Blockchain-based battery pole piece quality data traceability method and system

Abstract

The application relates to the technical field of data processing methods, in particular to a battery pole piece quality data tracing method and system based on a blockchain, wherein the battery pole piece quality data tracing method based on the blockchain comprises the following steps: acquiring processing parameters and corresponding quality inspection data of each working procedure in a battery pole piece production process, and writing the processing parameters and the quality inspection data into a blockchain according to the production time sequence of the same pole piece product; determining a working procedure risk index of each working procedure; determining a risk exceeding index of a working procedure section; and determining an abnormal source working procedure. Through the combination of a three-layer progressive analysis mode of a blockchain credible evidence base and single-working-procedure responsibility risk quantification, risk accumulation transmission and prediction between working procedures, and abnormal source positioning driven by prediction deviation, the cumulative quality risk caused by cross-working-procedure parameter nonlinear coupling in the battery pole piece production process is identified, predicted and accurately traced, and the accuracy and reliability of quality tracing are improved.

Description

Technical Field

[0001] This invention relates to the field of data processing methods, specifically to a blockchain-based method and system for tracing battery electrode quality data. Background Technology

[0002] In the high-energy-density manufacturing of lithium-ion batteries, increasing electrode thickness and compaction density are effective ways to improve battery capacity. However, thick electrodes can lead to difficulties in electrolyte wetting, severely affecting the battery's rate performance and cycle life. To address this, high-end battery manufacturing has introduced laser-based hole-forming technology, which creates electrolyte penetration channels by processing specific blind hole structures on the electrode. This technology requires not only extremely high micro-dimensional control but also stringent safety requirements: the inner wall of the hole needs high roughness to retain the electrolyte, while the outer surface roughness of the electrode must be extremely low, absolutely prohibiting any metal spatter or protrusions caused by laser melting. The electrode production process typically involves multiple independent manufacturing stages, such as coating, cold pressing, laser drilling, and final quality inspection, in series. The process parameters and test results of each stage are recorded in the local manufacturing execution system database of each workshop, and are macroscopically transferred and traced through batch numbers or roll numbers. In recent years, blockchain technology, with its immutable and chain-based evidence storage characteristics, has been gradually introduced into the field of industrial quality traceability to ensure the credibility and integrity of cross-process data.

[0003] However, existing technologies using blockchain to record pass / fail quality management behaviors with discrete Boolean values, while ensuring that process parameters for each step are independently compliant within their respective tolerance zones, cannot identify the cumulative quality risks caused by the nonlinear coupling of parameters across multiple steps, such as coating thickness and laser energy. Consequently, after the occurrence of defective products, existing traceability processes rely solely on batch numbers or roll numbers for macroscopic discrete backtracking, failing to accurately reconstruct the processing environment at the time based on the parameter coupling relationships between steps, nor to precisely locate the true source step causing the implicit coupling anomaly. Ultimately, this affects the accuracy of production control and quality traceability for electrode products. Summary of the Invention

[0004] This invention provides a blockchain-based method and system for tracing battery electrode quality data to solve existing problems.

[0005] The present invention provides a blockchain-based method and system for tracing battery electrode quality data, which adopts the following technical solution: In a first aspect, one embodiment of the present invention provides a blockchain-based method for tracing battery electrode quality data. The method includes: acquiring processing parameters and corresponding quality inspection data for each process in the battery electrode production process; writing the processing parameters and the quality inspection data into a blockchain according to the production sequence of the same electrode product; and establishing a correlation between the processing parameters, the quality inspection data, and the production sequence; determining a process risk index for each process based on the degree of deviation of the processing parameters of each process from the process benchmark and the degree of deterioration of the process quality inspection results relative to the qualified standard; wherein the process quality inspection results are determined based on the quality inspection data; accumulating the process risk indices of each process along the production sequence, and determining the risk exceedance index of the process segment by combining the absorption and correction capabilities of subsequent processes for the accumulated risks of the previous process; and determining the abnormal source process by combining the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, and the relative contribution of the process risk index of each process within the process segment.

[0006] Furthermore, the step of determining the anomaly source process based on the deviation between the actual defect level detected by the end-of-line quality inspection and the cumulative risk exceeding index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, includes: for each process within the process segment, calculating the deviation between the actual defect level detected by the end-of-line quality inspection and the risk exceeding index up to the current process; determining the anomaly contribution of each process based on the relative contribution of the process risk index of each process within the process segment; determining the anomaly coupling degree of each process based on the deviation and the anomaly contribution degree; and determining the anomaly source process from the process segment based on the anomaly coupling degree.

[0007] Furthermore, determining the abnormal contribution of each process based on the relative contribution of the process risk index of each process within the process segment includes: calculating the variance of the process risk index of each process within the process segment; using the variance as an amplification factor, and combining it with the proportion of the process risk index of each process to the sum of the process risk indices of all processes within the process segment, to determine the abnormal contribution of each process.

[0008] Furthermore, the end-of-line quality inspection includes using a three-dimensional morphology inspection device to inspect the aperture, blind hole depth, surface roughness, and spatter size of the electrode product. The actual degree of defect is determined based on the ratio of the detected spatter size to the normal allowable size.

[0009] Furthermore, along the production sequence, the process risk index of each process is accumulated, and combined with the absorption and correction capability of subsequent processes to the accumulated risks of the previous period, the risk excess index of the process segment is determined. This includes: along the production sequence, amplifying and mapping the process risk indices of each process within the process segment according to the process distance, and then superimposing them to determine the accumulated abnormal risk of the process segment; determining the absorption and correction capability of the subsequent processes to the accumulated risks of the previous period based on the proportion of the duration of each subsequent process in the process segment to the total process duration and the historical risk index of the subsequent processes; and determining the risk excess index of the process segment based on the accumulated abnormal risk and the absorption and correction capability.

[0010] Furthermore, the method further includes: determining whether the risk exceedance index of the process segment exceeds a preset threshold; wherein the preset threshold is determined based on the absorption and correction capability of subsequent processes for accumulated risks in the early stage; in response to the risk exceedance index exceeding the preset threshold, outputting risk warning information before the accumulated abnormal risk of the process segment actually exceeds the qualification standard corresponding to the quality inspection data.

[0011] Furthermore, the step of writing the processing parameters and the quality inspection data into the blockchain according to the production timeline of the same electrode product includes: generating a current block according to the production timeline, the current block containing the processing parameters, the quality inspection data, and the hash value of the previous block; connecting the current blockchain to the previous block to form a hash chain structure; wherein, when the data contained in the current block is modified, the hash value of the current block changes, causing the hash value of the current block stored in subsequent blocks to change.

[0012] Furthermore, the method also includes: extracting the processing parameters and quality inspection data corresponding to the abnormal source process from the blockchain; performing multi-dimensional visualization comparison of the processing parameters and the quality inspection data to render the formation process of the implicit coupling defect; packaging the traceability evidence, including the visualization comparison results, the responsibility sharing ratio, and the source parameters of the abnormal source process, into an abnormal event token, and writing the abnormal event token into the blockchain for storage.

[0013] Furthermore, the acquisition of processing parameters and corresponding quality inspection data for each process in the battery electrode production process includes: using the electrode production length coordinate as the index reference for the production sequence, determining each displacement event in which the electrode raw material moves by one finished electrode length as a collection moment; at each collection moment, collecting the processing parameters and corresponding quality inspection data for each process; and establishing the positional association between the processing parameters and the quality inspection data and the same electrode product according to the continuous order of the electrode production length coordinate.

[0014] Secondly, another embodiment of the present invention provides a blockchain-based battery electrode quality data traceability system, including a host computer and inspection equipment, wherein: The inspection equipment is deployed at each process stage of the battery electrode production line to extract the processing parameters and corresponding quality inspection data of each process, and to send the processing parameters and the quality inspection data to the host computer. The host computer is used to acquire the processing parameters and corresponding quality inspection data of each process in the battery electrode production process, write the processing parameters and quality inspection data into the blockchain according to the production sequence of the same electrode product, and establish the association between the processing parameters, the quality inspection data and the production sequence; based on the deviation of the processing parameters of each process from the process benchmark and the deterioration of the process quality inspection results from the qualified standard, the process risk index of each process is determined; wherein, the process quality inspection results are determined based on the quality inspection data; along the production sequence, the process risk index of each process is accumulated, and combined with the absorption and correction ability of subsequent processes to the accumulated risks of the previous process, the risk exceedance index of the process segment is determined; based on the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, the abnormal source process is determined.

[0015] The beneficial effects of the technical solution of the present invention are: In this embodiment of the invention, processing parameters and corresponding quality inspection data of each process in the battery electrode production process are obtained. The processing parameters and quality inspection data are written into the blockchain according to the production sequence of the same electrode product, and the association between the processing parameters and quality inspection data and the production sequence is established. Based on the degree of deviation of the processing parameters of each process from the process benchmark and the degree of deterioration of the process quality inspection results from the qualified standard, the process risk index of each process is determined. The process quality inspection results are determined based on the quality inspection data. Along the production sequence, the process risk index of each process is accumulated, and combined with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process, the risk exceedance index of the process segment is determined. Based on the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, the abnormal source process is determined.

[0016] This invention, through a three-tiered progressive analysis model combining a blockchain-based trusted evidence storage foundation with single-process responsibility risk quantification, inter-process risk accumulation and transmission prediction, and anomaly source localization driven by prediction deviation, achieves the identification, prediction, and precise traceability of cumulative quality risks caused by nonlinear coupling of cross-process parameters in battery electrode production, thus improving the accuracy and reliability of quality traceability. Furthermore, by quantifying the coupling relationship between the deviation of each process's processing parameters from the process baseline and the degree of deterioration of the process inspection results relative to the pass / fail standard, it can identify hidden quality risks even when single-process parameters are within the compliance tolerance zone, overcoming the limitations of traditional discrete Boolean value pass / fail records in characterizing parameters. The limitations of deviation from the limit and the trend of quality deterioration; on the other hand, by accumulating the risk index of a single process along the electrode production sequence and combining it with the absorption and correction capabilities of subsequent processes for the accumulated risks of the previous stage, it is possible to proactively perceive the superposition and amplification effect of nonlinear coupling of multi-process parameters in the flow process, and to provide early warning before the accumulated abnormal risks actually exceed the qualified standards; on the other hand, by comparing the deviation between the degree of abnormality detected by the actual quality inspection at the end and the predicted value of the risk accumulation in the process segment, and combining the relative contribution of the risk index of each process in the process segment, it is possible to effectively screen out the processing operation of subsequent processes on abnormal electrodes from the source of abnormality, and accurately locate the real abnormal source process that causes the hidden coupling defects. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the blockchain-based battery electrode quality data traceability method provided in this application embodiment; Figure 2 A flowchart illustrating a blockchain-based battery electrode quality data traceability method in a specific application scenario provided in this application embodiment; Figure 3 This is a schematic diagram of the architecture of a blockchain-based battery electrode quality data traceability system provided in an embodiment of this application. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, provides a detailed account of the specific implementation methods, structures, features, and effects of the method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] The following description, in conjunction with the accompanying drawings, details the specific scheme of the blockchain-based battery electrode quality data traceability method provided by this invention.

[0022] like Figure 1 As shown in the embodiment of this application, a method for tracing battery electrode quality data based on blockchain is provided. The method includes: Step S110: Obtain the processing parameters and corresponding quality inspection data of each process in the battery electrode production process, write the processing parameters and quality inspection data into the blockchain according to the production time sequence of the same electrode product, and establish the association relationship between the processing parameters and quality inspection data and the production time sequence.

[0023] The aforementioned battery electrode production process typically includes the following steps: In the front-end workshop, electrode slurry is coated and cold-pressed to form an initial electrode with a specific thickness and compaction density. The electrode roll is then transferred to the middle workshop, where laser processing equipment performs high-speed drilling based on set parameters such as wavelength, pulse width, and single-pulse energy to create specific blind hole structures on the electrode. Afterward, the electrode enters subsequent processing steps, including rolling and drying, which are used to control the electrode surface morphology and adjust the coating size, respectively. Finally, at the end of the production line, a three-dimensional morphology inspection device is used to perform random or full inspections of the hole diameter, blind hole depth, and surface roughness, completing the final quality verification of the electrode product.

[0024] In the battery electrode production process, the processing parameters of the front-end coating and cold pressing process, including coating thickness and compaction roller pressure, can be extracted in real time by the production equipment control system in the front-end workshop. The corresponding quality inspection data, including electrode thickness and compaction density, can be obtained by online thickness gauges and density detection devices. The processing parameters of the mid-stage laser drilling process, including laser power, wavelength, pulse width, and single pulse energy, can be directly read by the control unit of the laser processing equipment. The corresponding quality inspection data, including hole morphology and hole diameter, can be collected in real time by the visual contour initial inspection system built into the laser station. The processing parameters of the subsequent rolling, smoothing, and drying processes, including rolling pressure and drying temperature, can be obtained by the sensors and controllers of the corresponding equipment. The corresponding quality inspection data, including surface flatness and coating size, can be collected by online surface inspection instruments and dimensional measurement devices. The quality inspection data of the final quality inspection process, including hole diameter, blind hole depth, surface roughness, and spatter size, can be obtained by the three-dimensional morphology inspection equipment at the end of the production line through sampling or full inspection. After obtaining the corresponding data, the above-mentioned equipment will send it to the electronic equipment that executes the above-mentioned blockchain-based battery electrode quality data traceability method.

[0025] The above scheme uses the electrode production length coordinates as the index reference and triggers data acquisition based on the displacement events of finished electrode products. This achieves a precise correlation between processing parameters and quality inspection data relative to the position of the same electrode product, thus solving the technical problem that traditional discrete batch traceability cannot locate the specific processing environment of a single electrode product at each process stage. Specifically, the above acquisition of processing parameters and corresponding quality inspection data for each process in battery electrode production includes: using the electrode production length coordinates as the index reference for the production time sequence, determining each displacement event of the electrode raw material moving one finished electrode length as a data acquisition moment; at each data acquisition moment, collecting the processing parameters and corresponding quality inspection data for each process; and establishing the positional correlation between processing parameters and quality inspection data and the same electrode product according to the continuous order of the electrode production length coordinates.

[0026] The aforementioned electrode production length coordinates refer to one-dimensional spatial coordinates based on the physical displacement of the electrode roll on the production line. Typically, the origin is taken as the front end of the roll or the starting point of the first finished electrode, and the coordinates increase along the electrode travel direction. The finished electrode length refers to the nominal dimension of a single finished electrode along the longitudinal direction of the roll, which is determined by the electrode design specifications. The displacement event refers to the physical displacement action of the electrode raw material moving one finished electrode length along the travel direction under the drive of the production line conveyor mechanism. The acquisition time refers to the data sampling time point defined with the displacement event as the trigger condition, representing that the electrode product at the corresponding physical location has entered the detectable state. The electrode production length coordinates can be obtained in real time through rotary encoders, laser rangefinders, or machine vision positioning devices deployed on coating machines or unwinding mechanisms. The rotary encoder measures the number of pulses that drive the roller to rotate and converts this into the electrode travel length by combining the roller diameter parameters. The laser rangefinder directly measures the spatial displacement of the electrode edge or marked points. The machine vision positioning device performs absolute position calibration by recognizing preset tab marks or QR codes on the electrode surface. The determination of displacement events can be achieved by the programmable logic controller generating a hardware interrupt trigger signal when the cumulative travel length reaches an integer multiple of the finished electrode length, or by outputting a trigger command based on the real-time roll diameter calculation model when the arc length corresponding to the circumferential rotation angle of the roll material is equal to the finished electrode length. At each acquisition moment, the inspection equipment at each process end responds to the above trigger signals, synchronously executes the reading of processing parameters and the acquisition of quality inspection data, and transmits the acquired data to the host computer after labeling it with the current length coordinate.

[0027] Using the electrode production length coordinate as the index benchmark for the production timeline essentially maps the time-dimensional data sequence to the spatial-dimensional position sequence. Since the transmission speeds of electrodes differ across processes such as coating, cold pressing, laser drilling, and final quality inspection, relying solely on absolute timestamps cannot ensure that data for the same electrode product at different processes corresponds to the same physical entity. However, the length coordinate is strictly bound to the physical spatial location of the electrode product, eliminating time drift errors caused by speed fluctuations or buffering between processes. Using a displacement event equal to the length of a finished electrode as the acquisition interval ensures that the acquisition granularity is consistent with the final... Product unit alignment ensures that the data set corresponding to each acquisition moment can be traced back to a single finished electrode, avoiding the mixing of multiple electrode data under macro-batch number tracing. The positional association between processing parameters and quality inspection data is established according to the continuous order of electrode production length coordinates. In essence, this reconstructs the spatial-data mapping of electrode products along the production line at the data level, enabling the host computer to associate the front-end coating thickness, mid-section laser energy, and end-end morphology detection results with the same longitudinal position of the same electrode product based on a unified length coordinate index. This provides a precise spatial alignment basis for cross-process parameter coupling analysis and anomaly source location.

[0028] It is understandable that in the laser drilling production scenario for battery electrodes, although the process parameters of each process are independently compliant within their respective tolerance zones, the nonlinear coupling of multiple process parameters such as coating thickness and laser energy may cause the electrode quality to gradually deteriorate and exceed safety thresholds during the transfer process. If the processing parameters and quality inspection data of each process are only stored discretely in the local manufacturing execution system database of each workshop, it is impossible to establish a temporal correspondence between the parameters and quality inspection results of the same electrode product at different process positions. Consequently, it is impossible to trace and analyze the cross-process parameter coupling effect, and it is also impossible to accurately reconstruct the processing environment of the electrode at each process position and locate the real source of the anomaly after an abnormal quality product is produced. Therefore, writing the processing parameters and quality inspection data of each process into the blockchain according to the production timeline of the same electrode product can leverage the temporal evidence characteristics and immutability of the blockchain to establish a temporal association chain of parameters and quality inspection data for the entire process of the same electrode product, from coating, cold pressing, laser drilling to final quality inspection. This provides a reliable data foundation for cross-process coupling risk identification, accurate location of abnormal processes, and quality traceability.

[0029] It is understandable that the above scheme provides tamper-proof, time-series evidence for cross-process parameters and quality inspection data through the hash chain structure of the blockchain. This ensures that any unauthorized modification of historical data in any process will trigger a chain verification anomaly, thereby guaranteeing the authenticity and integrity of the quality traceability data throughout the entire process. Specifically, the above-mentioned method of writing processing parameters and quality inspection data into the blockchain according to the production timeline of the same electrode product includes: generating the current block according to the production timeline, the current block containing processing parameters, quality inspection data, and the hash value of the previous block; connecting the current blockchain to the previous block to form a hash chain structure; wherein, when the data contained in the current block is modified, the hash value of the current block changes, causing the hash value of the current block stored in subsequent blocks to change.

[0030] The current block refers to the most recently generated data storage unit in the blockchain according to the production sequence of the electrode sheets. It encapsulates the processing parameters and quality inspection data corresponding to the acquisition time, and carries the hash value of the previous block as a chain association identifier. The previous block refers to the adjacent block that was recorded on the chain before the current block was generated; it is the direct parent node of the current block. The hash value is a fixed-length digest string calculated from the block header data using a cryptographic hash algorithm, possessing uniqueness and collision resistance. The hash chain structure refers to a singly linked list data structure formed by connecting multiple blocks sequentially through their hash values. Modification of any block will break its hash association with subsequent blocks.

[0031] In practical implementation, the generation of the current block can be triggered by the displacement event of the electrode raw material moving by the length of a finished electrode each time. The inspection equipment deployed at each process end packages the collected processing parameters and quality inspection data into block body data, and extracts the hash digest of the previous block header data and writes it into the current block header, thereby establishing a time-series link. The hash value of the previous block can be calculated using cryptographic hash algorithms such as SHA-256 or the national cryptographic SM3. The calculation object can cover the previous block header field, timestamp, data fingerprint, and the previous block's own hash value, to ensure the unforgeability of the chain association. Blockchain connection operations can be implemented by setting the previous block hash pointer field in the current block header. This pointer field directly stores the hash value of the previous block, making the current block and the previous block form an explicit parent-child relationship.

[0032] When the processing parameters or quality inspection data contained in the current block are illegally modified, due to the avalanche effect of the cryptographic hash algorithm, even if the data undergoes bit-level changes, the hash value recalculated by the current block will produce a completely different output result. The header of subsequent blocks still stores the original hash value of the current block before it was modified. This causes inconsistencies when subsequent blocks verify the hash value of their predecessor, thus triggering a chain verification failure. This verification mechanism based on the hash chain structure requires that the tampering behavior must start from the tampered block and recalculate the hash values ​​of all blocks one by one, and synchronously modify the hash pointers of subsequent blocks. In a blockchain network with multi-node consensus, a single node cannot complete the above operation without triggering a network-wide consensus verification anomaly, thus ensuring the authenticity and integrity of the time-series evidence of cross-process parameters and quality inspection data from a technical perspective.

[0033] In addition to single-chain blockchains with hash chain structures, the above solutions can also adopt distributed ledgers based on directed acyclic graph structures, using multiple parallel branches to record data flows from different processes or different electrode rolls, thereby improving the efficiency of concurrent writing by multiple nodes; alternatively, a consortium blockchain architecture can be adopted, with the front-end coating workshop, the mid-stage laser drilling workshop, and the final quality inspection workshop acting as authorized nodes to form a consensus network, ensuring data immutability while enabling cross-departmental data sharing with controllable permissions; or a sharded blockchain architecture can be adopted, distributing the massive time-series data of each process across different shards and verifying it through cross-shard protocols to reduce the data storage pressure on a single chain.

[0034] After the data is uploaded to the blockchain, the relationship between processing parameters and quality inspection data and the production timeline can be established by embedding production timeline identifiers in the block structure. Specifically, when generating the current block, the length coordinate label, timestamp, and unique identifier of the electrode product corresponding to the collection time are written into the metadata field of the current block, and the processing parameters and quality inspection data of each process collected at the same collection time are encapsulated in the storage data body of the current block, so that the processing parameters and quality inspection data within the same block share the same production timeline identifier. Then, by retrieving the corresponding current block based on the production timeline identifier through the blockchain query interface, the processing parameters and quality inspection data corresponding to the time of the production timeline identifier can be extracted. In addition, since the current block forms a chain link with the previous block through the hash value of the previous block, the chain association between adjacent blocks implicitly constitutes a continuous association relationship between processing parameters and quality inspection data along the production timeline, enabling the host computer to realize cross-process data timeline traceability and association query on the blockchain based on the production timeline identifier.

[0035] Step S120: Determine the process risk index for each process based on the degree of deviation of the processing parameters of each process from the process baseline and the degree of deterioration of the process quality inspection results from the qualified standard; wherein, the process quality inspection results are determined based on the quality inspection data.

[0036] The aforementioned process quality inspection results refer to the quantitative indicators obtained by the inspection equipment deployed at each process stage when the electrode products are transferred to each process position. These indicators represent the degree of deviation of the product quality status at the current processing stage from the qualified standard. The indicators are determined by comparing the actual test values ​​with the qualified standard of the process based on the quality inspection data of the corresponding process. For example, in the laser drilling process, the visual contour initial inspection system of the laser station can extract the hole morphology data and compare it with the hole morphology standard to calculate the amount of hole morphology degradation. Alternatively, the three-dimensional morphology inspection equipment can obtain geometric morphology data such as hole diameter, blind hole depth and surface roughness and calculate the deviation with the corresponding standard value. The aforementioned degradation amount or deviation is then determined as the process quality inspection result to quantify the deterioration trend of the electrode product quality at the current process position.

[0037] The aforementioned process risk index is a comprehensive assessment indicator used to quantify the implicit quality risks existing in the current process under the premise of compliance with processing parameters. It quantifies the risk level that may affect the quality of the current process position by coupling two dimensions: deviation of processing parameters and quality inspection deterioration. Among them, the process benchmark refers to the optimal processing parameter setting value that makes the electrode product quality reach the best state under specific process conditions, and can be used as a reference center for assessing the degree of deviation of actual processing parameters; the degree of deviation refers to the ratio of the absolute deviation of the actual processing parameters relative to the process benchmark to the length of the maximum allowable range of the parameters, which is used to characterize the relative degree to which the parameters are pushed to the limit within the tolerance zone; the acceptance standard refers to the quality acceptance limit set for the local initial inspection indicators of each process; the degree of deterioration refers to the amount of deterioration of the local initial inspection indicators in the process quality inspection results that are worse than the acceptance standard, and it only participates in the risk coupling calculation when the amount of deterioration is positive.

[0038] When determining the process risk index, the process benchmark can be selected based on the optimal distribution center value of parameters obtained from historical production big data statistics, or the preferred process parameter settings explicitly specified in the process specification, or the optimal benchmark value can be dynamically calculated based on the design specifications of the current batch of electrode products in a multi-variety, small-batch production scenario. The degree of deviation can be calculated by dividing the absolute value of the difference between the actual processing parameters and the process benchmark by the length of the maximum allowable range to obtain the dimensionless relative deviation, or it can be characterized by a standard deviation multiple or a relative percentage. The degree of degradation can be calculated for different local initial inspection indicators such as hole morphology, surface roughness, or hole diameter, by calculating the absolute value of the difference between the actual detection value and the upper limit of the acceptable standard, or by calculating the relative deviation rate of the actual value relative to the standard value. The coupling method between processing parameter deviation and quality inspection degradation can use a multiplicative exponential coupling model, or a weighted sum model or a piecewise threshold model. However, the multiplicative exponential coupling model can produce a risk amplification effect when parameter deviation and quality inspection degradation coexist, and is more suitable for identifying implicit coupling risks. The following provides one method for calculating the process risk index:

[0039] in, This represents the process risk index at the current process location, and is dimensionless. The actual processing parameters of the current process are extracted from the blockchain, and their dimensions are consistent with the specific process parameter types (e.g., the dimension of laser energy is joule, and the dimension of compaction roller pressure is megapascal). The optimal reference values ​​for the current process parameters are the preferred process parameters, with dimensions equal to... Consistent; The maximum allowable range length of the current processing parameters, with units of... Consistent; This represents the deterioration of the local initial inspection index at the current location relative to the pass standard. Its dimension is consistent with that of the corresponding quality inspection index (e.g., the dimension of surface roughness deterioration is micrometers, and the dimension of pore size deviation is millimeters). Before participating in the exponential calculation, it is usually normalized to a dimensionless quantity. This means that the deterioration amount is taken only when the local initial inspection index is worse than the standard; otherwise, it is taken as zero. This ensures that if the quality inspection result is better than or equal to the standard, it does not amplify the risk index. The calculation principle of the above process risk index is that deviations in processing parameters within the tolerance band do not necessarily lead to defective products, because parameter deviations may be reasonable adjustments to adapt to the properties of the electrode material. Therefore, it is necessary to identify whether the current parameter deviation is a normal process adaptation or an abnormal parameter drift through the process quality inspection results. When the processing parameters deviate from the process baseline and local initial inspection indicators deteriorate, the index term... This will generate an amplification factor greater than 1, making the process risk index significantly higher than the simple parameter deviation, thereby forcibly amplifying the coupled abnormal characteristics of parameter deviation and quality inspection deterioration occurring simultaneously. Conversely, if the processing parameters deviate but the quality inspection results do not deteriorate, the index term is 1, and the risk index only reflects the baseline level of parameter deviation. This achieves differentiated quantification of hidden quality risks in a single process, providing a single-point risk assessment basis for subsequent cross-process risk accumulation and anomaly source location.

[0040] Step S130: Along the production sequence, accumulate the process risk index of each process, and combine it with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process to determine the risk exceeding the index of the process segment.

[0041] The aforementioned process segment refers to a continuous evaluation window consisting of multiple adjacent processes along the production sequence in the battery electrode production process. This window includes not only several preceding processes to be evaluated, but also a set of subsequent processes following the preceding processes. It is used to superimpose and accumulate the process risk index of the preceding processes along the time sequence to form the cumulative abnormal risk of the process segment. Combined with the absorption and correction capabilities of the subsequent process set for the accumulated risks of the previous period, the net risk exceeding the limit within the window is evaluated, thereby providing an evaluation unit for predicting the process position where the accumulated risk is about to exceed the correction capability threshold of the subsequent processes.

[0042] It is understandable that the above scheme can amplify and map the process risk indices of each process within the process segment according to the process distance along the electrode production sequence, and then superimpose them to form a risk accumulation state as the process evolves. Combined with the absorption and correction capabilities of subsequent processes based on their duration proportion and historical risk performance, the net excess degree of accumulated abnormal risk within the process segment is assessed. Specifically, step S130 includes: amplifying and mapping the process risk indices of each process within the process segment according to the process distance along the production sequence, and then superimposing them to determine the accumulated abnormal risk of the process segment; determining the absorption and correction capabilities of subsequent processes for the accumulated risk based on the duration proportion of each subsequent process in the process segment to the total process time and the historical risk index of the subsequent processes; and determining the risk excess index of the process segment based on the accumulated abnormal risk and the absorption and correction capabilities.

[0043] The aforementioned process distance refers to the difference between the index position of a process within a process segment and the index position of the process at the window termination boundary. It is used to quantify the spatial attenuation or amplification effect of risk during process flow. Amplification mapping refers to a mathematical operation that uses process distance as the independent variable and a specific mapping function to weight and amplify the process risk index, characterizing the physical process of early, small risks gradually accumulating and amplifying as the electrode moves. Cumulative abnormal risk refers to the sum of the process risk indices of all processes within a process segment after amplification mapping, representing the total quality deviation caused by the coupling of parameter deviations in multiple preceding processes. The set of subsequent processes refers to the set of several consecutive processes located after the window termination boundary of a process segment that have a mitigating or corrective effect on the cumulative abnormal risks generated by preceding processes. Absorption and correction capability refers to the comprehensive fault tolerance capability of the subsequent process set in digesting, compensating for, or suppressing the accumulated abnormal risks from the previous period based on its own process characteristics and historical risk performance. The risk excess index refers to the difference between the cumulative abnormal risk and the absorption and correction capability, used to quantify the degree to which the net risk within a process segment exceeds the fault tolerance threshold of subsequent processes.

[0044] The boundaries of the process segment window can be dynamically set according to the physical coupling characteristics of the electrode production process. For example, coating, cold pressing, and laser drilling, three strongly coupled processes, can be divided into the same process segment, or adjacent processes with significant thermo-mechanical coupling effects can be included in the same evaluation window. Alternatively, a fixed window length strategy can be adopted, that is, to evaluate each segment along the production sequence in a sliding window of every three or five consecutive processes. The amplification mapping can use an exponential distance decay model, which uses a natural exponential function to weight the process risk index by distance. A linear decay model or a power function decay model can also be used, but the exponential model can more sensitively amplify the cumulative effect of early process risks after long-distance transfer, and is more suitable for nonlinear processes with small parameter deviations in electrode production scenarios. The risk amplification characteristic; the definition of the subsequent process set can be based on the range of subsequent fault-tolerant processes clearly specified in the process specification, or adaptively selected by analyzing the success rate of each process in correcting previous anomalies in historical production data; the proportion of duration to total process time can be determined by statistically analyzing the ratio of the single-piece processing time of each subsequent process under standard production cycle time to the total process time, or it can be characterized by the equivalent conversion ratio of equipment running time, energy consumption ratio, or material dwell time of each process; the historical risk index can be obtained by retrieving the process risk index sequence of the subsequent process stored on the blockchain within the past preset time window and calculating its statistical mean or weighted moving average to reflect the typical risk level of the process in long-term operation. The following introduces one method for calculating the risk exceedance index:

[0045] In the formula, The risk of continuous process segment windows exceeds the exponential level, which is dimensionless; The set of processes encompassed by the current process segment window; For process segment windows A specific process within; For process segment windows The termination boundary process index is dimensionless. For process At the electrode position The process risk index at the location is dimensionless. Based on the natural constant and the process distance It is an exponential amplification mapping function, dimensionless, whose value varies with the process. Termination boundary process The exponential growth in distance between them is used to characterize the physical amplification effect of early process risks during the transfer process; For the current process segment window The set of subsequent processes; For subsequent process sets A specific subsequent process in the process; For subsequent processes The duration of the process is a dimensionless percentage of the total process time. For subsequent processes The historical risk index is dimensionless.

[0046] The calculation principle of the aforementioned risk excess index is as follows: the deviation risk generated by a single process does not disappear as the process proceeds, but is transmitted to subsequent processes as the electrode moves and may gradually amplify. Therefore, it is necessary to amplify and map the risk indices of each preceding process within the process segment according to the process distance and then superimpose them along the time sequence to simulate the superposition and physical amplification effect of multiple early minor risks such as slight deviations in coating thickness and high laser energy during the transfer process, thereby obtaining the total quality deviation generated by the process segment. At the same time, subsequent processes have a certain fault tolerance capability. For example, the rolling and smoothing process can redistribute the stress field on the electrode surface through mechanical pressure, and the drying process can bring the coating size back to the standard range through thermal field control. Therefore, it is necessary to extract the ability of subsequent processes to mitigate previous anomalies in order to determine whether the cumulative anomaly risk generated by the current process segment has exceeded the threshold that subsequent processes can correct and absorb. When the cumulative anomaly risk is greater than the absorption and correction capability of subsequent processes, the risk excess index shows a high positive value, indicating that the current local process risk is strong. Based on this, the process position where the cumulative risk is about to break through the correction capability threshold of subsequent processes can be predicted, thereby providing a basis for advance perception and early intervention of process coupling risks between processes.

[0047] Step S140: Based on the deviation between the actual degree of defects detected by the end quality inspection and the cumulative risk exceeding the index of the process segment, and combined with the relative contribution of the process risk index of each process within the process segment, determine the source process of the anomaly.

[0048] It is understandable that the above scheme can construct an anomaly coupling degree index by comparing the deviation between the actual degree of anomaly detected by the final quality inspection and the theoretical cumulative risk prediction value at the end of each process segment, and by combining the relative contribution of the process risk index of each process within the process segment. This effectively filters out the processing operations of subsequent processes on the abnormal electrode from the source of anomalies, so as to accurately locate the real source process of the anomaly that causes the latent coupling defect. Specifically, the above step S140 includes: for each process within the process segment, calculating the deviation between the actual degree of defect detected by the final quality inspection and the risk exceeding index at the end of the current process; determining the anomaly contribution of each process based on the relative contribution of the process risk index of each process within the process segment; determining the anomaly coupling degree of each process based on the deviation and the anomaly contribution degree; and identifying the source process of the anomaly from the process segment based on the anomaly coupling degree.

[0049] The aforementioned actual defect severity refers to the quantitative indicator characterizing the severity of defects obtained by the final quality inspection process through full or sampling inspection of the final quality status of electrode products using three-dimensional morphology detection equipment; deviation refers to the absolute difference between the actual defect severity detected by the final quality inspection and the theoretically predicted risk exceeding index at the end of the process segment, used to quantify the degree of deviation between the theoretically predicted risk of the process segment and the actual inspection results at the end; anomaly contribution refers to the weighted proportion of the relative contribution of each process's process risk index within the process segment after variance amplification, used to characterize the responsibility share of the process in quality anomalies within the window; anomaly coupling degree refers to the comprehensive index obtained by multiplying the above deviation and anomaly contribution degree, used to assess whether the current process is a coupling anomaly point that leads to a sharp increase in risk within the process segment; anomaly source process refers to the process extracted from the process segment through anomaly coupling degree that plays a leading role in triggering implicit coupling defects.

[0050] The aforementioned end-of-line quality inspection utilizes 3D topography inspection equipment to perform end-of-line geometric topography inspection on the aperture, blind hole depth, surface roughness, and spatter size of electrode products. This captures the final defect state of the electrode products after the entire processing cycle. The ratio of spatter size to the normal allowable size is used as a quantitative representation of the actual defect severity, providing an end-of-line verification benchmark for comparing the deviation between theoretically predicted risks and actual inspection results in anomaly source location. Specifically, the aforementioned end-of-line quality inspection includes using 3D topography inspection equipment to inspect the aperture, blind hole depth, surface roughness, and spatter size of electrode products. The actual defect severity is determined based on the ratio of the detected spatter size to the normal allowable size.

[0051] The aforementioned three-dimensional morphology inspection equipment refers to a high-precision non-contact surface measurement instrument based on the principles of optical interference, confocal microscopy, or structured light projection. It reconstructs the microscopic geometric morphology by acquiring three-dimensional point cloud data of the electrode surface and pore structure, enabling high-precision quantitative detection of parameters such as pore diameter, blind hole depth, surface roughness, and spatter size. Pore diameter refers to the diameter of the blind hole opening formed by laser-induced pore formation on the electrode surface, a key geometric parameter characterizing the cross-sectional area of ​​the electrolyte permeation channel. Blind hole depth refers to the vertical distance from the bottom of the blind hole to the electrode surface; its process requirement is to reach the current collector without damaging it. Ensure the electrolyte wets the current collector interface; surface roughness refers to the quantitative index of the unevenness of the micro-geometric morphology of the outer surface of the electrode, usually characterized by the arithmetic mean deviation or root mean square deviation, reflecting the smoothness of the outer surface of the electrode after laser processing; spatter size refers to the geometric size of the protrusions formed by the splashing and solidification of metal droplets on the surface of the electrode during laser thermal melting, including the protrusion height and equivalent diameter; actual defect degree refers to the index used by the final quality inspection process to quantitatively assess the severity of the final quality abnormality of the electrode product, which is defined in this scheme as the ratio of the detected spatter size to the normal allowable size. Three-dimensional morphology inspection equipment can employ white light interferometers, laser confocal microscopes, or structured light 3D scanners. White light interferometers are suitable for high-precision measurement of sub-micron surface roughness, laser confocal microscopes are suitable for 3D reconstruction of the internal morphology of deep holes, and structured light 3D scanners are suitable for rapid morphology acquisition of large-area electrode surfaces. Aperture and blind hole depth can be obtained by extracting the hole edge contour from 3D point cloud data and calculating using least-squares circle fitting or depth slicing algorithms. Surface roughness can be obtained by calculating the arithmetic mean deviation or root mean square deviation of the surface height distribution from 3D point cloud data of the electrode's outer surface. The size of the spatter can be determined by... By performing blob analysis or morphological processing on the three-dimensional topography data, the surface protrusion areas are extracted and their maximum height or equivalent sphere diameter is calculated; the final quality inspection can adopt a full inspection mode to scan each electrode sheet one by one, or a sampling inspection mode to sample and inspect specific locations of the electrode sheet roll at a preset sampling frequency; the actual defect degree can be calculated by the ratio of the maximum height of the detected spatter to the normal maximum allowable height, or by the ratio of the equivalent volume of the spatter to the normal allowable equivalent volume. However, using the size ratio can eliminate the dimensional influence caused by the difference in specifications of different batches of electrode sheets, making the defect degree a dimensionless relative quantity.The principle behind the selection of the aforementioned end-of-life quality inspection parameters is as follows: pore size and blind hole depth directly determine the effectiveness of electrolyte permeation channels and are fundamental quality characteristics of the laser-induced hole formation process; surface roughness reflects the microscopic morphology of the electrode's outer surface after laser processing. Excessive surface roughness increases contact friction with the separator during winding or stacking, but the surface roughness of the electrode itself is not a direct cause of internal micro-short circuits; spatter is a hard protrusion formed by the splashing and solidification of molten metal droplets on the electrode surface during laser melting. It is highly likely to pierce the separator after electrode winding or stacking, and is a direct physical defect that can cause internal micro-short circuits or even thermal runaway explosions in the battery; therefore, using the ratio of spatter size to the normal allowable size as the actual defect severity can quantify the severity of end-of-life quality inspection anomalies into a dimensionless index directly related to the electrode safety red line. This allows the actual defect severity to be compared across dimensions with the risk exceedance index theoretically predicted by the process segment, providing an end-of-life verification benchmark directly corresponding to physical safety risks for anomaly source location.

[0052] The above scheme, when calculating the abnormal contribution of each process, introduces the variance of the process risk index within the process segment as an amplification factor. This amplifies the proportion of each process's risk index to the total risk index of the process segment, thereby forcibly highlighting the abnormal characteristics of the truly abnormal processes that cause a sharp increase in risk within the process segment. Therefore, even if anomalies generated by preceding processes are masked by normal processing of the already abnormal electrode by subsequent processes, the abnormal process can still be effectively identified through the amplified abnormal contribution. Specifically, determining the abnormal contribution of each process based on the relative contribution of its process risk index within the process segment includes: calculating the variance of the process risk index of each process within the process segment; using the variance as an amplification factor, and combining it with the proportion of each process's process risk index to the total process risk index of all processes within the process segment, to determine the abnormal contribution of each process.

[0053] The aforementioned anomaly contribution is a weighted evaluation index used to quantify the responsibility share of each process within a process segment in the overall quality anomaly. It combines the proportion of the process risk index of a single process to the total of the process segments with the dispersion of the process risk index distribution within the process segment to achieve differentiated amplification of the abnormal characteristics of abnormal processes. Variance refers to the squared average of the deviations of the process risk index of all processes within a process segment from its arithmetic mean, used to characterize the dispersion of the risk level of each process within the process segment. Amplification factor refers to the value of an exponential function with the natural constant as the base and the variance of the process risk index of the process segment as the exponent, used to nonlinearly amplify the proportional term. Proportion refers to the ratio of the process risk index of the current process to the sum of the process risk indices of all processes within the process segment, representing the basic responsibility share of the process under unamplified conditions. Variance can be calculated using the population variance formula, which is the average of the sum of squares of the differences between the process risk index and the arithmetic mean of each process segment within the process segment. Alternatively, it can be calculated using the sample variance formula with an unbiased estimation method. In a sliding window scenario, moving variance can be used to reflect the time-varying characteristics of risk distribution within the process segment. The base of the amplification factor, besides the natural constant, can also be a logarithmic linear transformation with base 2 or 10. However, the natural exponential function produces a more sensitive amplification effect as variance increases, making it more suitable for identifying outliers with sharp increases in risk within the process segment. The proportion can be calculated directly... The arithmetic ratio of the current process's risk index divided by the sum of the process risk indices of all processes within the process segment can be used, or a weighted ratio can be used, where the weights can be set according to the proportion of each process in the total process time or the process importance coefficient. The determination of abnormal contribution can be achieved by first substituting the variance calculation result into an exponential function to obtain an amplification factor, then multiplying this amplification factor by the current process's risk index and dividing by the sum of the process segment's risk indices. Alternatively, a logarithmic-linear combination or a power function combination can be used, but the multiplicative exponential coupling method can produce a synergistic effect between variance amplification and proportional amplification. The following describes one method for calculating abnormal contribution: In the formula, For process At the electrode position The process risk index at the location is dimensionless. For process segment The variance of the process risk index for each process is dimensionless. It is a dimensionless amplification factor with the natural constant as the base and variance as the exponent. For process segment The sum of the process risk indices of all processes, dimensionless, where the summation subscript is... Traversing process segments All processes within the process segment. The calculation principle of the above-mentioned abnormal contribution is as follows: when the process risk index of each process within a process segment is relatively evenly distributed and no significant abnormalities occur, the variance value is small, and the amplification factor approaches 1. At this time, the abnormal contribution of each process is approximately equal to the basic proportion of its process risk index to the total of the process segment, and the responsibility share among processes is naturally distributed. When a certain process within a process segment generates an abnormally high process risk index due to the coupling of parameter limit deviation and quality inspection deterioration, this abnormal value will significantly increase the dispersion of the risk index within the process segment, causing the variance to increase sharply, and thus leading to an amplification factor much greater than 1. Exponential growth; because the amplification factor acts on the current process risk index, the abnormal contribution of this process will be forcibly amplified to a level far higher than its base ratio. Even if the absolute value of the risk index of this process does not show extreme abnormality under the correction effect of subsequent processes, its amplified abnormal contribution can still dominate the abnormal coupling degree. Thus, it is possible to break through the abnormal masking effect caused by the normal processing of abnormal electrodes by subsequent processes, effectively extract the previous abnormal process that actually caused the sharp increase in risk from the process segment, and avoid its abnormal characteristics being buried by the normalization trend of the overall window.

[0054] When calculating deviations, for each process segment, the actual degree of defects detected by the end quality inspection can be compared point by point with the risk exceeding index at the point of that process, and the absolute value of the difference between the two can be taken as the predicted deviation of the process position. When determining the anomaly contribution, the variance of the process risk index of each process in the process segment can be calculated as an amplification factor. Combined with the proportion of the process risk index of each process to the sum of the process risk indices of all processes in the process segment, the anomaly contribution of that process can be obtained. The variance amplification factor is used to forcibly amplify the abnormal characteristics of processes that lead to a sharp increase in risk. When determining the anomaly source process from the process segment, the anomaly coupling degree can be normalized to a value range of [0,1] using the sigmoid function, and processes with a normalized coupling degree greater than a preset threshold can be extracted as anomaly source processes. The preset threshold can be set according to the historical anomaly statistical distribution of the production line.

[0055] The following describes one method for calculating abnormal coupling degree:

[0056] In the formula, For the current process The anomalous coupling degree is dimensionless; The actual degree of defect detected by the end-of-line quality inspection is dimensionless and is defined as the ratio of the size of the spatter actually detected by the three-dimensional morphology detection equipment in the end-of-line quality inspection process to the normal allowable size. For process segment End of process The risk calculated at that time exceeds the index and is dimensionless; This means that the cumulative summation of process segment risks exceeding the index is truncated to the process segment level. The mapping operator.

[0057] In the above scheme, the principle of anomaly source localization is as follows: When an anomaly occurs in a preceding process within a process segment, and subsequent processes continue normal processing on the anomalous electrode, the processing of subsequent processes may mask the anomaly of the preceding process, causing the overall risk of the process segment to exceed the index and instead show a normal trend. In this case, if the judgment is based solely on the cumulative risk prediction value, it will be impossible to distinguish the true source of the anomaly. By introducing the degree of defects detected by the actual quality inspection at the end as a reverse verification benchmark and calculating the deviation between it and the theoretical risk exceedance index up to each process, it is possible to identify the location nodes within the process segment where the theoretical prediction deviates significantly from the actual inspection. Furthermore, by combining the anomaly contribution value to forcibly amplify the risk increase characteristics of the process at that location, even if the anomaly of the preceding process is masked by subsequent processes, the process can still be effectively extracted through high anomaly coupling. Thus, the processing situation of subsequent processes on the anomalous electrode can be screened out from the anomaly source, accurately locating the true anomaly source process that causes the hidden coupling defect, providing a location basis for subsequent quality traceability and responsibility attribution.

[0058] It is understandable that the above scheme dynamically compares the risk exceeding the index of a process segment with a preset threshold determined based on the absorption and correction capabilities of subsequent processes. It outputs risk warning information before the accumulated abnormal risk actually exceeds the quality inspection pass standard, thereby achieving proactive perception of risks coupled across processes. This allows for the reservation of a process intervention window before irreparable defects form in the electrode product, enabling operators to adjust processing parameters in a timely manner to eliminate the root cause of risk. Specifically, the above-mentioned blockchain-based battery electrode quality data traceability method also includes: determining whether the risk exceeding the index of a process segment exceeds a preset threshold; wherein the preset threshold is determined based on the absorption and correction capabilities of subsequent processes for previously accumulated risks; and, in response to the risk exceeding the preset threshold, outputting risk warning information before the accumulated abnormal risk of the process segment actually exceeds the pass standard corresponding to the quality inspection data.

[0059] The aforementioned preset threshold refers to the critical threshold value set by the index for the risk of a process segment. It is used to determine whether the cumulative abnormal risk of the current process segment has approached or exceeded the safety boundary of the absorption and correction capacity of subsequent processes. This threshold is determined based on the absorption and correction capacity of subsequent processes for the accumulated risks of the previous period, and represents the lower limit warning value of the fault tolerance margin of subsequent processes. Risk warning information refers to the notification signal generated before the accumulated abnormal risk actually exceeds the qualified standard corresponding to the quality inspection data. It is used to remind operators or the upper control system that there is a trend of coupled risk getting out of control in the current process segment. Its content may include key information such as the risk exceeding the current value of the index, the predicted location of the process exceeding the limit, and suggested intervention measures.

[0060] The setting of preset thresholds can adopt various strategies: the threshold can be directly set as a fixed percentage of the absorption and correction capability of subsequent processes, such as taking 80% or 90% as a safety margin threshold; the interpolation result between the absorption and correction capability and the historical maximum risk exceedance index can be used as a dynamic threshold; or the standard deviation of the absorption and correction capability can be calculated based on long-term production line operation data through statistical process control methods, and the mean minus three times the standard deviation can be used as the lower limit threshold; whether the risk exceedance index of a process segment exceeds the preset threshold can be determined by numerical comparison calculation, that is, the difference between the risk exceedance index of the current process segment and the preset threshold can be calculated in real time, and when the difference is positive, it is determined to exceed; the output form of risk warning information can include pushing warning work orders to the manufacturing execution system, sending audible and visual alarm signals to the process operation terminal, sending speed reduction or shutdown instructions to the production line controller, or writing the warning information into the blockchain as a pre-stored record of the abnormal event to be confirmed; the warning triggering time is strictly limited to before the accumulated abnormal risk actually exceeds the qualified standard, that is, before the electrode product becomes an irreparable waste, reserving a process intervention window.

[0061] The working principle of the aforementioned proactive early warning mechanism is as follows: the ability of subsequent processes to absorb and correct accumulated risks from earlier processes is not unlimited. When small parameter deviations in multiple preceding processes are gradually amplified through nonlinear coupling, the accumulated abnormal risk will gradually approach the correction limit of subsequent processes along the production sequence. A preset threshold serves as a safety margin warning value for this correction limit, triggering an early warning when the accumulated abnormal risk has exceeded the safety margin but has not actually broken the acceptance standard. This allows the above scheme to issue intervention signals before the electrode products flow to the final quality inspection, or even at the current processing stage of the preceding abnormal process. As a result, operators can adjust the processing parameters of the current process segment in real time based on the early warning information, such as reducing laser energy or adjusting coating thickness, thereby eliminating the root cause of risk before the coupling defect solidifies, preventing abnormal electrodes from continuing to flow to subsequent processes and causing a wider range of quality losses, and achieving proactive perception and prevention of cross-process coupling risks in electrode production.

[0062] It is understandable that the above solution, after accurately locating the source process of the anomaly, extracts the original processing parameters and quality inspection data corresponding to that process from the blockchain. Through multi-dimensional visualization comparison, it intuitively reconstructs the entire process from the generation to the manifestation of the latent coupling defect. The complete traceability evidence, including the visualization comparison results, the responsibility sharing ratio, and the source parameters, is packaged into an anomaly event token and written back into the blockchain. This immutable on-chain notarization method enables full-chain traceability and responsibility attribution for electrode product quality events. Specifically, the above blockchain-based battery electrode quality data traceability method also includes: extracting the processing parameters and quality inspection data corresponding to the source process of the anomaly from the blockchain; performing multi-dimensional visualization comparison of the processing parameters and quality inspection data to render the formation process of the latent coupling defect; packaging the traceability evidence, including the visualization comparison results, the responsibility sharing ratio, and the source parameters of the source process of the anomaly into an anomaly event token, and writing the anomaly event token into the blockchain for notarization.

[0063] The "originating process" refers to the process extracted from the process segment based on its abnormal coupling degree, which plays a leading role in triggering implicit coupling defects. "Tracing evidence" refers to the complete set of data and reports extracted and integrated from the blockchain after locating the originating process, used to reconstruct the full picture of the abnormal event. It includes at least visual comparison results, responsibility allocation ratios, and source parameters. "Visual comparison results" refers to an analysis report that simultaneously displays the fluctuations in processing parameters and the deterioration of quality inspection indicators of the originating process using multi-dimensional graphics. "Responsibility allocation ratio" refers to the quantitative ratio, calculated based on the abnormal contribution of each process within the abnormal process segment, representing the share of responsibility each process bears for the final defect. "Source parameters" refers to the actual processing parameter settings corresponding to the originating process at the time the abnormality occurred. "Abnormal event token" refers to a blockchain digital certificate with unique identification and immutability, formed by standardizing and encapsulating the above-mentioned tracing evidence.

[0064] Extracting processing parameters and quality inspection data corresponding to the source process of an anomaly from the blockchain can be achieved through on-chain retrieval based on the electrode production length coordinates or collection time identifiers: according to the process number of the source process of the anomaly and the time index of when the abnormal electrode entered the process, the current block containing the data of that time is located in the distributed ledger of the blockchain, and the processing parameter field and quality inspection data field are parsed from the stored data body of the block. At the same time, the associated block data of the electrode at the adjacent process position is traced and extracted to construct a complete processing environment dataset of the abnormal electrode at the source process of the anomaly.

[0065] Multi-dimensional visualization comparison of extracted processing parameters and quality inspection data can be performed. This can be achieved by comparing risk accumulation line charts and actual parameter fluctuation charts on the same screen, overlaying the cumulative risk index curves of each process within the process segment with the time-series fluctuation curves of key processing parameters such as laser energy and coating thickness on the same coordinate system. Alternatively, a heat map can be used to render the spatial distribution of abnormal coupling degree of each process within the process segment, or a three-dimensional scatter plot can be used to map the coupling relationship between processing parameter deviation and quality inspection deterioration degree in phase space, thereby intuitively presenting the evolution path of latent coupling defects from parameter deviation to quality deterioration.

[0066] When packaging traceability evidence into anomaly tokens and writing them into the blockchain for evidence storage, the visual comparison results, responsibility sharing ratios, and source parameters can be structured and encoded, and the timestamp of the anomaly event, the identifier of the anomaly source process, the unique number of the involved electrode products, and the digital fingerprint of the anomaly token can be attached. Subsequently, the encoded traceability evidence is submitted to the multi-node consensus through the smart contract of the blockchain network. After consensus verification, it is written into a new block, so that the anomaly token obtains the same immutability and time-series evidence storage effect as the underlying blockchain.

[0067] In the above scheme, the location result of the abnormal source process only provides the spatial coordinates of the abnormality, while complete quality traceability requires reconstructing the generation environment and evolution process of the abnormality in the temporal dimension. By extracting the original processing parameters and quality inspection data of the abnormal source process from the blockchain, it is possible to ensure that the traceability data is consistent with the actual state of the production site, avoiding data tampering or record loss that may exist in traditional local databases. Multi-dimensional visualization comparison transforms the abstract parameter coupling relationship into an intuitive graphical language, enabling quality management personnel to quickly understand the formation mechanism of implicit coupling defects. By packaging traceability evidence into abnormal event tokens and storing them on the blockchain again, the immutability of the blockchain is used to establish a permanent chain of evidence for determining responsibility for quality events, so that subsequent quality audits, process improvements, or supply chain accountability can all be based on trusted on-chain data, thereby improving the accuracy and legal effect of electrode product quality traceability.

[0068] To facilitate understanding of the working principle of the above-described blockchain-based battery electrode quality data traceability method, this application also provides specific application examples of this method in a certain application scenario. For example... Figure 2 As shown, in this application scenario, the above-mentioned blockchain-based battery electrode quality data traceability method mainly includes: Step 1: Extract the processing parameters of each step using the inspection equipment deployed on the production line, store them in real time, and upload them to the blockchain; The production equipment deployed on the electrode production line has pre-defined parameter ranges (laser power, compaction roller pressure, etc.) during production. Using the length coordinates of the electrode production line as a distinguishing reference, each time the raw material moves to the position of a finished electrode is considered a data acquisition point, thus recording the data at each position. All of them can be collected in the current process Multiple internal parameters ,as well as Parameter range Furthermore, according to position The system sequentially assigns a marker to each output value, storing the result at the current marked position. This completes the on-chain storage of parameters in the output results, achieving chained storage of the current data. Each block contains the hash value of the previous block and the stored data of the current block. Chained storage, built upon the blockchain structure, cannot be modified externally. Modifying the stored data of the current block would cause changes to the hash value of the previous block in subsequent blocks, leading to error propagation and thus ensuring the security of the collected operational data.

[0069] Step 2: Evaluate the matching degree between the processing parameters of the process and the process quality inspection results to obtain the process risk index of the current process. The raw data on the blockchain cannot directly indicate the execution result of the current process. Because the process parameters are within acceptable fluctuation ranges, tracing the parameters back to their normal range does not directly reveal an abnormal electrode product at the current process stage. Even if the parameters are at the "tolerance limit" (e.g., high laser energy), if the electrode itself allows it, it won't immediately produce defective products. However, if the parameters in the local quality inspection results show an abnormal fluctuation trend within the acceptable range, it indicates that the quality of the electrode itself has deteriorated in the current processing stage. The processing parameter input sequence on the blockchain is retrieved and matched with the microscopic quality inspection results of the current process (e.g., the initial visual contour inspection from the laser station) to calculate the deviation between the theoretical expected output and the actual quality inspection output. If the parameters are set extremely high but the quality inspection results are mediocre, or if the parameters are normal but the quality inspection fluctuates drastically, this mismatch between the current processing parameter changes and the quality inspection parameter changes is quantified to obtain a risk assessment value describing the potential quality impact at the current process stage. Therefore, based on the changes in the degree of deterioration of product quality inspection indicators at the current process stage, and the deviations of the processing equipment parameters from the compliance range at the current process stage, the risk index for the current process is calculated. Calculate the process risk index for the current process position: .

[0070] Step 3: Calculate the risk exceedance index for consecutive process segment windows; The deviation risk generated by a single process does not disappear as the process continues, but is transmitted to subsequent processes as the electrode moves. Therefore, by assessing the cumulative changes in the risk index between processes, it is possible to determine at which process locations the electrode product might exhibit abnormal quality exceeding the allowable range. This allows for early detection and judgment of risk anomalies formed by the accumulation of process risk indices in the processing steps, enabling early detection of process coupling risks between processes and effectively extracting anomalies in the electrode production process. This is based on the flow sequence (movement position) of the electrode during processing. The process risk indices of each independent process are superimposed along the chain to form an abnormal risk accumulation state. Based on the likelihood of risk exceeding the limits of subsequent processes, a risk accumulation curve is obtained as the process evolves. When the accumulated amount is about to exceed the risk correction capability of subsequent steps, the location of the process where a substantial anomaly is about to occur can be predicted. Therefore, the risk exceeding index of the continuous process segment window is calculated, and the risk exceeding index is assessed based on the risk correction capability (risk value reduction) of subsequent steps within the window. Calculation of the risk exceeding index of the continuous process segment window: .

[0071] Step 4: Based on the difference between the predicted location of the abnormal process and the actual location of the abnormal process, and in conjunction with the cumulative amount of the process risk index, locate the source process of the abnormality. When there is a discrepancy between the predicted location of an abnormal process and the actual location of the abnormality, it indicates that the current risk accumulation process differs from the scale of risk generation in the process. This suggests that subsequent processes have strong risk handling capabilities, and also implies that the abnormal processing risk generated by the abnormal process segment is within the scope of preceding processes. In other words, in actual production, a certain process concealed a highly irregular operation (e.g., the electrode sheet is too thick in the coating process, or the laser drilling process uses high energy compensation, but the subsequent drying process adjusts the electrode sheet coating size to a suitable range). Therefore, based on the distribution results of actual quality inspection on the production line, high-risk process segments in the current process segment window are extracted. This allows for the extraction of the process segments in the continuous process segment window that truly cause electrode sheet abnormalities. Situations where subsequent normal processes process abnormal electrodes, thus causing electrode sheet abnormalities, are eliminated, thereby truly locating the real abnormal process in the current process flow. The abnormal coupling degree of each local process segment in the current window is calculated: Therefore, after normalization with a value range of [0,1] by using the sigmoid function, the processes with a coupling degree A greater than 0.6 are extracted and treated as the source of the anomaly.

[0072] Step 5: Extract and visualize the source process of the anomaly and the process segments that make a significant contribution to the anomaly of the source process, and complete the on-chain storage of the anomaly event. After extracting the current source process of the anomaly, the location of the anomaly is retrieved from the current electrode based on the data stored on the blockchain of the source process. And the location and timing of the current electrode input anomaly source process. The corresponding processing and quality inspection data are extracted and packaged from the blockchain for display, enabling quality traceability of the current electrode production process. Through multi-dimensional graphics (such as a side-by-side comparison of risk accumulation line charts and actual parameter fluctuation charts), the entire process of defects caused by implicit coupling is visually rendered. Subsequently, the complete chain of evidence for determining responsibility, including visual reports, responsibility sharing ratios (risk ratios of each process in the abnormal process segment), and source parameters, is packaged into an abnormal event token. After consensus among multiple nodes on the blockchain network, this token is permanently stored on the blockchain. Due to the immutability of the blockchain itself, accurate traceability of electrode products is achieved.

[0073] like Figure 3 As shown, based on the same inventive concept, this application also provides a blockchain-based battery electrode quality data traceability system 200, including a host computer 210 and an inspection device 220, wherein: Inspection equipment 220 is deployed at each process end of the battery electrode production line to extract the processing parameters and corresponding quality inspection data of each process, and send the processing parameters and quality inspection data to the host computer 210. The host computer 210 is used to acquire the processing parameters and corresponding quality inspection data of each process in the battery electrode production process. It writes the processing parameters and quality inspection data into the blockchain according to the production sequence of the same electrode product and establishes a correlation between the processing parameters, quality inspection data, and the production sequence. Based on the deviation of the processing parameters of each process from the process baseline and the degree of deterioration of the process quality inspection results relative to the qualified standard, it determines the process risk index of each process. The process quality inspection results are determined based on the quality inspection data. Along the production sequence, the process risk index of each process is accumulated, and combined with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process, the risk exceedance index of the process segment is determined. Based on the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, the source process of the anomaly is determined.

[0074] The aforementioned inspection equipment 220 is deployed at each process stage of the battery electrode production line, including an online thickness gauge and pressure sensor at the front-end coating and cold pressing process, a laser processing equipment control unit and visual contour preliminary inspection system at the middle-stage laser drilling process, and a three-dimensional morphology detection device at the end-stage quality inspection process. It is used to collect data at each time the electrode raw material moves by the length of a finished electrode. At each collection time, the corresponding processing parameters and quality inspection data are extracted in real time. The processing parameters of the front-end process include coating thickness and compaction roller pressure, and the corresponding quality inspection data includes electrode thickness and compaction density. The processing parameters of the middle-stage process include laser power, wavelength, pulse width and single pulse energy, and the corresponding quality inspection data includes hole morphology and hole diameter. The end-stage quality inspection data includes hole diameter, blind hole depth, surface roughness and spatter size. The inspection equipment 220 sends the extracted processing parameters and quality inspection data to the host computer 210 in real time.

[0075] It is understood that the host computer 210 described above can realize any one of the functions of the blockchain-based battery electrode quality data traceability method provided in the embodiments of this application. The method embodiment section describes the way each function is implemented and the working principle. The system embodiment section will not repeat the description.

[0076] This invention is now complete.

[0077] In summary, in this embodiment of the invention, the processing parameters and corresponding quality inspection data of each process in the battery electrode production process are obtained. The processing parameters and quality inspection data are written into the blockchain according to the production sequence of the same electrode product, and the association between the processing parameters and quality inspection data and the production sequence is established. Based on the degree of deviation of the processing parameters of each process from the process benchmark and the degree of deterioration of the process quality inspection results from the qualified standard, the process risk index of each process is determined. The process quality inspection results are determined based on the quality inspection data. Along the production sequence, the process risk index of each process is accumulated, and combined with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process, the risk exceedance index of the process segment is determined. Based on the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, the abnormal source process is determined. This invention employs a three-tiered progressive analysis model that combines a blockchain-based trusted evidence storage foundation with single-process responsibility risk quantification, inter-process risk accumulation and transmission prediction, and anomaly source localization driven by prediction deviation. This model enables the identification, prediction, and precise traceability of accumulated quality risks caused by nonlinear coupling of cross-process parameters in battery electrode production, improving the accuracy and reliability of quality traceability. Furthermore, by quantifying the coupling relationship between the deviation of each process's processing parameters from the process baseline and the degree of deterioration of the process's quality inspection results from the pass / fail standard, it can identify hidden quality risks even when single-process parameters are within the compliance tolerance zone. This overcomes the limitations of traditional discrete Boolean value pass / fail records in characterizing parameters. Limitations of extreme deviations and quality deterioration trends; on the other hand, by accumulating the risk index of a single process along the electrode production sequence and combining it with the absorption and correction capabilities of subsequent processes for the accumulated risks of the previous stage, it is possible to proactively perceive the superposition and amplification effects of nonlinear coupling of multi-process parameters in the flow process, and to provide early warning before the accumulated abnormal risks actually exceed the qualified standards; on the other hand, by comparing the deviation between the degree of abnormality detected by the actual quality inspection at the end and the predicted value of the risk accumulation in the process segment, and combining the relative contribution of the risk index of each process in the process segment, it is possible to effectively screen out the processing operation of subsequent processes on abnormal electrodes from the source of abnormality, and accurately locate the real abnormal source process that causes the hidden coupling defects.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A blockchain-based method for tracing battery electrode quality data, characterized in that, The method includes: The processing parameters and corresponding quality inspection data of each process in the battery electrode production process are obtained. The processing parameters and the quality inspection data are written into the blockchain according to the production time sequence of the same electrode product, and the association relationship between the processing parameters, the quality inspection data and the production time sequence is established. Based on the degree of deviation of the processing parameters of each process from the process baseline and the degree of deterioration of the process quality inspection results from the qualified standard, the process risk index of each process is determined; wherein, the process quality inspection results are determined based on the quality inspection data. Along the production sequence, the process risk index of each process is accumulated, and combined with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process, the risk exceedance index of each process segment is determined. Based on the deviation between the actual degree of defects detected by the final quality inspection and the cumulative risk exceeding the index of the process segment, and combined with the relative contribution of the process risk index of each process within the process segment, the source process of the anomaly is determined.

2. The blockchain-based battery electrode quality data traceability method according to claim 1, characterized in that, The deviation between the actual defect severity detected by the final quality inspection and the cumulative risk exceeding the index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, determines the source process of the anomaly, including: For each process within the aforementioned process segment, the deviation between the actual degree of defect detected by the final quality inspection and the risk exceedance index up to the current process is calculated. The abnormal contribution of each process is determined based on the relative contribution of the process risk index within the process segment. Based on the deviation and the abnormal contribution, the abnormal coupling degree of each process is determined; Based on the abnormal coupling degree, the abnormal source process is determined from the process segment.

3. The blockchain-based battery electrode quality data traceability method according to claim 2, characterized in that, The determination of the abnormal contribution of each process based on the relative contribution of the process risk index within the process segment includes: Calculate the variance of the process risk index for each process within the process segment; Using the variance as an amplification factor, and combining it with the proportion of the process risk index of each process to the sum of the process risk indices of all processes within the process segment, the abnormal contribution of each process is determined.

4. The blockchain-based battery electrode quality data traceability method according to claim 2, characterized in that, The final quality inspection includes using a three-dimensional morphology inspection device to detect the aperture, blind hole depth, surface roughness, and spatter size of the electrode product. The actual defect level is determined based on the ratio of the detected spatter size to the normal allowable size.

5. The blockchain-based battery electrode quality data traceability method according to claim 1, characterized in that, The process risk index of each process is accumulated along the production sequence, and combined with the ability of subsequent processes to absorb and correct the accumulated risks of the previous process, the risk exceeding the index of the process segment is determined, including: Along the production sequence, the process risk index of each process within the process segment is amplified and mapped according to the process distance and then superimposed to determine the cumulative abnormal risk of the process segment; Based on the proportion of the duration of each subsequent process in the process segment to the total process duration and the historical risk index of the subsequent process, the ability of the subsequent process to absorb and correct the accumulated risks in the early stage is determined. Based on the cumulative abnormal risk and the absorption and correction capability, the risk exceedance index of the process segment is determined.

6. The blockchain-based battery electrode quality data traceability method according to any one of claims 1-5, characterized in that, The method further includes: Determine whether the risk exceedance index of the process segment exceeds a preset threshold; wherein the preset threshold is determined based on the ability of subsequent processes to absorb and correct the risks accumulated in the early stage; In response to the risk exceeding the preset threshold, a risk warning message is output before the cumulative abnormal risk in the process segment actually exceeds the qualification standard corresponding to the quality inspection data.

7. The blockchain-based battery electrode quality data traceability method according to any one of claims 1-5, characterized in that, The step of writing the processing parameters and the quality inspection data into the blockchain according to the production sequence of the same electrode product includes: The current block is generated according to the production sequence, and the current block includes the processing parameters, the quality inspection data, and the hash value of the previous block; The current blockchain is connected to the previous block to form a hash chain structure; Specifically, when the data contained in the current block is modified, the hash value of the current block changes, causing the hash value of the current block stored in subsequent blocks to change.

8. The blockchain-based battery electrode quality data traceability method according to any one of claims 1-5, characterized in that, The method further includes: Extract the processing parameters and quality inspection data corresponding to the abnormal source process from the blockchain; The processing parameters and the quality inspection data are compared in a multi-dimensional visualization to illustrate the formation process of latent coupling defects. The traceability evidence, including the visual comparison results, the responsibility sharing ratio, and the source parameters of the abnormal source process, is packaged into an abnormal event token, and the abnormal event token is written into the blockchain for storage.

9. The blockchain-based battery electrode quality data traceability method according to any one of claims 1-5, characterized in that, The acquisition of processing parameters and corresponding quality inspection data for each step in the battery electrode production process includes: Using the electrode production length coordinate as the index reference for the production time sequence, each displacement event in which the electrode raw material moves by one finished electrode length is determined as a collection moment. At each acquisition moment, the processing parameters of each process and the corresponding quality inspection data are acquired. Establish the positional correlation between the processing parameters and the quality inspection data and the same electrode product according to the continuous sequence of the electrode production length coordinates.

10. A blockchain-based battery electrode quality data traceability system, characterized in that, Includes a host computer and testing equipment, among which: The inspection equipment is deployed at each process stage of the battery electrode production line to extract the processing parameters and corresponding quality inspection data of each process, and to send the processing parameters and the quality inspection data to the host computer. The host computer is used to acquire the processing parameters and corresponding quality inspection data of each process in the battery electrode production process, write the processing parameters and quality inspection data into the blockchain according to the production sequence of the same electrode product, and establish the association between the processing parameters, the quality inspection data and the production sequence; based on the deviation of the processing parameters of each process from the process benchmark and the deterioration of the process quality inspection results from the qualified standard, the process risk index of each process is determined; wherein, the process quality inspection results are determined based on the quality inspection data; along the production sequence, the process risk index of each process is accumulated, and combined with the absorption and correction ability of subsequent processes to the accumulated risks of the previous process, the risk exceedance index of the process segment is determined; based on the deviation between the actual defect degree detected by the final quality inspection and the accumulated risk exceedance index of the process segment, combined with the relative contribution of the process risk index of each process within the process segment, the abnormal source process is determined.

Images