An intelligent generator for fully automatic production report generation

Abstract

The present invention relates to the field of report generation, and discloses an intelligent generator for fully automatic generation of production reports, which is used to integrate complex system theory and information security technology to build a non-machine learning driven fully automatic generation system for production reports. The original time series data is fractally cut and encrypted and stored in edge nodes, a cellular automaton network is constructed, cellular state transition rules are generated based on network traffic and access logs, a system dynamics model is established, the coupling relationship between production data flow and network traffic is evaluated, real-time risk confidence is calculated, and production reports are generated in combination with access rights and risk confidence. The intelligent generation method for fully automatic generation of production reports proposed in the present invention not only improves data processing efficiency and security, but also enhances the flexibility of access control and the accuracy of risk assessment, providing strong technical support for the intelligent management of modern production systems.

Description

Technical Field

[0001] The present invention relates to the field of report generation, and in particular to an intelligent generator for fully automatic generation of production reports. Background Art

[0002] With the rapid development of Industry 4.0 and intelligent manufacturing, modern production systems are becoming increasingly complex, and data volumes are growing exponentially. Production reports, a crucial basis for enterprise management and decision-making, are directly impacted by their accuracy and real-time performance, impacting production efficiency and operational quality. However, traditional methods for generating production reports rely primarily on manual compilation and data analysis, which is not only time-consuming and labor-intensive but also struggles to cope with the large-scale, high-frequency processing of time series data. In the Industrial Internet, in particular, time series data such as equipment operating parameters, production records, and process indicators are massive, multi-source, and heterogeneous, posing significant challenges to the generation of production reports.

[0003] The existing production report generation technology has the following main shortcomings:

[0004] Low data processing efficiency: Traditional methods often require complex data cleaning, integration, and analysis processes when processing large-scale time series data. These steps usually rely on manual operations or simple automated scripts and are difficult to meet real-time requirements.

[0005] Insufficient data security and privacy protection: Production data contains a large amount of sensitive information, such as equipment operating status and output data. Traditional methods lack effective encryption and distributed storage mechanisms during data storage and transmission, which easily leads to data leakage risks.

[0006] Inflexible access control: With the diversification of user roles and the dynamic changes in access requirements, traditional access control mechanisms find it difficult to implement fine-grained permission management and cannot effectively prevent unauthorized access and data abuse.

[0007] Weak risk assessment and early warning capabilities: Traditional production reports mainly focus on data display and statistics, lacking in-depth analysis of the coupling relationship between production data flows and network traffic, making it difficult to timely discover and warn of potential security risks.

[0008] Poor data visualization and tamper resistance: Traditional reports have limitations in data visualization and tamper resistance, making it difficult to intuitively display data trends and risk factors, and lack effective data integrity verification mechanisms.

[0009] Therefore, we propose an intelligent generator for fully automatic production report generation to solve the above problems. Summary of the Invention

[0010] The present invention provides an intelligent generator for fully automatic generation of production reports, which is used to integrate complex system theory and information security technology to build a fully automatic production report generation system that is not driven by machine learning.

[0011] The first aspect of the present invention provides an intelligent generation method for fully automatic generation of production reports, the intelligent generation method for fully automatic generation of production reports comprising: obtaining original time series data, the original time series data including equipment operating parameters, output records and process indicators, determining fractal cutting dimensions based on the original time series data, cutting the original time series data into fractal data sub-blocks with self-similarity, encrypting each fractal data sub-block to generate encrypted fractal sub-blocks, and distributing and storing the encrypted fractal sub-blocks to multiple edge storage nodes; constructing a cellular automaton network, generating cellular state transition rules based on real-time network traffic data and historical access logs, and upon receiving a user-initiated When a report access request is made, access permission verification is performed based on the current state of the corresponding user account cell in the cellular automaton network to generate an access permission verification result; a system dynamics model is established, which defines the coupling relationship between the production data flow and the network traffic, and a real-time risk confidence index is calculated based on the equipment operating parameters and the access permission verification result; the encrypted fractal sub-block related to the current report request is retrieved from the edge storage node to obtain the restored data trend, and the restored data trend is dynamically desensitized based on the access permission verification result, and the production report content is generated based on the desensitized data trend and the real-time risk confidence index.

[0012] Optionally, in a first implementation of the first aspect of the present invention, the method includes: calculating the Hurst index of the original time series data, generating a fractal dimension value D=2-H, and determining the number of fractal cutting layers according to the fractal dimension value. Where N is an integer and 3≤N≤8; multi-level fractal cutting is performed on the original time series data according to the number of cutting layers N to generate 2 N fractal data sub-blocks, adding fractal tree encoding to each fractal data sub-block; independently generating encryption keys for each fractal data sub-block, and linking sub-block keys at the same fractal level through a hash chain, randomly distributing the encrypted fractal data sub-blocks to at least three edge storage nodes, and recording the storage location mapping table, generating a fractal storage verification code for each edge storage node.

[0013] Optionally, in a second implementation method of the first aspect of the present invention, it includes: mapping each user account and device access terminal to an independent cell in a cellular automaton network, assigning an initial state label to each cell, defining the cell neighborhood range as network nodes with a topological distance of no more than 3 hops, and generating a cellular adjacency relationship matrix; counting abnormal request features in real-time network traffic data, and generating a cellular state transition rule set based on the abnormal request features and the cellular adjacency relationship matrix; based on the cellular state transition rule set, periodically updating the current state of all cells in the cellular automaton network, and when a user initiates a report access request, extracting the cell state and neighborhood abnormality ratio corresponding to the user, and generating a dynamic permission label.

[0014] Optionally, in a third implementation of the first aspect of the present invention, the method includes: defining a first core variable of the system dynamics model as a normal traffic change rate, which is dynamically adjusted by the difference between the production interface call frequency and the abnormal request detection rate; defining a second core variable as a risk value change rate, which is calculated by the weighted product of the vulnerability exposure rate and the security patch coverage rate, where the weight distribution is a function of the vulnerability severity level; and calculating the real-time vulnerability exposure rate based on the production equipment firmware version and the port opening status:

[0015] Vulnerability exposure rate = (Σ vulnerability weight × corresponding service activity) / total number of services;

[0016] Among them, unpatched high-risk vulnerabilities are given a weight of 3 times, medium-risk vulnerabilities are given a weight of 2 times, and low-risk vulnerabilities are given a weight of 1 times; the patch coverage rate is calculated based on the security patch deployment log; the normal traffic change rate and the risk value change rate are normalized to generate a dynamic risk index, and the dynamic risk index is corrected based on the access permission verification results to output a real-time risk confidence indicator.

[0017] Optionally, in a fourth implementation method of the first aspect of the present invention, it includes: generating fractal tree coding retrieval conditions based on the time range and data type of the current report request, and matching the target encrypted fractal sub-blocks from the edge storage node; decrypting at least 30% of the target encrypted fractal sub-blocks, and restoring the data trend curve through a fractal interpolation algorithm based on the hierarchical identification and parent block association relationship of the fractal tree code; parsing the access permission verification result to obtain the desensitized security data trend, mapping the real-time risk confidence index into a visual warning level label, aligning the warning level label with the desensitized security data trend according to the time axis, and generating production report content with risk annotations; performing symbolic dynamics conversion on the production report content with risk annotations to generate a symbolic data fingerprint, associating the symbolic data fingerprint with the report content, and writing the root hash into a preset blockchain node to output an anti-tampering verification identifier.

[0018] Optionally, in a fifth implementation of the first aspect of the present invention, the restoration of the data trend curve by the fractal interpolation algorithm includes the following sub-steps: performing self-similarity verification on the decrypted fractal data sub-block, calculating the morphological similarity between its fractal tree encoding and the parent block encoding, and marking it as an abnormal sub-block if the similarity is lower than a preset threshold and triggering re-retrieval; according to the hierarchical position of the missing sub-block, selecting the self-similarity coefficient of the adjacent level for interpolation, and the coefficient is dynamically adjusted by the fractal dimension value, and the interpolation formula is:

[0019] V 插值 =α·V 父块 +(1-α)·V 兄弟子块

[0020] Among them, α is the self-similarity weight factor of the current level, and its value range is 0.6-0.9. The mutation point in the interpolated data trend curve is detected. If the mutation amplitude exceeds 3 times the standard deviation of the historical mean, a smoothing correction is performed based on fractal self-similarity to retain the trend morphological characteristics.

[0021] The second aspect of the present invention provides an intelligent generator for fully automatic generation of production reports, which includes: an acquisition module for acquiring original time series data, the original time series data including equipment operating parameters, output records and process indicators, determining fractal cutting dimensions based on the original time series data, cutting the original time series data into fractal data sub-blocks with self-similarity, encrypting each fractal data sub-block to generate encrypted fractal sub-blocks, and distributing the encrypted fractal sub-blocks to multiple edge storage nodes; a processing module for constructing a cellular automaton network, generating cellular state transition rules based on real-time network traffic data and historical access logs, and processing the cellular state transition rules when receiving a report initiated by a user. When a table access request is made, access permission verification is performed based on the current state of the corresponding user account cell in the cellular automaton network to generate an access permission verification result; a setting module is used to establish a system dynamics model, which defines the coupling relationship between the production data flow and the network traffic, and calculates the real-time risk confidence index based on the equipment operating parameters and the access permission verification result; an allocation module is used to retrieve the encrypted fractal sub-block related to the current report request from the edge storage node, obtain the restored data trend, and dynamically desensitize the restored data trend based on the access permission verification result, and generate production report content based on the desensitized data trend and the real-time risk confidence index.

[0022] The third aspect of the present invention provides an intelligent generation device for fully automatic generation of production reports, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory to enable the intelligent generation device for fully automatic generation of production reports to execute the above-mentioned intelligent generation method for fully automatic generation of production reports.

[0023] A fourth aspect of the present invention provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the above-mentioned intelligent generation method for fully automatic generation of production reports.

[0024] The mechanism of the present invention is as follows:

[0025] Combining fractal dimension computing with distributed storage security achieves dual protection at the physical and logical layers, addressing the vulnerability of traditional encrypted storage to pattern recognition attacks; the local rules of cellular automata drive global behavior characteristics, which are applied for the first time to real-time permission management of industrial reporting systems, replacing the static role authorization model and significantly reducing the risk of unauthorized access; the system dynamics model is innovatively introduced into the joint analysis of production and network security to achieve real-time prediction of cross-domain risks and address the blind spots of traditional independent assessment models; fractal data morphology verification combined with symbolic dynamics compression reduces 90% of computing overhead while ensuring anti-tampering capabilities, breaking through the performance bottleneck of traditional full hash verification.

[0026] Beneficial effects

[0027] Through fractal cutting, the original time series data is divided into smaller blocks that are easier to manage and process, improving data processing efficiency. Each fractal data sub-block is independently encrypted and then distributedly stored, enhancing data security and preventing illegal access or tampering.

[0028] A cellular automaton network is constructed to verify access rights based on the user account and the cellular state of the device access terminal, achieving fine-grained access control. Through dynamic updates of the cellular state, user access behavior and permission changes can be reflected in real time, improving the flexibility and security of the system.

[0029] A system dynamics model is established, taking into account multiple factors such as production equipment firmware version, port open status, vulnerability exposure rate, and security patch coverage. It dynamically generates risk confidence indicators. Through real-time risk assessment, potential security risks can be discovered in a timely manner, providing risk warning capabilities for production reports.

[0030] The restored data trends are dynamically desensitized to protect data privacy, and the production report content is converted into symbolic dynamics and stored in blockchain to ensure the integrity and traceability of the report content and prevent data from being tampered with or forged. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of an embodiment of an intelligent generation method for fully automatic generation of production reports according to an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram of another embodiment of the intelligent generation method for fully automatic generation of production reports according to an embodiment of the present invention;

[0033] Figure 3 A schematic diagram of an embodiment of an intelligent generator for fully automatic production report generation according to an embodiment of the present invention;

[0034] Figure 4 This is a schematic diagram of an embodiment of an intelligent generation device for fully automatic generation of production reports in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] An embodiment of the present invention provides an intelligent generator for fully automatic generation of production reports, which is used to integrate complex system theory and information security technology to build a non-machine learning driven fully automatic generation system for production reports. The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0036] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 An embodiment of the intelligent generation method for fully automatic generation of production reports in an embodiment of the present invention includes:

[0037] 101. Obtain original time series data, which includes equipment operating parameters, production records, and process indicators; determine the fractal cutting dimension based on the Hurst exponent of the original time series data, and cut the original time series data into fractal data sub-blocks with self-similarities; perform AES-256 encryption on each fractal data sub-block to generate encrypted fractal sub-blocks, and distribute the encrypted fractal sub-blocks to multiple edge storage nodes;

[0038] It is understandable that the execution subject of the present invention can be an intelligent generator for fully automatic production report generation, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking the server as the execution subject as an example.

[0039] For example, in March 2025, production line data from a precision manufacturing company was collected over a 30-day period. Minute-by-minute data was recorded for equipment operating parameters (temperature accuracy within ±0.1°C, pressure fluctuation within ±5kPa), production output (unit: pieces / minute), and process indicators (qualification rate within ±0.3% fluctuation), generating 43,200 time-series data items. This data was cleaned using an Industrial IoT gateway, removing 3.2% of invalid data due to sensor anomalies (e.g., temperature spikes to 1000°C). This ultimately resulted in a standardized data matrix (timestamp, parameter category, and value triplet).

[0040] Using the R / S analysis method, the data was divided into 10-minute windows, and the Hurst exponent was calculated for each window. For one window, H = 0.78 (a value > 0.5 indicates long-range correlation), and the fractal dimension D = 2-H = 1.22. Based on the fractal linearity indicator Δ < 0.1, the cutting dimension was determined to be 2^1 = 2 (optimizing storage for integer powers). The full data was then divided into 216 self-similar sub-blocks, each containing 2 hours of data (120 records). Verification showed that the variance of the Hurst exponent between sub-blocks was 0.03, meeting the self-similarity requirement.

[0041] Key management: A hardware security module (HSM) is used to generate a 256-bit master key, which is then combined with a timestamp to derive sub-keys (each sub-block has an independent key). The key update cycle is 24 hours.

[0042] Encryption process: Each fractal sub-block is encrypted in CBC mode, with an initialization vector (IV) generated using HMAC-SHA256. A temperature data sub-block (original value: 35.2°C ±0.5 fluctuation) is encrypted to generate a 512-byte ciphertext (Base64 encoding example: k3Gh9dL...).

[0043] Storage strategy: Encrypted sub-blocks are distributed to three edge nodes (East China, South China, and North China data centers) according to the hash consistency algorithm. Each node stores 50% of the full data and redundantly backs up 30% to achieve data availability and resistance to single point failures.

[0044] Encryption Strength: Verified by the NIST test suite, ciphertext passed five tests, including the frequency test (P-value>0.01) and the runs test, meeting the FIPS140-2 standard.

[0045] Storage efficiency: Fractal cutting reduces storage space by 42% (original data 1.2GB → encrypted 696MB), and edge node response latency is <50ms (supported by edge computing architecture).

[0046] 102. Construct a cellular automaton network. Each cell in the cellular automaton network corresponds to a user account or device access terminal. Cell states include normal access state, unauthorized request state, and isolated state. Generate cellular state transition rules based on real-time network traffic data and historical access logs. The state transition rules include: when the proportion of adjacent cells of a target cell in the unauthorized request state exceeds 30%, update the state of the target cell to the isolated state. When a report access request is received from a user, perform access permission verification based on the current state of the corresponding user account cell in the cellular automaton network and generate an access permission verification result. The access permission verification result may include allowing access, restricting access to certain fields, or denying access.

[0047] It should be noted that a one-dimensional cellular space containing 250 cells is constructed based on 200 user accounts (production supervisors, quality inspectors, etc.) and 50 equipment access terminals (MES system terminals, PLC controllers) within the enterprise. Each cell is initially in a normal access state and is assigned the following attributes:

[0048] User account cell: binds role permissions (production supervisors can access all fields, quality inspectors can only view quality indicators); equipment terminal cell: records IP address and access frequency (a PLC terminal requests data 5 times per hour); historical access log: counts the number of unauthorized requests for each cell in the past 30 days (the quality inspector account attempted to access process parameters 3 times, triggering an alarm).

[0049] Define dynamic rules based on real-time network traffic (sampling with 1-second granularity) and historical logs:

[0050] Rule 1 (Neighbor Infection Isolation): If the proportion of unauthorized request status in the adjacent cells of a cell (the first 5 and the last 5 cells) is ≥ 30%, the target cell status is changed to isolated. If 4 out of 10 cells around a certain MES terminal trigger unauthorized requests (illegal IP access), the system automatically isolates the terminal for 12 hours.

[0051] Rule 2 (Behavior pattern correction): When a cell's access requests comply with the permission rules for three consecutive times, the weight of its historical violation record is reduced by 50% (after the production supervisor account accesses normally, its isolation status triggered by the neighbor is restored).

[0052] Real-time permission verification process, taking a report request on March 15, 2025 as an example:

[0053] Scenario: Quality inspector A initiates a "process indicator trend analysis" request through the terminal.

[0054] Status detection: Among the adjacent cells of this user cell, two device terminals were in an unauthorized request state due to abnormal port scanning (accounting for 20%). The isolation rules were not triggered, and the cell status remained normal.

[0055] Dynamic desensitization: Based on role permissions, the system hides the alloy ratio field (sensitive field) in the process parameters and only displays the temperature fluctuation curve (±0.5℃ accuracy) and the pass rate (85.6% → desensitized to "≥85%").

[0056] Verification result: Access to some fields is restricted, allowing viewing of trend charts but blocking specific values.

[0057] In response to an abnormal event, on March 20, a device terminal (cell number #153) triggered the following behavior:

[0058] Anomaly detection: 50 high-frequency requests for "production data" within 10 minutes (10 times the baseline value) are marked as unauthorized requests;

[0059] Chain Isolation: Five adjacent cells (including three quality inspector accounts) are automatically isolated because the neighbor's unauthorized usage rate reaches 40%, blocking their data access channels.

[0060] Recovery mechanism: Two hours later, terminal #153 is confirmed to have no malicious behavior through security scanning, the cell status is reset to normal, and the neighbor isolation is simultaneously lifted.

[0061] 103. Establish a system dynamics model that defines the coupling relationship between production data flows and network traffic, including the following variables: the normal traffic change rate, which is dynamically adjusted by the production interface call frequency and the abnormal request detection rate; the risk value change rate, which is calculated from the vulnerability exposure rate and security patch coverage rate of the production system; and the real-time risk confidence index calculated based on the device operating parameters and access rights verification results.

[0062] It should be noted that the production data flow variables are: equipment operating parameters: temperature (±0.2℃ fluctuation), pressure (±10kPa accuracy), spindle speed (2000±50rpm), and the collection frequency is 1 second / time; production records: the number of good products per hour (356 pieces were produced from 8:00 to 9:00 on March 15, with a yield rate of 98.3%).

[0063] Network traffic variables: Production interface call frequency: MES system interface request rate per second (baseline value 20 times / second, peak value 150 times / second); abnormal request detection rate: based on real-time detection of traffic characteristics (unconventional port access), a total of 12,000 abnormal requests were intercepted in March, with an average daily detection efficiency of 95.6%.

[0064] Risk variables: Vulnerability exposure rate: the percentage of unpatched vulnerabilities in the production line PLC control system (5 new vulnerabilities were discovered in March, with a patch coverage rate of 80%); Security patch coverage rate: the patch progress based on the CVE vulnerability library (critical vulnerability patch delay ≤ 24 hours).

[0065] Coupling relationship modeling, normal flow rate change equation:

[0066]

[0067] Here, α = 0.8 (production data flow weight), and β = 1.2 (security suppression coefficient). At 2:00 PM on March 20th, the number of interface calls suddenly increased to 120 calls per second, and the anomaly detection rate was simultaneously increased to 30 calls per second. After dynamic balancing, the normal traffic change rate stabilized at +12.6% per minute.

[0068] The risk value change rate equation is:

[0069]

[0070] γ = 0.5 is the risk amplification factor. On March 5th, due to an unpatched vulnerability (exposure rate 25%, patch coverage rate 70%), the risk value increased by 8.3% per hour, triggering a yellow alert.

[0071] Real-time risk confidence calculation, input parameters:

[0072] Abnormal equipment fluctuations: On March 12, the temperature sensor exceeded the threshold (set value ±2°C, measured value ±3.5°C) three times in a row, with a weight of 40%;

[0073] Access verification result: Quality inspector account #205 has access restrictions on some fields due to the unauthorized neighboring cells exceeding 25%, with a weight of 30%.

[0074] Network load pressure: The period of time when the interface response delay is greater than 500ms accounts for 15% and the weight accounts for 30%.

[0075] Fusion formula: Risk confidence = 0.4·Device Anomaly Index + 0.3·Access Risk Index + 0.3·Network Load Index. At 10:00 on March 18, the Device Anomaly Index was 72 (out of 100), the Access Risk Index was 55, and the Network Load Index was 60. The final risk confidence was 64.1 (orange warning threshold ≥ 60), triggering the automated frequency reduction strategy.

[0076] 104. Retrieve the encrypted fractal sub-blocks related to the current report request from the edge storage node, decrypt at least 30% of the encrypted fractal sub-blocks, and restore the data trend based on fractal self-similarity; based on the access permission verification result, dynamically desensitize the restored data trend to hide the sensitive fields in the process indicators; integrate the desensitized data trend with the real-time risk confidence indicator to generate the production report content; generate a data fingerprint based on symbolic dynamics for the production report content, associate the data fingerprint with the production report content and store it in the blockchain node, and output an anti-tampering verification mark.

[0077] It should be noted that the fractal sub-blocks generated by the enterprise production line on March 15, 2025, are stored in three edge nodes in East China, South China, and North China, including process indicators such as temperature fluctuation (±0.1°C), pressure curve (10-minute particle size), and alloy ratio (accuracy 0.01%). When the production supervisor initiates a request for "Process trend analysis from March 10 to 20":

[0078] Retrieval strategy: Locate the relevant sub-blocks (12 sub-blocks from 8:00-12:00 on March 10) based on the timestamp hash value, and retrieve data from 3 nodes in parallel, with a redundancy matching rate of 98%.

[0079] Decryption rules: Four sub-blocks (33%) are randomly selected for AES-256 decryption. The key is dynamically derived by the hardware security module (HSM) and combined with the timestamp to generate the initialization vector (IV). The ciphertext of a pressure sub-block (Base64 encoded as k3Gh9dL...) is decrypted and restored to the original data sequence (pressure value: 35.2kPa → 34.8kPa fluctuation).

[0080] Fractal self-similarity data restoration and trend reconstruction: Based on the fractal dimension D = 1.22 (calculated by the Hurst exponent H = 0.78), the decrypted 4 sub-blocks were linearly interpolated and expanded. The 2 sub-blocks (120 records) from 10:00 to 11:00 on March 10 were expanded into complete 4-hour data (480 records) through self-similarity. The root mean square error (RMSE) was ≤ 0.05 kPa, verifying the data continuity.

[0081] Anomaly correction: An abnormal temperature surge (40°C → 50°C) was detected at 9:30 AM on March 12. Using trend smoothing of adjacent sub-blocks, the temperature was corrected to 45.2°C (99% matching the device log).

[0082] Dynamic desensitization and sensitive field protection, permission-driven desensitization: Production supervisors (full permissions) can view complete data; quality inspectors (partial permissions) trigger dynamic desensitization rules:

[0083] Masking treatment: alloy ratio "15.8%" is desensitized to "≥15%";

[0084] Truncation processing: Precision machining parameters (laser power 2000W±50) are displayed as "2000W±**".

[0085] Hiding process indicators: Hide the "Heat treatment cooling rate" field (enterprise core technology) and replace it with the classification label "Grade A process stability".

[0086] Report integration and data fingerprint generation, along with risk indicator fusion, correlates real-time risk confidence (risk value 64.1 on March 15th, indicating an orange alert) with de-identified data trends. During periods of pressure fluctuation exceeding thresholds (2:00 PM to 4:00 PM on March 18th), a risk alert indicating a 20% increase in vulnerability exposure was added. Symbolic dynamic fingerprinting: A 16-bit data fingerprint (#A3F9B2C7) is generated using the Lempel-Ziv complexity algorithm to uniquely identify report content. The fingerprint is associated with the report and stored in a Hyperledger Fabric blockchain node, generating a tamper-resistant identifier (Block#5821TxHash:0x9a4f...).

[0087] In the embodiment of the present invention, by fractal cutting, the original time series data is divided into small blocks that are easier to manage and process, thereby improving the efficiency of data processing. The encrypted fractal data sub-blocks are distributed and stored in multiple edge nodes, achieving redundant backup and efficient access to data, and improving the availability and reliability of data. The implementation of the cellular automaton network makes access control more fine-grained and dynamic, and can automatically adjust access rights according to real-time network traffic and historical access logs. Through dynamic desensitization processing, sensitive data is protected from access by unauthorized users, thereby enhancing data security. The establishment of the system dynamics model takes into account the coupling relationship between production data flow and network traffic, and can more accurately assess real-time risks. The calculation of real-time risk confidence indicators provides risk warning capabilities for production reports, enabling enterprises to promptly discover and respond to potential security risks. By generating a data fingerprint based on symbolic dynamics and associating the fingerprint with the report content and storing it in the blockchain node, the integrity and immutability of the report content are ensured. The output of the tamper-resistant verification mark provides a reliable basis for tracing and verifying the report content. In summary, the present invention proposes a new method for fully automatic generation of production reports by integrating multiple advanced technologies. This method not only improves data processing and storage efficiency, enhances access control and security, but also improves risk assessment and early warning capabilities, and ensures the integrity and traceability of report content.

[0088] See also Figure 2 Another embodiment of the intelligent generation method for fully automatic generation of production reports in the embodiment of the present invention includes:

[0089] 201. Obtain original time series data, which includes equipment operating parameters, production records, and process indicators; determine the fractal cutting dimension based on the Hurst exponent of the original time series data, and cut the original time series data into fractal data sub-blocks with self-similarities; perform AES-256 encryption on each fractal data sub-block to generate encrypted fractal sub-blocks, and distribute the encrypted fractal sub-blocks to multiple edge storage nodes;

[0090] Specifically, the Hurst index of the original time series data is calculated based on the rescaled range analysis method (R / S analysis) to generate a fractal dimension value D = 2-H; the number of fractal cutting layers is determined according to the fractal dimension value. Where N is an integer and 3≤N≤8; perform multi-level fractal cutting on the original time series data according to the number of cutting layers N to generate 2 N fractal data sub-blocks, and each sub-block maintains morphological self-similarity with its parent block; a fractal tree code is added to each fractal data sub-block, and the code includes a hierarchical identifier and a parent block association relationship;

[0091] An AES-256 encryption key is independently generated for each fractal data sub-block, and the sub-block keys at the same fractal level are linked through a hash chain; the encrypted fractal data sub-blocks are randomly distributed to at least three edge storage nodes, and the storage location mapping table is recorded; a fractal storage verification code is generated for each edge storage node, and the verification code is calculated by concatenating the fractal tree code and the node's geographic location hash value.

[0092] It should be noted that the 12-hour equipment operating parameters collected from a manufacturing workshop (with a 1-minute interval) include:

[0093] Spindle speed (unit: rpm): [1500,1523,1498,...,1620] (720 data points);

[0094] Temperature index (unit: °C): [45.2, 46.8, 44.5, ..., 49.7];

[0095] Production records (unit: piece): hourly cumulative value [320,655,983,...,3840];

[0096] The Hurst index was calculated using the R / S analysis method: the spindle speed series was divided into 12 60-minute subsequences, and the ratio of the range R to the standard deviation S of each subsequence was calculated. A linear regression of log(R / S) and log(n) was fitted, resulting in a slope H = 0.78 and a fractal dimension D = 2-H = 1.22.

[0097] The number of cutting layers is determined according to the formula However, due to the constraint condition 3≤N≤8, N=6 layers are finally taken, generating 2^6=64 sub-blocks.

[0098] Fractal self-similar cutting, multi-level cutting process:

[0099] Layer 1: Divide the 720-point data into 2 blocks (360 points / block);

[0100] Second layer: Each block is divided into 2 blocks → 4 blocks in total (180 points / block); ...

[0102] Level 6: Final 64 blocks (11.25 points / block);

[0103] In the tree coding example, the second sub-block of the third layer is encoded as L3-P2-F1, which means: L3: the third layer; P2: the parent block is the first sub-block of the second layer; F1: the first sub-block of the current layer;

[0104] Dynamic key generation: Layer 1 sub-block key: Key1 = SHA256 (master key + "L1-P0-F1"); Layer 2 sub-block key: Key2_1 = SHA256 (Key1 + "L2-P1-F1") (hash chain association);

[0105] The encrypted 64 sub-blocks are randomly distributed to three edge nodes: Node A (Shanghai): stores 22 sub-blocks; Node B (Shenzhen): stores 21 sub-blocks; Node C (Chengdu): stores 21 sub-blocks;

[0106] Verification code generation: the verification code of the L3-P2-F1 sub-block stored in node A = SHA3 (encoded L3-P2-F1 || node A latitude and longitude 31.23N / 121.47E).

[0107] 202. Construct a cellular automaton network, where each cell in the cellular automaton network corresponds to a user account or device access terminal. Cell states include normal access state, unauthorized request state, and isolated state. Generate cellular state transition rules based on real-time network traffic data and historical access logs. The state transition rules include: when the proportion of adjacent cells of a target cell in the unauthorized request state exceeds 30%, update the state of the target cell to the isolated state. When a report access request is received from a user, perform access permission verification based on the current state of the corresponding user account cell in the cellular automaton network and generate an access permission verification result. The access permission verification result may include allowing access, restricting access to certain fields, or denying access.

[0108] Specifically, each user account and device access terminal is mapped to an independent cell in a cellular automaton network, and each cell is assigned an initial state label based on the security rating in historical access logs. The cellular neighborhood is defined as network nodes with a topological distance of no more than three hops, and a cellular adjacency matrix is ​​generated. Abnormal request features in real-time network traffic data are counted, including high-frequency access attempts, unconventional time operations, and sensitive interface call behaviors.

[0109] Based on the abnormal request characteristics and the cell adjacency matrix, a cell state transition rule set is generated. The cell state transition rule set includes:

[0110] If the proportion of neighboring cells of the target cell in the unauthorized request state exceeds 30%, the target cell state is triggered to transition to the isolated state;

[0111] If the target cell does not trigger abnormal characteristics in 5 consecutive access requests, its status will be restored from isolated to normal access status;

[0112] According to the set of cellular state transition rules, the current states of all cells in the cellular automaton network are periodically updated. When a user initiates a report access request, the cell state and neighborhood anomaly ratio corresponding to the user are extracted to generate dynamic permission labels. The dynamic permission labels include: Allow access: the cell state is normal and the neighborhood anomaly ratio is ≤10%; Restrict access to some fields: the cell state is normal but the neighborhood anomaly ratio is greater than 10% and less than 30%; Deny access: the cell state is isolated or the neighborhood anomaly ratio is greater than 30%.

[0113] It should be noted that a certain automotive parts manufacturing workshop needs to implement dynamic access control for production reports. The workshop contains five user accounts (AE) and three equipment terminals (X, Z). The historical access log shows:

[0114] User A (high security rating): No unauthorized operations in the past three months;

[0115] User C (low security rating): triggered two sensitive interface calls;

[0116] Device X (safety rating: medium): Operations were recorded at unusual times during the early morning hours.

[0117] Cell mapping and initial state allocation,map 5 user accounts and 3 device terminals into independent cells,generating 8 cells (U_A to U_E represent users, D_X to D_Z represent devices).

[0118] Assign initial states based on security ratings: User A → Normal state (green); User C → Unauthorized request state (red); Device X → Normal state (green);

[0119] The adjacency matrix is ​​constructed to define neighborhoods with topological distance ≤ 3 hops (user A connects to devices X and Y; user B connects to devices Y and Z).

[0120] Generate an adjacency matrix (partial example):

[0121] Cell Neighbor Cells U_A D_X,U_B,D_Y U_C U_D,D_Z

[0122] Abnormal feature statistics,real-time network traffic detected that user D initiated 15 calls to the process parameter query interface within 10 minutes (threshold: 10 times / 10 minutes).

[0123] Device Z triggers the device control command at 02:30 (non-routine operation time).

[0124] Rule triggering example, isolation rule: among user D's neighbor cells (U_C, D_Z), U_C is in an unauthorized state (accounting for 50%>30%), triggering user D's status to be isolated; recovery rule: user B has no abnormalities after 5 consecutive visits, and the status is restored from isolation to normal.

[0125] Periodic status updates: Scan the cellular network every 5 minutes and update the status: User C remains in the overauthorized state due to an abnormality in device Z in the neighborhood. User E's neighborhood abnormality ratio drops from 25% to 8%, and the status returns to normal.

[0126] Dynamic permission determination, user A requests report access: Cell status: Normal; Neighborhood abnormality ratio: 8% (equipment X has no recent abnormalities); Permission label: Access allowed (display all process indicators);

[0127] User D requests access: Cell status: Isolated; Permission label: Access denied (connection blocked);

[0128] Device X requests data export: Neighborhood anomaly ratio: 22% (user B previously triggered high-frequency access); Permission tag: Partial field restrictions (hide device accuracy parameters).

[0129] 203. Establish a system dynamics model that defines the coupling relationship between production data flows and network traffic, including the following variables: the normal traffic change rate, which is dynamically adjusted by the production interface call frequency and the abnormal request detection rate; the risk value change rate, which is calculated from the vulnerability exposure rate and security patch coverage rate of the production system; and the real-time risk confidence index calculated based on the device operating parameters and access permission verification results.

[0130] Specifically, the first core variable of the system dynamics model is defined as the normal traffic change rate, which is dynamically adjusted by the difference between the production interface call frequency and the abnormal request detection rate. Specifically, when the production interface call frequency increases by more than 10 times per minute, the normal traffic baseline value is proportionally increased; when the abnormal request detection rate reaches 80% of the historical peak, the normal traffic change rate attenuation coefficient is triggered;

[0131] The second core variable is defined as the risk value change rate, which is calculated as the weighted product of the vulnerability exposure rate and the security patch coverage rate, with the weight assigned as a function of the vulnerability severity level;

[0132] Calculate the real-time vulnerability exposure rate based on the production equipment firmware version and port open status, specifically:

[0133] Vulnerability exposure rate = (Σ vulnerability weight × corresponding service activity) / total number of services;

[0134] Unpatched high-risk vulnerabilities are given a weight of 3 times, medium-risk vulnerabilities are given a weight of 2 times, and low-risk vulnerabilities are given a weight of 1 times; based on the security patch deployment log, the patch coverage rate is calculated as the number of fixed vulnerabilities / (number of fixed vulnerabilities + number of unfixed vulnerabilities); the dynamic risk index is generated by normalizing the normal traffic change rate and the risk value change rate; the dynamic risk index is corrected based on the access permission verification result: if there is a verification result of "access restricted to some fields", the risk index increases by 20%; if there is a "denied access" result, the risk index increases by 50%; the real-time risk confidence index is output, and the real-time risk confidence index is a normalized probability value of 0-100%.

[0135] It's worth noting that an automotive parts workshop deployed 12 CNC machine tools, using the Industrial Internet of Things to collect real-time data on equipment parameters such as spindle temperature (40-60°C), tool wear (0-100%), and production output (200-300 pieces / hour). The workshop network generated over 5,000 production interface calls daily, and approximately 50 abnormal requests (illegal access to the process database) were detected.

[0136] Normal traffic change rate calculation, baseline value setting: the historical interface call peak is 60 times / minute, and the initial baseline value is 40 times / minute.

[0137] Dynamic adjustment example: When the interface call frequency suddenly increases from 40 times / minute to 55 times / minute during a certain period (an increase of more than 10 times / minute), the system proportionally increases the baseline value to 45 times / minute.

[0138] If the abnormal request detection rate reaches 80% (ie 40 times / day) of the historical peak (50 times / day), the attenuation coefficient is triggered to be 0.8, and the normal traffic change rate = 45×0.8 = 36 times / minute.

[0139] Calculation of risk value change rate and vulnerability weight allocation: There are 3 high-risk vulnerabilities (weight 3), 2 medium-risk vulnerabilities (weight 2), and 1 low-risk vulnerability (weight 1) in the workshop equipment.

[0140] Service activity: The spindle control service associated with the high-risk vulnerability has an activity of 0.9 (running 18 hours a day), and the tool calibration service associated with the medium-risk vulnerability has an activity of 0.6.

[0141] Vulnerability exposure rate calculation, the formula is: Vulnerability exposure rate = (3 × 0.9 + 2 × 0.6 + 1 × 0.3) / 6 = (2.7 + 1.2 + 0.3) / 6 ≈ 0.7 (70%);

[0142] Patch coverage: 4 vulnerabilities have been fixed (3 high-risk + 1 medium-risk), and 2 vulnerabilities have not been fixed (1 medium-risk + 1 low-risk). Coverage = 4 / (4+2) = 66.7%.

[0143] Risk value change rate synthesis: Risk value change rate = vulnerability exposure rate × (1-patch coverage rate) × vulnerability level correction factor (1.5 for high risk);

[0144] Calculation: 0.7 × (1 - 0.667) × 1.5 ≈ 0.35 (35% / hour);

[0145] Normalization processing, the normal flow rate change rate (36 times / minute) and the risk value change rate (35% / hour) are normalized to the range of 0-1: flow factor = 36 / 60 = 0.6; risk factor = 35 / 100 = 0.35;

[0146] Dynamic risk index = 0.6 × 0.5 + 0.35 × 0.5 = 0.475 (47.5%);

[0147] Permission modification: If the log shows two "Partial field access restricted" events, the risk index increases by 20%: 47.5% × 1.2 = 57%; if there is one "Access denied" event, the risk index increases by another 50%: 57% × 1.5 = 85.5%;

[0148] The output results show that the final real-time risk confidence index is 85.5%, which is mapped to a high risk level and triggers a red warning label in the production report.

[0149] 204. Retrieve the encrypted fractal sub-blocks related to the current report request from the edge storage node, decrypt at least 30% of the encrypted fractal sub-blocks, and restore the data trend based on fractal self-similarity; based on the access permission verification result, dynamically desensitize the restored data trend to hide the sensitive fields in the process indicators; integrate the desensitized data trend with the real-time risk confidence indicator to generate the production report content; generate a data fingerprint based on symbolic dynamics for the production report content, associate the data fingerprint with the production report content and store it in the blockchain node, and output an anti-tampering verification mark.

[0150] Specifically, based on the time range and data type of the current report request, a fractal tree encoding retrieval condition is generated to match the target encrypted fractal sub-block from the edge storage node; the integrity and storage location authenticity of the target encrypted fractal sub-block are verified based on the storage verification code, and sub-blocks with mismatched hash values ​​are eliminated;

[0151] Decrypt at least 30% of the target encrypted fractal sub-blocks, and restore the data trend curve using the fractal interpolation algorithm based on the hierarchical identification and parent block association relationship of the fractal tree encoding; for the data intervals of the missing blocks, complete the trend continuity based on the self-similarity characteristics of the adjacent levels; parse the access permission verification results. If the result is "access restricted to some fields", remove the chemical composition ratio and equipment accuracy parameters in the process indicators from the reconstructed data trend; if the result is "access allowed", only hide the core formula number field in the process indicators; map the real-time risk confidence index to a visual warning level (low risk, medium risk, high risk); align the warning level label with the desensitized security data trend along the timeline to generate production report content with risk annotations;

[0152] Perform symbolic dynamics conversion on the risk-labeled report content, map the numerical sequence into a symbolic sequence ("↑" indicates an upward trend, "↓" indicates a downward trend), and generate a symbolic data fingerprint; associate the symbolic data fingerprint with the report content, calculate the Merkle tree root hash value, and write the root hash to the preset blockchain node; output an anti-tampering verification mark containing the blockchain transaction ID for users to check the integrity of the report.

[0153] It should be noted that a workshop needs to generate a production report for April 3, 2025, covering spindle temperature, output fluctuations, and core process parameters. Sensitive fields include recipe number (CX-203), equipment accuracy (±0.01μm), and chemical composition ratio (Al 85%).

[0154] Fractal data retrieval and integrity verification, retrieval condition generation: time range: 2025-04-03 08:00 to 18:00; data type: spindle temperature (unit: °C); fractal tree encoding: L4-P3-F2 (indicates the fourth level, the parent block is the third child block of the third layer);

[0155] Edge node retrieval: matching target sub-blocks from three edge nodes in Shanghai, Shenzhen, and Chengdu;

[0156] Storage verification code verification: Shanghai node sub-block verification code = SHA3(L4-P3-F2||31.23N / 121.47E); 2 sub-blocks with hash mismatches are eliminated, and finally 22 of the 64 sub-blocks pass verification;

[0157] Partial data decryption: AES-256 decryption of 22 decrypted sub-blocks (34% of the total) using the SHA256 key chain (master key + "L4-P3-F2").

[0158] Trend restoration and completion: the missing period (14:00-15:00) data is interpolated using the parent block (L3-P2-F1) and the sibling block (L4-P3-F1): interpolated temperature = 0.8 × parent block mean (52.3°C) + 0.2 × sibling block mean (54.1°C) = 52.7°C;

[0159] Self-similarity verification: the morphological similarity between layers reached 88.6% (threshold 85%);

[0160] Permission-driven dynamic desensitization, permission parsing:

[0161] User A's permission tag: Access to some fields is restricted (neighborhood anomaly ratio 25%);

[0162] Perform desensitization: remove chemical composition ratios (the Al content field is displayed as **); hide equipment accuracy parameters (displayed as ±**μm); retain temperature trend and output fluctuation data;

[0163] Risk-data fusion report generation, risk indicator mapping, real-time risk confidence: 85.5% (calculated from step 203) → high risk (red alert);

[0164] Visual alignment: High temperature periods (13:00-14:00) are marked with red warning bars; the production fluctuation curve is superimposed with risk level color bands;

[0165] Symbolic fingerprint generation and blockchain evidence storage, symbolic conversion:

[0166] Temperature trend symbol sequence: ↑↑→↓↓↑→↑ (one symbol every 30 minutes);

[0167] Output fluctuation symbol: ↓→↑↑↑;

[0168] Blockchain evidence storage:

[0169] Merkle tree root hash: 0x3a7d...f1c2;

[0170] Blockchain transaction ID: TXID-20250403-085214;

[0171] Tamper-resistant verification identification: Users can enter TXID through the blockchain browser to check data integrity.

[0172] 205. Restoring the data trend curve using the fractal interpolation algorithm includes the following sub-steps:

[0173] The decrypted fractal data sub-block is self-similarity verified, and the morphological similarity between its fractal tree encoding and the parent block encoding is calculated. If the similarity is lower than the preset threshold (≥85%), it is marked as an abnormal sub-block and triggers re-retrieval;

[0174] According to the hierarchical position of the missing sub-block, the self-similarity coefficients of the adjacent levels are selected for interpolation. The coefficients are dynamically adjusted by the fractal dimension value. The interpolation formula is:

[0175] V 插值 =α·V 父块 +(1-α)·V 兄弟子块

[0176] Where α is the self-similarity weight factor of the current level, ranging from 0.6 to 0.9;

[0177] The mutation point in the interpolated data trend curve is detected. If the mutation amplitude exceeds 3 times the standard deviation of the historical mean, a smoothing correction is performed based on fractal self-similarity to retain the trend morphological characteristics.

[0178] It should be noted that a certain workshop needs to restore the spindle temperature data trend on April 3, 2025. The original data is cut into 64 sub-blocks through 6 layers of fractal cutting. After encrypted storage, the 4th layer L4-P3-F2 sub-block (corresponding to 14:00-15:00 data) is missing.

[0179] Fractal hierarchical integrity check, self-similarity verification, hierarchical association: the parent block of the missing sub-block L4-P3-F2 is L3-P2-F1, and its sibling sub-blocks are L4-P3-F1 and L4-P3-F3.

[0180] Morphological similarity calculation: The mean temperature of the parent block L3-P2-F1 is 52.3°C, with a standard deviation of ±1.2°C;

[0181] Data of adjacent time periods of missing sub-blocks (L4-P3-F1: 54.1℃±0.9℃; L4-P3-F3: 53.8℃±1.1℃);

[0182] Similarity assessment: The similarity of fluctuation amplitude is 87% (>85% threshold), but the similarity of data distribution shape is only 82% (marked as abnormal).

[0183] Exception handling triggers the edge node to re-retrieve and obtain the backup sub-block L4-P3-F2' from the Chengdu node, and the verification similarity is improved to 89%.

[0184] Dynamic fractal interpolation optimization, interpolation parameter configuration:

[0185] Fractal dimension value D = 1.22 (Hurst index H = 0.78);

[0186] Self-similarity weight factor α=0.8 (level 4 weight range 0.6-0.9).

[0187] Interpolation calculation, formula application:

[0188] V interpolation = 0.8 × parent block mean 52.3°C + 0.2 × sibling block mean 53.8°C = 52.7°C;

[0189] Trend completion: The interpolated temperatures per hour from 14:00 to 15:00 were 52.7℃, 53.1℃, and 52.9℃, which connected smoothly with the adjacent time periods.

[0190] Mutation point detection, historical mean: 12-hour temperature mean 51.5°C, standard deviation ±2.3°C; anomaly threshold: 51.5 + 3 × 2.3 = 58.4°C; interpolated data: The interpolated temperature at 14:30 is 53.1°C (normal), but the original sub-block at 15:00 has an outlier value of 59.2°C (triggering smoothing).

[0191] Smoothing correction, fractal self-similarity correction:

[0192] The fluctuation pattern of the parent block L3-P2-F1 (mean 52.3°C, ±1.2°C) was adopted;

[0193] Corrected 59.2°C to 53.5°C (based on the 53.8°C trend of the sibling sub-block L4-P3-F3).

[0194] In the embodiment of the present invention, through fractal cutting and distributed storage, the pressure of data storage and processing is effectively reduced, and the access speed and processing efficiency of data are improved; multiple mechanisms such as encrypted storage, cellular automaton network access control and system dynamics model risk assessment work together on the system to form a comprehensive security protection system, which effectively prevents data leakage and illegal access; fractal interpolation algorithm and data trend restoration technology ensure the integrity and accuracy of the data, making the generated production reports more reliable, providing strong support for the decision-making of the enterprise; symbolic dynamics data fingerprint and blockchain evidence storage technology ensure the non-tamperability of the production reports, enhance the credibility and transparency of the data, and help to establish trust relationships between enterprises. In summary, this embodiment realizes the fully automatic generation and intelligent management of production reports by combining a variety of advanced theories and technical means, improves data storage and processing efficiency, enhances system security, improves data accuracy and reliability, and enhances data credibility and transparency, providing strong support for the efficient, safe and intelligent operation of enterprises.

[0195] The above describes the intelligent generation method for fully automatic generation of production reports in the embodiment of the present invention. The following describes the intelligent generator for fully automatic generation of production reports in the embodiment of the present invention. Figure 3 In one embodiment of the present invention, an intelligent generator for fully automatic production report generation includes: an acquisition module 301 for acquiring original time series data, the original time series data including equipment operating parameters, production records and process indicators, determining the fractal cutting dimension based on the original time series data, cutting the original time series data into fractal data sub-blocks with self-similarity, encrypting each fractal data sub-block to generate encrypted fractal sub-blocks, and distributing the encrypted fractal sub-blocks to multiple edge storage nodes; a processing module 302 for constructing a cellular automaton network, generating cellular state transition rules based on real-time network traffic data and historical access logs, and upon receiving a report access request initiated by a user, based on The current state of the corresponding user account cell in the cellular automaton network is used to perform access permission verification and generate an access permission verification result; a setting module 303 is used to establish a system dynamics model, which defines the coupling relationship between the production data flow and the network traffic, and calculates the real-time risk confidence index based on the equipment operating parameters and the access permission verification result; an allocation module 304 is used to retrieve the encrypted fractal sub-block related to the current report request from the edge storage node, obtain the restored data trend, and dynamically desensitize the restored data trend based on the access permission verification result, and generate the production report content based on the desensitized data trend and the real-time risk confidence index.

[0196] In the embodiments of the present invention, by combining advanced technologies such as fractal theory, cellular automata, system dynamics, and edge storage, efficient, secure, and intelligent processing and report generation of production data are achieved, the efficiency and accuracy of production management are improved, and strong support is provided for the production decision-making of enterprises.

[0197] above Figure 3 The intelligent generator for fully automatic generation of production reports in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The intelligent generation device for fully automatic generation of production reports in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0198] Figure 4 This is a schematic diagram of the structure of an intelligent generation device for fully automatic generation of production reports provided by an embodiment of the present invention. The intelligent generation device 400 for fully automatic generation of production reports may have relatively large differences due to different configurations or performances. The device 400 includes a transmitter 401, a receiver 402, and a processor 403. The processor 403 may also be a controller. Figure 4 denoted as “controller / processor 403 ”. Optionally, the device 400 may further include a modem processor 405 , wherein the modem processor 405 may include an encoder 406 , a modulator 407 , a decoder 408 , and a demodulator 409 .

[0199] In one example, transmitter 401 conditions (e.g., performs analog-to-analog conversion, filtering, amplification, and frequency upconversion) the output samples and generates an uplink signal, which is transmitted via an antenna to an access network device. On the downlink, the antenna receives the downlink signal transmitted by the access network device. Receiver 402 conditions (e.g., performs filtering, amplification, frequency downconversion, and digitization) the signal received from the antenna and provides input samples. Within modem processor 405, encoder 406 receives traffic data and signaling messages to be transmitted on the uplink and processes them (e.g., formats, encodes, and interleaves them). Modulator 407 further processes (e.g., performs symbol mapping and modulation) the encoded traffic data and signaling messages and provides output samples. Demodulator 409 processes (e.g., demodulates) the input samples and provides symbol estimates. Decoder 408 processes (e.g., deinterleaves and decodes) the symbol estimates and provides decoded data and signaling messages for transmission to device 400. The encoder 406, modulator 407, demodulator 409, and decoder 408 can be implemented by the combined modem processor 405. These units perform processing based on the radio access technology (e.g., LTE and other evolved system access technologies) used by the radio access network. It should be noted that when the device 400 does not include the modem processor 405, the above functions of the modem processor 405 can also be performed by the processor 403.

[0200] Processor 403 controls and manages the actions of device 400, and is configured to execute the processing performed by device 400 in the above-described embodiments of the present disclosure. For example, processor 403 is also configured to execute the various steps of the sending device or receiving device in the above-described method embodiments, and / or other steps of the technical solutions described in the embodiments of the present disclosure.

[0201] Furthermore, the device 400 may further include a memory 404 , and the memory 404 is used to store program codes and data for the device 400 .

[0202] It is understandable that Figure 4 Only a simplified design of the device 400 is shown. In actual applications, the device 400 may include any number of transmitters, receivers, processors, modem processors, memories, etc., and all devices that can implement the embodiments of the present disclosure are within the scope of protection of the embodiments of the present disclosure.

[0203] The present invention also provides an intelligent generation device for fully automatic generation of production reports. The intelligent generation device for fully automatic generation of production reports includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the intelligent generation method for fully automatic generation of production reports in the above-mentioned embodiments.

[0204] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions. When the instructions are run on a computer, the computer executes the steps of the intelligent generation method for fully automatic generation of production reports.

[0205] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0206] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0207] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent generation method for fully automatic generation of production reports, characterized in that: The intelligent generation method for fully automatic generation of production reports includes: Obtaining original time series data, the original time series data including equipment operating parameters, production records, and process indicators, determining a fractal cutting dimension based on the original time series data, cutting the original time series data into fractal data sub-blocks with self-similarities, encrypting each fractal data sub-block to generate encrypted fractal sub-blocks, and distributively storing the encrypted fractal sub-blocks in multiple edge storage nodes; Constructing a cellular automaton network, generating cellular state transition rules based on real-time network traffic data and historical access logs, and upon receiving a report access request initiated by a user, performing access permission verification based on the current state of the corresponding user account cell in the cellular automaton network, and generating an access permission verification result; Establishing a system dynamics model, wherein the system dynamics model defines a coupling relationship between production data flow and network traffic, and calculating a real-time risk confidence index based on the device operating parameters and the access permission verification result; The encrypted fractal sub-blocks related to the current report request are retrieved from the edge storage node to obtain the restored data trend. The restored data trend is dynamically desensitized based on the access permission verification result, and the production report content is generated based on the desensitized data trend and the real-time risk confidence indicator.

2. The intelligent generation method for fully automatic generation of production reports according to claim 1 is characterized in that: include: Calculate the Hurst index of the original time series data to generate a fractal dimension value D = 2-H, and determine the number of fractal cutting layers based on the fractal dimension value Where N is an integer and 3≤N≤8; Perform multi-level fractal cutting on the original time series data according to the number of cutting layers N to generate 2 N fractal data sub-blocks, adding fractal tree coding to each fractal data sub-block; An encryption key is generated independently for each fractal data sub-block, and the sub-block keys of the same fractal level are associated through a hash chain. The encrypted fractal data sub-blocks are randomly distributed to at least three edge storage nodes, and the storage location mapping table is recorded. A fractal storage verification code is generated for each edge storage node.

3. The intelligent generation method for fully automatic generation of production reports according to claim 1 is characterized in that: include: Map each user account and device access terminal to an independent cell in the cellular automaton network, assign an initial state label to each cell, define the cellular neighborhood range as network nodes with a topological distance of no more than 3 hops, and generate a cellular adjacency matrix; Counting abnormal request features in real-time network traffic data, and generating a set of cellular state transition rules based on the abnormal request features and a cellular adjacency matrix; According to the cellular state transition rule set, the current states of all cells in the cellular automaton network are periodically updated. When a user initiates a report access request, the cell state and neighborhood anomaly ratio corresponding to the user are extracted to generate a dynamic permission label.

4. The intelligent generation method for fully automatic generation of production reports according to claim 1 is characterized in that: include: The first core variable of the system dynamics model is defined as the normal traffic change rate, which is dynamically adjusted by the difference between the production interface call frequency and the abnormal request detection rate. The second core variable is defined as the risk value change rate, which is calculated by the weighted product of the vulnerability exposure rate and the security patch coverage rate, and the weight distribution is a function of the vulnerability severity level. Calculate the real-time vulnerability exposure rate based on the production equipment firmware version and port open status: Vulnerability exposure rate = (Σ vulnerability weight × corresponding service activity) / total number of services; Among them, unpatched high-risk vulnerabilities are given a weight of 3 times, medium-risk vulnerabilities are given a weight of 2 times, and low-risk vulnerabilities are given a weight of 1 times; Calculate patch coverage based on security patch deployment logs; Normalization is performed based on the normal traffic change rate and the risk value change rate to generate a dynamic risk index. Based on the access permission verification results, the dynamic risk index is corrected and a real-time risk confidence indicator is output.

5. The intelligent generation method for fully automatic generation of production reports according to claim 1 is characterized in that: include: Generate fractal tree encoding retrieval conditions based on the time range and data type of the current report request, and match the target encrypted fractal sub-blocks from the edge storage nodes; Decrypt at least 30% of the target encrypted fractal sub-blocks, and restore the data trend curve using the fractal interpolation algorithm based on the hierarchical identification and parent block association relationship of the fractal tree encoding; Analyze the access permission verification result to obtain the desensitized security data trend, map the real-time risk confidence index into a visual warning level label, align the warning level label with the desensitized security data trend along the timeline, and generate production report content with risk annotations; The production report content with risk annotation is converted into symbolic dynamics to generate a symbolic data fingerprint, which is associated with the report content, and the root hash is written into the preset blockchain node to output an anti-tampering verification mark.

6. The intelligent generation method for fully automatic generation of production reports according to claim 5 is characterized in that: Restoring the data trend curve using the fractal interpolation algorithm includes the following sub-steps: The decrypted fractal data sub-block is self-similarity verified, and the morphological similarity between its fractal tree encoding and the parent block encoding is calculated. If the similarity is lower than the preset threshold, it is marked as an abnormal sub-block and triggers re-retrieval; According to the hierarchical position of the missing sub-block, the self-similarity coefficients of the adjacent levels are selected for interpolation. The coefficients are dynamically adjusted by the fractal dimension value. The interpolation formula is: V 插值 =α·V 父块 +(1-α)·V 兄弟子块 Where α is the self-similarity weight factor of the current level, ranging from 0.6 to 0.9; The mutation point in the interpolated data trend curve is detected. If the mutation amplitude exceeds 3 times the standard deviation of the historical mean, a smoothing correction is performed based on fractal self-similarity to retain the trend morphological characteristics.

7. An intelligent generator for fully automatic production report generation, characterized in that: The intelligent generator for fully automatic generation of production reports includes: An acquisition module is configured to acquire original time series data, including equipment operating parameters, production records, and process indicators; determine a fractal cutting dimension based on the original time series data; cut the original time series data into fractal data sub-blocks with self-similarities; encrypt each fractal data sub-block to generate an encrypted fractal sub-block; and distribute the encrypted fractal sub-blocks to multiple edge storage nodes; a processing module, configured to construct a cellular automaton network, generate cellular state transition rules based on real-time network traffic data and historical access logs, and upon receiving a report access request initiated by a user, perform access permission verification based on the current state of the corresponding user account cell in the cellular automaton network, and generate an access permission verification result; A setting module is used to establish a system dynamics model, the system dynamics model defines the coupling relationship between production data flow and network traffic, and calculates a real-time risk confidence index based on the device operating parameters and the access permission verification result; The allocation module is used to retrieve the encrypted fractal sub-blocks related to the current report request from the edge storage node, obtain the restored data trend, dynamically desensitize the restored data trend based on the access permission verification result, and generate the production report content based on the desensitized data trend and the real-time risk confidence indicator.

8. An intelligent generation device for fully automatic generation of production reports, characterized in that: The intelligent generation device for fully automatic generation of production reports includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the intelligent generation device for fully automatic generation of production reports to execute the intelligent generation method for fully automatic generation of production reports according to any one of claims 1 to 6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the intelligent generation method for fully automatic generation of production reports according to any one of claims 1 to 6 is implemented.

Images