Aluminum processing equipment health status assessment method and system

Abstract

The present invention belongs to the field of distributed evaluation technology, and in particular relates to a method and system for evaluating the health status of aluminum processing equipment. The method first combines cross-factory equipment attributes and operating information with a Merkle tree and an evaluation algorithm to construct a distributed fault assessment forest; secondly, the tree is used with a preset cross-factory real-time monitoring network to carry out distributed edge interaction status evaluation to obtain evaluation results of single equipment and process-related equipment; thirdly, based on these results and a root cause tracing warning rule and strategy generation library, fault tracing is performed, and maintenance strategies are generated and shared. At the same time, with the help of a traceability warning visualization mapping space, the entire tracing and sharing process is visualized, and ultimately a visualized fault assessment warning forest is formed, thereby realizing accurate evaluation of the status of cross-factory aluminum processing equipment, fault tracing, and effective generation and sharing of maintenance strategies.

Description

Technical Field

[0001] The present invention belongs to the field of distributed evaluation technology, and in particular relates to a health status evaluation method and system for aluminum processing equipment. Background Art

[0002] As the aluminum processing industry moves toward intelligent and large-scale development, the importance of cross-plant equipment health status assessment has become increasingly prominent. Each plant has a large number of equipment and complex operating conditions. Comprehensive and accurate understanding of equipment status is of great significance to ensuring production continuity and improving product quality. However, existing technologies have defects. On the one hand, data silos are serious, making it difficult to share data across plants and integrate multi-source information to build a comprehensive model. On the other hand, under a centralized learning model, centralized data transmission is prone to privacy leakage risks, and facing the heterogeneity of equipment in different plants, traditional methods are difficult to effectively integrate data and accurately assess. These problems restrict the accuracy and security of cross-plant aluminum processing equipment status assessment. To this end, the present invention provides a method and system for aluminum processing equipment health status assessment. Summary of the Invention

[0003] In response to the shortcomings of the existing technology, the present invention proposes a health status assessment method and system for aluminum processing equipment. The method first combines cross-factory equipment attributes and operating information with Merkle trees and evaluation algorithms to construct a distributed fault assessment forest; secondly, the distributed fault assessment forest and a preset cross-factory real-time monitoring network are used to carry out distributed edge interaction status assessment to obtain assessment results of single equipment and process-related equipment; thirdly, based on these results and the root cause tracing warning rules and strategy generation library, fault tracing is carried out, and maintenance strategies are generated and shared. At the same time, with the help of the traceability warning visualization mapping space, the entire tracing and sharing process is visualized, and finally a visual fault assessment warning forest is formed, which realizes the accurate assessment of the status of aluminum processing equipment across factories, fault tracing, and the effective generation and sharing of maintenance strategies.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for assessing the health status of aluminum processing equipment, comprising:

[0006] Based on cross-plant equipment attributes and operating information combined with Merkle trees and evaluation algorithms, a distributed fault assessment forest is obtained;

[0007] Based on the distributed fault assessment forest and the preset cross-plant real-time monitoring network, distributed edge interactive status assessment is performed to obtain single equipment assessment results and process-related equipment assessment results;

[0008] Based on the evaluation results of single equipment and process-related equipment, combined with the root cause tracing warning rules and strategy generation library configured in a distributed fault assessment forest, fault tracing and maintenance strategy generation and sharing are carried out. The tracing and sharing process is visualized in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning forest.

[0009] Specifically, the distributed fault assessment forest includes M regional distributed fault assessment trees; the regional distributed fault assessment tree includes N distributed fault assessment subtrees; and the construction process of the regional distributed fault assessment tree includes:

[0010] Based on the monitoring equipment locations of all equipment configured in a single plant, the physical properties of the equipment, the single-state operation information, and the data flow relationship between each device, the monitoring nodes and directed link relationship of the real-time monitoring subnet corresponding to a single plant are constructed;

[0011] Obtaining an evaluation node corresponding to each device in the current factory based on a one-to-one mapping of each monitoring node in the real-time monitoring subnet;

[0012] Based on the data flow relationship between each device, the corresponding production process operation information between different devices, and the historical equipment related fault information, the correlation algorithm combined with the Bayesian algorithm is used to obtain the causal correlation degree and corresponding confidence level between the faults of each device;

[0013] Based on the inverse of the fault causal correlation between each device, a directed causal connection between each device is constructed. The directed fault causal correlation between each device is visualized using the corresponding confidence level and the visualization mapping constructed using the preset HSV space to obtain a visualized directed causal connection.

[0014] Based on the evaluation nodes corresponding to each device and the visualized directed causal connections between the devices combined with the blockchain, a horizontal production process association evaluation chain based on production process association is obtained.

[0015] Specifically, the construction process of the regional distributed fault assessment tree also includes:

[0016] The evaluation node corresponding to each device is used as the generation root node of each distributed fault assessment subtree. Based on the device status information and interaction information obtained in real time by the monitoring node corresponding to the generation root node, the time information hash block sequence corresponding to each generation root node is obtained through the Merkle tree algorithm built into the monitoring node;

[0017] Each time information hash block corresponding to each generation root node is used to store the device operation status information, physical attribute information and interaction information of each monitoring node within a preset operation time interval and the corresponding feature hash value;

[0018] Based on the time information hash block sequence corresponding to each generation root node, the device status evaluation score of each generation root node under each time information hash block is obtained through the evaluation algorithm built into the generation root node;

[0019] The device status evaluation score of each generated root node under each time information hash block is combined with the preset fault discrimination score threshold and fault type library to obtain whether the corresponding device under the current time information hash block has a fault;

[0020] If it exists, obtain the fault level and fault type corresponding to the current time information hash block, and use the fault level and fault type corresponding to the current time information hash block to construct the abnormal state node of the corresponding distributed fault assessment subtree under the current time information hash block. If it does not exist, construct the normal state node;

[0021] Repeat the state node construction process to obtain the abnormal state node set and normal state node set of each generated root node under different time information hash blocks.

[0022] Specifically, the status node is used to store the operating status and characteristic hash value corresponding to each time information hash block of the corresponding device; the operating status includes normal status and fault status and the corresponding fault timestamp; the fault status includes fault type and the corresponding abnormality score and level; the normal state is represented by 0; the fault state is represented by the corresponding fault type and fault level value;

[0023] Specifically, the construction process of the regional distributed fault assessment tree also includes:

[0024] If the current abnormal state node set has only one fault type, the generation root node is used as the initial abnormal state node, and the longitudinal time connection between the two adjacent abnormal state nodes corresponding to each generation root node is constructed through the total length of the running time interval corresponding to the normal state nodes contained between the two adjacent abnormal state nodes and the corresponding feature hash value;

[0025] If there is more than one fault type in the current abnormal state node set, the longitudinal time connection between adjacent abnormal state nodes corresponding to the new fault type is constructed based on the total length of the running time interval of all abnormal state nodes and normal state nodes corresponding to non-new fault types between the first abnormal state node corresponding to the new fault type and the generated root node and the corresponding feature hash value.

[0026] Specifically, the construction process of the regional distributed fault assessment tree also includes:

[0027] When the current fault type occurs again, the corresponding fault type abnormal state node and the corresponding longitudinal time connection are constructed under the first abnormal state node;

[0028] When a new fault type appears again and the process of generating longitudinal time connections with more than one fault type is repeated, a new abnormal state node and longitudinal time connection are generated under the current generation root node;

[0029] Based on the abnormal state nodes corresponding to different fault types and the corresponding longitudinal time connections, the corresponding distributed fault assessment subtree under the current generated root node is constructed;

[0030] Based on the distributed fault assessment subtrees corresponding to different generation root nodes, the horizontal production process association evaluation chains corresponding to different time information hash blocks, and the verification hash between the state nodes with interactive relationships between different distributed fault assessment subtrees, a regional distributed fault assessment tree is constructed through a distributed framework.

[0031] Specifically, the construction process of the regional distributed fault assessment tree also includes:

[0032] Based on the regional distributed fault assessment tree, the real-time time information hash blocks obtained from different monitoring nodes in the current plant are combined with the preset confidence threshold and fault discrimination score threshold corresponding to the fault causal correlation between each device and the fault type library for training to obtain the regional distributed fault assessment tree trained in the current plant;

[0033] When a distributed fault assessment subtree corresponding to a new device appears, the above process is repeated to update and train the distributed fault assessment tree in the current area.

[0034] Specifically, the construction process of the distributed fault assessment forest includes:

[0035] The feature hash values ​​and operating status values ​​of all status nodes in the regional distributed fault assessment tree corresponding to each plant area, as well as the verification hashes between the status nodes with interactive relationships between different distributed fault assessment subtrees, are stored in the interactive storage block constructed by combining the root node of each regional distributed fault assessment tree with the blockchain. At the same time, based on the interactive storage block corresponding to each regional distributed fault assessment tree and the interactive relationship between the production process flows corresponding to different plant areas, a directed block connection is constructed;

[0036] Based on the characteristic hash value and verification hash stored in the interactive storage block corresponding to each regional distributed fault assessment tree, a zero-knowledge proof algorithm is used to obtain cross-region traceability verification proofs between interactive storage blocks with interactive relationships.

[0037] The cross-regional traceability verification proof is mapped to the block directed connection between the corresponding interactive storage blocks to construct a cross-regional directed block traceability verification chain.

[0038] Specifically, the construction process of the distributed fault assessment forest also includes:

[0039] Each regional distributed fault assessment tree is used as a federated interactive edge node. The cross-regional directed block traceability verification chain is used as the cross-regional security traceability verification path, and the visual mapping in the visual directed causal connection is mapped to the corresponding block directed connection.

[0040] Based on the federated interactive edge nodes, cross-regional security traceability verification path, preset verification attack library, preset traceability accuracy threshold and security verification threshold, the federated algorithm is trained to obtain a trained distributed fault assessment forest.

[0041] Specifically, the construction process of the visual fault assessment and warning forest includes:

[0042] Based on the trained distributed fault assessment forest and the real-time monitoring network, all plant equipment is monitored and assessed in real time to obtain a distributed fault assessment tree for areas with abnormal conditions.

[0043] Based on the horizontal production process association evaluation chain and distributed fault assessment subtree in the regional distributed fault assessment tree with the current abnormal state, the root cause of the abnormality and the associated abnormal nodes within the factory are traced. Based on the corresponding fault causal correlation and the corresponding confidence level between the root cause of the abnormality and the associated abnormal nodes, the corresponding level of traceability warning is performed in combination with the preset warning level, and the traceability process is visualized based on the visual directed causal connection;

[0044] At the same time, based on the trained distributed fault assessment forest, the distributed fault assessment tree with the current abnormal state and the associated distributed fault assessment tree with a fault causal relationship are traced across the plant area for safety, and the horizontal production process associated assessment chain and the distributed fault assessment subtree in the block directed connection and associated distributed fault assessment tree combined with visual mapping are used to trace the associated fault equipment across the plant area and visualize the tracing process. The fault causal correlation degree and the corresponding confidence level of each connection relationship in the corresponding traceability path are combined with the preset warning level to perform cross-plant difference warning, and obtain a visual fault assessment warning forest;

[0045] The warning level includes a main warning level and an associated warning level; the main warning level corresponds to the main fault device monitored at the current moment; the associated warning level corresponds to the associated fault device at the current moment;

[0046] The main warning level is issued according to the fault level corresponding to the main fault device; the associated fault device is issued according to the fault level corresponding to the associated fault device and the fault causal correlation and confidence between the associated fault device and the main fault device.

[0047] An aluminum processing equipment health status assessment system includes: an assessment tree module, a monitoring and assessment module, and a reasoning mapping module;

[0048] The evaluation tree module obtains a distributed fault evaluation tree based on cross-plant equipment attributes and operation information combined with a Merkle tree and an evaluation algorithm;

[0049] The monitoring and evaluation module performs distributed edge interaction status evaluation based on a distributed fault evaluation tree combined with a preset cross-plant real-time monitoring network to obtain single equipment evaluation results and process-related equipment evaluation results;

[0050] The reasoning mapping module generates and shares fault tracing and maintenance strategies based on the single equipment assessment results and process-related equipment assessment results combined with the root cause tracing warning rules and strategy generation library configured in the distributed fault assessment forest. It also visualizes the tracing and sharing process in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning tree.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] In response to the shortcomings of the existing technology, the present invention constructs a trusted storage structure for cross-factory equipment status data through a blockchain-based Merkle tree and distributed evaluation algorithm, and uses time information hash block sequences and feature hash values ​​to achieve verifiability and tamper-proof capabilities for equipment operating status; combines the Bayesian association algorithm with a fault causal correlation visualization mapping mechanism to establish a horizontal production process correlation evaluation chain, effectively quantifies the fault propagation path and confidence between equipment, and enhances the causal reasoning accuracy of multi-device collaborative monitoring; integrates regional distributed fault assessment trees and cross-regional directed block traceability verification chains through a federated learning framework to achieve cross-factory collaborative modeling and security verification under data privacy protection, and uses zero-knowledge proof to ensure trusted traceability of cross-domain interaction processes; based on HSV spatial visualization mapping and multi-level warning rules, dynamically maps fault nodes, causal connections, and traceability paths into a multi-dimensional visualization topology, combines longitudinal time connections with spatiotemporal correlation analysis of abnormal state node sets, and constructs a spatiotemporal coupling representation system for the fault evolution process, significantly improving the early identification and cross-domain root cause location capabilities of hidden faults in cross-factory aluminum processing equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of a method for assessing the health status of aluminum processing equipment according to Example 1 of the present invention;

[0054] Figure 2 This is a module diagram of an aluminum processing equipment health status assessment system according to Example 2 of the present invention. DETAILED DESCRIPTION

[0055] Example 1

[0056] See also Figure 1 The present invention provides an embodiment of a method for evaluating the health status of aluminum processing equipment, comprising the following steps:

[0057] A method for assessing the health status of aluminum processing equipment, comprising:

[0058] S1. Based on the cross-plant equipment attributes and operation information combined with the Merkle tree and evaluation algorithm, a distributed fault assessment forest is obtained;

[0059] S2. Based on the distributed fault assessment forest and the preset cross-plant real-time monitoring network, a distributed edge interactive status assessment is performed to obtain the assessment results of single equipment and process-related equipment;

[0060] S3. Based on the single equipment assessment results and process-related equipment assessment results, combined with the root cause tracing warning rules and strategy generation library configured in the distributed fault assessment forest, fault tracing and maintenance strategy generation and sharing are performed. The tracing and sharing process is visualized in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning forest.

[0061] Furthermore, the strategy generation library in this embodiment is based on historical fault data and maintenance strategies, and is constructed through deep text reasoning generation models and knowledge graphs;

[0062] Furthermore, the distributed fault assessment forest in this embodiment includes M regional distributed fault assessment trees; the regional distributed fault assessment tree includes N distributed fault assessment subtrees;

[0063] Furthermore, the process of constructing the regional distributed fault assessment tree in this embodiment includes:

[0064] Based on the monitoring equipment locations of all equipment configured in a single plant, the physical properties of the equipment, the single-state operation information, and the data flow relationship between each device, the monitoring nodes and directed link relationship of the real-time monitoring subnet corresponding to a single plant are constructed;

[0065] Furthermore, in this embodiment, the specific variable information corresponding to the monitoring device location points, device physical properties, single state operation information, and data flow relationship between each device configured for all devices includes:

[0066] The monitoring equipment locations for all equipment within a single plant. For example, a plant may contain a melting furnace, a casting mill, a cold rolling mill, and other equipment. The melting furnace is equipped with temperature and pressure sensors; the casting mill is equipped with vibration sensors and speed sensors. These locations are recorded as coordinates, accurate to the specific installation location within the equipment.

[0067] Device physical properties, such as model, size, material, and rated power, describe the basic physical characteristics of the device and provide valuable reference for subsequent analysis of the device's operating status.

[0068] Single-state operating information, such as the current temperature and pressure of the melting furnace; the current vibration amplitude and operating speed of the casting and rolling mill, etc., these information are the real-time operating parameters of the equipment at a certain moment.

[0069] The data flow relationship between various devices. For example, the molten aluminum produced by the smelting furnace is transported to the casting and rolling mill through a pipeline. This indicates that there is a data flow from the smelting furnace to the casting and rolling mill, which can be expressed as smelting furnace → casting and rolling mill.

[0070] For example, the detailed example process of establishing the relationship between monitoring nodes and directed links of the real-time monitoring subnet in this embodiment includes:

[0071] For each monitoring device location, a monitoring node is created; for example, a monitoring node N_A1 is created based on the temperature sensor location A1 of the smelting furnace. The node contains the sensor location information, the device information, and the initial data structure for subsequent data storage.

[0072] Add relevant attribute tags to each monitoring node based on the physical properties of the device. For example, add attributes such as device model and material to monitoring node N_A1 to distinguish the monitoring data characteristics of different devices later.

[0073] Based on the single state operation information, the data value of the monitoring node is updated in real time. For example, when the temperature sensor detects a new temperature value T2, the temperature data in the monitoring node N_A1 is updated to T2.

[0074] Directed links are constructed based on the data flow between devices. For example, for data flow from a smelting furnace to a casting mill, directed edges are created from the monitoring nodes associated with the smelting furnace (such as N_A1 and N_A2) to the monitoring nodes associated with the casting mill (such as N_B1 and N_B2), forming a directed link relationship. N_B1 is the monitoring node created based on the speed sensor location B2, and the construction of other monitoring nodes is not detailed here. These directed edges not only indicate the data flow direction but also carry information such as the data transmission rate and latency.

[0075] Furthermore, the output of this embodiment includes:

[0076] The monitoring node set of the real-time monitoring subnet includes the monitoring nodes corresponding to all equipment monitoring points in a single factory, such as {N_A1, N_A2, N_B1, N_B2, ...}. Each node has detailed equipment location, attributes and real-time data information.

[0077] Directed link relationship set: A set of directed edges that represent the direction of data flow between devices, such as {(N_A1, N_B1), (N_A2, N_B2), ...}. Each directed edge carries the necessary information related to data transmission.

[0078] Obtaining an evaluation node corresponding to each device in the current factory based on a one-to-one mapping of each monitoring node in the real-time monitoring subnet;

[0079] Based on the data flow relationship between each device, the corresponding production process operation information between different devices, and the historical equipment related fault information, the correlation algorithm combined with the Bayesian algorithm is used to obtain the causal correlation degree and corresponding confidence level between the faults of each device;

[0080] For example, the detailed example process of calculating the fault causal correlation and confidence level in this embodiment includes:

[0081] Obtain the data flow relationship between each device and the corresponding production process associated operation information between different devices, such as the data flow information from the melting furnace to the casting and rolling mill, to clarify the associated path between the devices. In this embodiment, the corresponding production process associated operation information between different devices includes the process correlation between the temperature and pressure of the melting furnace and the aluminum liquid quality and casting and rolling speed of the casting and rolling mill.

[0082] In this embodiment, when the temperature of the melting furnace is too high, it may cause quality problems in the aluminum strip produced by the casting and rolling mill. Such related information includes the mutual influence relationship between process parameters and the description of the degree of influence, as well as historical equipment-related fault information. For example, it records relevant information when equipment failures occurred in the past, such as the abnormal increase in the temperature of the melting furnace at a certain moment, followed by the aluminum strip breakage failure of the casting and rolling mill, as well as data such as the time and frequency of the failure.

[0083] Based on historical equipment-related fault information, a priori probability model of fault occurrence is constructed. For example, the number of aluminum strip breakage failures that occurred in the casting and rolling mill after the abnormal increase in the melting furnace temperature over the past period of time, as well as the total number of abnormal increases in the melting furnace temperature, are counted. The priori probability P of aluminum strip breakage failures occurring in the casting and rolling mill under the condition of abnormally increased melting furnace temperature is calculated. τ (Casting and rolling mill failure | Abnormal melting furnace temperature).

[0084] The data flow relationships between various devices and the associated operational information of the production process are used to update the probability model of failure. For example, considering the impact of factors such as pipeline pressure on the quality of molten aluminum during its transfer from the smelting furnace to the casting and rolling mill, and the impact of molten aluminum quality on the operating status of the casting and rolling mill, the Bayesian formula is combined with new evidence, such as pipeline pressure data, to update the probability of casting and rolling mill failure.

[0085] For each pair of equipment, calculate the degree of causal correlation between them. This degree of causal correlation can be determined by calculating the change in the probability that a failure in one equipment will cause another equipment failure under various conditions. For example, if a temperature anomaly occurs in a smelting furnace, the probability of a casting mill failure increases from P3 to P4. The causal correlation can be defined as (P4 - P3) / P3.

[0086] At the same time, the confidence level of the fault causal correlation is determined by analyzing historical data and model calculation results. The confidence level reflects the reliability of the calculated fault causal correlation. For example, the confidence level of the current calculation result can be determined by counting the proportion of calculated fault causal correlations in similar past situations that match the actual situation.

[0087] Furthermore, the corresponding output results of this embodiment include: a fault causal correlation matrix between each device and a fault causal correlation confidence matrix between each device. In this embodiment, the fault causal correlation matrix between each device comprises a two-dimensional matrix, with the rows and columns representing different devices, and the matrix elements representing the fault causal correlation between the corresponding devices. For example, for device A and device B, the element [A, B] in the matrix represents the fault causal correlation degree that a fault in device A causes a fault in device B.

[0088] The fault causal relationship confidence matrix between devices in this embodiment is also a two-dimensional matrix, corresponding to the fault causal relationship matrix. The matrix elements represent the confidence levels of the fault causal relationships between the corresponding devices. For example, the element [A, B] in the matrix represents the confidence level of the fault causal relationship that the fault of device A causes the fault of device B.

[0089] Based on the inverse of the fault causal correlation between each device, a directed causal connection between each device is constructed. The directed fault causal correlation between each device is visualized using the corresponding confidence level and the visualization mapping constructed using the preset HSV space to obtain a visualized directed causal connection.

[0090] For example, the example process of constructing a visual directed causal connection in this embodiment includes:

[0091] Obtain the fault causal correlation matrix between each device. For example, the matrix describing the fault causal correlation between devices obtained in the previous step uses a preset visualization mapping rule constructed using the HSV space. For example, the HSV (Hue, Saturation, Value) space is used to map numerical information into visual colors.

[0092] Furthermore, in this embodiment, the higher the fault causal correlation is, the more the hue leans towards red, and the saturation and brightness are based on a preset rule of positive correlation with the hue to show the strength of the correlation.

[0093] For the fault causal correlation matrix between each device, its inverse is calculated. Because the higher the fault causal correlation, the smaller its inverse is, which makes it easier to perform numerical mapping in subsequent visualization.

[0094] Based on the inverse of the fault causal correlation, the visualization mapping rules constructed using the preset HSV space are used to assign colors and other visualization attributes to the directed causal connection between each pair of devices. For example, for the connection between device A and device B with an inverse of the fault causal correlation of a1, the hue is set to a value close to red according to the HSV mapping rules. The saturation and brightness are determined according to specific rules. At the same time, the line thickness is set according to the size of the inverse of the correlation. The smaller the inverse of the correlation (that is, the higher the fault causal correlation), the thicker the line.

[0095] The corresponding confidence level can be used to further adjust the visualization properties. For example, if the confidence level of the causal relationship between device A and device B is low, the transparency of the connecting line color can be appropriately reduced to indicate that its reliability is high but not absolutely certain.

[0096] Furthermore, the output of this embodiment includes: a set of visualized directed causal connections, wherein the set of visualized directed causal connections includes a visual representation of the directed causal connections between all devices, and each connection has visual attributes such as color, line thickness, and transparency determined according to the inverse of the fault causal correlation and the confidence level, intuitively showing the strength and reliability of the fault causal relationship between devices.

[0097] Based on the evaluation nodes corresponding to each device and the visual directed causal connection between each device and the blockchain, a horizontal production process correlation evaluation chain based on production process correlation is obtained;

[0098] The evaluation node corresponding to each device is used as the generation root node of each distributed fault assessment subtree. Based on the device status information and interaction information obtained in real time by the monitoring node corresponding to the generation root node, the time information hash block sequence corresponding to each generation root node is obtained through the Merkle tree algorithm built into the monitoring node;

[0099] Illustratively, in this embodiment, the exemplary process of obtaining the root node related information includes:

[0100] Obtain a set of evaluation nodes corresponding to each device. Each evaluation node in the set corresponds to a specific device within the plant and contains various attributes and associated information about the device. Monitoring nodes also acquire real-time device status and interaction information. For example, monitoring nodes acquire real-time status information such as the temperature and pressure of the smelting furnace, as well as data interaction information with other devices, such as the amount and time of aluminum liquid transferred. This information is updated in real time to reflect the current operating status of the device.

[0101] For each evaluation node, the associated monitoring node collects real-time equipment status and interaction information. For example, for the generation root node corresponding to the smelting furnace, the monitoring node continuously collects data from temperature and pressure sensors, as well as data interaction information with the casting and rolling mill.

[0102] Using the monitoring node's built-in Merkle tree algorithm, time information hash blocks are generated at preset time intervals. Each time information hash block stores the device operating status, physical property information, and interaction information acquired by the monitoring node during that time period, along with the characteristic hash value calculated from this information. For example, information such as the melting furnace's temperature, pressure, and aluminum liquid transfer volume over a period of m minutes is combined into a data block, and its SHA-256 hash value is calculated to form the characteristic hash value in the time information hash block.

[0103] Generate a sequence of time information hash blocks corresponding to each generation root node in chronological order. For example, starting from time 0, generate a time information hash block every m minutes, and arrange them in sequence to form the sequence {HB1, HB2, HB3, ..., HBn}, where HBn represents the nth time information hash block.

[0104] Each time information hash block corresponding to each generation root node is used to store the device operation status information, physical attribute information and interaction information of each monitoring node within a preset operation time interval and the corresponding feature hash value;

[0105] Based on the time information hash block sequence corresponding to each generation root node, the device status evaluation score of each generation root node under each time information hash block is obtained through the evaluation algorithm built into the generation root node;

[0106] The device status evaluation score of each generated root node under each time information hash block is combined with the preset fault discrimination score threshold and fault type library to obtain whether the corresponding device under the current time information hash block has a fault;

[0107] If it exists, obtain the fault level and fault type corresponding to the current time information hash block, and use the fault level and fault type corresponding to the current time information hash block to construct the abnormal state node of the corresponding distributed fault assessment subtree under the current time information hash block. If it does not exist, construct the normal state node;

[0108] Repeat the state node construction process to obtain the abnormal state node set and normal state node set of each generated root node under different time information hash blocks;

[0109] Illustratively, the example process of fault judgment and status node construction in this embodiment includes:

[0110] Obtain the device status evaluation score set and fault type library for each generated root node under each time information hash block, where the evaluation score set contains the set of device status evaluation scores for each time period; the fault type library contains various possible equipment fault types and their corresponding feature descriptions, fault causes and other information. For example, the fault type library records "overtemperature fault", which is characterized by the temperature exceeding the normal range to a certain extent, and the possible cause is a cooling system failure, etc.

[0111] Each device status evaluation score for each generated root node is compared with a preset fault discrimination score threshold. For example, for the device status evaluation score Sn_melting for the smelting furnace generated root node, if Sn_melting is less than the preset fault discrimination threshold, the device is determined to have a fault in that time period.

[0112] If a fault is determined, the fault level and type are determined based on the device status assessment score and the fault type library. Based on the fault determination result, an abnormal status node is constructed. If the device is in a normal state (the device status assessment score is greater than or equal to the fault discrimination score threshold), a normal status node is constructed, recording the operating state as "0" (indicating a normal state) and obtaining the characteristic hash value of the corresponding time information hash block. If the device is faulty, an abnormal status node is constructed, recording the operating state as the fault type and fault level, the fault timestamp, the abnormality score (e.g., obtained by mapping the device status assessment score to the severity of the corresponding fault type in the fault type library), and the characteristic hash value of the corresponding time information hash block.

[0113] Repeat the above process to construct the corresponding abnormal state node set and normal state node set for each generated root node under different time information hash blocks.

[0114] Furthermore, the state node in this embodiment is used to store the operating state and the corresponding characteristic hash value corresponding to each time information hash block of the corresponding device; the operating state includes a normal state and a fault state and the corresponding fault timestamp; the fault state includes a fault type and a corresponding abnormality score and level; the normal state is represented by 0; the fault state is represented by the corresponding fault type and fault level value;

[0115] If the current abnormal state node set has only one fault type, the generation root node is used as the initial abnormal state node, and the longitudinal time connection between the two adjacent abnormal state nodes corresponding to each generation root node is constructed through the total length of the running time interval corresponding to the normal state nodes contained between the two adjacent abnormal state nodes and the corresponding feature hash value;

[0116] If there is more than one fault type in the current abnormal state node set, the longitudinal time connection between adjacent abnormal state nodes corresponding to the new fault type is constructed based on the total length of the running time interval of all abnormal state nodes and normal state nodes corresponding to non-new fault types between the first abnormal state node corresponding to the new fault type and the generated root node and the corresponding feature hash value.

[0117] When the current fault type occurs again, the corresponding fault type abnormal state node and the corresponding longitudinal time connection are constructed under the first abnormal state node;

[0118] When a new fault type appears again, repeat the above process of generating longitudinal time connections when there is more than one fault type, and generate a new abnormal state node and longitudinal time connection under the current generation root node;

[0119] Based on the abnormal state nodes corresponding to different fault types and the corresponding longitudinal time connections, the corresponding distributed fault assessment subtree under the current generated root node is constructed;

[0120] Illustratively, the example process of constructing a longitudinal time connection in this embodiment includes:

[0121] Obtain all abnormal state nodes that appear in the generation root node under different time information hash blocks and all normal state nodes that appear in the generation root node under different time information hash blocks;

[0122] Using the generated root node as the initial abnormal state node, for each of two adjacent abnormal state nodes, calculate the total length of the runtime intervals corresponding to the normal state nodes between them. For example, if there are two normal state nodes, N1_normal_melting and N2_normal_melting, between abnormal state nodes A1_melting and A2_melting, and their runtime intervals are recorded as [t1, t2] and [t3, t4], respectively, the total length is (t2 - t1) + (t4 - t3).

[0123] At the same time, the feature hash values ​​corresponding to the two adjacent abnormal state nodes are obtained. For example, the feature hash value of A1_melting is H1, and the feature hash value of A2_melting is H2.

[0124] Based on the above information, a longitudinal time connection is constructed between each generated root node and the two adjacent abnormal state nodes. The connection records information such as the total length of the time interval, the feature hash value of the starting abnormal state node, and the feature hash value of the ending abnormal state node.

[0125] For the first abnormal state node corresponding to the new fault type, set to A_new1, calculate the total runtime length of all abnormal state nodes and normal state nodes corresponding to non-new fault types within the interval between it and the generated root node. For example, before A_new1, there are A1_melting (which belongs to the old fault type and the duration of the old fault type is obtained), as well as two normal state nodes, N1_normal_melting and N2_normal_melting. Calculate the total runtime length of their runtime intervals.

[0126] Based on the nodes corresponding to the new fault type and the abnormal state nodes corresponding to all non-new fault types, as well as the total length of the running time interval corresponding to the normal state nodes, combined with the hash algorithm, the characteristic hash value of the node corresponding to the new fault type is obtained. For example, the characteristic hash value of A_new1 is H_new1, the characteristic hash value of A1_melting is H1, the characteristic hash value of N1_normal_melting is H_N1, and so on.

[0127] Based on the characteristic hash values ​​of the nodes corresponding to the new fault type and the abnormal state nodes corresponding to all non-new fault types, as well as the normal state nodes, a longitudinal time connection is constructed between adjacent abnormal state nodes corresponding to the new fault type; the connection also records information such as the total length of the time interval and the characteristic hash values ​​of the involved nodes.

[0128] When the current fault type reappears, an abnormal state node of the corresponding fault type and the corresponding longitudinal time connection are constructed under the first abnormal state node. For example, if the abnormal state node A3_melting of the same fault type as A1_melting reappears, a longitudinal time connection is constructed under A1_melting, and the total length of the time interval between A1_melting and A3_melting and the related feature hash value and other information are calculated.

[0129] When a new fault type appears again, the above process is repeated, generating a new abnormal state node and a longitudinal time connection at the current generation root node. For example, if an abnormal state node A_new2 of a new fault type appears, the total length of the time interval and the characteristic hash value between A_new2 and the previously related nodes are calculated to construct a longitudinal time connection; in this embodiment, the previously related nodes include the generation root node, the existing abnormal state node, and the normal state node.

[0130] Furthermore, in this embodiment, the longitudinal time connection sets corresponding to different fault types include longitudinal time connection sets corresponding to different fault types. These connections reflect the order in which the equipment fails at different times, changes in fault types, and information about normal operation time periods.

[0131] Based on the distributed fault assessment subtrees corresponding to different generation root nodes, the horizontal production process association evaluation chains corresponding to different time information hash blocks, and the verification hash between the state nodes with interactive relationships between different distributed fault assessment subtrees, a regional distributed fault assessment tree is constructed through a distributed framework.

[0132] Based on the regional distributed fault assessment tree, the real-time time information hash blocks obtained from different monitoring nodes in the current plant are combined with the preset confidence threshold and fault discrimination score threshold corresponding to the fault causal correlation between each device and the fault type library for training to obtain the trained regional distributed fault assessment tree corresponding to the current plant;

[0133] When a distributed fault assessment subtree corresponding to a new device appears, the above process is repeated to update and train the distributed fault assessment tree in the current area.

[0134] This process builds a cross-level fault assessment system through the integration of multi-dimensional technologies. First, a real-time monitoring subnet is established based on the relationship between the physical properties of the equipment and the data flow direction. The equipment topology is accurately mapped with coordinate-level monitoring nodes and directed links. The holographic mapping of physical space and data space is realized by combining the dynamic update mechanism of equipment status parameters. The Bayesian algorithm is used to integrate historical fault data and process correlation information, quantify the causal correlation and confidence matrix between equipment faults, and establish a fault propagation model based on probability evolution, breaking through the isolated judgment limitations of traditional threshold alarms. The inverse of the fault causal correlation is innovatively combined with the visualization mapping rule of HSV space to dynamically convert the correlation strength and confidence into a visualization connection of hue-saturation-lightness multi-parameter coupling, realizing an intuitive topological expression of complex causal relationship networks. With the help of blockchain technology, the evaluation nodes and causal connections are connected The system constructs a horizontal production process correlation evaluation chain based on information, generates a time information hash block sequence through the Merkle tree and embeds the characteristic hash value to form a spatiotemporal fingerprint database of the equipment's operating status throughout its life cycle; in the fault assessment link, the distributed fault assessment subtree based on the generated root node constructs an abnormal state node set through longitudinal time connection, and combines the normal node operating interval length and the characteristic hash value to establish a spatiotemporal correlation model of fault evolution, effectively capturing the spatiotemporal coupling characteristics of intermittent faults and compound faults; when the distributed framework integrates multiple subtrees, it adopts verification hash verification to check the cross-subtree interaction relationship, and combines the federated learning framework to realize the dynamic expansion of the regional evaluation tree and cross-domain security verification, forming a three-dimensional evaluation system with both equipment-level fine-grained evaluation and cross-factory collaborative traceability, which significantly improves the early identification accuracy and multi-level root cause location efficiency of hidden faults in complex manufacturing systems.

[0135] Furthermore, the construction process of the distributed fault assessment forest in this embodiment includes:

[0136] The feature hash values ​​and operating status values ​​of all status nodes in the regional distributed fault assessment tree corresponding to each plant area, as well as the verification hashes between the status nodes with interactive relationships between different distributed fault assessment subtrees, are stored in the interactive storage block constructed by combining the root node of each regional distributed fault assessment tree with the blockchain. At the same time, based on the interactive storage block corresponding to each regional distributed fault assessment tree and the interactive relationship between the production process flows corresponding to different plant areas, a directed block connection is constructed;

[0137] Based on the characteristic hash value and verification hash stored in the interactive storage block corresponding to each regional distributed fault assessment tree, a zero-knowledge proof algorithm is used to obtain cross-region traceability verification proofs between interactive storage blocks with interactive relationships.

[0138] The cross-regional traceability verification proof is mapped to the block directed connection between the corresponding interactive storage blocks to construct a cross-regional directed block traceability verification chain.

[0139] Each regional distributed fault assessment tree is used as a federated interactive edge node, and the cross-regional directed block traceability verification chain is used as the cross-regional security traceability verification path. The visual mapping in the visual directed causal connection is mapped to the corresponding block directed connection.

[0140] Based on the federated interactive edge nodes, cross-regional security traceability verification path, preset verification attack library, preset traceability accuracy threshold and security verification threshold, combined with the federated algorithm for training, a trained distributed fault assessment forest is obtained.

[0141] Furthermore, the process of constructing the visual fault assessment and warning forest in this embodiment includes:

[0142] Based on the trained distributed fault assessment forest and the real-time monitoring network, all plant equipment is monitored and assessed in real time to obtain a distributed fault assessment tree for areas with abnormal conditions.

[0143] Based on the horizontal production process association evaluation chain and distributed fault assessment subtree in the regional distributed fault assessment tree with the current abnormal state, the root cause of the abnormality and the associated abnormal nodes within the factory are traced. Based on the corresponding fault causal correlation and the corresponding confidence level between the root cause of the abnormality and the associated abnormal nodes, the corresponding level of traceability warning is performed in combination with the preset warning level, and the traceability process is visualized based on the visual directed causal connection;

[0144] At the same time, based on the trained distributed fault assessment forest, the distributed fault assessment tree with the current abnormal state and the associated distributed fault assessment tree with a fault causal relationship are traced across the plant area for safety, and the horizontal production process associated assessment chain and the distributed fault assessment subtree in the block directed connection and associated distributed fault assessment tree combined with visual mapping are used to trace the associated fault equipment across the plant area and visualize the tracing process. The fault causal correlation degree and the corresponding confidence level of each connection relationship in the corresponding traceability path are combined with the preset warning level to perform cross-plant difference warning, and obtain a visual fault assessment warning forest;

[0145] Furthermore, the warning level in this embodiment includes a main warning level and an associated warning level; the main warning level corresponds to the main fault device monitored at the current moment; the associated warning level corresponds to the associated fault device at the current moment;

[0146] Furthermore, in this embodiment, the primary warning level is determined by the fault level of the primary faulty device. Associated faulty devices are assigned corresponding warning levels based on the fault levels of the associated faulty devices and the degree of causal correlation and confidence between the associated faulty devices and the primary faulty device. In this embodiment, the primary faulty device is the device that directly triggers the abnormality alarm or the core influencing device in the fault causal chain.

[0147] Furthermore, in this embodiment, the maintenance strategy generation and sharing process is as follows: based on the functional attributes of the corresponding faulty equipment, the generated maintenance strategy is shared to the production root node corresponding to the equipment with the same production function through the distributed fault assessment forest.

[0148] This process achieves multi-dimensional technological collaborative innovation by deeply integrating blockchain technology with a distributed fault assessment tree and combining it with a zero-knowledge proof algorithm to construct a cross-regional directed block traceability verification chain. Specifically, by embedding device state feature hashes and interactive verification hashes into the blockchain's interactive storage blocks, the blockchain's immutable nature ensures data traceability throughout its entire lifecycle. A zero-knowledge proof algorithm is used to generate cross-regional traceability verification certificates, enabling trusted verification of cross-site interactions without exposing sensitive data, effectively resolving the conflict between data privacy and verification validity in traditional methods. By integrating distributed fault assessment trees across multiple sites through a federated learning framework, the system preserves data privacy across each site while achieving global optimization of model parameters, significantly improving the generalization of fault assessment. A dynamic mapping mechanism between visual directed causal connections and block directed connections transforms complex multi-level fault correlations into intuitive visual topological structures, significantly improving fault location efficiency. In particular, through a combined constraint mechanism of a pre-set verification attack library and dual thresholds, a dynamic security verification system is constructed during the federated training process. This not only enhances the model's defenses against security threats such as data tampering and man-in-the-middle attacks, but also enables model self-optimization through a traceability accuracy feedback mechanism. The systematic integration of these technical means enables the constructed distributed fault assessment forest to have core capabilities such as cross-regional data security sharing, multi-source heterogeneous data fusion analysis, and complex fault correlation reasoning, providing reliable, efficient, and explainable intelligent decision-making support for the collaborative operation and maintenance of large-scale industrial equipment groups.

[0149] Example 2

[0150] See also Figure 2 , another embodiment provided by the present invention: a health status assessment system for aluminum processing equipment, comprising: an assessment tree module, a monitoring and assessment module, and a reasoning mapping module;

[0151] The evaluation tree module obtains a distributed fault evaluation tree based on cross-plant equipment attributes and operating information combined with the Merkle tree and evaluation algorithm;

[0152] The monitoring and evaluation module performs distributed edge interaction status evaluation based on a distributed fault assessment tree combined with a preset cross-plant real-time monitoring network, obtaining evaluation results for both single equipment and process-related equipment.

[0153] The reasoning mapping module generates and shares fault tracing and maintenance strategies based on the single equipment assessment results and process-related equipment assessment results combined with the root cause tracing warning rules and strategy generation library configured in the distributed fault assessment forest. It also visualizes the tracing and sharing process in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning tree.

[0154] Example 3

[0155] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements a health status assessment method for aluminum processing equipment when executing the computer program.

[0156] A computer-readable storage medium stores computer instructions, which, when executed, execute a method for evaluating the health status of aluminum processing equipment.

[0157] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are protected by the present invention.

Claims

1. A method for assessing the health status of aluminum processing equipment, characterized in that: include: Based on cross-plant equipment attributes and operating information combined with Merkle trees and evaluation algorithms, a distributed fault assessment forest is obtained; Based on the distributed fault assessment forest and the preset cross-plant real-time monitoring network, distributed edge interactive status assessment is performed to obtain single equipment assessment results and process-related equipment assessment results; Based on the evaluation results of single equipment and process-related equipment, combined with the root cause tracing warning rules and strategy generation library configured in the distributed fault assessment forest, fault tracing and maintenance strategy generation and sharing are carried out. The tracing and sharing process is visualized in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning forest. The distributed fault assessment forest includes M regional distributed fault assessment trees; The regional distributed fault assessment tree includes N distributed fault assessment subtrees; The construction process of the regional distributed fault assessment tree includes: Based on the trained distributed fault assessment forest and the real-time monitoring network, all plant equipment is monitored and assessed in real time to obtain a distributed fault assessment tree for areas with abnormal conditions. Based on the horizontal production process association evaluation chain and distributed fault assessment subtree in the regional distributed fault assessment tree with abnormal status, the root causes of abnormalities and related abnormal nodes within the factory are traced, and traceability warnings are issued according to the fault causal correlation, confidence and preset warning level, and the traceability process is visualized through visual directed causal connections; at the same time, the associated distributed fault assessment tree with a fault causal relationship with the abnormal assessment tree is traced across factories safely through a cross-regional directed block traceability verification chain, and the horizontal production process association evaluation chain and distributed fault assessment subtree of the block directed connection and association evaluation tree combined with visual mapping are used to trace the associated fault equipment across factories and visualize the traceability process. The fault causal correlation, confidence and preset warning level of each connection relationship in the traceability path are combined to issue cross-factory difference warnings to obtain a visual fault assessment warning forest.

2. The method for evaluating the health status of aluminum processing equipment according to claim 1, wherein: The construction process of the regional distributed fault assessment tree includes: Based on the monitoring equipment locations of all equipment configured in a single plant, the physical properties of the equipment, the single-state operation information, and the data flow relationship between each device, the monitoring nodes and directed link relationship of the real-time monitoring subnet corresponding to a single plant are constructed; Obtaining an evaluation node corresponding to each device in the current factory based on a one-to-one mapping of each monitoring node in the real-time monitoring subnet; Based on the data flow relationship between each device, the corresponding production process operation information between different devices, and the historical equipment related fault information, the correlation algorithm combined with the Bayesian algorithm is used to obtain the causal correlation degree and corresponding confidence level between the faults of each device; Based on the inverse of the fault causal correlation between each device, a directed causal connection between each device is constructed. The directed fault causal correlation between each device is visualized using the corresponding confidence level and the visualization mapping constructed using the preset HSV space to obtain a visualized directed causal connection. Based on the evaluation nodes corresponding to each device and the visualized directed causal connections between the devices combined with the blockchain, a horizontal production process association evaluation chain based on production process association is obtained.

3. The method for evaluating the health status of aluminum processing equipment according to claim 2, wherein: The construction process of the regional distributed fault assessment tree further includes: The evaluation node corresponding to each device is used as the generation root node of each distributed fault assessment subtree. Based on the device status information and interaction information obtained in real time by the monitoring node corresponding to the generation root node, the time information hash block sequence corresponding to each generation root node is obtained through the Merkle tree algorithm built into the monitoring node. Each time information hash block corresponding to each generation root node is used to store the device operation status information, physical attribute information and interaction information of each monitoring node within a preset operation time interval and the corresponding feature hash value; Based on the time information hash block sequence corresponding to each generation root node, the device status evaluation score of each generation root node under each time information hash block is obtained through the evaluation algorithm built into the generation root node; The device status evaluation score of each generated root node under each time information hash block is combined with the preset fault discrimination score threshold and fault type library to obtain whether the corresponding device under the current time information hash block has a fault; If it exists, obtain the fault level and fault type corresponding to the current time information hash block, and use the fault level and fault type corresponding to the current time information hash block to construct the abnormal state node of the corresponding distributed fault assessment subtree under the current time information hash block. If it does not exist, construct the normal state node; Repeat the state node construction process to obtain the abnormal state node set and normal state node set of each generated root node under different time information hash blocks.

4. The method for evaluating the health status of aluminum processing equipment according to claim 3, wherein: The state node is used to store the operating state and characteristic hash value corresponding to each time information hash block of the corresponding device; The operating status includes a normal state and a fault state and corresponding fault timestamps; the fault state includes a fault type and corresponding abnormality score and level; The normal state is represented by 0; The fault state is represented by the corresponding fault type and fault level value; The construction process of the regional distributed fault assessment tree further includes: If the current abnormal state node set has only one fault type, the generation root node is used as the initial abnormal state node. The longitudinal time connection between the two adjacent abnormal state nodes corresponding to each generation root node is constructed by the total length of the running time interval corresponding to the normal state nodes contained between the two adjacent abnormal state nodes and the corresponding feature hash value; If there is more than one fault type in the current abnormal state node set, the longitudinal time connection between adjacent abnormal state nodes corresponding to the new fault type is constructed based on the total length of the running time interval of all abnormal state nodes and normal state nodes corresponding to non-new fault types between the first abnormal state node corresponding to the new fault type and the generated root node and the corresponding feature hash value.

5. The method for evaluating the health status of aluminum processing equipment according to claim 4, wherein: The construction process of the regional distributed fault assessment tree further includes: When the current fault type occurs again, the corresponding fault type abnormal state node and the corresponding longitudinal time connection are constructed under the first abnormal state node; When a new fault type appears again and the process of generating longitudinal time connections with more than one fault type is repeated, a new abnormal state node and longitudinal time connection are generated under the current generation root node; Based on the abnormal state nodes corresponding to different fault types and the corresponding longitudinal time connections, the corresponding distributed fault assessment subtree under the current generated root node is constructed; Based on the distributed fault assessment subtrees corresponding to different generation root nodes, the horizontal production process association evaluation chains corresponding to different time information hash blocks, and the verification hash between the state nodes with interactive relationships between different distributed fault assessment subtrees, a regional distributed fault assessment tree is constructed through a distributed framework.

6. The method for evaluating the health status of aluminum processing equipment according to claim 5, wherein: The construction process of the regional distributed fault assessment tree further includes: Based on the regional distributed fault assessment tree, the real-time time information hash blocks obtained from different monitoring nodes in the current plant are combined with the preset confidence threshold and fault discrimination score threshold corresponding to the fault causal correlation between each device and the fault type library for training to obtain the regional distributed fault assessment tree trained in the current plant; When a distributed fault assessment subtree corresponding to a new device appears, the above process is repeated to update and train the distributed fault assessment tree in the current area.

7. The method for evaluating the health status of aluminum processing equipment according to claim 6, wherein: The construction process of the distributed fault assessment forest includes: The feature hash values ​​and operating status values ​​of all status nodes in the regional distributed fault assessment tree corresponding to each plant area, as well as the verification hashes between the status nodes with interactive relationships between different distributed fault assessment subtrees, are stored in the interactive storage block constructed by combining the root node of each regional distributed fault assessment tree with the blockchain. At the same time, based on the interactive storage block corresponding to each regional distributed fault assessment tree and the interactive relationship between the production process flows corresponding to different plant areas, a directed block connection is constructed; Based on the characteristic hash value and verification hash stored in the interactive storage block corresponding to each regional distributed fault assessment tree, a zero-knowledge proof algorithm is used to obtain cross-region traceability verification proofs between interactive storage blocks with interactive relationships. The cross-region traceability verification proof is mapped to the block directed connection between the corresponding interactive storage blocks to construct a cross-regional directed block traceability verification chain.

8. The method for evaluating the health status of aluminum processing equipment according to claim 7, wherein: The construction process of the distributed fault assessment forest also includes: Each regional distributed fault assessment tree is used as a federated interactive edge node, and the cross-regional directed block traceability verification chain is used as the cross-regional security traceability verification path. The visual mapping in the visual directed causal connection is mapped to the corresponding block directed connection. Based on the federated interactive edge nodes, cross-regional security traceability verification path, preset verification attack library, preset traceability accuracy threshold and security verification threshold, combined with the federated algorithm for training, a trained distributed fault assessment forest is obtained.

9. The method for evaluating the health status of aluminum processing equipment according to claim 8, wherein: The construction process of the visual fault assessment and early warning forest includes: Based on the trained distributed fault assessment forest and combined with the real-time monitoring network, all plant equipment is monitored and evaluated in real time to obtain a distributed fault assessment tree for areas with abnormal conditions; Based on the horizontal production process association evaluation chain and distributed fault assessment subtree in the regional distributed fault assessment tree with the current abnormal state, the root cause of the abnormality and the associated abnormal nodes within the factory are traced. Based on the corresponding fault causal correlation and the corresponding confidence level between the root cause of the abnormality and the associated abnormal nodes, the corresponding level of traceability warning is performed in combination with the preset warning level, and the traceability process is visualized based on the visual directed causal connection; At the same time, based on the trained distributed fault assessment forest, the distributed fault assessment tree with the current abnormal state and the associated distributed fault assessment tree with a fault causal relationship are traced across the plant area for safety, and the horizontal production process associated assessment chain and the distributed fault assessment subtree in the block directed connection and associated distributed fault assessment tree combined with visual mapping are used to trace the associated fault equipment across the plant area and visualize the tracing process. The fault causal correlation degree and the corresponding confidence level of each connection relationship in the corresponding traceability path are combined with the preset warning level to perform cross-plant difference warning, and obtain a visual fault assessment warning forest; The warning level includes a main warning level and an associated warning level; the main warning level corresponds to the main fault device monitored at the current moment; the associated warning level corresponds to the associated fault device at the current moment; The main warning level is issued according to the fault level corresponding to the main fault device; the associated fault device is issued according to the fault level corresponding to the associated fault device and the fault causal correlation and confidence between the associated fault device and the main fault device.

10. A health status assessment system for aluminum processing equipment, used to implement the health status assessment method for aluminum processing equipment according to any one of claims 1 to 9, characterized in that: include: Evaluation tree module, monitoring and evaluation module, and reasoning mapping module; The evaluation tree module obtains a distributed fault evaluation tree based on cross-plant equipment attributes and operation information combined with a Merkle tree and an evaluation algorithm; The monitoring and evaluation module performs distributed edge interaction status evaluation based on a distributed fault evaluation tree combined with a preset cross-plant real-time monitoring network to obtain single equipment evaluation results and process-related equipment evaluation results; The reasoning mapping module generates and shares fault tracing and maintenance strategies based on the single equipment assessment results and process-related equipment assessment results combined with the root cause tracing warning rules and strategy generation library configured in the distributed fault assessment forest. It also visualizes the tracing and sharing process in combination with the traceability warning visualization mapping space to obtain a visual fault assessment warning tree.

Images