Iot-driven real-time production process traceability monitoring method and system

Abstract

The application discloses an Internet of Things driven real-time production process tracing and monitoring method and system, relates to the technical field of manufacturing process management, and comprises the following steps: monitoring production lines in real time according to the Internet of Things, and obtaining various production line monitoring blocks; detecting product abnormalities of each production line according to the blocks, and obtaining a product abnormality feature set of each production line; tracing and constructing a production line abnormality tracing map from equipment, environment and other factors; and finally, managing production line abnormalities according to the three types of maps, realizing production process tracing and abnormality management. The application solves the technical problem that, in manufacturing process management, traditional methods are difficult to accurately trace product abnormalities of multiple production lines in a production workshop from equipment, environment and other factors, leading to incomplete and untimely production abnormality management, and achieves the technical effects of accurately positioning product abnormalities from equipment, environment and other factors, and improving the comprehensiveness and timeliness of production abnormality management.

Description

Technical Field

[0001] This invention relates to the field of manufacturing process management technology, and in particular to a method and system for real-time production process traceability and monitoring driven by the Internet of Things. Background Technology

[0002] In the field of manufacturing process management, real-time monitoring of the production process and accurate tracing of anomalies are crucial for ensuring product quality and production efficiency. Current technologies often rely on manual inspections or single-dimensional data collection (such as focusing solely on equipment operating parameters) to detect product anomalies and trace equipment failures using fixed thresholds. While these methods are applicable to closed production lines with stable environments and simple process parameters, their limitations are becoming increasingly apparent as smart manufacturing demands greater transparency and refined management of the production process. Summary of the Invention

[0003] This application provides an IoT-driven real-time production process traceability and monitoring method and system to solve the technical problem that traditional methods are difficult to accurately trace product anomalies across multiple production lines in the manufacturing process, resulting in incomplete and untimely production anomaly management.

[0004] The first aspect of this application provides an IoT-driven real-time production process traceability and monitoring method, the method comprising: real-time monitoring of each production line in an IoT production workshop to obtain monitoring blocks for each production line; performing product anomaly detection on each production line based on the monitoring blocks to obtain a set of product anomaly features for each production line; tracing equipment factors based on the monitoring blocks to construct a first production line anomaly traceability map; tracing environmental factors based on the monitoring blocks to construct a second production line anomaly traceability map; tracing other factors based on the monitoring blocks to construct a third production line anomaly traceability map; and managing production line anomalies based on the first, second, and third production line anomaly traceability maps.

[0005] A second aspect of this application provides an IoT-driven real-time production process traceability and monitoring system, the system comprising: a production line monitoring block acquisition module, used for real-time monitoring of each production line in an IoT-enabled production workshop to obtain monitoring blocks for each production line; an anomaly feature set acquisition module, used for detecting product anomalies on each production line based on the monitoring blocks to obtain anomaly feature sets for each production line; a first map construction module, used for tracing equipment factors based on the monitoring blocks to construct a first map for production line anomaly traceability; a second map construction module, used for tracing environmental factors based on the monitoring blocks to construct a second map for production line anomaly traceability; a third map construction module, used for tracing other factors based on the monitoring blocks to construct a third map for production line anomaly traceability; and a production line anomaly management module, used for managing production line anomalies based on the first, second, and third maps for production line anomaly traceability.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application relates to the field of manufacturing process management. It collects multi-source data (equipment operating status, product characteristics, environmental parameters, etc.) from monitoring blocks on various production lines in an IoT-enabled production workshop. This data is then processed through blockchain encapsulation and machine learning modeling to obtain abnormal feature data. A multi-dimensional traceability map of equipment, environment, and other factors is constructed. Combined with dynamic matching of production control schemes and reinforcement learning causal analysis, the root causes of production process anomalies are accurately located, making the detection and traceability results in manufacturing process management more comprehensive and reliable. Specifically, a dynamic anomaly detection space is constructed by real-time monitoring of production line data. Key features are screened using a support-confidence model, and a causal model fused with reinforcement learning is used to associate multi-source data, forming an interpretable anomaly causal network. This solves the problem of insufficient traceability accuracy in multi-factor coupled scenarios using traditional methods, providing technical support for intelligent and transparent management of the manufacturing process. It achieves precise location of multiple factors such as equipment and environment causing product anomalies, improving the comprehensiveness and timeliness of production anomaly management. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0009] Figure 1This is a flowchart illustrating the IoT-driven real-time production process traceability and monitoring method provided in the embodiments of this application.

[0010] Figure 2 This is a schematic diagram of the structure of the IoT-driven real-time production process traceability and monitoring system provided in the embodiments of this application.

[0011] Figure labeling: Production line monitoring block acquisition module 1, abnormal feature set acquisition module 2, first map construction module 3, second map construction module 4, third map construction module 5, production line abnormal management module 6. Detailed Implementation

[0012] This application provides an IoT-driven real-time production process traceability and monitoring method and system to solve the technical problem that traditional methods are difficult to accurately trace product anomalies across multiple production lines in the manufacturing process, resulting in incomplete and untimely production anomaly management.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, an IoT-driven real-time production process traceability and monitoring method includes:

[0016] Step A100: Real-time monitoring of each production line in the IoT production workshop to obtain the monitoring blocks for each production line.

[0017] In this embodiment, the IoT production workshop uses IoT technology to monitor each production line in real time, collects multi-source data such as equipment operation, product characteristics, and environmental parameters, and forms production line monitoring blocks through data cleaning and blockchain encapsulation, thereby realizing an intelligent production site for product anomaly detection, multi-factor traceability, and anomaly management.

[0018] Specifically, based on the production line monitoring dataset obtained from real-time monitoring of each production line in the IoT production workshop, the data is cleaned to obtain the monitoring standard dataset, and then processed through blockchain encapsulation to obtain the monitoring blocks for each production line. The specific steps are explained in detail in A110-A130.

[0019] Step A200: Perform product anomaly detection on each production line according to the monitoring blocks of each production line to obtain the product anomaly feature set of each production line.

[0020] Optionally, based on the product feature data identified in the monitoring blocks of each production line and the production control scheme of the corresponding production line, the product anomaly detection space of each production line is constructed through product feature mining. Anomalies are then detected in the product feature data of each production line to generate anomaly feature sets for each production line. The specific steps are explained in detail in A210-A240.

[0021] Step A300: Based on the monitoring blocks of each production line, trace the abnormal feature set of each production line product by equipment factors and construct the first map of production line abnormality tracing.

[0022] In one embodiment of this application, the monitoring data of the equipment in each production line monitoring block is identified and anomalies are identified. The association between each production line equipment and the product anomaly feature set is analyzed by tracing the cause. After the graph is sorted out, the first graph of production line anomaly tracing is generated. The specific steps are described in detail in A310-A350.

[0023] Step A400: Based on the monitoring blocks of each production line, trace the environmental factors of the abnormal product feature set of each production line to construct the second map of production line abnormality traceability.

[0024] Specifically, based on the environmental monitoring data of each production line monitoring block, anomalies are identified and analyzed. The correlation between the environmental and product anomaly feature sets of each production line is analyzed through cause tracing. After the anomalies are visualized, a second anomaly tracing map of the production line is generated. The specific steps are explained in detail in A410-A440.

[0025] Step A500: Based on the monitoring blocks of each production line, trace other factors of the abnormal feature set of each production line product to construct a third map of production line abnormality tracing.

[0026] Specifically, based on the monitoring blocks of each production line, other monitoring data are identified and anomalies are identified. By analyzing the correlation between other factors of each production line and the product anomaly feature set, the third anomaly traceability map of the production line is generated after being visualized. The specific steps are explained in detail in A510-A540.

[0027] Step A600: Perform production line anomaly management based on the first production line anomaly tracing map, the second production line anomaly tracing map, and the third production line anomaly tracing map.

[0028] Specifically, the first, second, and third maps of integrated production line anomaly tracing are used to generate anomaly alarms for each production line. The specific steps are explained in detail in A610.

[0029] Furthermore, step A200 in the method provided in this application embodiment includes:

[0030] A210: Based on the monitoring blocks of each production line, product feature identification is performed to obtain product feature data of the first production line, product feature data of the second production line, ..., product feature data of the Pth production line, where P is a positive integer.

[0031] A220: Read the production control schemes for each production line to obtain the production control scheme for the first production line, the production control scheme for the second production line, ... the production control scheme for the Pth production line.

[0032] A230: Based on the production control scheme of the first production line, the production control scheme of the second production line, ... the production control scheme of the Pth production line, perform product feature mining to construct the first product anomaly detection space, the second product anomaly detection space, ... the Pth product anomaly detection space.

[0033] A240: Based on the first product anomaly detection space, the second product anomaly detection space, ... the Pth product anomaly detection space, perform anomaly detection on the first production line product feature data, the second production line product feature data, ... the Pth production line product feature data respectively, and generate the product anomaly feature sets for each production line.

[0034] In this embodiment, product features are information used to characterize the product's state. The production control scheme refers to the current control parameters of each production line. The anomaly detection space is a model space constructed by mining product features based on the production control schemes of each production line.

[0035] Specifically, firstly, based on the established monitoring blocks for each production line (trusted data encapsulated by blockchain), product feature identification is performed on each production line. Key feature data of products during the production process are collected in real time through IoT sensors. For example, physical parameters such as dimensional accuracy and surface roughness are collected for the first production line, while electrical performance indicators (such as voltage and current fluctuations) are collected for the second production line. This forms structured product feature data for the first production line, the second production line, and so on, up to the Pth production line (where P is the total number of production lines), achieving a comprehensive digital representation of the product status of each production line.

[0036] Secondly, when reading the production control scheme for each production line, the system obtains the currently effective set of control parameters in real time from the corresponding control parameter storage nodes (such as databases or edge computing devices) through the IoT interface, forming a structured production control scheme. For example, for production lines 1 to 1P, the system reads the production control schemes for each production line, which include data such as equipment operating parameters (e.g., spindle speed, temperature threshold), process standards (e.g., welding time, assembly accuracy), and quality specifications (e.g., dimensional tolerances, surface roughness requirements). A specific production line control scheme might include real-time control parameters such as the rotation speed of equipment A being 1500 r / min, the temperature being controlled between 20℃ and 25℃, and the dimensional tolerance of product B being ±0.05 mm, providing a benchmark for subsequent product anomaly detection.

[0037] Next, based on the production control scheme of each production line, the normal product characteristic record set is retrieved. After support evaluation, confidence analysis and confidence evaluation constraint selection, the first product anomaly detection space is constructed. The specific steps are explained in detail in A231-A234. Similarly, the second product anomaly detection space...the Pth product anomaly detection space can be further constructed.

[0038] Finally, the real-time product feature data from each production line is input into the corresponding anomaly detection space for comparative analysis. By calculating the deviation between the real-time data and the normal features in the detection space, features exceeding the allowable range are identified, generating anomaly feature sets for each production line. Each production line's anomaly feature set includes a first product anomaly feature set, a second product anomaly feature set, and so on up to the Pth product anomaly feature set. For example, if the dimensional parameter of a product on the first production line exceeds the ±0.1mm tolerance range, it is marked as an anomaly feature and included in the first product anomaly feature set.

[0039] Through a closed-loop process of real-time production line data acquisition, precise reading of control schemes, dynamic construction of feature space, and intelligent identification of abnormal features, dynamic anomaly detection based on production line personalized standards has been achieved, significantly improving the accuracy and comprehensiveness of product anomaly feature identification and providing a reliable anomaly data foundation for subsequent multi-factor traceability.

[0040] Furthermore, step A230 in the method provided in this application embodiment includes:

[0041] A231: Perform normal product feature retrieval based on the production control scheme of the first production line to obtain a product feature record set.

[0042] A232: Perform a support evaluation based on the product feature record set to obtain a product feature support evaluation set.

[0043] A233: Based on the product feature support evaluation set, perform confidence analysis on the product feature record set to obtain the product feature confidence evaluation set.

[0044] A234: Based on confidence evaluation constraints, the product feature record set is selected according to the product feature confidence evaluation set to obtain the first product anomaly detection space.

[0045] In this embodiment, support is a quantitative evaluation of the frequency of occurrence of each feature in the product feature record set, used to measure the universality of the feature. Confidence is the probability that the product is qualified when a feature appears, used to filter features that are strongly correlated with the product's qualification status.

[0046] Optionally, firstly, a normal product feature retrieval is performed based on the production control plan for the first production line. The feature range of qualified products (such as dimensional tolerances, performance index thresholds, etc.) is extracted from the process standards and quality specifications included in the plan. Using this as a condition, product feature data conforming to the standards is retrieved from historical production data to form a product feature record set. For example, if a production line control plan specifies that the product length must be within the range of 100±0.5mm, product features with lengths within this range are retrieved from historical data to form the record set for that production line.

[0047] Secondly, a support evaluation is performed on the product feature record set. The prevalence of a feature is quantified by statistically analyzing its frequency of occurrence in the record set (i.e., support). For example, if a feature with a length of 100mm ± 0.5mm appears 200 times in the record set, with a total of 250 records, its support is 80%, forming a product feature support evaluation set used to select frequently occurring core features.

[0048] Next, confidence scores are analyzed on the record set based on the support evaluation set. Confidence score measures the strength of the association between a feature and the product's pass / fail status, and is calculated by determining the probability that the product is pass / fail when the feature appears. For example, if the product pass rate is 95% when the feature with surface roughness Ra ≤ 1.6 μm appears, then its confidence score is 95%, forming a product feature confidence evaluation set. High-frequency features that are weakly associated with the pass / fail status (such as non-critical features that appear occasionally) are then removed.

[0049] Finally, based on preset confidence evaluation constraints (such as requiring a confidence level ≥ 90%), features in the record set are selected. Only features that meet both support and confidence standards are retained to construct the first product anomaly detection space. For example, a feature with a support of 85% and a confidence level of 92% meets the dual threshold requirement and is included in the detection space as a normal feature benchmark.

[0050] By employing a dual-threshold screening process and utilizing a support-confidence quantification evaluation system, key features that are highly compatible with production line processes are dynamically selected. This constructs an anomaly detection space that better meets actual production requirements, significantly improving the accuracy and reliability of product anomaly detection and providing a scientific basis for subsequent anomaly localization based on reliable data.

[0051] Furthermore, step A300 in the method provided in this application embodiment includes:

[0052] A310: The abnormal feature sets of each production line include the first product abnormal feature set, the second product abnormal feature set, ... the Pth product abnormal feature set.

[0053] A320: Based on the monitoring data of each production line monitoring block, identify the equipment monitoring area to obtain the first production line equipment monitoring area, the second production line equipment monitoring area, ... the Pth production line equipment monitoring area.

[0054] A330: Based on the first production line equipment monitoring area, the second production line equipment monitoring area...the Pth production line equipment monitoring area, anomaly identification is performed to obtain the first equipment anomaly feature set, the second equipment anomaly feature set...the Pth equipment anomaly feature set.

[0055] A340: Based on the first equipment anomaly feature set, the second equipment anomaly feature set...the Pth equipment anomaly feature set, the first product anomaly feature set, the second product anomaly feature set...the Pth product anomaly feature set are analyzed to obtain multiple product anomaly equipment cause sets.

[0056] A350: The abnormal feature sets of each production line and the cause-finding sets of multiple abnormal equipment for each product are graphically sorted to generate the first abnormal traceability map of the production line.

[0057] In this embodiment of the application, graphing refers to the structured organization of the abnormal feature sets of products from each production line and the abnormal feature sets of equipment / environment / other factors and their causal relationships, with nodes representing abnormal features and directed edges representing causal relationships, to construct a visual graph.

[0058] Specifically, firstly, equipment monitoring data is identified based on the monitoring blocks (encapsulated trusted data) of each production line. Real-time parameters of equipment operation are extracted from the production line monitoring data stored on the blockchain, such as the vibration frequency, temperature, and rotational speed of equipment on the first production line, forming a monitoring area for the first production line containing real-time data from multiple devices. Similarly, monitoring areas for the second to Pth production lines are obtained. For example, a monitoring area for a certain production line may contain 10 operating parameters for 3 key pieces of equipment, reflecting the dynamic status of the equipment in real time.

[0059] Secondly, anomaly identification is performed on the monitoring data of each production line equipment. By using preset equipment operating thresholds (such as vibration amplitude within the normal range of ≤50dB and temperature range of 20℃-40℃), parameter values ​​exceeding the normal range are detected, generating the first set of equipment anomaly features (such as abnormal vibration frequency of equipment A and excessive temperature of equipment B). Similarly, the equipment anomaly feature sets of other production lines are obtained.

[0060] Next, the causes of equipment anomaly features and product anomaly features are analyzed to form a product anomaly equipment cause set. The specific steps are explained in detail in A341-A346.

[0061] Finally, the product anomaly feature sets and equipment cause-of-fact sets for each production line are graphically organized. The process is as follows: First, the product anomaly feature sets for each production line (e.g., the first to Pth product anomaly feature sets) and the product anomaly equipment cause-of-fact sets obtained through cause-of-fact analysis are summarized (each product anomaly feature set corresponds to one equipment cause-of-fact set). Then, using product anomaly features (e.g., surface cracks, dimensional deviations) and equipment anomaly features (e.g., abnormal vibration frequency of equipment A, spindle speed fluctuations) as graph nodes, and the causal relationships determined by cause-of-fact analysis (e.g., abnormal equipment vibration frequency → product surface cracks) as directed edges, are structured according to the production line dimension. For example, if the temperature exceedance of equipment B identified from the equipment monitoring area of ​​the first production line is related to the insufficient fracture strength of the product in the product anomaly feature set of that production line, after confirming the causal relationship through cause-of-fact analysis, a directed edge is established in the graph pointing from the equipment B temperature exceedance node to the product fracture strength insufficient node, and attributes such as the association confidence level are labeled. This process transforms the complex relationships between equipment and product anomalies into a visual map, clearly showing the impact path of equipment factors on product quality, forming the first map for tracing production line anomalies, and providing an intuitive basis for quickly locating the root cause of production anomalies.

[0062] Through the above steps, the precise location of product anomalies due to equipment factors is achieved. Compared with existing technologies, this solution significantly improves the efficiency and accuracy of equipment tracing through reliable data support and intelligent cause-finding models, providing a visualized decision-making basis for rapid response to production anomalies and preventive maintenance of equipment.

[0063] Furthermore, step A340 in the method provided in this application embodiment includes:

[0064] A341: Extract the Kth feature of the product anomaly based on the first product anomaly feature set, where K is a positive integer.

[0065] A342: Based on the Kth feature of the product anomaly, perform associated data extraction on the first equipment anomaly feature set to obtain the Kth associated equipment anomaly feature.

[0066] A343: Supervised training is performed based on the product abnormal equipment cause-finding record set to obtain multiple product abnormal equipment cause-finding models.

[0067] A344: Based on the aforementioned multiple product abnormal equipment cause-finding models, establish a cause-finding first node model and a cause-finding second node model.

[0068] A345: Based on the first node model of cause tracing, the second node model of cause tracing is fused with reinforcement learning to obtain the cause tracing channel for abnormal product equipment.

[0069] A346: Input the Kth feature of the product anomaly and the Kth associated device anomaly feature into the product anomaly device tracing channel to obtain the Kth product anomaly device tracing result, and add the Kth product anomaly device tracing result to the first product anomaly device tracing set.

[0070] Specifically, firstly, the Kth product anomaly feature (K is a positive integer) is extracted from the first product anomaly feature set. For example, after extracting the Kth product anomaly feature (such as surface cracks) from the first product anomaly feature set, data mining is performed on the first equipment anomaly feature set using an association rule algorithm. The support and confidence of the equipment operating parameters (such as vibration frequency, temperature, etc.) with the product anomaly feature are calculated. Equipment anomaly parameters with a correlation higher than a preset threshold are selected, thereby obtaining the Kth associated equipment anomaly feature (such as abnormal vibration frequency of equipment A), and realizing the correlation analysis between product anomalies and equipment operating status.

[0071] Secondly, when using the random forest algorithm to build a product anomaly equipment cause-finding model, the input is a historically accumulated set of product anomaly equipment cause-finding records. This set contains verified causal pairs between equipment anomalies and product anomalies, specifically covering equipment anomaly characteristics (such as abnormal parameters like equipment vibration frequency, temperature, and rotation speed) and corresponding product anomaly characteristics (such as surface cracks and dimensional deviations). Feature engineering processing (such as normalization and discretization) is required to adapt the data to the algorithm's input requirements. The algorithm process is as follows: multiple subsets are randomly selected from the original data using .bootstrap sampling. A decision tree is built for each subset. When splitting at a decision tree node, some features are randomly selected to calculate the Gini index or information gain to determine the splitting rule, ultimately generating a forest model composed of multiple decision trees. The output consists of multiple product anomaly equipment cause-finding models (basic models). These models, by learning the feature mapping relationships in historical data, have the ability to predict the impact of equipment anomalies on product quality.

[0072] Next, these basic models are used as the first-node models (base models) for causal analysis, enabling them to make preliminary predictions about the correlation between single device anomalies and product anomalies, outputting preliminary causal correlation probabilities or conclusions. Then, a second-node model (meta-model) is constructed, integrating the prediction results of multiple base models through ensemble learning methods (such as voting and weighted averaging). For complex scenarios where multiple device anomalies synergistically affect product quality, the comprehensive reasoning of the meta-model improves the accuracy of causal relationship judgments. This process, through hierarchical modeling, achieves a progression from single-feature correlation analysis to comprehensive reasoning in complex scenarios, laying the foundation for subsequently generating reliable device causal analysis channels for product anomalies.

[0073] Then, the two node models are fused and optimized using a reinforcement learning algorithm. First, the preliminary predictions from the base model (such as the probability of association between a single device anomaly and a product anomaly) are used as input data for the meta-model. Then, a reinforcement learning reward function is designed to assign positive rewards to accurate causal results (such as verified causal pairs where a device anomaly leads to a product anomaly), and negative rewards to incorrect predictions or results with weak association strength. Through continuous iterative training, the parameter weights of the meta-model are dynamically adjusted using the reward mechanism, strengthening high-confidence association rules and weakening low-confidence rules, ultimately forming an adaptive product anomaly device causal channel. This channel can dynamically optimize association rules based on real-time input device and product anomaly characteristics, achieving accurate derivation of the causal relationship between device and product anomalies.

[0074] Finally, the extracted Kth feature of the product anomaly and the associated equipment anomaly features are input into the product anomaly equipment cause-finding channel. The model outputs the Kth product anomaly equipment cause-finding result (e.g., abnormal vibration frequency of equipment A is the main cause of surface cracks in the product), and this result is added to the first product anomaly equipment cause-finding set, completing the cause-finding process for a single anomaly. Repeat the above steps to obtain multiple product anomaly equipment cause-finding sets from the first to the Pth.

[0075] Through multi-model training and intelligent causal reasoning, the system enables intelligent deduction of the causal relationship between equipment anomalies and product anomalies, significantly improving the efficiency and accuracy of root cause localization in complex production environments.

[0076] Furthermore, step A400 in the method provided in this application embodiment includes:

[0077] A410: Based on the environmental monitoring data identification of each production line monitoring block, obtain the first production line environmental monitoring area, the second production line environmental monitoring area, ... the Pth production line environmental monitoring area.

[0078] A420: Based on the first production line environmental monitoring area, the second production line environmental monitoring area...the Pth production line environmental monitoring area, anomaly identification is performed to obtain the first environmental anomaly feature set, the second environmental anomaly feature set...the Pth environmental anomaly feature set.

[0079] A430: Based on the first environmental anomaly feature set, the second environmental anomaly feature set...the Pth environmental anomaly feature set, the first product anomaly feature set, the second product anomaly feature set...the Pth product anomaly feature set are analyzed to obtain multiple product anomaly environmental cause sets.

[0080] A440: The abnormal feature sets of each production line and the multiple abnormal environment cause sets of each product are graphically sorted out to generate the second abnormal traceability map of the production line.

[0081] Specifically, firstly, based on reliable data from each production line monitoring area, environmental monitoring data is identified. Through IoT terminals such as temperature and humidity sensors and air quality monitoring equipment deployed in the production workshop, environmental parameters of each production line (such as temperature, humidity, and dust particle concentration in the first production line) are collected in real time, forming structured environmental monitoring data for the first production line. Similarly, environmental monitoring data for the second to Pth production lines are obtained. For example, a certain production line's environmental monitoring area includes real-time data such as temperature (20℃-30℃), humidity (40%-60%RH), and dust concentration (≤1000ppm), covering key indicators of the production environment.

[0082] Secondly, anomaly identification is performed on the environmental monitoring data of each production line. Those skilled in the art compare real-time data with the preset environmental standard thresholds for the actual production line (e.g., a process requirement temperature of 25℃±2℃) to detect environmental parameters exceeding the allowable range, generating a first set of environmental anomaly features (e.g., excessive temperature, abnormally high humidity). Similarly, sets of environmental anomaly features for other production lines are obtained. For example, if the real-time temperature of a production line is 35℃, exceeding the standard upper limit of 30℃, it is marked as a temperature anomaly.

[0083] Next, a causal analysis was performed on the environmental anomaly feature set and the product anomaly feature set. A correlation analysis was conducted between the first environmental anomaly feature set (e.g., exceeding temperature limits) and the first product anomaly feature set (e.g., blistering on product surfaces). A causal model was trained using historical data. The model establishment and training process is as follows:

[0084] The input consists of causal pairs composed of historically accumulated environmental monitoring data and product anomaly data (such as corresponding records of abnormal environmental parameters such as temperature and humidity and abnormalities such as blistering and corrosion on the product surface). The data is trained through regression analysis and association rule algorithms (such as Apriori). Regression analysis is used to model the numerical correlation between environmental parameters and product anomalies, while association rule algorithms are used to discover frequent co-occurrence patterns between environmental anomaly features and product anomaly features.

[0085] First, environmental monitoring data (such as continuous parameters like temperature and humidity) and product anomaly data from historical production data are preprocessed. Environmental parameter values ​​are normalized, and product anomaly characteristics are discretized and encoded to form a structured dataset. For regression analysis, linear or nonlinear regression models (such as multinomial regression) are selected, with environmental parameters as independent variables and quantitative indicators of product anomalies (such as foaming rate and defect rate) as dependent variables. The model parameters are optimized using least squares or gradient descent methods to establish a numerical correlation model between environmental parameters and product anomalies. For example, a regression equation can be fitted showing that the foaming rate on the product surface increases by 30% for every 5°C increase in temperature. For the Apriori association rule algorithm, a support threshold (e.g., ≥80%) and a confidence threshold (e.g., ≥90%) are first set. Frequent itemset mining is then performed on the discretized environmental anomaly features (e.g., excessive temperature, abnormal humidity) and product anomaly features to generate association rules. For example, excessive temperature and abnormal humidity → blistering on product surface (support 85%, confidence 92%). Cross-validation is used to ensure the reliability of the rules, and finally a causal rule set of environmental anomalies and product anomalies is formed for subsequent cause analysis.

[0086] The trained model can output corresponding product anomaly features and causal relationship strength (e.g., temperature exceeding limits → blistering on product surface, confidence level 95%) based on the input set of environmental anomaly features (e.g., temperature exceeding limits → blistering on product surface, confidence level 95%), forming multiple sets of product anomaly environmental attribution.

[0087] Finally, the abnormal feature sets and environmental cause sets of each production line are graphically organized (this step is similar to the previous steps and will not be repeated due to limitations in the manual). Using environmental and product abnormal features as nodes and causal relationships as directed edges, a second production line abnormality tracing graph is constructed. For example, the humidity abnormality node in the graph is connected to the product corrosion node via directed edges, visually demonstrating the impact chain of environmental factors on product quality.

[0088] By employing a process of real-time environmental data collection, intelligent identification of abnormal features, causal relationship modeling and analysis, and visualization of graphs, this approach overcomes the lag and one-sidedness of existing technologies in tracing environmental factors. It significantly improves the efficiency and accuracy of tracing the impact of environmental factors on product quality during the production process, providing a scientific basis for targeted adjustment of environmental control parameters and optimization of production processes.

[0089] Furthermore, step A500 in the method provided in this application embodiment includes:

[0090] A510: Identify other monitoring data based on the monitoring blocks of each production line to obtain other monitoring areas of the first production line, other monitoring areas of the second production line, ... other monitoring areas of the Pth production line.

[0091] A520: Based on the other monitoring areas of the first production line, the other monitoring areas of the second production line, ... the other monitoring areas of the Pth production line, anomaly identification is performed to obtain the first set of other anomaly features, the second set of other anomaly features, ... the Pth set of other anomaly features.

[0092] A530: Based on the first set of other abnormal features, the second set of other abnormal features, ... the Pth set of other abnormal features, the first set of product abnormal features, the second set of product abnormal features, ... the Pth set of product abnormal features are analyzed to obtain multiple sets of other traceable causes for product abnormalities.

[0093] A540: The abnormal feature set of each production line and the other cause set of multiple product abnormalities are analyzed in a graph-like manner to generate the third graph for tracing the abnormality of the production line.

[0094] In one embodiment, firstly, based on reliable data from each production line's monitoring area, other monitoring data is identified. Multi-source heterogeneous data, excluding equipment and environmental data, is collected via the Internet of Things (IoT). This includes raw material composition indicators (such as the carbon content of a batch of steel), process parameter adjustment records (such as temporary extensions of welding time), and operator permission change logs, forming structured data for other monitoring areas of the first production line. Similarly, data for other monitoring areas of production lines two through P are obtained. This data covers non-equipment / environmental factors that may affect product quality during the production process, forming a comprehensive traceability data source.

[0095] Secondly, anomaly identification is performed on data from other monitoring areas of each production line. Anomalies are identified by comparing real-time data with preset process standards or historical data distribution models (such as acceptable ranges for raw material composition and normal fluctuation ranges for process parameters). For example, if the sulfur content of raw materials in a production line exceeds the standard upper limit by 0.05%, or if operators on a shift do not follow standard procedures, these are both marked as other abnormal features, generating the first set of other abnormal features. Similarly, other abnormal feature sets for other production lines are obtained.

[0096] Next, the causes of other abnormal feature sets and the product abnormal feature set are analyzed. Data mining algorithms (such as Apriori association rules, the process is the same as described above) are used to analyze the correlation between other abnormal features and product abnormal features. For example, through association analysis, it was found that the confidence level of excessive sulfur content in raw material batch A and insufficient fracture strength of the product reached 90%, forming a set of other causes for product abnormalities, clarifying the impact paths of specific other factors on product quality.

[0097] Finally, the abnormal feature sets and other cause sets of each production line are graphically analyzed (similar to the steps described above, and will not be repeated here due to limitations of the manual). Using other abnormal features (such as process parameter adjustments and raw material batch anomalies) and product abnormal features as nodes, and causal relationships as directed edges, a third-party graph for tracing production line anomalies is constructed. For example, in the graph, the node indicating excessive sulfur content in raw materials points to the node indicating insufficient fracture strength in the product, and the correlation probability and historical frequency are marked, visually presenting the impact chain of non-equipment / environmental factors.

[0098] Through the process of integrating multi-source data, intelligent detection of abnormal features, in-depth mining of causal relationships, and visualization of graphs, it fills the gap in existing technologies for tracing other factors in the production process, significantly improves the efficiency and accuracy of locating complex anomalies in the manufacturing process, and provides comprehensive data support for achieving multi-factor collaborative management and preventive process adjustments.

[0099] Furthermore, step A100 in the method provided in this application embodiment includes:

[0100] A110: Perform real-time monitoring of each production line to obtain monitoring datasets for each production line.

[0101] A120: Perform data cleaning based on the monitoring datasets of each production line to obtain the monitoring standard datasets.

[0102] A130: The monitoring standard datasets are encapsulated and processed according to the blockchain to obtain the monitoring blocks for each production line.

[0103] Optionally, firstly, real-time monitoring is implemented on each production line within the IoT-enabled production workshop. This is achieved by deploying IoT sensors (such as equipment operation status sensors, product feature acquisition devices, and environmental parameter monitoring devices) at various stages of the production line to continuously collect multi-dimensional information, including equipment operation data, real-time product feature data, and environmental temperature and humidity, forming raw monitoring datasets for each production line. These datasets contain a large amount of real-time dynamic data, such as the rotational speed of equipment on a particular production line, the frequency of temperature fluctuations, or real-time measurements of product dimensions, providing a fundamental data source for subsequent analysis.

[0104] Next, the original monitoring datasets for each production line are cleaned. This involves denoising algorithms to remove outliers (such as extreme data from temporary sensor malfunctions), format standardization (unifying data formats across different equipment interfaces), and missing value imputation (filling in missing values ​​based on historical data or data from adjacent nodes). This process filters out invalid or erroneous data, resulting in standardized and reliable monitoring datasets with consistent structure. For example, outliers in the vibration data of equipment on a certain production line that exceed the normal fluctuation range are marked and corrected to ensure the data accurately reflects the actual production status.

[0105] Finally, blockchain technology is used to encapsulate the cleaned monitoring standard datasets. A unique hash value is generated for each standard dataset using a hash algorithm, and combined with timestamp information, the data is packaged into blocks and linked to the blockchain in chronological order, forming immutable and traceable monitoring blocks for each production line. The distributed ledger characteristic of blockchain ensures that once data is on the chain, it cannot be unilaterally modified, and the traceability information of each block can accurately pinpoint the time of data collection, the equipment used, and the original state.

[0106] By implementing a process of real-time monitoring to collect raw data, data cleaning to improve data quality, and blockchain encapsulation to ensure data trustworthiness, we have achieved end-to-end management from production site data collection to trusted data storage. This provides a reliable data foundation for subsequent product anomaly detection, multi-factor traceability, and anomaly management based on trusted data.

[0107] Furthermore, step A600 in the method provided in this application embodiment includes:

[0108] A610: Generate anomaly alarms for each production line based on the first production line anomaly tracing map, the second production line anomaly tracing map, and the third production line anomaly tracing map.

[0109] In one embodiment, firstly, data from monitoring blocks of each production line are collected in real time to obtain the current abnormal feature set of each production line's products (such as abnormal features such as dimensional deviations and surface cracks appearing in the first production line).

[0110] Then, query the three traceability graphs respectively to locate the root cause corresponding to the abnormal features:

[0111] In the equipment factor map, we searched for the associated edges of nodes with out-of-tolerance dimensions and found a causal relationship of 95% confidence level between the spindle speed fluctuation and the abnormal equipment feature, confirming the influence of equipment factors. In the environmental factor map, we verified whether there was a correlation between abnormal environmental features and current product abnormalities, such as temperature exceeding limits and surface cracks with a confidence level of 90%. In other factor maps, we investigated whether there was a correlation between raw material batch fluctuations or process parameter adjustments and abnormal features, such as excessive sulfur content in raw material A → insufficient product fracture strength with a confidence level of 88%.

[0112] Based on the tracing results of the three graphs above, and through preset anomaly priority rules (such as equipment factors taking precedence over environmental factors), the main anomaly root causes and impact paths were determined. For example, the main cause of dimensional deviations on the first production line was identified as spindle speed fluctuations (equipment factor), and the secondary cause was excessive temperature (environmental factor).

[0113] Finally, based on the anomaly level (e.g., emergency, warning), corresponding production line anomaly alarms are generated and pushed to responsible departments such as equipment maintenance and process adjustment, along with causal chain data from a traceability graph (e.g., spindle speed fluctuation → dimensional deviation, 95% confidence level), supporting rapid response. Other production lines follow the same steps to generate their own anomaly alarms.

[0114] By integrating causal relationship data from multi-dimensional traceability maps and combining them with real-time anomaly detection results, the system achieves fully automated management of the entire process, from anomaly feature identification to root cause location, impact analysis, and alarm generation. This significantly improves the efficiency and accuracy of production line anomaly management and provides a scientific basis for real-time monitoring and preventative maintenance of the production process.

[0115] In summary, the IoT-driven real-time production process traceability and monitoring method provided in this application has the following technical effects:

[0116] This application utilizes monitoring equipment deployed on each production line in an IoT-enabled production workshop to collect real-time monitoring datasets from each line. After data cleaning and blockchain encapsulation, monitoring blocks are formed for each production line. By identifying product features and reading production control schemes, a product anomaly detection space is constructed and anomaly detection is performed, generating anomaly feature sets for each production line. Combining monitoring data from equipment, environment, and other factors, a production line anomaly traceability map is constructed through cause analysis and graph-based organization, thereby generating anomaly alarms. This achieves precise traceability and anomaly management of the production process, making manufacturing process management more intelligent and reliable. It achieves precise location of multiple factors related to equipment and environment in product anomalies, improving the comprehensiveness and timeliness of production anomaly management.

[0117] Example 2, as Figure 2As shown, based on the same inventive concept as in Embodiment 1 above, this application provides an Internet of Things-driven real-time production process traceability and monitoring system, the system comprising:

[0118] Production line monitoring block acquisition module 1 is used to obtain each production line monitoring block by performing real-time monitoring of each production line in the IoT production workshop.

[0119] Anomaly feature set acquisition module 2 is used to perform product anomaly detection on each production line according to the monitoring blocks of each production line, and obtain the product anomaly feature set of each production line.

[0120] The first map construction module 3 is used to trace equipment factors based on the abnormal feature set of each production line product according to the monitoring blocks of each production line, and construct the first map of production line abnormality tracing.

[0121] The second map construction module 4 is used to trace environmental factors of the abnormal feature set of each production line product based on the monitoring blocks of each production line, and construct a second map of production line abnormality tracing.

[0122] The third map construction module 5 is used to trace other factors of the abnormal feature set of each production line product based on the monitoring blocks of each production line, and construct the third map of production line abnormality tracing.

[0123] The production line anomaly management module 6 is used to manage production line anomalies based on the first production line anomaly traceability map, the second production line anomaly traceability map, and the third production line anomaly traceability map.

[0124] Furthermore, the anomaly feature set acquisition module 2 is used to perform the following steps:

[0125] Product feature identification is performed based on the monitoring blocks of each production line to obtain product feature data for the first production line, the second production line, ..., the Pth production line, where P is a positive integer. Production control schemes are read from each production line to obtain production control schemes for the first, second, ..., the Pth production line. Product feature mining is performed based on these schemes to construct a first product anomaly detection space, a second product anomaly detection space, ..., the Pth product anomaly detection space. Anomaly detection is then performed on the product feature data for the first, second, ..., the Pth production line, based on these schemes, generating anomaly feature sets for each production line.

[0126] Furthermore, the anomaly feature set acquisition module 2 is used to perform the following steps:

[0127] Normal product features are retrieved according to the production control scheme of the first production line to obtain a product feature record set; support evaluation is performed on the product feature record set to obtain a product feature support evaluation set; confidence analysis is performed on the product feature record set according to the product feature support evaluation set to obtain a product feature confidence evaluation set; based on confidence evaluation constraints, the product feature record set is selected according to the product feature confidence evaluation set to obtain the first product anomaly detection space.

[0128] Furthermore, the first map construction module 3 is used to perform the following steps:

[0129] The abnormal feature sets of each production line product include a first product abnormal feature set, a second product abnormal feature set, ..., a Pth product abnormal feature set; equipment monitoring data is identified based on the monitoring blocks of each production line to obtain the first production line equipment monitoring area, the second production line equipment monitoring area, ..., the Pth production line equipment monitoring area; anomalies are identified based on the first production line equipment monitoring area, the second production line equipment monitoring area, ..., the Pth production line equipment monitoring area to obtain a first equipment abnormal feature set, a second equipment abnormal feature set, ..., the Pth equipment abnormal feature set; the first product abnormal feature set, the second product abnormal feature set, ..., the Pth product abnormal feature set are analyzed for their respective causes to obtain multiple product abnormal equipment cause sets; the product abnormal feature sets of each production line and the multiple product abnormal equipment cause sets are graphically organized to generate the first production line anomaly traceability graph.

[0130] Furthermore, the first map construction module 3 is used to perform the following steps:

[0131] Extract the Kth product anomaly feature from the first product anomaly feature set, where K is a positive integer; extract associated data from the first equipment anomaly feature set based on the Kth product anomaly feature to obtain the Kth associated equipment anomaly feature; perform supervised training based on the product anomaly equipment cause-finding record set to obtain multiple product anomaly equipment cause-finding models; establish a first cause-finding node model and a second cause-finding node model based on the multiple product anomaly equipment cause-finding models; perform reinforcement learning fusion on the second cause-finding node model based on the first cause-finding node model to obtain a product anomaly equipment cause-finding channel; input the Kth product anomaly feature and the Kth associated equipment anomaly feature into the product anomaly equipment cause-finding channel to obtain the Kth product anomaly equipment cause-finding result, and add the Kth product anomaly equipment cause-finding result to the first product anomaly equipment cause-finding set.

[0132] Furthermore, the second map construction module 4 is used to perform the following steps:

[0133] Environmental monitoring data is identified based on the monitoring blocks of each production line to obtain the first production line environmental monitoring area, the second production line environmental monitoring area, ..., the Pth production line environmental monitoring area; anomalies are identified based on the first production line environmental monitoring area, the second production line environmental monitoring area, ..., the Pth production line environmental monitoring area to obtain the first environmental anomaly feature set, the second environmental anomaly feature set, ..., the Pth environmental anomaly feature set; the causes of the first product anomaly feature set, the second product anomaly feature set, ..., the Pth product anomaly feature set are analyzed based on the first environmental anomaly feature set, the second environmental anomaly feature set, ..., the Pth product anomaly feature set to obtain multiple product anomaly environmental cause sets; the product anomaly feature sets of each production line and the multiple product anomaly environmental cause sets are graphically organized to generate the second production line anomaly traceability map.

[0134] Furthermore, the third map construction module 5 is used to perform the following steps:

[0135] Based on the monitoring blocks of each production line, other monitoring data are identified to obtain other monitoring areas of the first production line, other monitoring areas of the second production line, ..., other monitoring areas of the Pth production line; based on the other monitoring areas of the first production line, other monitoring areas of the second production line, ..., other monitoring areas of the Pth production line, anomalies are identified to obtain a first set of other anomaly features, a second set of other anomaly features, ..., other set of Pth anomaly features; based on the first set of other anomaly features, the second set of other anomaly features, ..., other set of Pth anomaly features, cause analysis is performed on the first product anomaly feature set, the second product anomaly feature set, ..., product anomaly feature set, respectively, to obtain multiple sets of other cause-of-facts for product anomalies; the product anomaly feature sets of each production line and the multiple sets of other cause-of-facts for product anomalies are graphically organized to generate the third graph for production line anomaly tracing.

[0136] Furthermore, the production line monitoring block acquisition module 1 is used to perform the following steps:

[0137] Each production line is monitored in real time to obtain a monitoring dataset for each production line; data cleaning is performed on the monitoring datasets to obtain a standard monitoring dataset; and the standard monitoring datasets are encapsulated using blockchain technology to obtain a monitoring block for each production line.

[0138] Furthermore, the production line anomaly management module 6 is used to perform the following steps:

[0139] Based on the first production line anomaly tracing map, the second production line anomaly tracing map, and the third production line anomaly tracing map, anomaly alarms for each production line are generated.

[0140] The IoT-driven real-time production process traceability and monitoring system provided in this embodiment of the invention can execute the IoT-driven real-time production process traceability and monitoring method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0141] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0142] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A real-time production process traceability and monitoring method driven by the Internet of Things, characterized in that, The method includes: Real-time monitoring of each production line in the IoT production workshop is conducted to obtain monitoring blocks for each production line. Based on the monitoring blocks of each production line, product anomaly detection is performed on each production line to obtain the product anomaly feature set of each production line. Based on the monitoring blocks of each production line, the equipment factors of the abnormal feature set of each production line are traced to construct the first map of production line abnormality tracing. Based on the monitoring blocks of each production line, environmental factors are traced to the abnormal feature set of each production line product to construct a second map of production line abnormality tracing. Based on the monitoring blocks of each production line, other factors are traced to the abnormal feature set of each production line product to construct a third map of production line abnormality tracing. Production line anomaly management is carried out based on the first production line anomaly tracing map, the second production line anomaly tracing map, and the third production line anomaly tracing map. Specifically, based on the monitoring blocks of each production line, equipment factors are traced to identify the abnormal product feature sets of each production line, and a first production line abnormality traceability map is constructed, including: The product anomaly feature sets for each production line include the first product anomaly feature set, the second product anomaly feature set, ..., the Pth product anomaly feature set; Based on the equipment monitoring data of each production line monitoring block, the first production line equipment monitoring area, the second production line equipment monitoring area, ... the Pth production line equipment monitoring area are obtained; Anomaly identification is performed based on the first production line equipment monitoring area, the second production line equipment monitoring area...the Pth production line equipment monitoring area to obtain the first equipment anomaly feature set, the second equipment anomaly feature set...the Pth equipment anomaly feature set; Based on the first equipment anomaly feature set, the second equipment anomaly feature set...the Pth equipment anomaly feature set, the first product anomaly feature set, the second product anomaly feature set...the Pth product anomaly feature set are analyzed to obtain multiple product anomaly equipment cause sets; The abnormal feature sets of each production line and the multiple abnormal equipment cause sets of the products are graphically sorted to generate the first abnormal traceability map of the production line. Specifically, based on the first equipment anomaly feature set, the second equipment anomaly feature set...the Pth equipment anomaly feature set, a cause-tracing analysis is performed on the first product anomaly feature set, the second product anomaly feature set...the Pth product anomaly feature set, including: Extract the Kth feature of the product anomaly from the first product anomaly feature set, where K is a positive integer; Based on the Kth feature of the product anomaly, the first equipment anomaly feature set is correlated with data to obtain the Kth associated equipment anomaly feature; Supervised training was conducted based on the product abnormal equipment cause-finding record set to obtain multiple product abnormal equipment cause-finding models; Based on the aforementioned multiple product abnormal equipment cause-finding models, establish a cause-finding first node model and a cause-finding second node model; Based on the first node model of cause tracing, the second node model of cause tracing is fused with reinforcement learning to obtain the cause tracing channel for abnormal product equipment. Input the Kth feature of the product anomaly and the Kth associated device anomaly feature into the product anomaly device cause tracing channel to obtain the Kth product anomaly device cause tracing result, and add the Kth product anomaly device cause tracing result to the first product anomaly device cause tracing set; The first-node model of cause attribution makes a preliminary prediction of the correlation between a single device anomaly and a product anomaly. The second-node model of cause attribution integrates the prediction results of multiple first-node models of cause attribution through an ensemble learning method.

2. The IoT-driven real-time production process traceability and monitoring method as described in claim 1, characterized in that, Based on the monitoring blocks of each production line, product anomaly detection is performed on each production line to obtain a set of product anomaly features for each production line, including: Based on the monitoring blocks of each production line, product feature identification is performed to obtain product feature data of the first production line, product feature data of the second production line, ..., product feature data of the Pth production line, where P is a positive integer; The production control schemes of each production line are read to obtain the production control schemes of the first production line, the second production line, ..., the Pth production line. Based on the production control scheme of the first production line, the production control scheme of the second production line, ... the production control scheme of the Pth production line, product feature mining is performed to construct the first product anomaly detection space, the second product anomaly detection space, ... the Pth product anomaly detection space; Based on the first product anomaly detection space, the second product anomaly detection space, ... the Pth product anomaly detection space, anomaly detection is performed on the first production line product feature data, the second production line product feature data, ... the Pth production line product feature data, respectively, to generate the product anomaly feature sets for each production line.

3. The IoT-driven real-time production process traceability and monitoring method as described in claim 2, characterized in that, Product feature mining is performed based on the production control schemes for the first production line, the second production line, ..., the production control scheme for the Pth production line, including: Based on the production control scheme of the first production line, normal product feature retrieval is performed to obtain a product feature record set; Based on the product feature record set, a support evaluation is performed to obtain a product feature support evaluation set. Based on the product feature support evaluation set, the confidence level of the product feature record set is analyzed to obtain the product feature confidence evaluation set; Based on confidence evaluation constraints, the product feature record set is selected according to the product feature confidence evaluation set to obtain the first product anomaly detection space.

4. The IoT-driven real-time production process traceability and monitoring method as described in claim 1, characterized in that, Based on the monitoring blocks of each production line, environmental factors are traced to identify the abnormal product feature sets of each production line, and a second production line anomaly tracing map is constructed, including: Based on the environmental monitoring data of each production line monitoring block, the first production line environmental monitoring area, the second production line environmental monitoring area, ... the Pth production line environmental monitoring area are obtained; Anomalies are identified based on the first production line environmental monitoring area, the second production line environmental monitoring area, ... the Pth production line environmental monitoring area to obtain the first environmental anomaly feature set, the second environmental anomaly feature set, ... the Pth environmental anomaly feature set; Based on the first environmental anomaly feature set, the second environmental anomaly feature set...the Pth environmental anomaly feature set, the first product anomaly feature set, the second product anomaly feature set...the Pth product anomaly feature set are analyzed to obtain multiple product anomaly environmental cause sets; The abnormal feature sets of each production line and the multiple abnormal environment cause sets of each product are graphically analyzed to generate the second abnormal traceability map of the production line.

5. The IoT-driven real-time production process traceability and monitoring method as described in claim 1, characterized in that, Based on the monitoring blocks of each production line, other factors are traced to the abnormal feature sets of products of each production line to construct a third map for production line abnormality tracing, including: Based on the monitoring blocks of each production line, other monitoring data are identified to obtain other monitoring areas of the first production line, other monitoring areas of the second production line, ... other monitoring areas of the Pth production line; Anomalies are identified based on the other monitoring areas of the first production line, the other monitoring areas of the second production line, ... the other monitoring areas of the Pth production line to obtain the first set of other anomalies, the second set of other anomalies, ... the Pth set of other anomalies; Based on the first set of other abnormal features, the second set of other abnormal features, ... the Pth set of other abnormal features, the cause analysis of the first set of product abnormal features, the second set of product abnormal features, ... the Pth set of product abnormal features is performed to obtain multiple sets of other cause analysis for product abnormalities; The abnormal feature sets of each production line and the other cause sets of the multiple product abnormalities are analyzed in a graph-like manner to generate the third graph for tracing the abnormalities of the production lines.

6. The IoT-driven real-time production process traceability and monitoring method as described in claim 1, characterized in that, Based on real-time monitoring of each production line in the IoT production workshop, monitoring blocks for each production line are obtained, including: Real-time monitoring of each production line is performed to obtain monitoring datasets for each production line; Data cleaning is performed on the monitoring datasets of each production line to obtain the monitoring standard datasets; The monitoring standard datasets are encapsulated and processed according to the blockchain to obtain the monitoring blocks for each production line.

7. The IoT-driven real-time production process traceability and monitoring method as described in claim 1, characterized in that, Based on the first production line anomaly tracing map, the second production line anomaly tracing map, and the third production line anomaly tracing map, anomaly alarms for each production line are generated.

8. An IoT-driven real-time production process traceability and monitoring system, characterized in that, For implementing the IoT-driven real-time production process traceability and monitoring method according to any one of claims 1-7, the system comprises: The production line monitoring block acquisition module is used to obtain the monitoring blocks of each production line by real-time monitoring of each production line in the IoT production workshop. The abnormal feature set acquisition module is used to perform product abnormality detection on each production line according to the monitoring blocks of each production line, and obtain the product abnormality feature set of each production line. The first map construction module is used to trace equipment factors based on the abnormal feature set of each production line product according to the monitoring blocks of each production line, and construct the first map of production line abnormality tracing. The second map construction module is used to trace environmental factors based on the abnormal feature set of each production line product according to the monitoring blocks of each production line, and construct the second map of production line abnormality tracing. The third map construction module is used to trace other factors of the abnormal feature set of each production line product based on the monitoring blocks of each production line, and construct the third map of production line abnormality tracing. The production line anomaly management module is used to manage production line anomalies based on the first production line anomaly traceability map, the second production line anomaly traceability map, and the third production line anomaly traceability map.

Images