A process control system, method, device and medium of a positive electrode material

Abstract

This application provides a process control system, method, equipment, and medium for cathode materials, effectively solving the problems of information silos, standard disconnect, and delayed response in existing production processes for cathode materials of lithium-ion batteries. The system includes a production acquisition module for real-time acquisition of various process parameter data; a parameter management module for responding to process parameter changes, automatically generating new versions, and updating a pre-set process parameter metadata model based on the new versions; a production alarm module for triggering a distribution channel to push alarm reports to target communication software based on the processing results obtained from processing the various process parameter data according to the updated process parameter metadata model; and a production analysis module for automatically analyzing various process parameter data based on natural query commands to obtain interpretable analysis results.

Description

Technical Field

[0001] This application relates to the field of lithium-ion battery production technology, and more specifically, to a process control system, method, equipment, and medium for a cathode material. Background Technology

[0002] In the preparation of lithium-ion battery cathode materials (such as lithium nickel cobalt manganese oxide and lithium nickel cobalt aluminum oxide), process parameters have a decisive impact on product quality and production efficiency. However, existing technologies generally suffer from the following shortcomings: First, information is severely fragmented, lacking a unified knowledge entry point. Process-related knowledge is scattered across multiple channels such as MES, SCADA, process control platforms, paper SOPs, and corporate policy documents, lacking a unified "knowledge platform." When quality anomalies occur, engineers need to search across systems for original data, standards, and handling methods. The lack of information linkage and correlation leads to low analysis efficiency and delayed decision-making.

[0003] Secondly, there is a disconnect between process standards and their implementation. The criteria for judging process parameters and the methods for handling anomalies are scattered in equipment programs, paper documents, or spreadsheets. The standards retrieved cannot be directly linked to real-time data, and the closed loop of "data comparison - anomaly identification - method recommendation" cannot be automatically completed. This results in a disconnect between standards and implementation, making it difficult to quickly support on-site problem handling.

[0004] Third, there is insufficient data value mining and a lack of lightweight analytical capabilities for frontline staff. Existing systems' CPK and trend analysis functions are largely geared towards technical personnel, are complex to operate, and are difficult for frontline staff to use directly. A large amount of process data is only used as a basis for "single-point qualification judgment," and cannot be automatically generated with simple queries to produce analytical results such as process stability trends. Therefore, the value of the data cannot be quickly transformed into a basis for process optimization.

[0005] Fourth, the lack of linkage between alarms and knowledge dissemination hinders rapid response to anomalies. The existing alarm mechanism can only push notifications of threshold exceeding limits, and cannot automatically link them to corresponding process standards, anomaly handling SOPs, and historical handling cases. After receiving an alarm, on-site personnel still need to manually query relevant specifications, resulting in long response times and potential mishandling due to incomplete information, making it difficult to form a closed loop of "alarm-knowledge push-handling guidance".

[0006] In addition, frontline staff need to query multiple independent systems to obtain information, which is lengthy, inefficient, time-consuming, and prone to information omissions or misunderstandings.

[0007] In summary, existing technologies cannot achieve comprehensive digital control of process parameters, intelligent version management, one-click deployment, and integrated intelligent alarms, nor can they form a closed-loop process knowledge service system encompassing "collection—integration—output—application." Therefore, there is an urgent need for an integrated and intelligent process control solution that can address the aforementioned issues. Summary of the Invention

[0008] In view of this, the purpose of this application is to provide a process control system, method, equipment and medium for cathode materials, which effectively solves the problems of information silos, standard disconnect and response lag in the existing production processes of cathode materials for lithium-ion batteries.

[0009] In a first aspect, embodiments of this application provide a process control system for cathode materials, the system comprising a production acquisition module, a production alarm module, a parameter management module, and a production analysis module: The production acquisition module is used to collect various process parameter data of the cathode material in real time during the production process, and send the various process parameter data to the production alarm module and the production analysis module. The parameter management module is used to respond to process parameter change operations for various process parameter data, automatically generate a new version, and update the preset process parameter metadata model according to the new version; the process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module is used to trigger the distribution channel to push alarm reports to the target communication software by processing the various process parameter data according to the updated process parameter metadata model. The production analysis module is used to automatically analyze the various process parameter data based on the user's natural query instructions, through preset analysis strategies and updated process parameter metadata models, to obtain interpretable analysis results.

[0010] In conjunction with the first aspect, this application provides a first possible implementation of the first aspect, wherein the production analysis module is used to automatically analyze the various process parameter data according to the user's natural query instructions, through a preset analysis strategy and an updated process parameter metadata model, including: The natural query command is semantically parsed to obtain the semantic parsing result, and an automated query task is generated based on the semantic parsing result; For the automated query task, multiple parameter standards in the updated process parameter metadata model are invoked to execute the analysis strategy.

[0011] In conjunction with the first aspect, this application provides a second possible implementation of the first aspect, wherein the production analysis module, for calling multiple parameter standards in the updated process parameter metadata model to execute the analysis strategy, includes: Based on the multiple parameter standards in the updated process parameter metadata model, a multi-dimensional impact assessment standard is constructed for the target parameters in automated query tasks. Based on the aforementioned analysis strategy, the impact assessment results corresponding to the multi-dimensional impact assessment criteria are dynamically integrated to obtain the analysis results.

[0012] In conjunction with the first aspect, this application provides a third possible implementation of the first aspect, wherein the parameter management module automatically generates a new version in response to process parameter change operations involving multiple process parameter data, including: Extract the key change information of the process parameter change operation, and perform consistency verification on the key change information to obtain the verification result; Based on the verification results, a version generation process is triggered to generate a new version with a process change history.

[0013] In conjunction with the first aspect, this application provides a fourth possible implementation of the first aspect, wherein the parameter management module, in addition to triggering the version generation process to generate a new version with a process change history, includes: The new version is controlled to be the production version, and the version before the process parameter change operation is marked as the old version; According to the target update mode, the process parameters on the production line are updated by replacing the old version of the process parameters with the process parameters of the current production version.

[0014] In conjunction with the first aspect, this application provides a fifth possible implementation of the first aspect, wherein the production alarm module is used to push alarm reports to the target communication software, including: Extract the product model from the processing result and automatically match it in the pre-built group chat library of the target communication software according to the product model; A communication connection is established with the matched target communication group chat through the distribution channel, so as to send the alarm report to the target communication group chat.

[0015] In conjunction with the first aspect, this application provides a sixth possible implementation of the first aspect, wherein the production alarm module is used to trigger a distribution channel based on the processing result obtained by processing the various process parameter data according to the updated process parameter metadata model, including: For unstructured data in various process parameter data, extract multiple key indicators from the unstructured data; Based on the updated process parameter metadata model, the various key indicators are compared and predicted one by one to obtain the processing results and trigger the distribution channel.

[0016] Secondly, embodiments of this application provide a method for process control of cathode materials, the method comprising: The production acquisition module collects various process parameter data of the cathode material in real time during the production process and sends the data to the production alarm module and the production analysis module. The parameter management module responds to process parameter changes in various process parameter data, automatically generates a new version, and updates the preset process parameter metadata model based on the new version; the process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module processes the various process parameter data according to the updated process parameter metadata model, and the resulting processing triggers the distribution channel to push an alarm report to the target communication software. The production analysis module automatically analyzes various process parameter data based on the user's natural query commands, using preset analysis strategies and updated process parameter metadata models to obtain interpretable analysis results.

[0017] Thirdly, embodiments of this application provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the process control method for a cathode material are performed.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the process control method for a cathode material.

[0019] This application provides a process control system for cathode materials. The system includes a production acquisition module, a production alarm module, a parameter management module, and a production analysis module. The production acquisition module collects various process parameter data during the cathode material production process in real time and sends this data to the production alarm module and the production analysis module. The parameter management module responds to process parameter changes in the various process parameter data, automatically generates a new version, and updates a pre-set process parameter metadata model based on the new version. The process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module triggers a distribution channel to push an alarm report to the target communication software based on the processing results obtained from processing the various process parameter data according to the updated process parameter metadata model. The production analysis module automatically analyzes the various process parameter data according to the user's natural query commands, using a pre-set analysis strategy and the updated process parameter metadata model, to obtain interpretable analysis results. The cathode material process control system provided in this application effectively addresses the pain points of the background technology, such as fragmented information, disconnected standards, isolated alarms, and inefficient analysis, through the synergistic effect of four modules: production data acquisition, parameter management, production alarms, and production analysis. The parameter management module automates the generation of process parameter versions and updates the metadata model, ensuring close linkage between process standards and execution, and guaranteeing traceability and controllability of parameter changes. The production alarm module significantly shortens anomaly response time, avoiding the lag and omissions of manual cross-system queries. The production analysis module lowers the barrier to entry for frontline personnel, enabling rapid conversion of process data value. The system described in this application comprehensively connects the entire chain from data acquisition, standard updates, alarm linkage to intelligent analysis, constructing an integrated closed-loop process knowledge service system of "acquisition—integration—output—application," significantly improving the accuracy, response speed, and decision-making efficiency of process control. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This paper shows a structural block diagram of a process control system for a cathode material provided in an embodiment of this application; Figure 2 A flowchart illustrating the production alarm module provided in an embodiment of this application is shown; Figure 3A flowchart illustrating the production analysis module provided in an embodiment of this application is shown; Figure 4 A schematic flowchart of a process control method for a cathode material provided in an embodiment of this application is shown. Figure 5 A structural block diagram of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0023] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0024] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0025] In cathode material manufacturing processes, knowledge is scattered across multiple systems and documents, lacking a unified central platform; standards and implementation are disconnected, hindering the automatic completion of data comparison, anomaly identification, and method recommendation loops; analytical tools are complex and difficult for frontline workers to use; alarms are not linked to handling specifications, resulting in delayed responses; and cross-system queries are inefficient. Overall, an integrated, closed-loop service system for process knowledge has not been established.

[0026] Based on this, the present application provides a process control system, method, equipment and medium for cathode materials, which are described below through embodiments.

[0027] Example 1 To facilitate understanding of this embodiment, a process control system for a cathode material disclosed in this application will first be described in detail. For example... Figure 1 The diagram shows a structural block diagram of a process control system for cathode materials. This application provides a process control system for cathode materials, which includes a production acquisition module, a production alarm module, a parameter management module, and a production analysis module. The production acquisition module 101 is used to acquire multiple process parameter data of the cathode material in real time during the production process, and send the multiple process parameter data to the production alarm module and the production analysis module. The parameter management module 102 is used to respond to process parameter change operations of various process parameter data, automatically generate a new version, and update the preset process parameter metadata model according to the new version; the process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module 103 is used to trigger the distribution channel to push an alarm report to the target communication software by processing the multiple process parameter data according to the updated process parameter metadata model. The production analysis module 104 is used to automatically analyze the various process parameter data according to the user's natural query instructions, through preset analysis strategies and updated process parameter metadata models, to obtain interpretable analysis results.

[0028] The system described in this application is based on a three-layer decoupled process AI intelligent agent architecture, specifically consisting of a user interface and view presentation layer, a core business logic and AI scheduling layer, and an industrial data asset and computing layer. The user interface and view presentation layer includes a production analysis module, the core business logic and AI scheduling layer includes a parameter management module and a production alarm module, and the industrial data asset and computing layer includes a production acquisition module. There are no code-level dependencies between the three layers, and upgrading or replacing any layer (such as replacing with a better LLM, introducing a new algorithm, or connecting to a new data source) will not affect the operation of the other two layers, significantly improving the system's resilience, maintainability, and long-term evolution capabilities.

[0029] In the production acquisition module 101, various process parameter data of the cathode material during production are collected in real time through Industrial Internet of Things (IIoT) technology. Specifically, this is achieved by uniformly accessing a multi-source heterogeneous data system through a standardized data interface. This includes real-time PLC operation data of the kiln process (such as sintering temperature, furnace pressure, gas flow rate, oxygen concentration, conveyor belt speed, etc.), temperature and humidity data collected by environmental monitoring equipment (covering silos, clean rooms, and key workstations), physicochemical parameters output by raw material and intermediate testing equipment (such as nickel, cobalt, and manganese content, pH value, solid content, particle size distribution D10 / D50 / D90, and moisture content), and structured and semi-structured data such as batch numbers, work instructions, inspection records, and anomaly markers generated from MES, SCADA, LIMS, and inspection systems including production, quality, equipment, and process systems. The various process parameter data are then encapsulated and simultaneously sent to the production alarm module and the production analysis module. All the multiple process parameter data are encrypted and verified, then marked according to a four-level dimension of "workshop—production line—process—timestamp," and synchronously written into the system's storage module. This module does not rely on manual export or secondary entry, completely eliminating the risks of lag, fragmentation, and inconsistency caused by traditional methods such as paper records, offline forms, and manual integration across systems, providing a real, complete, reliable, and traceable data foundation for subsequent intelligent management and control.

[0030] If, during the production of cathode materials for lithium-ion batteries, there are process parameter change operations performed by process engineers, such as modifying the upper limit of the sintering temperature of a kiln on a certain production line or adjusting the alarm threshold for the moisture content of a certain type of cathode material, then the parameter management module 102, in response to the process parameter change operations of multiple process parameter data, automatically generates a new version of the process parameters and updates the preset process parameter metadata model according to the updated parameter factors in the new version. The process parameter metadata model is constructed based on the process parameter factors corresponding to multiple process parameters. The process parameter metadata model adopts a four-level attachment system of "workshop-production line-process-specific control project" oriented towards business scenarios. It is the core data structure for realizing the digital management of the entire life cycle of process parameters. In this model, each process parameter is abstracted into a structured digital asset entity, namely a process parameter factor, which contains a standard parameter value, a specification upper limit (USL) and a specification lower limit (LSL), and an alarm upper limit (Warning Upper Limit) and an alarm lower limit (Warning Lower Limit). The process parameter metadata model includes multiple dimensions such as Limit, alarm triggering methods (e.g., audible and visual alerts, message pushes, automatic shutdown linkage), judgment logic type (e.g., single-point threshold comparison, trend slope judgment, CPK interval evaluation), applicable environmental conditions (e.g., temperature and humidity range, raw material batch type), associated detection methods, and corresponding SOP numbers. The model also possesses strong extensibility, thus supporting the addition of new process parameter factors (e.g., introducing carbon footprint factors and energy consumption factors) without requiring a reconstruction of the underlying architecture.

[0031] In some embodiments, the parameter management module automatically generates a new version in response to process parameter change operations involving multiple process parameter data, including: Extract the key change information of the process parameter change operation, and perform consistency verification on the key change information to obtain the verification result; Based on the verification results, a version generation process is triggered to generate a new version with a process change history.

[0032] In this embodiment, after receiving the process parameter change operation, such as manual submission via interface, API call, or AI agent suggestion adoption, the parameter management module extracts key change information. The key change information includes, but is not limited to: parameter names and values ​​before and after the change, the process and equipment number, the effective workshop and production line identifier, the expected effective time, the operator's identity and permission level, the associated product model and batch range, the basis for the change (such as quality anomaly report number, customer specification update notification, environmental monitoring exceedance record), and contextual semantic elements such as whether cross-process linkage adjustment is involved. The system initiates multi-dimensional consistency checks on the key change information: First, business rule checks, verifying whether the changed values ​​fall within industry-standard safety boundaries (e.g., sintering temperature must not exceed the equipment's rated upper limit) and whether they meet the process matching constraints of upstream and downstream processes (e.g., changes in moisture content in the front-end mixing process must be synchronously verified against the downstream drying temperature control logic); second, data integrity checks, confirming that all required factors (e.g., alarm upper and lower limits, judgment methods) have been assigned values ​​and are logically consistent (e.g., USL must not be less than LSL); and third, permission and process compliance checks, verifying whether the operator possesses the qualifications to revise parameters for the corresponding process and confirming whether the necessary countersigning has been completed (e.g., electronic approval records from quality, process, and equipment parties). Only after all checks pass will the parameter management module officially trigger the version generation process, such as automatically creating a new version number, archiving a full snapshot of the old version, writing a structured process change history (including timestamps, operators, verification conclusions, and links to associated original vouchers), and binding the process change history with all subsequent production batch data in real time to obtain the final new version. The process change history not only facilitates forward tracing of "which version of parameters a certain batch was produced according to", but also supports the realization of production analysis of "how many batches and quality results were actually affected by a certain parameter change".

[0033] In some embodiments, the parameter management module, in addition to triggering the version generation process to generate a new version with a process change history, includes: The new version is controlled to be the production version, and the version before the process parameter change operation is marked as the old version; According to the target update mode, the process parameters on the production line are updated by replacing the old version of the process parameters with the process parameters of the current production version.

[0034] In this embodiment, the parameter management module implements a version management mode of "dual-state coexistence, single-state effectiveness" for each process parameter: when a new version passes the consistency verification and is officially released, the system automatically sets its status to "production version" and marks the corresponding parameter instance that was originally being executed as "historical version" and archives it into a process knowledge base built based on multiple historical versions. The historical versions in the process knowledge base are prohibited from being called by any business process or distributed to field equipment. All historical versions retain all their metadata, change reason explanations, approval records and associated quality verification reports, and support multi-dimensional retrieval and comparison by time, workshop, product model and other dimensions. The parameter management module offers three update modes to adapt to different process sensitivities and production rhythm requirements: First, the "Instant Effect Mode," suitable for non-critical processes or well-verified fine-tuning parameters. The new version is synchronized to the corresponding production line's edge computing nodes and PLC control system within seconds of release, achieving automatic parameter overlay. Second, the "Batch Trigger Mode," automatically activates the new version only when a new batch of a specified product model is fed, ensuring that older batches continue to use the original standards and avoiding batch mixing risks. Third, the "Manual Confirmation Mode," requiring the production line supervisor to confirm via the HMI interface before switching, and simultaneously pushing graphic and textual prompts containing key change points and precautions to the on-site operation terminals. The parameter management module determines the target update mode based on the number of parameter changes in the process parameter change operation. Regardless of the update mode used, the parameter management module strictly ensures that "only one effective version runs on the same production line at the same physical moment," preventing execution chaos caused by multiple versions running concurrently. The entire update process is recorded, including the release time, receiving device ID, execution result feedback (success / failure / timeout), and the actual production batch number of the first application of this version. All information is written to the ODS layer in real time and pushed synchronously to the management dashboard and quality traceability system, thereby building an end-to-end trusted logical chain from parameter definition, approval, release, distribution to actual execution.

[0035] When the new version is updated on the production line based on the target update mode, the production alarm module 103 deeply integrates process knowledge and real-time data stream to construct an integrated intelligent alarm mechanism of "perception-judgment-decision-reach". The processing result obtained by processing the multiple process parameter data according to the multiple process parameter factors of the updated process parameter metadata model triggers the distribution channel. For example, when the real-time kiln temperature exceeds the alarm limit configured in the metadata model for the "BN17 model nickel-cobalt-manganese lithium oxide sintering process" for 3 consecutive seconds, or when the moisture content of a batch of raw materials falls into the "high humidity environment sensitive range" marked by the model and simultaneously triggers the particle size Span coefficient to exceed the tolerance, the production alarm generates a structured alarm event containing the abnormal parameter name, measured value, standard threshold, deviation range, occurrence time, location production line and equipment number, associated batch number, and preliminary root cause label (such as "temperature field fluctuation" or "silo condensation"). After the structured alarm event is rendered by a lightweight rich text engine, a visual alarm report is automatically generated containing red highlighted abnormal values, trend screenshots, historical comparison curves, and recommended handling SOP links, and a distribution channel is triggered to push the alarm report to the target communication software such as DingTalk or WeChat. That is, through the distribution channel, the production alarm module can establish instant communication with communication software such as DingTalk or WeChat to send the alarm report to the communication software such as DingTalk or WeChat.

[0036] In some embodiments, the production alarm module is used to push alarm reports to the target communication software, such as... Figure 2 As shown, it includes: Extract the product model from the processing result and automatically match it in the pre-built group chat library of the target communication software according to the product model; A communication connection is established with the matched target communication group chat through the distribution channel, so as to send the alarm report to the target communication group chat.

[0037] In this embodiment, after the production alarm module generates an alarm event, it first intelligently parses and extracts the key business identifier, i.e., the product model, from the structured processing results. This product model specifically originates from the work order master data synchronized by the MES system, the batch header information uploaded by the equipment, or the product binding fields (such as "BN17", "NCM811-A", "NCA-2025", etc.) preset in the process parameter metadata model associated with the alarm trigger, ensuring accurate, unique, and semantically meaningful model identification. The production alarm module then calls the built-in communication group chat mapping service to access the pre-configured and dynamically maintained "communication group chat library." This library stores the mapping relationship between product models and target communication groups in key-value pairs, such as "BN17 → DingTalk Group ID: chat_bn17_process", "NCM811-A → Enterprise WeChat Group ID: wecom_ncm811_alert", and "General Environment Anomaly → DingTalk Group ID: chat_env_safety." It also supports multi-dimensional tag combination matching by workshop, production line, and functional role (such as "kiln engineer" or "IPQC supervisor"). The matching process is based not only on exact string matching but also supports fuzzy error tolerance (such as automatically identifying "BN17-2" as belonging to the "BN17" product family) and inheritance relationship resolution (such as sub-models inheriting the parent model's group strategy by default). Upon successful matching, the distribution channel immediately loads the corresponding group's authentication credentials, message template, and permission context, establishing a secure connection with the target communication software API. Based on this, the rendered alarm report (including anomaly parameter snapshots, trend charts, SOP links for handling suggestions, and shortcut buttons to the platform details page) is encapsulated into a message body conforming to the target platform's protocol format, asynchronously delivered via a message queue, and its delivery status confirmed. The entire process is fully automated with zero human intervention, low response latency, and supports retrying on failure, message receipt tracking, and delivery statistical analysis. This ensures that every alarm is accurately delivered to the most relevant and demanding frontline personnel and management units, significantly improving anomaly response efficiency and collaborative handling quality.

[0038] In some embodiments, the production alarm module is used to trigger a distribution channel based on the processing result obtained by processing the various process parameter data according to the updated process parameter metadata model, including: For unstructured data in various process parameter data, extract multiple key indicators from the unstructured data; Based on the updated process parameter metadata model, the various key indicators are compared and predicted one by one to obtain the processing results and trigger the distribution channel.

[0039] In this embodiment, during the cathode material production process, a large amount of key quality data exists in unstructured form, such as Excel reports exported from particle size analyzers (containing dozens of indicators and charts such as D10 / D50 / D90 / SPAN / BET), PDF inspection reports generated by XRF elemental scanning, original CSV time-series logs of the kiln temperature control system, and infrared thermal imaging spectrum files. The production alarm module uses an embedded parsing engine to perform semantic recognition and pattern matching on the above files: automatically locating table areas and recognizing column header semantics, such as mapping "Median Diameter" to D50, "Span=(D90)", etc. The parameters "D10" / "D50" are identified as broadening coefficients. Values ​​and units are extracted, and data validity is verified, such as excluding blank lines, abnormal symbols, and dimensional misalignments. Finally, a structured set of standard key indicators is output, including but not limited to particle size distribution (Dmin, Dmax, D10, D50, D90, Span, specific surface area), elemental content (Ni%, Co%, Mn%, Al%, O%, impurity Fe / Cu / Na content), and thermal parameters (peak temperature duration, heating slope, and temperature fluctuation range in the holding section). After extraction, the production alarm module immediately calls the currently effective process parameter metadata model, according to... Based on the predefined index-standard mapping relationship for each process (e.g., the D50 standard value for "BN17 sintered material" is 12.5±0.8μm, with an alarm upper limit of 13.3μm; its Span standard upper limit is 1.8, with an alarm upper limit of 2.0), independent and parallel compliance judgments are performed on each key indicator. These key indicators are those affecting the production quality of cathode materials, including sintering temperature, oxygen concentration, etc. The system can also automatically call the CPK calculation engine to perform predictive analysis on each key indicator, specifically by extracting the minimum value Dmin, minimum value Dmax, standard deviation σ, and mean of the particle size distribution. In addition to other statistical quantities, the corresponding CPK value is calculated or the process capability status is determined using the SPC method. For example, if CPK < 1.33, it is marked as insufficient capability. Based on this, the production alarm module also integrates historical trends, environmental disturbance factors (such as daily workshop temperature and humidity fluctuations > ±5%RH or ±2℃) and multi-process coupling relationships, and calls a lightweight time series prediction model (such as exponentially weighted moving regression) to extrapolate deviations and identify potential over-limit risks (such as "sintering temperature is expected to deviate by +3.2℃ after 4 hours"). If any indicator in the processing result exceeds the alarm threshold configured in the process parameter metadata model or has an over-limit risk, the production alarm module generates a corresponding alarm event and automatically triggers the distribution channel to push a structured alarm report containing the indicator name, measured value, standard limit, deviation direction, and visual comparison chart to the target communication group chat of the target communication software according to the aforementioned product model routing strategy.

[0040] In the production process of cathode materials for lithium-ion batteries, if a user inputs a natural query command via a web, mobile device, or voice interaction interface, such as: "Why did the kiln temperature fluctuate greatly in the BN17 production line last month?", "D50 has been too high for three consecutive batches, is it related to humidity?", "Compare the changes in CPK under the parameters of versions V1.2.0 and V1.3.0", or "Please analyze the trend and influencing factors of the particle size broadening coefficient of the most recent ten batches of NCM811-A", then based on the production analysis module 104 of the system, through AI intelligent scheduling of various tools such as intelligent mind map recognition engine, CPK calculation engine, and semantic parsing engine, the intelligent mind map recognition engine is first called to perform deep semantic parsing on the input statement, accurately extracting the time range, product model, process link, target parameters, analysis type (such as attribution analysis, trend judgment, version comparison, correlation test) and implicit business intent, and transforming it into a structured automated query task. Subsequently, based on the characteristics of the automated query task, the appropriate analytical capability components are dynamically loaded and scheduled: if process capability assessment is involved, the CPK calculation engine is automatically invoked, combining the upper and lower limits of the specifications defined in the metadata model with the actual distribution of the current batch, and outputting CPK values ​​with confidence interval labels and a classification conclusion of "capability sufficient / critical / insufficient"; if it is trend analysis, the time series data of the ODS layer is aggregated to generate a smooth trend line, mark key inflection points and abnormal fluctuation segments, and correlate with contextual events such as ambient temperature and humidity and raw material batch changes during the same period; if root cause mining is required, a multidimensional correlation analysis process is initiated, automatically comparing the correlation strength of potential factors such as the slope of the kiln temperature curve, the stability of atmospheric oxygen concentration, and the uniformity of the front-end mixing, and presenting them in order of influence weight. All analysis processes are strictly executed according to the parameter semantics, constraints, and business rules defined in the updated process parameter metadata model, ensuring that the results are process interpretable—not only outputting numerical conclusions, but also simultaneously providing judgment criteria (such as "D50 is too high because version V1.3.0 relaxed the upper limit of the drying section wind speed, leading to increased powder agglomeration"), links to the original data sources, references to corresponding SOP clauses, and visualization charts (heat maps, scatter matrices, comparison bar charts, etc.). The final analysis results are returned in the form of rich text reports with rich graphics, accurate terminology, and clear hierarchy. The analysis results support one-click export, sharing, and embedding into the knowledge base, truly transforming complex data analysis into process explanations that front-line engineers can understand, verify, and act upon, thereby achieving intelligent process analysis that is "accessible to everyone, answers questions instantly, and what you see is what you get."

[0041] In some embodiments, the production analysis module is used to automatically analyze the various process parameter data based on the user's natural query instructions, using preset analysis strategies and an updated process parameter metadata model, such as... Figure 3 As shown, it includes: The natural query command is semantically parsed to obtain the semantic parsing result, and an automated query task is generated based on the semantic parsing result; For the automated query task, multiple parameter standards in the updated process parameter metadata model are invoked to execute the analysis strategy.

[0042] In this embodiment, when a user inputs colloquial, unstructured commands such as "Is the recent five batches of D90 exceeding the standard on the BN17 production line related to the high temperature in July?" or "Compare the CPK stability under the two sintering parameter versions V1.2.0 and V1.3.0", the production analysis module first activates a semantic parsing engine customized for industrial scenarios. This engine integrates the intent understanding capabilities of a large language model with a knowledge graph of the process domain, deconstructing the command layer by layer: identifying the core query object (such as "D90", "CPK"), the limited scope (such as "the recent five batches", "July", "V1.2.0 / V1.3.0" for the BN17 production line), and the time dimension (…). The analysis includes absolute time, relative time period, batch interval, logical relationships (causal inference, comparative analysis, trend tracking), and implicit business objectives (such as "investigating the root cause of anomalies," "evaluating version effectiveness," and "verifying environmental impact"). The parsed results form a standardized automated query task description package, containing structured fields: target parameter identifier, associated process path, time window, comparison benchmark, analysis type label, and contextual constraints. It also dynamically calls the updated process parameter metadata model to accurately extract multiple parameter standards strongly related to the task—including not only basic specification limits (such as LSL / USL for D90) and alarm thresholds (Warning...). The limits also include environmental adaptation rules such as "when the workshop temperature > 32℃, the D90 alarm limit is automatically tightened by 0.3μm", version binding relationships such as "version V1.3.0 explicitly adjusts the heating rate of the kiln insulation section from 5℃ / min to 4.2℃ / min", and process coupling constraints such as "D90 anomalies require simultaneous verification of the amplitude of the front-end crusher and the mesh size of the screen," as well as other in-depth process semantic information. These standards serve as rigid bases for the execution of analysis strategies, driving subsequent modules to complete data filtering, index calculation, multi-dimensional correlation, and visualization rendering. For example, when performing the "high temperature impact analysis" task, the production analysis module automatically correlates meteorological monitoring data with the kiln... The furnace temperature field log calls the "temperature-granularity sensitivity mapping table" defined in the process parameter metadata model, performs segmented correlation tests, and outputs attributable process explanations. In the "version comparison" task, it strictly extracts the measured data streams of the corresponding batches according to the parameter sets bound to each version in the process parameter metadata model, and performs statistical significance tests and process capability decay analysis. The entire process is fully automated, requiring no user knowledge of SQL, statistical principles, or the system's underlying logic. All analysis conclusions are accompanied by standard source annotations and traceable data paths, ensuring that the results are authoritative, transparent, and verifiable, effectively supporting frontline personnel in rapid response, scientific decision-making, and continuous improvement.

[0043] In some embodiments, the production analysis module, when invoking multiple parameter standards from the updated process parameter metadata model to execute the analysis strategy, includes: Based on the multiple parameter standards in the updated process parameter metadata model, a multi-dimensional impact assessment standard is constructed for the target parameters in automated query tasks. Based on the aforementioned analysis strategy, the impact assessment results corresponding to the multi-dimensional impact assessment criteria are dynamically integrated to obtain the analysis results.

[0044] In this embodiment, in the process parameter metadata model, each key parameter not only defines static specification limits such as D50=12.5±0.8μm, but also includes rich dynamic semantic standards: including environmental coupling rules, such as "when the ambient humidity is >65%RH, the upper limit of D50 is automatically lowered by 0.4μm", equipment status association standards such as "if the vibration value of the crusher bearing is >5.2mm / s, the D90 tolerance band is narrowed to ±0.6μm", upstream process dependence standards such as "for every 0.3% decrease in the solid content of the mixing section, the probability of the upward trend of D10 after sintering increases by 37%", and historical quality feedback mapping relationships such as "in the 12 cases of D90 exceeding the standard in the past six months, 9 cases also showed that the O2 concentration of the kiln tail gas was >1.8%". Therefore, once the automated query task locks onto a target parameter such as "current BN17 batch D90=14.2μm", the production analysis module, based on its process, occurrence time, associated equipment ID, and environmental snapshot, activates and loads all applicable multi-dimensional impact assessment standards in real time from the process parameter metadata model, forming an assessment framework covering five dimensions: environment, equipment, materials, operation, and historical experience. Subsequently, the analysis is performed according to the preset analysis strategy: first, it is determined whether a basic alarm has been triggered; then, it is verified whether the standards of each dimension have been exceeded. For example, if the current humidity is confirmed to be 68%RH, the humidity coupling rule is triggered, and it is determined that D90 has actually exceeded the dynamic tightening limit; simultaneously, the daily vibration monitoring data of the crusher is retrieved to identify abnormal fluctuation segments and activate the equipment-related standards; further, the solid content records of the previous mixing material of this batch are correlated with the similar abnormal handling database of the past three months to match high-probability root cause combinations. Finally, the production analysis module semantically weights and fuses the independent assessment conclusions from various dimensions, such as "Environmental Contribution: High," "Equipment Status Risk: Medium," "Material Batch Consistency: Low," and "Historical Similarity: 82%," generating a structured analysis result. This result not only indicates that "the high D90 is due to the superposition of multiple factors," but also clearly lists the ranking of the dominant factors and the action paths of each factor. For example, "High humidity causes the powder to absorb moisture, leading to a decrease in flowability and insufficient crushing, thus increasing D90," with corresponding SOP guidelines and a visual evidence chain diagram. The analysis result possesses strong process interpretability, operability, and knowledge reusability, directly supporting rapid on-site location, precise intervention, and closed-loop improvement.

[0045] The following is an example of the system provided in this application: 1. In 2024, a company in Chengdu successfully completed the hardware construction of the aforementioned system, achieving real-time data acquisition for kiln processes. Rigorous verification and testing ensured the system's stability and accuracy. Based on this, the company initially established a process parameter control, statistical analysis, and alarm platform mechanism, and implemented control charts, trend charts, and comparative analysis functions for key indicators such as process trends, standard deviation, and Cpk. Simultaneously, through the construction and implementation of the kiln process early warning analysis platform, in-depth communication with process engineers and extensive collection of requirements laid a solid foundation for the successful construction of a cathode material process control and analysis system.

[0046] 2. In 2025, the above companies successfully implemented one-click issuance and alarm functions for cathode material process parameters, built a continuously updated process knowledge base, and achieved full-chain integrated management and equipment process parameter (SPC) analysis by systematically sorting and optimizing big data platform signals, comprehensively covering process parameters of each process, establishing a unified management process for process parameter versions, developing a parameter issuance process and one-click issuance function, and developing factor control and traceability analysis functions. These innovative measures greatly promoted knowledge sharing and process innovation, significantly improved production efficiency and product quality, and laid a solid foundation for the company's intelligent manufacturing transformation.

[0047] 3. This application's system, based on a big data platform and data stream processing technology, comprehensively incorporates process parameters, environmental temperature and humidity, and raw material and product parameters into a digital management and control system, ensuring the real-time nature and accuracy of the data. Through intelligent parameter alarm functions, the system monitors parameter status in real time, automatically triggering alarms when parameters deviate from preset ranges, thus improving the control precision and safety of the process. Simultaneously, the system also features a one-click distribution mechanism, enabling rapid dissemination of the latest process parameters to relevant departments and personnel, improving management efficiency. Furthermore, the system constructs an scalable process analysis knowledge base system. By automatically comparing and analyzing newly generated process parameters and intelligently tagging them in conjunction with environmental factors, it continuously updates and improves professional knowledge and analysis results related to process parameters, promoting efficient knowledge sharing and in-depth innovation in the process field.

[0048] The following are comparative examples of the systems provided in this application: 1. Traditional cathode material production process control relies primarily on manual operation and paper records, resulting in limited data coverage and difficulties in ensuring real-time performance and accuracy. Key information such as environmental temperature and humidity, and raw material parameters are often overlooked, leading to insufficient precision in process control. Furthermore, process parameter updates and management are typically conducted through paper documents or simple spreadsheets, which are inefficient and prone to errors. The lack of intelligent alarm and early warning mechanisms, relying mainly on manual inspections and post-event analysis, makes it difficult to promptly detect and address anomalies in the process. The accumulation and management of process knowledge also often depend on personal experience, lacking a systematic and intelligent knowledge base, resulting in inefficient knowledge sharing and hindering in-depth innovation in the process field.

[0049] By comparison, it can be concluded that this application achieves comprehensive digital control of process parameter factors, intelligent management of process parameter versions and knowledge base construction, and integration of one-click distribution and intelligent parameter alarm functions. This not only improves the management efficiency of process parameters but also ensures precise control and safe operation of the process, significantly improving production efficiency and product quality. Simultaneously, the implementation of this system promotes efficient knowledge sharing and in-depth innovation in the process field, providing strong support for the digital transformation of the cathode material production industry.

[0050] ① Through the application of the system, comprehensive digital control of process parameters has been achieved, including real-time collection, storage and analysis of process parameters of each process, environmental temperature and humidity, raw material and product parameters, which greatly improves the accuracy and completeness of the data and provides a solid foundation for the fine control of the process.

[0051] ② An intelligent management and knowledge base construction mechanism for process parameter versions has been introduced. This not only enables real-time alarms and intelligent comparative analysis for parameter modifications, but also intelligently marks new version process parameters based on environmental factors, continuously updating and improving the process knowledge base, and promoting efficient knowledge sharing and in-depth innovation in the process field.

[0052] ③ The integration of one-click data transmission and intelligent parameter alarm functions enables the latest process parameters to be quickly transmitted to relevant departments and personnel. At the same time, the parameter status is monitored in real time, and an alarm is automatically triggered once it deviates from the preset range, ensuring precise control and safe operation of the process, and significantly improving production efficiency and product quality.

[0053] ④ By constructing a three-layer decoupled process AI intelligent agent architecture, unified access, integration, and governance of multi-source heterogeneous industrial data are achieved, effectively breaking down information silos; relying on natural language understanding and AI intelligent scheduling, the threshold for process knowledge and data query is significantly reduced, and the efficiency of personnel operations and the speed of anomaly response are significantly improved; at the same time, the entire process of process analysis such as CPK, trend, and regression is automated, reducing manual intervention and improving the accuracy of analysis; the modular and loosely coupled design enables the system to have good scalability and reusability, ultimately realizing the transformation from traditional manual decision-making to data-driven, AI-empowered intelligent process control, comprehensively improving the stability of production processes, quality traceability capabilities, and overall management level.

[0054] Example 2 This application also provides a method for process control of cathode materials, such as... Figure 4 The diagram shown is a flowchart illustrating a process control method for a cathode material. The steps implemented in this process control method correspond to the functional effects of the aforementioned process control system for a cathode material executed on a terminal device. The process control method for a cathode material described in this application includes: S401. Through the production acquisition module, various process parameter data of the cathode material are collected in real time during the production process, and the various process parameter data are sent to the production alarm module and the production analysis module. S402. Through the parameter management module, respond to process parameter change operations of various process parameter data, automatically generate a new version, and update the preset process parameter metadata model according to the new version; the process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. S403. The production alarm module processes the data of the various process parameters according to the updated process parameter metadata model, and triggers the distribution channel to push the alarm report to the target communication software. S404. Through the production analysis module, based on the user's natural query instructions, the various process parameter data are automatically analyzed using preset analysis strategies and the updated process parameter metadata model to obtain interpretable analysis results.

[0055] Example 3 This application also provides an electronic device, such as Figure 5 As shown, it includes: a processor 501, a memory 502, and a bus 503. The memory 502 stores machine-readable instructions that can be executed by the processor 501. When the electronic device is running, the processor 501 and the memory 502 communicate through the bus 503. When the machine-readable instructions are executed by the processor 501, the steps of the process control method for a cathode material are performed.

[0056] Example 4 This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of a process control method for a cathode material.

[0057] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0058] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0059] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0060] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a platform server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0061] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A process control system for a cathode material, characterized in that, The system includes a production acquisition module, a production alarm module, a parameter management module, and a production analysis module. The production acquisition module is used to collect various process parameter data of the cathode material in real time during the production process, and send the various process parameter data to the production alarm module and the production analysis module. The parameter management module is used to respond to process parameter change operations of various process parameter data, automatically generate a new version, and update the preset process parameter metadata model according to the new version. The process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module is used to trigger the distribution channel to push alarm reports to the target communication software by processing the various process parameter data according to the updated process parameter metadata model. The production analysis module is used to automatically analyze the various process parameter data based on the user's natural query instructions, through preset analysis strategies and updated process parameter metadata models, to obtain interpretable analysis results.

2. The system according to claim 1, characterized in that, The production analysis module is used to automatically analyze various process parameter data based on the user's natural query instructions, using preset analysis strategies and an updated process parameter metadata model, including: The natural query command is semantically parsed to obtain the semantic parsing result, and an automated query task is generated based on the semantic parsing result; For the automated query task, multiple parameter standards in the updated process parameter metadata model are invoked to execute the analysis strategy.

3. The system according to claim 2, characterized in that, The production analysis module, in order to execute the analysis strategy by calling multiple parameter standards from the updated process parameter metadata model, includes: Based on the multiple parameter standards in the updated process parameter metadata model, a multi-dimensional impact assessment standard is constructed for the target parameters in automated query tasks. Based on the aforementioned analysis strategy, the impact assessment results corresponding to the multi-dimensional impact assessment criteria are dynamically integrated to obtain the analysis results.

4. The system according to claim 1, characterized in that, The parameter management module automatically generates new versions in response to process parameter changes involving various process parameter data, including: Extract the key change information of the process parameter change operation, and perform consistency verification on the key change information to obtain the verification result; Based on the verification results, a version generation process is triggered to generate a new version with a process change history.

5. The system according to claim 4, characterized in that, The parameter management module, in addition to triggering the version generation process to generate a new version with a process change history, includes: The new version is controlled to be the production version, and the version before the process parameter change operation is marked as the old version; According to the target update mode, the process parameters on the production line are updated by replacing the old version of the process parameters with the process parameters of the current production version.

6. The system according to claim 1, characterized in that, The production alarm module is used to push alarm reports to the target communication software, including: Extract the product model from the processing result and automatically match it in the pre-built group chat library of the target communication software according to the product model; A communication connection is established with the matched target communication group chat through the distribution channel, so as to send the alarm report to the target communication group chat.

7. The system according to claim 1, characterized in that, The production alarm module is used to trigger the distribution channel based on the processing result obtained from processing the various process parameter data according to the updated process parameter metadata model, including: For unstructured data in various process parameter data, extract multiple key indicators from the unstructured data; Based on the updated process parameter metadata model, the various key indicators are compared and predicted one by one to obtain the processing results and trigger the distribution channel.

8. A method for process control of a cathode material, characterized in that, The method includes: The production acquisition module collects various process parameter data of the cathode material in real time during the production process and sends the data to the production alarm module and the production analysis module. The parameter management module responds to process parameter changes in various process parameter data, automatically generates a new version, and updates the preset process parameter metadata model based on the new version; the process parameter metadata model is constructed based on process parameter factors corresponding to various process parameters. The production alarm module processes the various process parameter data according to the updated process parameter metadata model, and the resulting processing triggers the distribution channel to push an alarm report to the target communication software. The production analysis module automatically analyzes various process parameter data based on the user's natural query commands, using preset analysis strategies and updated process parameter metadata models to obtain interpretable analysis results.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the process control method for a cathode material as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the process control method for a cathode material as described in claim 8.

Images