Determine the application scenario based on the terminal product model specified in the production order, and execute production by calling the production control parameter set corresponding to the application scenario.

2. The method for production management of high-entropy alloy electrodes based on quality traceability according to claim 1, characterized in that, In step S1, the specific process for forming the process parameter curve indexed by the batch identifier is as follows: By deploying data acquisition interfaces in each production unit, the original operating condition signals of the production equipment during the corresponding batch production period are acquired in real time; the original operating condition signals are converted from analog to digital and timestamped to generate discrete process parameter data points with time series. The data points are resampled and smoothed at equal intervals according to a preset sampling period to form a continuous sequence of process parameter data. The data sequence is bound to the batch identifier assigned to the corresponding production batch to obtain the process parameter curve with the batch identifier as the primary key index.

3. The method for production management of high-entropy alloy electrodes based on quality traceability according to claim 1, characterized in that, In S3, the specific process for calculating the failure lifetime is as follows: Receive a list provided by a downstream manufacturer containing the serial numbers of terminal products and their corresponding models; read the production start date and final failure date corresponding to the serial number of the terminal product; calculate the difference between the final failure date and the production start date to obtain the failure life of the terminal product; Based on the assembly association between the terminal product serial number and the electrode batch identifier, the failure life of the terminal product is classified into the corresponding electrode batch identifier; the failure life of all terminal products belonging to the same electrode batch identifier is calculated by arithmetic mean to obtain the statistical failure life of the electrode batch identifier. Establish a record linking the electrode batch identifier, the terminal product model, and the statistical failure lifetime.

4. The method for production management of high-entropy alloy electrodes based on quality traceability according to claim 1, characterized in that, In step S4, the specific process for obtaining the quantified fluctuation characteristics is as follows: Read each process parameter curve in the analysis sample set; divide the process parameter curve into time windows with a preset duration; within each time window, extract the corresponding process parameter data segment and calculate the mean, maximum and minimum values ​​of each process parameter data segment within the window; The time sequence identifier of the time window, the mean, the maximum value and the minimum value are combined into a quadruple data structure; All the quadruple data structures corresponding to a process parameter curve are sorted according to the time sequence identifier to form the quantitative fluctuation characteristics of the process parameter curve.

5. The method for managing the production of high-entropy alloy electrodes based on quality traceability according to claim 4, characterized in that, In step S4, the specific process of correlating the quantitative fluctuation characteristics with the statistical failure lifetime of the corresponding batch is as follows: Using the quantitative fluctuation characteristics identified by each electrode batch as input variables and the statistical failure lifetime identified by that batch as output variables, a set of data pairs for correlation analysis is constructed; the values ​​of all quantitative fluctuation characteristics in the set of data pairs are standardized. Using a linear regression model, a correlation analysis is performed on the standardized quantitative fluctuation characteristics and the statistical failure lifetime to obtain the regression coefficients and p-values ​​corresponding to the mean, maximum, and minimum values ​​at each time position in the quantitative fluctuation characteristics; mean, maximum, or minimum values ​​and their time positions with p-values ​​lower than a preset correlation threshold are selected, and the time positions are grouped into continuous time intervals according to the continuity of the time axis; The regression coefficients corresponding to the mean, maximum, or minimum values ​​selected within the time series interval are recorded together with the time series interval as a list of correlation relationships.

6. The method for production management of high-entropy alloy electrodes based on quality traceability according to claim 5, characterized in that, In step S5, the specific process for obtaining the optimized process features is as follows: Based on the aforementioned list of relationships, filter the time series intervals where the regression coefficient of each electrode batch identifier is positive, and the corresponding mean, maximum, and minimum values. For each time point within each time series interval, electrode batch identifiers with positively correlated regression coefficients at that time point are selected from the analysis sample set, and the batch identifier with the largest statistical failure lifetime value is selected from these batch identifiers. Extract the mean, maximum, and minimum values ​​of the process parameter curves for the selected electrode batch identifier at the corresponding time points, and use them as the reference mean, reference maximum, and reference minimum values ​​for that time point, respectively. A reference mean sequence is formed based on the reference mean at all time points; a reference maximum sequence is formed based on the reference maximum at all time points; and a reference minimum sequence is formed based on the reference minimum at all time points. The mean optimization interval boundary value for each time point in the time series interval is determined according to the reference mean sequence, the upper limit benchmark value for maximum optimization for each time point in the time axis is determined according to the reference maximum sequence, and the lower limit benchmark value for minimum optimization for each time point in the time axis is determined according to the reference minimum sequence. By summing the boundary values ​​of the mean optimization interval, the upper limit benchmark value of the maximum optimization, and the lower limit benchmark value of the minimum optimization, the optimized process characteristics are obtained.

7. The method for production management of high-entropy alloy electrodes based on quality traceability according to claim 6, characterized in that, In S5, the specific process of setting the production control parameter set is as follows; Read the mean optimization interval boundary value, maximum optimization upper limit benchmark value, and minimum optimization lower limit benchmark value recorded in the optimized process features; and calibrate the mean optimization interval boundary value as the target process parameter at the corresponding time point; The maximum value optimization upper limit benchmark value is directly set as the upper boundary of the operating range of the process parameters at the corresponding time point; The minimum value optimization lower limit benchmark is directly set as the lower boundary of the operating range of the process parameters at the corresponding time point; The target process parameters, the upper and lower boundaries of the process parameter operating range are combined to form a closed-loop control parameter set for the corresponding time point; the closed-loop control parameter sets for each time point on the time axis are arranged and encapsulated according to the time sequence to form the production control parameter set bound to the application scenario.

8. A high-entropy alloy electrode production management system based on quality traceability, used to implement the high-entropy alloy electrode production management method based on quality traceability as described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to assign a batch identifier to each production batch of high-entropy alloy electrodes and collect process parameters during the production process to form a process parameter curve indexed by the batch identifier. The scenario classification module is used to obtain the terminal product model assembled by the batch identifier, associate the batch identifier with the obtained terminal product model, classify application scenarios according to the terminal product model, and classify and store the process parameter curves according to the application scenarios. The sample acquisition module is used to obtain the statistical failure lifetime of electrode batches corresponding to the terminal product model, and extract the process parameter curves of relevant batches from the corresponding scenarios based on the terminal product model to form an analysis sample set. The data analysis module is used to extract quantitative fluctuation characteristics from the process parameter curves of the analysis sample set, and to perform correlation analysis between the quantitative fluctuation characteristics and the statistical failure lifetime of the corresponding batch. The parameter generation module is used to define, based on the correlation analysis, the numerical range in the application scenario where the quantitative fluctuation characteristics and the statistical failure lifetime are correlated and use it as an optimized process characteristic, and generate a corresponding set of production control parameters based on the optimized process characteristic. The production control module is used to determine the application scenario based on the terminal product model specified in the production order, and to execute production by calling the production control parameter set corresponding to that application scenario.

Images