Special equipment intelligent control method and system based on internet of things

Abstract

The application relates to the field of special equipment safety management, and discloses a special equipment intelligent control method and system based on an Internet of Things. The method collects facial images and fingerprint data of an operator through an industrial Internet of Things sensor, matches a preset biological recognition template to obtain preliminary verification information, calls personnel qualification grades and historical operation records from a database, processes the data through a random forest algorithm to determine a comprehensive qualification score, collects equipment temperature and pressure parameters in real time, marks an abnormal state to generate a state label if the parameters exceed a threshold value after detection, verifies the authenticity of the identity in multiple dimensions based on the state label and the qualification score, matches equipment operation qualification requirements, generates a matching degree, generates a temporary permission token if the matching degree meets the requirements, and otherwise, access is denied; the token is transmitted to an equipment control module after being encrypted, and the operation range is adjusted in real time to complete dynamic permission configuration. The method realizes accurate permission control, improves the operation safety and management efficiency of special equipment, and provides protection for the safe operation of industrial production.

Description

Technical Field

[0001] This invention relates to the field of special equipment safety management, and in particular to a method and system for intelligent control of special equipment based on the Internet of Things. Background Technology

[0002] Specialized equipment (such as boilers, pressure vessels, or lifting machinery) is a core support for industrial production. Its safe and stable operation directly impacts the safety of operators and the efficiency of enterprise production, playing an irreplaceable role in key sectors such as chemical and energy industries. With the rapid development of Industrial Internet of Things (IIoT) technology, intelligent management has become a core direction for improving the safety and operational efficiency of specialized equipment. The scientific control of operating permissions, as a key link in intelligent management, directly determines the effectiveness of the overall management system. Among these technologies, NB-IoT, with its wide coverage, low power consumption, and massive connectivity, provides reliable support for the real-time transmission of data from specialized equipment in industrial scenarios. It not only solves the problems of insufficient coverage and excessive power consumption of traditional wireless transmission in complex industrial environments but also creates conditions for the dynamic linkage between operating permissions, equipment status, and personnel qualifications. This drives the transformation of specialized equipment management from a traditional model relying on manual verification to a data-driven intelligent collaborative model.

[0003] Currently, in the field of special equipment management, traditional management models still primarily rely on manual verification of operator qualifications and fixed permission allocation, which is difficult to adapt to complex and ever-changing industrial operation scenarios. Manual verification is not only inefficient but also prone to omissions due to human error, posing safety hazards. Fixed permission allocation, once determined, remains unchanged for a long period, failing to adapt to dynamic needs such as operator job adjustments or skill updates, and also neglecting the status differences caused by changes in parameters such as temperature or pressure during special equipment operation. For example, allowing high-load operation with conventional permissions when equipment pressure exceeds limits can easily lead to malfunctions. While existing improvement methods attempt to introduce identity verification technology, most rely on single methods such as passwords or magnetic cards, which are prone to misuse or loss. Some biometric methods, such as facial recognition or fingerprint recognition, suffer from decreased recognition accuracy due to factors such as dust, high temperatures, or electromagnetic interference in industrial environments. More importantly, existing methods generally disconnect the connection between "personnel qualifications - equipment status - operating permissions," failing to utilize NB-IoT technology for real-time collaboration of multi-dimensional data. This results in a lack of dynamism and accuracy in permission allocation, failing to meet the actual needs of special equipment safety management.

[0004] Therefore, how to achieve highly reliable identity verification and real-time device status analysis in complex industrial environments by relying on NB-IoT technology, break down the information gap between "personnel-equipment-authority", and ultimately realize the dynamic and precise allocation of operating permissions has become a key issue in the intelligent control of special equipment. Summary of the Invention

[0005] This invention provides a method and system for intelligent control of special equipment based on the Internet of Things, so as to achieve precise control of permissions, improve the safety and management efficiency of special equipment operation, and provide a guarantee for the safe operation of industrial production.

[0006] In a first aspect, to address the aforementioned technical problems, the present invention provides a method for intelligent control of special equipment based on the Internet of Things, comprising:

[0007] The system collects facial images and fingerprint data of operators through industrial IoT sensors and matches them with preset biometric templates to obtain preliminary verified identity information.

[0008] Based on the initially verified identity information, the corresponding personnel's qualification level and historical operation records are retrieved from the preset database. A trained random forest algorithm is used for comprehensive evaluation to determine the comprehensive qualification score.

[0009] The operating parameters of special equipment are collected in real time through the industrial Internet of Things interface and compared with the preset safety threshold. If the safety threshold is exceeded, the equipment is marked as abnormal and the equipment status label is obtained.

[0010] A multi-dimensional verification mechanism is adopted to verify the authenticity of the initially verified identity information. At the same time, the corresponding operation qualification requirements are extracted from the device status label. The comprehensive qualification score is matched and verified with the operation qualification requirements. The results of the authenticity verification and matching verification are combined and the quantitative value is output as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied. This determines the basis for permission allocation.

[0011] A temporary permission token is extracted from the permission allocation basis and transmitted to the special equipment control module via an encryption protocol to adjust the range of operation instructions in real time, thereby completing the dynamic permission configuration.

[0012] The special equipment operating parameters include at least temperature and pressure values, and the temporary access token is a dynamic electronic certificate that binds personnel qualifications and equipment status, including the scope of operation and time limit.

[0013] In one optional implementation, the step of collecting facial images and fingerprint data of operators through industrial IoT sensors and matching them with a preset biometric template to obtain preliminary verified identity information includes:

[0014] Based on the identity verification requirements before operating special equipment, the industrial IoT sensors are activated to collect facial images and fingerprint data of the operators.

[0015] The facial image and fingerprint data are transmitted to the data processing module. Feature points are extracted from the acquired facial image and compared with authorized facial standard features. At the same time, features are extracted from the fingerprint data and compared with fingerprint baseline texture features. The biometric template includes authorized facial standard features and fingerprint baseline texture features.

[0016] After the comparison is completed, the comparison results are analyzed. If all of them meet the preset matching criteria, the comparison is deemed successful and preliminary identity information is generated. If any comparison fails to meet the criteria, the data collection is triggered again until the preliminary identity information is obtained or the identity verification is confirmed to have failed.

[0017] In one optional implementation, the step of retrieving the corresponding personnel's qualification level and historical operation records from a preset database based on the pre-verified identity information, and then using a trained random forest algorithm for comprehensive evaluation to determine the comprehensive qualification score includes:

[0018] Based on the initially verified identity information, a targeted query is initiated into the preset database to extract the corresponding personnel's qualification level and historical operation records;

[0019] The current operational requirements of the special equipment are extracted from the equipment status tags. The qualification level, historical operation records and extracted operational requirements are integrated to form a comprehensive dataset. The missing information and abnormal values ​​in the comprehensive dataset are cleaned.

[0020] A trained random forest algorithm is used to perform multi-dimensional evaluation on the cleaned comprehensive dataset. By constructing a preset number of decision trees, the suitability of qualification level and operation requirements, historical operation compliance rate and operation proficiency are evaluated respectively, and the evaluation results are output. The decision tree is the core component of the random forest algorithm.

[0021] The evaluation results of each decision tree are summarized and calculated to determine the comprehensive qualification score that reflects the overall operational ability of the corresponding personnel.

[0022] In one alternative implementation, after completing the dynamic permission configuration, the method further includes:

[0023] When an operation command is detected to be inconsistent with the device status label, an alarm mechanism is triggered, and real-time feedback data in the current scenario is collected.

[0024] The historical operation records and equipment status tags in the preset database are updated using the collected real-time feedback data, and the weights of the evaluation dimensions of the comprehensive qualification score are adjusted synchronously.

[0025] The operator's facial image and fingerprint data are re-collected and matched with the biometric template to generate optimized identity information;

[0026] For the optimized identity information and the current device status label, a multi-dimensional verification mechanism is initiated for fusion processing. Based on the processing results, an updated temporary permission token is generated to complete the continuous iteration of dynamic permission configuration.

[0027] The real-time feedback data includes the specific operation instructions triggered by the operator, the time information of instruction sending and execution, the fluctuation data of the core operating parameters of the equipment corresponding to the operation instructions, the response status of the equipment to the instructions and the state switching time, and the environmental interference parameters of the operation scenario.

[0028] In one alternative implementation, after completing the dynamic permission configuration, the method further includes:

[0029] Real-time monitoring of changes in the operation scenario corresponding to the temporary permission token. Once any change in the operation scenario is detected, the token status verification is automatically triggered immediately. If the operator fails to perform the operation for more than the preset time or the device changes from normal to abnormal status, the temporary permission token is immediately invalidated.

[0030] When a temporary permission token expires, the token expiration information is simultaneously transmitted to the special equipment control module and the preset database. Upon receiving the expiration information, the special equipment control module immediately clears the current operation permission scope, and the preset database synchronously records the token expiration time and the reason for expiration.

[0031] The changes in the operating scenarios include interruption of operator actions, switching of special equipment operating status, and abnormal operating environment parameters.

[0032] Secondly, the present invention also provides an intelligent control system for special equipment based on the Internet of Things, comprising:

[0033] Data acquisition module: Collects facial images and fingerprint data of operators through industrial IoT sensors, and matches them with preset biometric templates to obtain preliminary verified identity information;

[0034] Comprehensive evaluation module: Based on the initially verified identity information, retrieve the corresponding personnel's qualification level and historical operation records from the preset database, and use a trained random forest algorithm to conduct a comprehensive evaluation to determine the comprehensive qualification score;

[0035] Anomaly detection module: Collects special equipment operating parameters in real time through industrial IoT interface and compares them with preset safety thresholds. If the safety threshold is exceeded, the equipment is marked as abnormal and an equipment status label is obtained.

[0036] The permission allocation module uses a multi-dimensional verification mechanism to verify the authenticity of the initially verified identity information. At the same time, it extracts the corresponding operation qualification requirements from the device status tags, matches and verifies the comprehensive qualification score with the operation qualification requirements, integrates the results of authenticity verification and matching verification, and outputs the quantitative value as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied, thereby determining the basis for permission allocation.

[0037] Permission configuration module: Extracts temporary permission tokens from the permission allocation criteria, transmits them to the special equipment control module via an encryption protocol, adjusts the range of operation instructions in real time, and completes dynamic permission configuration;

[0038] The special equipment operating parameters include at least temperature and pressure values, and the temporary access token is a dynamic electronic certificate that binds personnel qualifications and equipment status, including the scope of operation and time limit.

[0039] In an optional implementation, an optimization and update module is further included, the optimization and update module comprising:

[0040] Alarm unit: When an operation command is detected to be mismatched with the device status label, an alarm mechanism is triggered, and real-time feedback data in the current scenario is collected at the same time;

[0041] Data update unit: Updates historical operation records and equipment status tags in the preset database using real-time feedback data collected, and simultaneously adjusts the weights of the evaluation dimensions of the comprehensive qualification score;

[0042] Secondary data acquisition unit: re-acquires the operator's facial image and fingerprint data, combines them with the biometric template to complete the matching, and generates optimized identity information;

[0043] Secondary permission configuration unit: For the optimized identity information and the current device status label, a multi-dimensional verification mechanism is launched for fusion processing, and an updated temporary permission token is generated based on the processing result to complete the continuous iteration of dynamic permission configuration;

[0044] The real-time feedback data includes the specific operation instructions triggered by the operator, the time information of instruction sending and execution, the fluctuation data of the core operating parameters of the equipment corresponding to the operation instructions, the response status of the equipment to the instructions and the state switching time, and the environmental interference parameters of the operation scenario.

[0045] In an optional implementation, a token expiration monitoring module is further included, the token expiration monitoring module comprising:

[0046] Failure monitoring unit: Real-time monitoring of changes in the operation scenario corresponding to the temporary permission token. Upon detecting any change in the operation scenario, the token status verification is immediately triggered. If the operator fails to perform the operation for more than the preset time or the equipment changes from normal to abnormal state, the temporary permission token is immediately invalidated.

[0047] Anomaly Response Unit: After the temporary permission token expires, the token expiration information is transmitted synchronously to the special equipment control module and the preset database. After receiving the expiration information, the special equipment control module immediately clears the current operation permission range, and the preset database synchronously records the token expiration time and expiration reason.

[0048] The changes in the operating scenarios include interruption of operator actions, switching of special equipment operating status, and abnormal operating environment parameters.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] (1) By collecting facial images and fingerprint data of operators through industrial IoT sensors and combining them with preset biometric templates, dual identity verification is completed. Compared with traditional single password or magnetic card verification, the reliability of identity recognition is greatly improved. This dual biometric verification method can effectively resist the influence of dust or electromagnetic interference in the industrial environment on single recognition, avoid misjudgment of qualifications caused by identity theft or loss, ensure that the operator's identity corresponds accurately with the actual qualifications, and lay a secure identity foundation for permission allocation.

[0051] (2) Based on the initially verified identity information, the personnel qualification level and historical operation records are retrieved, and a random forest algorithm is used for comprehensive evaluation to determine the comprehensive qualification score, rather than allocating permissions solely based on static qualifications. This algorithm can simultaneously consider the matching degree between qualifications and operational requirements, historical compliance rate, and operational proficiency, fully integrating multi-dimensional information related to personnel capabilities, improving the accuracy of the evaluation of the actual operational capabilities of operators, and providing a scientific basis for personnel capabilities for subsequent permission matching.

[0052] (3) Real-time acquisition of operating parameters such as temperature and pressure of special equipment through industrial IoT interfaces, comparison with preset safety thresholds to generate equipment status tags, and accurate capture of equipment dynamic status. The wide coverage and low power consumption of industrial IoT technology ensure the real-time and stable transmission of equipment parameters in complex industrial scenarios, avoid misjudgment of equipment status due to parameter delay or loss, provide a basis for permission allocation that fits the actual operating status of the equipment, and break the limitation of the separation between traditional permission management and equipment status.

[0053] (4) A multi-dimensional verification mechanism is adopted to integrate preliminary identity information, comprehensive qualification scores and equipment status tags. Temporary permission tokens are generated by calculating the matching degree, rather than fixed permission allocation. This multi-dimensional integration method realizes the linkage of "personnel qualification-equipment status-operation permission", so that the permission can adapt to the differences in personnel ability and changes in equipment status, avoids the problem of granting high-risk operation permissions when the equipment is abnormal, and improves the accuracy and flexibility of permission allocation.

[0054] (5) Temporary permission tokens are transmitted to the special equipment control module via an encryption protocol, and the range of operation instructions is adjusted in real time to complete dynamic permission configuration. Combined with the low latency characteristics of the Industrial Internet of Things, the response time for permission adjustment is greatly shortened. When the equipment status changes suddenly or the personnel qualifications are abnormal, the permission adjustment instructions can take effect quickly, avoiding security risks caused by the mismatch between operation and equipment status due to permission delay, and ensuring the real-time performance and security of permission control.

[0055] (6) By utilizing real-time feedback data collected when the operation instructions and equipment status do not match, the preset database is updated and the weights of the comprehensive qualification score evaluation dimensions are adjusted. At the same time, the temporary permission tokens are iteratively optimized. This feedback-based continuous iteration mechanism allows the permission management system to be dynamically optimized according to the equipment operation rules and personnel operation performance. With long-term use, the accuracy of permission allocation can be further improved, providing continuously adaptable permission management support for the long-term safe operation of special equipment, and enhancing the practicality and reliability of intelligent control of special equipment in industrial scenarios. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating an intelligent control method for special equipment based on the Internet of Things provided in an embodiment of the present invention.

[0057] Figure 2 This is a schematic diagram of the structure of an intelligent control system for special equipment based on the Internet of Things provided in an embodiment of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Reference Figure 1 This invention provides an intelligent control method for special equipment based on the Internet of Things, comprising the following steps:

[0060] S11 collects facial images and fingerprint data of operators through industrial IoT sensors and matches them with preset biometric templates to obtain preliminary verified identity information;

[0061] S12, based on the initially verified identity information, retrieve the corresponding personnel's qualification level and historical operation records from the preset database, and use the trained random forest algorithm to conduct a comprehensive evaluation to determine the comprehensive qualification score;

[0062] S13, collect the operating parameters of special equipment in real time through the industrial Internet of Things interface, and compare them with the preset safety threshold. If the safety threshold is exceeded, the equipment is marked as abnormal and the equipment status label is obtained.

[0063] S14, a multi-dimensional verification mechanism is used to verify the authenticity of the initially verified identity information. At the same time, the corresponding operation qualification requirements are extracted from the device status label. The comprehensive qualification score is matched and verified with the operation qualification requirements. The results of the authenticity verification and the matching verification are integrated and the quantitative value is output as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated. Otherwise, access is denied, thereby determining the basis for permission allocation.

[0064] S15, extract a temporary permission token from the permission allocation basis, transmit it to the special equipment control module via an encryption protocol, adjust the range of operation instructions in real time, and complete the dynamic permission configuration.

[0065] In step S11, facial images and fingerprint data of operators are collected by industrial IoT sensors and matched with preset biometric templates to obtain preliminary verified identity information.

[0066] In one implementation, this step takes the pre-operation identity verification of a 10t / h steam boiler (special equipment type: pressure vessel) in a chemical workshop as an example, and the specific implementation process is as follows:

[0067] Prior to implementation, an industrial-grade dual-function data acquisition terminal was deployed 1.5m in front of the boiler control panel. This terminal integrates two types of industrial IoT sensors. The facial image acquisition uses a 20-megapixel CMOS infrared camera, supporting a frame rate of 30fps and a resolution of 1920×1080. It is equipped with an 850nm infrared supplementary lighting module to adapt to dusty, strong light, or low light environments in the workshop. The acquisition angle covers the area where the operator stands (0.8m wide, 1.5-1.9m high). The fingerprint data acquisition uses a 512dpi capacitive fingerprint sensor, capable of extracting ≥200 feature points (such as ridge endpoints or bifurcation points) in a single acquisition. The sensor surface is covered with an oil-resistant coating to protect hands from smudges after mechanical operation in the workshop. In oil-stained scenarios, it supports press-type data collection with a collection time of ≤0.5 seconds per scan. Simultaneously, it constructs a biometric template for authorized operators in a pre-defined database. Authorized facial standard features are obtained by collecting three sets of facial images from different angles (front and 45° to the left and right sides) for each operator. A 128-dimensional facial feature vector is extracted using a convolutional neural network (CNN), and the mean of the three vectors is calculated. The coordinate ranges of key facial regions (corners of the eyes, nostrils, and corners of the mouth) are stored. Fingerprint baseline texture features are obtained by collecting three fingerprints from the operator's right index finger. A Minutiae-based algorithm is used to extract feature point sets for each fingerprint, and the feature points present in all three fingerprints (≥150) are selected and recorded. The system establishes topological relationships such as relative distances and angles between objects; it also sets preset matching standards, namely, facial feature comparison must meet the following requirements: the Euclidean distance between the collected facial feature vector and the "authorized facial standard feature" must be <0.6 and the coordinate deviation of the key facial region must be ≤5%; fingerprint feature comparison must meet the following requirements: the number of matching fingerprint feature points with the "fingerprint baseline texture feature" must be ≥30 and the topological similarity of the feature points must be ≥90%; when the authorized operator "Zhang San" (boiler operator certificate A level, ID: ZS001) clicks the "start operation" button at the operating console, the system detects the identity verification requirement, automatically starts the acquisition terminal, and the infrared camera starts and fills in light within 0.3 seconds, continuously acquiring 3 frames of facial images (discarding blurry frames and retaining the remaining frames). The image data (JPEG format, approximately 2MB) is transmitted to the data processing module via an industrial IoT gateway with a transmission delay of ≤50ms. Zhang San presses his right index finger on the fingerprint sensor, and fingerprint acquisition (generating a 500×500 pixel grayscale image) is completed within 0.5 seconds. The fingerprint data (BMP format, approximately 250KB) is then transmitted to the data processing module via the same gateway. After receiving the data, the data processing module performs feature comparison. The facial image undergoes denoising, grayscale normalization, and histogram equalization preprocessing to extract the coordinates of 68 feature points. After comparison with the "authorized facial standard features", the Euclidean distance is 0.48 (<0.6), and the key area coordinate deviation is 3.2% (≤5%), facial comparison is considered passed. Fingerprint data undergoes binarization, thinning, and anti-spoofing feature preprocessing, extracting 198 feature points. After comparison with the "fingerprint baseline texture features," 42 matching points are found (≥30), with a topological similarity of 92% (≥90%), thus fingerprint comparison is considered passed. When both types of features meet the preset standards, the system generates preliminary verified identity information: {"Operator ID": "ZS001", "Name": "Zhang San", "Identity Verification Result": "Passed", "Facial Comparison Settings";} "Reliability": 0.85, "Fingerprint Comparison Confidence": 0.92, "Verification Time": "2025-XX-XX 09:15:30"}, and transmit it to the next stage for qualification retrieval. If any feature fails to meet the standard (e.g., oily hands resulting in only 25 fingerprint matching points), a voice prompt ("Fingerprint collection is unclear, please clean your finger and press again") and a flashing red indicator light will trigger a re-collection. If three consecutive attempts fail, the collection terminal will be locked (unlocked after 10 minutes), and an alarm message will be pushed to the workshop management terminal.

[0068] In step S12, based on the initially verified identity information, the qualification level and historical operation records of the corresponding personnel are retrieved from the preset database, and a trained random forest algorithm is used for comprehensive evaluation to determine the comprehensive qualification score.

[0069] In one implementation, before implementation, a pre-set database stores personnel qualification level data (operator certificate level: A-level is 1.0, B-level is 0.75, and C-level is 0.5; certificate validity period: >6 months is 1.0, 3-6 months is 0.7, and <3 months is 0.3; specialized training duration: ≥200 hours is 1.0, 100-200 hours is 0.7, and <100 hours is 0.3) and historical operation record data (number of operations in the past year: ≥50 times is 1.0, 30-5...). 0 times is 0.7 and <30 times is 0.3; Compliance operation rate: ≥95% is 1.0, 90-95% is 0.8 and <90% is 0.5; Fault handling time: ≤1 hour is 1.0, 1-2 hours is 0.7 and >2 hours is 0.3. Simultaneously configure the random forest algorithm parameters: Total number of decision trees N=100 (including n1=40 decision trees for evaluating "fitness between qualification level and operational requirements", n2=30 decision trees for evaluating "historical operation compliance rate", and n3=30 decision trees for evaluating "operational compliance rate"). (The system obtains the preliminary identity information of "Zhang San" (ID: ZS001) and then retrieves his qualification data (Operator Certificate Level A, quantitative value 1.0; certificate validity until December 2026, 14 months from now, quantitative value 1.0; specialized training duration 220 hours, quantitative value 1.0) and historical operation data (58 operations in the past year, quantitative value 1.0; compliance operation rate 93%, quantitative value 1.0) from the database.) 0.8; average fault handling time 1.1 hours, quantized value 0.7), integrated into a comprehensive dataset {1.0, 1.0, 1.0, 1.0, 0.8, 0.7}, and removed one outlier in fault handling time caused by "sudden equipment power outage" (the original 5-hour outlier was replaced with an average of 1.1 hours), ensuring that the data was complete and anomalies-free; the trained random forest algorithm was used to classify the cleaned dataset, and all 40 decision trees evaluating qualification matching degree output evaluation scores S. 1i =1.0 (i=1 to 40, representing the evaluation result of each decision tree of this type), and all 30 decision trees evaluating compliance rate output an evaluation score S. 2j =0.8 (j=1 to 30), all 30 decision trees for evaluating proficiency output an evaluation score S. 3k =0.75 (k=1 to 30), then through the formula Calculate the comprehensive qualification score, where S is the final comprehensive qualification score (ranging from 0 to 1.0, with a higher score indicating stronger comprehensive operational ability). The sum of the evaluation scores for the 40 qualification matching decision trees. The sum of the evaluation scores for the 30 compliance rate decision trees. The sum of the evaluation scores of 30 proficiency decision trees is given by N=100, n1=40, and S 1i =1.0, n2=30, S2j =0.8, n3=30 and S 3k Substituting 0.75 into the formula, we get the final comprehensive qualification score S as 0.865. This score is then stored in the database for subsequent fusion with the device status tag to generate an access token.

[0070] In step S13, the operating parameters of the special equipment are collected in real time through the industrial Internet of Things interface and compared with the preset safety threshold. If the safety threshold is exceeded, the equipment is marked as abnormal, and the equipment status label is obtained.

[0071] In one implementation, the industrial IoT interface and sensors are deployed first, and safety thresholds and anomaly levels are set: For sensor deployment, thermocouple temperature sensors (range 0-400℃, accuracy ±0.5℃) are installed on the top of the boiler furnace, and strain gauge pressure sensors (range 0-4MPa, accuracy ±0.02MPa) are installed on the steam outlet pipe. Both types of sensors are connected to the workshop industrial IoT gateway via an RS485 industrial bus. The gateway then communicates with the data acquisition module via TCP / IP protocol, with the parameter acquisition frequency set to every 5... Once per second; the safety threshold and anomaly level are set as follows: normal temperature range 120℃-180℃ (above 180℃ is abnormal), normal pressure range 0.5MPa-1.5MPa (above 1.5MPa is abnormal), where temperature 180-200℃ / pressure 1.5-1.8MPa is defined as "mild anomaly", and temperature >200℃ / pressure >1.8MPa is defined as "severe anomaly"; after "Zhang San" completes identity verification, the system automatically triggers the equipment operating parameter acquisition process, and the industrial IoT gateway acquires data from the temperature sensor at a preset frequency. The current furnace temperature is 182.3℃ and the steam outlet pressure obtained from the pressure sensor is 1.58MPa. These parameters are encapsulated as JSON data {"deviceID":"GL-001","temp":182.3,"press":1.58,"collectTime":"2025-XX-XX09:18:25"} via the Industrial Internet of Things (IIoT) interface and transmitted to the data processing module with a transmission delay ≤100ms. Upon receiving the parameters, the data processing module first verifies the data integrity and validity (removing null values ​​and out-of-range values ​​caused by temporary sensor interference). It confirms that the temperature of 182.3℃ and the pressure of 1.58MPa are valid data. Then, it compares the data with preset safety thresholds. It finds that the temperature of 182.3℃ exceeds 180℃ (threshold) and the pressure of 1.58MPa exceeds 1.5MPa (threshold), but does not meet the "severe anomaly" standard. Based on this comparison result, the system generates a device status label: {"deviceID":"GL-001","status":"mild anomaly","abnormal"}. Param":{"temp":182.3,"press":1.58},"abnormalLevel":"Low Risk","labelTime":"2025-XX-XX09:18:30"}, and store this label in the database for subsequent integration with "Zhang San's" comprehensive qualification score to determine permissions.

[0072] In step S14, a multi-dimensional verification mechanism is used to verify the authenticity of the initially verified identity information. At the same time, the corresponding operation qualification requirements are extracted from the device status label, and the comprehensive qualification score is matched and verified with the operation qualification requirements. The results of the authenticity verification and the matching verification are merged and the quantitative value is output as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied, thereby determining the basis for permission allocation.

[0073] In one implementation, a multi-dimensional verification mechanism with verification rules and preset matching thresholds is configured before implementation: different device status labels correspond to different operational qualification requirements. In this embodiment, the minimum requirement for comprehensive qualification score (S0) is used as the quantitative form of operational qualification requirements. Identity authenticity verification is performed by comparing the SHA-256 hash value of the preliminary verification identity information with the value stored in the database. If it passes, A=1.0; if it fails, A=0. The device's "mildly abnormal" status corresponds to the minimum requirement for the operator's comprehensive qualification score S0=0.8, and "severely abnormal" corresponds to S0=0.9. The matching degree threshold is set to 0.85 (value 0-1.0, above the threshold indicates that the personnel and device status are compatible and permissions can be granted). At the same time, the identity authenticity verification weight W1=0.3 and the qualification score matching degree weight W2=0.7 are set. W1 and W2 represent the importance of the two verification dimensions in the comprehensive matching; after "Zhang San's" comprehensive qualification score S=0.865 and the boiler's "slightly abnormal" status label are synchronized to the data processing module, the system first verifies its identity (retrieving the SHA-256 hash value of the identity information "ZS001" in the database, which is completely consistent with the hash value calculated from the current preliminary verification information, resulting in A=1.0), then extracts the minimum qualification requirement S0=0.8 from the equipment status label, and then uses the formula Calculate the matching degree, where M is the matching degree of multi-dimensional verification (value 0-1.0, reflecting the comprehensive situation of the authenticity of the person's identity and the suitability of their qualifications). For qualification score matching degree Calculate the ratio of the actual score to the minimum requirement. If the ratio is ≥1.0, take 1.0, indicating that the qualification fully meets the requirements; if the ratio is <1.0, take the actual ratio, indicating that the qualification is not fully met. Substitute W1=0.3, A=1.0, W2=0.7, S=0.865, and S0=0.8 into the formula to calculate... Since 1.081 ≥ 1.0, min(1.081, 1.0) = 1.0. Therefore, M = 0.3 × 1.0 + 0.7 × 1.0 = 1.0. Since 1.0 > 0.85 (threshold), the system generates a temporary access token (bound to "Zhang San" ID: ZS001, boiler ID: GL-001, and the equipment's "minor abnormality" status, allowing low-risk operations such as "water supply adjustment" and "steam valve fine-tuning," valid for 30 minutes). If we assume the operator's comprehensive qualification score S = 0.7 (S0 = 0.8), then... Given min(0.875, 1.0) = 0.875, the matching degree M = 0.3 × 1.0 + 0.7 × 0.875 = 0.3 + 0.6125 = 0.9125. If S = 0.65, then... The matching degree M = 0.3 + 0.56875 = 0.86875. If S = 0.6, then M = 0.3 × 1.0 + 0.7 × (0.6 / 0.8) = 0.3 + 0.525 = 0.825 (0.825 < 0.85 (threshold), triggering an access denial instruction). Finally, a temporary permission token or access denial instruction is used as the basis for permission allocation in the current operation scenario.

[0074] In step S15, a temporary permission token is extracted from the permission allocation basis, transmitted to the special equipment control module via an encryption protocol, and the range of operation instructions is adjusted in real time to complete the dynamic permission configuration.

[0075] In one implementation, the encryption protocol and operation command range mapping rules are configured before implementation: the encryption protocol uses the AES-256 encryption algorithm (using a 32-byte random string key, updated monthly by the system backend), and is paired with the TLS1.3 protocol to build a token transmission encryption channel. A pre-installed decryption program in the special equipment control module (boiler PLC control system) is used to parse the encrypted token; the operation command range mapping rules are set as follows:

[0076] Boiler "normal state" (temperature 120℃-180℃, pressure 0.5MPa-1.5MPa): The whitelist of executable commands is "feed water adjustment (3-10m³ / h), steam valve adjustment (20%-80% opening) and fuel supply adjustment (5-15kg / h)", with no prohibition commands;

[0077] Boiler "slightly abnormal" state (temperature 180-200℃, pressure 1.5-1.8MPa): The whitelist of executable instructions is "adjustment of feedwater (5-8m³ / h), fine adjustment of steam valve (30%-50% opening)", and the list of prohibited instructions is "increase in fuel supply and increase in boiler load".

[0078] Boiler "severe abnormality" status (temperature > 200℃, pressure > 1.8MPa): The whitelist of executable instructions is limited to "emergency pressure relief operation", and all non-emergency instructions such as "adjustment of feedwater flow, increase of fuel supply and opening of steam valves" are prohibited;

[0079] The system extracts a temporary permission token containing the identity of "Zhang San," boiler ID, executable instruction code, and validity period from the permission allocation criteria. It then encrypts the token using the AES-256 encryption algorithm to generate a 256-bit encrypted string. Simultaneously, it calculates the token hash value using the SHA-256 algorithm (used by the control module to verify data integrity upon receipt). The encrypted token and hash value are transmitted to the boiler control module via the industrial IoT network with a transmission delay of ≤50ms. After receiving the data, the control module decrypts the token using a pre-set decryption program and compares its own calculated hash value with the received hash value. Once confirmed that the two are consistent and the data has not been tampered with, the control module... The system parses the executable instruction code in the token. Based on the parsed instruction code, the control module calls the operation instruction range mapping rules to adjust the permission configuration in real time. The operation interface only displays the "Water Supply Adjustment" and "Steam Valve Fine-tuning" buttons, hiding prohibited instructions such as "Fuel Supply Increase". When clicking "Water Supply Adjustment", only values ​​of 5-8 m³ / h are allowed to be entered. If the value exceeds this range, a pop-up window will prompt "Parameter exceeds the safe range, you need to re-enter". After the permission configuration is completed, the control module generates a permission configuration log containing the configuration time, operator ID, equipment status and a list of executable instructions, and synchronously sends it back to the system database for storage, which is used for subsequent permission traceability.

[0080] In one embodiment, after completing the dynamic permission configuration, the method further includes: triggering an alarm mechanism when a mismatch is detected between the operation command and the device status label, and simultaneously collecting real-time feedback data in the current scenario; updating historical operation records and device status labels in a preset database using the collected real-time feedback data, and synchronously adjusting the evaluation dimension weights of the comprehensive qualification score; re-collecting the operator's facial image and fingerprint data, and matching them with the biometric template to generate optimized identity information; and initiating a multi-dimensional verification mechanism to perform fusion processing on the optimized identity information and the current device status label, generating an updated temporary permission token based on the processing result, and completing the continuous iteration of the dynamic permission configuration; wherein, the real-time feedback data includes the specific operation command content triggered by the operator, the time information of command sending and execution, the fluctuation data of the core operating parameters of the device corresponding to the operation command, the device's response status to the command and the state switching time, and the environmental interference parameters of the operation scenario.

[0081] In one implementation, before implementation, configure the command-state matching rules, alarm mechanism, and data update rules:

[0082] Command-Status Matching Rules: If the boiler is in any of the normal, slightly abnormal, or severely abnormal states, and the operation command code is not in the whitelist of executable commands for the corresponding state, or the parameter exceeds the corresponding adjustment range, it is judged as a mismatch.

[0083] Alarm mechanism: Trigger an audible and visual alarm (red light flashing on the control panel and a 2kHz buzzer alarm), and simultaneously push alarm information to the workshop management terminal;

[0084] Data update rules: When an operation violates regulations, the weight of "historical operation compliance rate" in the comprehensive qualification score assessment increases from 0.3 to 0.4, while the weight of "qualification level" decreases to 0.3. Real-time feedback data collection dimensions include instruction content, instruction sending / execution time, equipment parameter fluctuations, equipment response status, and environmental interference parameters (dust concentration, humidity). It should be noted that the definition of an operation violation is: "the operation instruction is not in the whitelist of executable instructions corresponding to the current equipment status, or the parameter adjustment exceeds the preset safety range," and the compliance rate calculation process is explained (e.g., "originally, there were 58 operations in the past year, with 54 compliant, resulting in a compliance rate of 93%; after adding 1 violation instruction, the total number of operations is 59, with 54 compliant, resulting in a compliance rate of ≈91.5%").

[0085] During operation, "Zhang San" accidentally triggered the "increase fuel supply" command (a prohibited command). The system detected a mismatch between the command and the equipment's "minor anomaly" status label.

[0086] Trigger alarm mechanism: The red light on the control panel flashes and the buzzer sounds an alarm, pushing information to the management terminal (including "Zhang San" ID, violation command and trigger time "2025-XX-XX09:30:15");

[0087] Collect real-time feedback data: Record the instruction content "fuel supply increased" and the execution status "not executed", and collect boiler parameters (temperature 182.3℃, pressure 1.58MPa, no fluctuation) and environmental parameters (dust concentration 8mg / m³, humidity 65%) within 10 seconds.

[0088] The system will update the preset database with real-time feedback data:

[0089] One violation record was added to "Zhang San's" historical operation record, reducing the compliance rate from 93% to 91.8%.

[0090] The boiler status label has been supplemented with a note stating "A violation command was triggered".

[0091] The weighting of the comprehensive qualification score assessment dimensions has been adjusted: the weight of "historical operation compliance rate" is 0.4, and the weight of "qualification level" is 0.3.

[0092] After the data update is completed, the system prompts "Zhang San" to re-verify his identity: the sensor is activated to collect his facial image and fingerprint data, and combined with the biometric template matching to generate optimized identity information (verification confidence level 0.88); for the optimized identity information and the current boiler "slightly abnormal" status label, a multi-dimensional verification mechanism is activated to recalculate the comprehensive qualification score (due to the decrease in compliance rate, the score drops from 0.885 to 0.86), the matching degree = 1.0 × 0.3 + (0.86 / 0.8) × 0.7 = 0.3 + 0.753 = 1.053 (take 1.0), and an updated temporary permission token is generated (only the "water supply adjustment" command is retained), completing the dynamic permission configuration iteration.

[0093] In another embodiment, after completing the dynamic permission configuration, the method further includes: real-time monitoring of changes in the operation scenario corresponding to the temporary permission token; upon detecting any change in the operation scenario, immediately triggering token status verification; if the operator fails to perform an operation for more than a preset time, or if the equipment changes from a normal state to a slightly abnormal or severely abnormal state, the temporary permission token is immediately invalidated; after the temporary permission token is invalidated, the token invalidation information is simultaneously transmitted to the special equipment control module and the preset database; upon receiving the invalidation information, the special equipment control module immediately clears the current operation permission range, and the preset database synchronously records the token invalidation time and the reason for invalidation; wherein, the changes in the operation scenario include interruption of operator behavior, switching of special equipment operating status, and abnormal operation environment parameters.

[0094] In one implementation method, before implementation, configuration is performed to monitor changes in the operational scenario and to establish a preset threshold and a synchronization channel for failure information.

[0095] Monitoring rules: Infrared human body sensors (installed in front of the control panel) monitor whether the operator has left their post (determining "operational behavior interruption"); boiler temperature / pressure sensors track equipment status (determining "operation status switch"); workshop environmental sensors collect dust concentration (≥20mg / m³ is determined as "abnormal environmental parameters").

[0096] Preset thresholds: 10 minutes of inactivity by the operator; the criteria for determining whether the equipment is "normal to abnormal" are temperature > 180℃ (threshold) and pressure > 1.5MPa (threshold).

[0097] Failure information synchronization channel: The transmission link is built using industrial IoT technology to ensure that the delay in synchronizing failure information to the control module and database is ≤30ms.

[0098] After "Zhang San" completes the dynamic permission configuration, the system starts real-time monitoring of the operation scenario:

[0099] The infrared human body sensor collects personnel status every 2 seconds, the boiler temperature / pressure sensor transmits parameters every 5 seconds, and the environmental sensor uploads dust concentration every 10 seconds.

[0100] If "Zhang San" leaves the control panel to answer a phone call and the sensor does not detect any human activity for 10 consecutive minutes, it is determined that "operation behavior is interrupted", triggering token status verification. If the system verifies that no operation has been performed for more than 10 minutes, the temporary permission token will be invalidated immediately.

[0101] If the boiler temperature rises to 205℃ (exceeding the upper limit of 200℃ for 'mild anomaly'), the equipment will be classified as 'severe anomaly'.

[0102] Upon token expiration, the system immediately triggers token invalidation, retaining only the "emergency pressure relief operation" permission.

[0103] Transmitted to boiler control module: The control module immediately clears the operation permission scope, hides all command buttons, and pops up a message "Permission has expired, please re-verify identity and equipment status" to prohibit command input;

[0104] Transmitted to the preset database: The database records the failure details (associated personnel ID: ZS001, failure time: 2025-XX-XX10:05:20 and failure reason: operator did not operate within time limit / equipment status is severely abnormal), forming a traceable failure log.

[0105] In summary, this invention achieves highly reliable identity verification through dual biometric features, combines a trained random forest algorithm to quantify the operator's comprehensive qualification score, synchronously collects equipment operating parameters in real time to generate status tags, generates temporary permission tokens through multi-dimensional verification and dynamically adjusts the operating range, and monitors scene changes to trigger token expiration and permission iteration, forming a closed-loop control of "collection-analysis-configuration-monitoring-iteration". This achieves upgraded safety of special equipment operation, improved management efficiency, and adaptation to complex scenarios, providing technical support for intelligent management of special equipment.

[0106] refer to Figure 2 The second embodiment of the invention provides an intelligent control system for special equipment based on the Internet of Things, comprising:

[0107] Data acquisition module: Collects facial images and fingerprint data of operators through industrial IoT sensors, and matches them with preset biometric templates to obtain preliminary verified identity information;

[0108] Comprehensive evaluation module: Based on the initially verified identity information, retrieve the corresponding personnel's qualification level and historical operation records from the preset database, and use a trained random forest algorithm to conduct a comprehensive evaluation to determine the comprehensive qualification score;

[0109] Anomaly detection module: Collects special equipment operating parameters in real time through industrial IoT interface and compares them with preset safety thresholds. If the safety threshold is exceeded, the equipment is marked as abnormal and an equipment status label is obtained.

[0110] The permission allocation module uses a multi-dimensional verification mechanism to verify the authenticity of the initially verified identity information. At the same time, it extracts the corresponding operation qualification requirements from the device status tags, matches and verifies the comprehensive qualification score with the operation qualification requirements, integrates the results of authenticity verification and matching verification, and outputs the quantitative value as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied, thereby determining the basis for permission allocation.

[0111] Permission configuration module: Extracts temporary permission tokens from the permission allocation criteria, transmits them to the special equipment control module via an encryption protocol, adjusts the range of operation instructions in real time, and completes dynamic permission configuration;

[0112] The special equipment operating parameters include at least temperature and pressure values, and the temporary access token is a dynamic electronic certificate that binds personnel qualifications and equipment status, including the scope of operation and time limit.

[0113] It should be noted that the IoT-based intelligent control system for special equipment provided in this embodiment of the invention is used to execute all the process steps of the IoT-based intelligent control method for special equipment in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0114] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, or improvements made by those skilled in the art within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent control of special equipment based on the Internet of Things, characterized in that, include: The system collects facial images and fingerprint data of operators through industrial IoT sensors and matches them with preset biometric templates to obtain preliminary verified identity information. Based on the initially verified identity information, the corresponding personnel's qualification level and historical operation records are retrieved from the preset database. A trained random forest algorithm is used for comprehensive evaluation to determine the comprehensive qualification score. The operating parameters of special equipment are collected in real time through the industrial Internet of Things interface and compared with the preset safety threshold. If the safety threshold is exceeded, the equipment is marked as abnormal and the equipment status label is obtained. A multi-dimensional verification mechanism is adopted to verify the authenticity of the initially verified identity information. At the same time, the corresponding operation qualification requirements are extracted from the device status label. The comprehensive qualification score is matched and verified with the operation qualification requirements. The results of the authenticity verification and matching verification are combined and the quantitative value is output as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied. This determines the basis for permission allocation. Temporary permission tokens are extracted from the permission allocation criteria and transmitted to the special equipment control module via an encryption protocol. The range of operation instructions is adjusted in real time to complete dynamic permission configuration. After dynamic permission configuration is completed, the following steps are also included: triggering an alarm mechanism when a mismatch is detected between the operation instruction and the device status label, and simultaneously collecting real-time feedback data in the current scenario; updating historical operation records and device status labels in the preset database using the collected real-time feedback data, and synchronously adjusting the evaluation dimension weights of the comprehensive qualification score; re-collecting the operator's facial image and fingerprint data, and matching them with the biometric template to generate optimized identity information; and initiating a multi-dimensional verification mechanism to perform fusion processing on the optimized identity information and the current device status label, generating an updated temporary permission token based on the processing result, and completing the continuous iteration of dynamic permission configuration. The real-time feedback data includes the specific operation instruction content triggered by the operator, the time information of instruction sending and execution, the fluctuation data of the core operating parameters of the device corresponding to the operation instruction, the device's response status to the instruction, the state switching time, and the environmental interference parameters of the operation scenario. The special equipment operating parameters include temperature and pressure values, and the temporary access token is a dynamic electronic certificate that binds personnel qualifications and equipment status, including the scope of operation and time limit.

2. The intelligent control method for special equipment based on the Internet of Things according to claim 1, characterized in that, The process involves collecting facial images and fingerprint data of operators using industrial IoT sensors and matching them with a preset biometric template to obtain preliminary verified identity information, including: Based on the identity verification requirements before operating special equipment, the industrial IoT sensors are activated to collect facial images and fingerprint data of the operators. The facial image and fingerprint data are transmitted to the data processing module. Feature points are extracted from the acquired facial image and compared with authorized facial standard features. At the same time, features are extracted from the fingerprint data and compared with fingerprint baseline texture features. The biometric template includes authorized facial standard features and fingerprint baseline texture features. After the comparison is completed, the comparison results are analyzed. If all of them meet the preset matching criteria, the comparison is deemed successful and preliminary identity information is generated. If any comparison fails to meet the criteria, the data collection is triggered again until the preliminary identity information is obtained or the identity verification is confirmed to have failed.

3. The intelligent control method for special equipment based on the Internet of Things according to claim 1, characterized in that, The process involves retrieving the corresponding personnel's qualification level and historical operation records from a preset database based on the pre-verified identity information, and then using a trained random forest algorithm for comprehensive evaluation to determine a comprehensive qualification score, including: Based on the initially verified identity information, a targeted query is initiated into the preset database to extract the corresponding personnel's qualification level and historical operation records; The current operational requirements of the special equipment are extracted from the equipment status tags. The qualification level, historical operation records and extracted operational requirements are integrated to form a comprehensive dataset. The missing information and abnormal values ​​in the comprehensive dataset are cleaned. A trained random forest algorithm is used to perform multi-dimensional evaluation on the cleaned comprehensive dataset. By constructing a preset number of decision trees, the suitability between qualification level and operational requirements, historical operation compliance rate, and operational proficiency are evaluated, and the evaluation results are output. The decision tree is the core component of the random forest algorithm. The evaluation results of each decision tree are summarized and calculated to determine the comprehensive qualification score that reflects the overall operational ability of the corresponding personnel.

4. The intelligent control method for special equipment based on the Internet of Things according to claim 1, characterized in that, After completing the dynamic permission configuration, it also includes: Real-time monitoring of changes in the operation scenario corresponding to the temporary permission token. Once any change in the operation scenario is detected, the token status verification is automatically triggered immediately. If the operator fails to perform the operation for more than the preset time or the device changes from normal to abnormal status, the temporary permission token is immediately invalidated. When a temporary permission token expires, the token expiration information is simultaneously transmitted to the special equipment control module and the preset database. Upon receiving the expiration information, the special equipment control module immediately clears the current operation permission scope, and the preset database synchronously records the token expiration time and the reason for expiration. The changes in the operating scenarios include interruption of operator actions, switching of special equipment operating status, and abnormal operating environment parameters.

5. A special equipment intelligent control system based on the Internet of Things, characterized in that, The method for implementing the Internet of Things-based intelligent control of special equipment as described in any one of claims 1 to 4 includes: Data acquisition module: Collects facial images and fingerprint data of operators through industrial IoT sensors, and matches them with preset biometric templates to obtain preliminary verified identity information; Comprehensive evaluation module: Based on the initially verified identity information, retrieve the corresponding personnel's qualification level and historical operation records from the preset database, and use a trained random forest algorithm to conduct a comprehensive evaluation to determine the comprehensive qualification score; Anomaly detection module: Collects special equipment operating parameters in real time through industrial IoT interface and compares them with preset safety thresholds. If the safety threshold is exceeded, the equipment is marked as abnormal and an equipment status label is obtained. The permission allocation module uses a multi-dimensional verification mechanism to verify the authenticity of the initially verified identity information. At the same time, it extracts the corresponding operation qualification requirements from the device status tags, matches and verifies the comprehensive qualification score with the operation qualification requirements, integrates the results of authenticity verification and matching verification, and outputs the quantitative value as the matching degree. If the matching degree is higher than the preset matching threshold, a temporary permission token is generated; otherwise, access is denied, thereby determining the basis for permission allocation. Permission configuration module: Extracts temporary permission tokens from the permission allocation criteria, transmits them to the special equipment control module via an encryption protocol, adjusts the range of operation instructions in real time, and completes dynamic permission configuration; The special equipment operating parameters include temperature and pressure values, and the temporary access token is a dynamic electronic certificate that binds personnel qualifications and equipment status, including the scope of operation and time limit.

6. The IoT-based intelligent control system for special equipment according to claim 5, characterized in that, It also includes an optimization and update module, which includes: Alarm unit: When an operation command is detected to be mismatched with the device status label, an alarm mechanism is triggered, and real-time feedback data in the current scenario is collected at the same time; Data update unit: Updates historical operation records and equipment status tags in the preset database using real-time feedback data collected, and simultaneously adjusts the weights of the evaluation dimensions of the comprehensive qualification score; Secondary data acquisition unit: re-acquires the operator's facial image and fingerprint data, combines them with the biometric template to complete the matching, and generates optimized identity information; Secondary permission configuration unit: For the optimized identity information and the current device status label, a multi-dimensional verification mechanism is launched for fusion processing, and an updated temporary permission token is generated based on the processing result to complete the continuous iteration of dynamic permission configuration; The real-time feedback data includes the specific operation instructions triggered by the operator, the time information of instruction sending and execution, the fluctuation data of the core operating parameters of the equipment corresponding to the operation instructions, the response status of the equipment to the instructions and the state switching time, and the environmental interference parameters of the operation scenario.

7. The IoT-based intelligent control system for special equipment according to claim 5, characterized in that, It also includes a token expiration monitoring module, which comprises: Failure monitoring unit: Real-time monitoring of changes in the operation scenario corresponding to the temporary permission token. Upon detecting any change in the operation scenario, the token status verification is immediately triggered. If the operator fails to perform the operation for more than the preset time or the equipment changes from normal to abnormal state, the temporary permission token is immediately invalidated. Anomaly Response Unit: After the temporary permission token expires, the token expiration information is transmitted synchronously to the special equipment control module and the preset database. After receiving the expiration information, the special equipment control module immediately clears the current operation permission range, and the preset database synchronously records the token expiration time and expiration reason. The changes in the operating scenarios include interruption of operator actions, switching of special equipment operating status, and abnormal operating environment parameters.

Images