Maintenance and self-reliance safety information system

Abstract

The invention relates to the technical field of equipment state monitoring, in particular to a maintenance and self-reliance safety information system which comprises an equipment state monitoring module for monitoring tension of a steel wire rope and load of a lifting hook in real time, calculating current load and tension parameters, performing difference operation and deviation identification and generating a real-time equipment parameter deviation result based on temperature, vibration and rotating speed. According to the method, a matching mechanism of the personnel identity information and the equipment authority management data is adopted, the risk of misoperation is reduced, the safety of the equipment is improved, the operation parameters of the equipment are monitored and screened by adopting an isolated forest algorithm, the abnormal state can be more accurately marked, the accumulation of fault hidden dangers in the operation process of the equipment is prevented, and the safety of the equipment is improved. And the alarm mechanism is based on a deep Q network algorithm, and by judging triggering conditions of different alarm levels, the equipment can quickly enter a safety mode, so that the intelligent level of the system is improved, and equipment faults can be found and processed in time.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment status monitoring, and in particular to a maintenance and self-safety information system. Background Art

[0002] The field of equipment status monitoring technology ensures the safe and stable operation of equipment by real-time or periodic monitoring of the equipment's operating status to obtain the equipment's working parameters, operating status and potential fault information. Through real-time data collection, status analysis and automatic alarm mechanisms, it helps enterprises improve equipment operating efficiency, reduce failure rates and maintenance costs, and ensure the long-term reliability and safety of equipment.

[0003] The purpose of the maintenance and self-service safety information system is to ensure the stability, safety and efficiency of the equipment during operation, and to achieve equipment status monitoring and fault prediction through automated means, thereby reducing equipment failure rates, improving production efficiency and ensuring the safety of personnel and equipment.

[0004] Traditional systems rely on fixed authority rules, which leads to confusion in operating authority, increases the possibility of misoperation, and causes equipment failure or safety accidents. In terms of fault monitoring, traditional systems rely on logical judgments based on fixed thresholds and are unable to handle complex equipment parameters with multiple variables, resulting in underreporting or false alarms when identifying abnormal conditions. Accurate early warnings are impossible, and the lack of dynamic real-time analysis leads to the failure to predict equipment failures in advance, increasing equipment downtime and maintenance costs, and affecting production efficiency and the long-term reliability of equipment. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a maintenance and self-reliant safety information system.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: The maintenance and self-safety information system includes:

[0007] Condition monitoring module: Based on the operating parameters of the internal sensors of the equipment, the data of environmental and operating parameters are called, and the standard operating range of the equipment is compared with the real-time data, the difference value is calculated, and the equipment status is judged whether it exceeds the normal range. The abnormal parameters are marked and the equipment operating status information is generated;

[0008] Personnel identification module: Based on the camera capturing the facial information of the current operator, the facial features of the operator are matched with the data stored in the database, the identity is confirmed through the comparison results, and the conformity between the identity of the operator and the system authority is judged in combination with the current operating environment to generate the identity information of the operator;

[0009] Device authority verification module: Based on the operator's identity information, the device's authority management data is called to compare the current operation authority. Combined with the current device status, it is determined whether the operator is allowed to operate, whether the operation behavior meets the authority requirements, and the device operation authority result is generated;

[0010] Real-time operation monitoring module: Based on the equipment operation status information and the equipment operation authority results, the isolation forest algorithm is used to perform real-time comparison of equipment operation data, determine the stability of various parameters, screen abnormal states by setting thresholds, mark abnormal situations and record operation behaviors, and generate abnormal handling records;

[0011] Alarm control module: Based on the abnormal handling record, the deep Q network algorithm is used to call the alarm signal control system, determine the alarm triggering conditions, issue sound and light prompts for different alarm levels, control the operation interface of the equipment to lock, and execute equipment shutdown or safety mode to generate a safety control signal;

[0012] Status recovery module: Based on the safety control signal, the historical operation data and current status data of the equipment are called to perform fault self-check of the equipment, determine whether the equipment has returned to normal state, gradually restore the operation function of the equipment, and generate equipment recovery operation path information.

[0013] As a further solution of the present invention, the operator identity information includes personnel identity identification, authority level, and authorization validity period; the equipment operation authority result includes allowed operation status, operation authority level, and operation restriction conditions; the equipment recovery operation path information includes fault recovery steps, equipment restart sequence, and status check records.

[0014] As a further solution of the present invention, the state monitoring module includes a parameter acquisition submodule, a state calculation submodule, and an abnormal marking submodule, wherein:

[0015] Parameter collection submodule: Based on the operating parameters of the internal sensors of the equipment, the data of the environment and operating parameters are called. By acquiring real-time data from each sensor, the operating data of the equipment in the current working state is collected, and the data is synchronized to generate an operating parameter set.

[0016] State calculation submodule: based on the operating parameter set, compare each parameter with the standard operating range of the equipment, determine whether each parameter exceeds the allowable range through difference calculation, and classify and sort out the parameters exceeding the threshold to generate a state difference result;

[0017] Abnormal marking submodule: Based on the status difference results, the abnormal parameters are screened, the parameters are associated with the equipment status through the marking mechanism, and the category and occurrence time of the abnormal parameters are recorded to generate equipment operation status information.

[0018] As a further solution of the present invention, the personnel identification module includes a face capture submodule, an identity comparison submodule, and an authority verification submodule, wherein:

[0019] Facial capture submodule: Based on the device camera, the operator's facial features are captured, the facial key point information is recorded in real time through the camera, the operator's facial feature set is extracted, the integrity of all feature points is ensured and saved, and facial feature data is generated;

[0020] Identity comparison submodule: Based on the facial feature data, compare it with the facial information stored in the database, confirm the identity of the person by calculating the similarity of each feature point, mark the comparison result status, and generate the identity comparison result;

[0021] Permission verification submodule: Based on the identity comparison result, call the personnel permission data to perform permission verification under the current operating environment, judge the legitimacy of the operating permission by comparing the current permission level and operation requirements, and generate the operator identity information.

[0022] As a further solution of the present invention, the device authority verification module includes an identity authentication submodule, an authority comparison submodule, and an operation confirmation submodule, wherein:

[0023] Identity verification submodule: extracts the identity data of the operator based on the operator's identity information, confirms the validity of the current operator's identity by comparing the operator's identity information with the authorization data stored in the database, checks whether the data is consistent, and generates an identity verification result;

[0024] Permission comparison submodule: Based on the identity authentication result, the permission management data of the current device is extracted, and by comparing the operator's permission information and the device permission requirements, the permission level of the personnel is checked to see if it meets the operating standards, the permission compliance is confirmed, and the permission verification status is generated;

[0025] Operation confirmation submodule: Based on the permission verification status, extract the current operation status data of the device, determine whether the operation behavior is allowed by comparing the permission and the device operation status, verify the match between the operation permission and the device status, and generate the device operation permission result.

[0026] As a further solution of the present invention, the real-time operation monitoring module includes a status monitoring submodule, an abnormal screening submodule, and an operation recording submodule, wherein:

[0027] Status monitoring submodule: extracts the current operating parameters of the device based on the device operating status information, monitors the device status changes by comparing the parameters with the normal operating standards of the device, checks whether the parameters of the device remain within the normal range, and generates real-time status records;

[0028] Abnormal screening submodule: Based on the real-time status record and the equipment operation authority result, the isolation forest algorithm is used to screen the operation parameters of the equipment, check the abnormal parameters of the equipment through the set threshold, extract the status data beyond the normal range, screen and mark the abnormal status, and generate abnormal screening results;

[0029] Operation record submodule: based on the abnormal screening results, record the operation behavior of the equipment, mark the time of occurrence of the abnormal operation by timestamp, record the information of related operation behavior, and generate abnormal handling record.

[0030] As a further solution of the present invention, the isolation forest algorithm is according to the formula:

[0031]

[0032] Where: S′(x) is the improved anomaly score, h(x) is the path length from data point x to the leaf node in the isolated tree, C(n) is the average path length adjustment factor of sample size n, P(k) is the equipment operation authority weight, D(t) is the equipment operation time dimension, M(v) is the fluctuation range of equipment operation parameters, and w1 and w2 are weight coefficients.

[0033] As a further solution of the present invention, the alarm control module includes an alarm triggering submodule, a signal sending submodule, and a device control submodule, wherein:

[0034] Alarm trigger submodule: based on the abnormal handling record, extract abnormal data, check whether the alarm trigger conditions are met by comparing the equipment operation status with the preset alarm conditions item by item, determine the alarm level according to the equipment status information, and generate alarm trigger information;

[0035] Signal sending submodule: based on the alarm trigger information, by activating the sound and light alarm device, selecting the corresponding alarm form, sending the sound and light signal corresponding to the alarm level, and locking the device operation interface, generating alarm signal data;

[0036] Equipment control submodule: Based on the alarm signal data, the deep Q network algorithm is used to call the equipment control system, lock the equipment operation authority, and generate a safety control signal through the control operation of shutting down or switching to a safe mode.

[0037] As a further solution of the present invention, the deep Q network algorithm is according to the formula:

[0038] Q(s,a)=r+γ·max a′ [Q(s′,a′)·α+β·η(s,a)+θ·T(s,a)]

[0039] Where: Q(s,a) is the expected return value, r is the immediate return signal, γ is the discount factor, α is the weight coefficient, β is the equipment operation risk coefficient, η(s,a) is the environmental interference coefficient, θ is the execution time weight factor, and T(s,a) is the operation time consumption.

[0040] As a further solution of the present invention, the state recovery module includes a fault self-detection submodule, a state judgment submodule, and an operation recovery submodule, wherein:

[0041] Fault self-check submodule: Based on the safety control signal, the historical operation data and current status data of the equipment are extracted, and the difference between the current equipment status and the historical data is analyzed by comparing each equipment parameter item by item, so as to check whether there is any hidden fault in the equipment and generate the fault self-check result;

[0042] Status judgment submodule: Based on the fault self-check result, analyze the current device status, check whether the device has been restored to the standard operating state by judging each key parameter of the device item by item, and generate a status recovery judgment result;

[0043] Operation recovery submodule: based on the state recovery judgment result, gradually recover the operation function of the device, by gradually releasing the locked state of the device, reactivating the operation authority and operation interface of the device, and generating device recovery operation path information.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, a matching mechanism between personnel identity information and equipment authority management data is adopted to reduce the risk of misoperation and improve the safety of equipment. The isolation forest algorithm is used to monitor and screen equipment operating parameters, which can more accurately mark abnormal conditions, prevent potential faults from accumulating during equipment operation, and ensure the stability of equipment. The alarm mechanism is based on a deep Q network algorithm. By judging the trigger conditions of different alarm levels, the accuracy and timeliness of the alarm are improved, so that the equipment can quickly enter the safe mode, improve the intelligence level of the system, ensure that equipment failures can be discovered and handled in a timely manner, reduce equipment downtime, and improve production efficiency and long-term stability of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a system flow chart of the present invention;

[0047] Figure 2 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0049] Embodiment 1

[0050] See also Figure 1 The present invention provides a technical solution: the maintenance and self-reliance safety information system includes:

[0051] Condition monitoring module: Based on the operating parameters of the internal sensors of the equipment, the data of environmental and operating parameters are called, and the standard operating range of the equipment is compared with the real-time data, the difference value is calculated, and the equipment status is judged whether it exceeds the normal range. The abnormal parameters are marked and the equipment operating status information is generated;

[0052] Personnel identification module: Based on the camera capturing the facial information of the current operator, the facial features of the operator are matched with the data stored in the database, the identity is confirmed through the comparison results, and the conformity between the identity of the operator and the system authority is judged in combination with the current operating environment to generate the identity information of the operator;

[0053] Device authority verification module: Based on the operator's identity information, the device's authority management data is called to compare the current operation authority. Combined with the current device status, it is determined whether the operator is allowed to operate, whether the operation behavior meets the authority requirements, and the device operation authority result is generated;

[0054] Real-time operation monitoring module: Based on the equipment operation status information and the equipment operation authority results, the isolation forest algorithm is used to perform real-time comparison of equipment operation data, determine the stability of various parameters, screen abnormal states by setting thresholds, mark abnormal situations and record operation behaviors, and generate abnormal handling records;

[0055] Alarm control module: Based on the abnormal handling record, the deep Q network algorithm is used to call the alarm signal control system, determine the alarm triggering conditions, issue sound and light prompts for different alarm levels, control the operation interface of the equipment to lock, and execute equipment shutdown or safety mode to generate a safety control signal;

[0056] Status recovery module: Based on the safety control signal, the historical operation data and current status data of the equipment are called to perform fault self-check of the equipment, determine whether the equipment has returned to normal state, gradually restore the operation function of the equipment, and generate equipment recovery operation path information.

[0057] The operator identity information includes personnel identity, authority level, and authorization validity period; the device operation authority result includes allowed operation status, operation authority level, and operation restriction conditions; the device recovery operation path information includes fault recovery steps, device restart sequence, and status check records.

[0058] See also Figure 2 The state monitoring module includes a parameter acquisition submodule, a state calculation submodule, and an abnormal marking submodule, wherein:

[0059] Parameter collection submodule: Based on the operating parameters of the internal sensors of the equipment, the data of the environment and operating parameters are called. By acquiring real-time data from each sensor, the operating data of the equipment in the current working state is collected, and the data is synchronized to generate an operating parameter set.

[0060] State calculation submodule: based on the operating parameter set, compare each parameter with the standard operating range of the equipment, determine whether each parameter exceeds the allowable range through difference calculation, and classify and sort out the parameters exceeding the threshold to generate a state difference result;

[0061] Abnormal marking submodule: based on the status difference results, screen abnormal parameters, associate parameters with equipment status through a marking mechanism, record the category and occurrence time of abnormal parameters, and generate equipment operation status information;

[0062] Parameter acquisition submodule: Based on the operating parameters of the internal sensors of the equipment, the Modbus protocol is used for communication control. The function code 0x03 is sent to read multiple data parameters such as temperature, pressure, and speed, and the data is stored in a 16-bit register. After the call is completed, the collected data is packaged and transmitted to the server using the TCP / IP protocol for synchronous data update. At the same time, the synchronization flag in the synchronization mechanism is called to confirm the integrity of the data transmission, and the data timestamp is uniformly corrected to generate an operating parameter set.

[0063] State calculation submodule: Based on the operating parameter set, using the difference calculation method, using Boolean operations to perform logical judgment on each parameter, by calling the upper and lower limit threshold parameters in the standard operating range of the equipment, to make a judgment of exceeding the range, calling the actual operating value in the parameter set in the judgment operation and comparing it with the standard threshold bit by bit, if it exceeds the upper limit, the overflow flag is triggered, if it is lower than the lower limit, the underflow flag is triggered, and the triggered abnormal state is recorded through bitwise operation, and the parameters exceeding the threshold are binary coded according to the type of parameters to generate state difference results;

[0064] Abnormal marking submodule: Based on the status difference result, a hash mapping mechanism is used to perform parameter marking processing, a unique identifier is generated for the abnormal parameter through the CRC32 check algorithm, a bit offset processing is performed on the abnormal parameter to record the starting address and category of the parameter, and the abnormal parameter is associated with the equipment operation status through a chained hash table. The timestamp data in the equipment status table is called in the marking operation to record the time when the abnormality occurred. At the same time, a classified hash table is used to hash the parameter category and store it in the abnormal parameter table to generate equipment operation status information.

[0065] See also Figure 2 The personnel identification module includes a face capture submodule, an identity comparison submodule, and an authority verification submodule, wherein:

[0066] Facial capture submodule: Based on the device camera, the operator's facial features are captured, the facial key point information is recorded in real time through the camera, the operator's facial feature set is extracted, the integrity of all feature points is ensured and saved, and facial feature data is generated;

[0067] Identity comparison submodule: Based on the facial feature data, compare it with the facial information stored in the database, confirm the identity of the person by calculating the similarity of each feature point, mark the comparison result status, and generate the identity comparison result;

[0068] Permission verification submodule: Based on the identity comparison result, the personnel permission data is called to perform permission verification under the current operating environment. By comparing the current permission level and the operation requirements, the legitimacy of the operation permission is judged and the operator identity information is generated;

[0069] Facial capture submodule: Based on the device camera, the facial key point detection algorithm is used to extract features through the three-level networks of P-Net, R-Net, and O-Net in sequence. At the same time, the facial information of the operator is captured in real time through the camera, and the P-Net network is called to generate rough candidate face frames, and non-maximum suppression is performed to remove redundant frames. Then, the candidate frames are refined through R-Net to determine the facial area, and then the facial key points are located through O-Net. The key point coordinates are extracted through bilinear interpolation, and all facial key point information is recorded to generate facial feature data;

[0070] Identity matching submodule: Based on the facial feature data, a similarity comparison is performed using the Euclidean distance calculation algorithm. First, the facial features stored in the database are normalized, and the coordinates of each feature point are normalized using the L2 norm. The obtained coordinates are matched with the currently captured facial features, and the distance between the current feature point and the feature point stored in the database is calculated. By setting a similarity threshold, feature points with a distance less than the preset threshold are considered to be successfully matched. At the same time, a hash algorithm is called during the comparison process to generate a unique identifier to record the comparison result status, and finally an identity matching result is generated;

[0071] Permission verification submodule: Based on the identity comparison result, the personnel permission data is called, and the current operation permission is verified through the RBAC permission control model. First, the permission level data of the operator's identity is called, and Boolean operations are used to determine whether the current operation permission is lower than the operation requirements. If the conditions are met, the system's built-in permission mapping table is called, and the operation requirements and personnel permissions are compared through the permission mapping function. The verification results are recorded in the hash table to generate the operator's identity information.

[0072] See also Figure 2 The device authority verification module includes an identity authentication submodule, an authority comparison submodule, and an operation confirmation submodule, wherein:

[0073] Identity verification submodule: extracts the identity data of the operator based on the operator's identity information, confirms the validity of the current operator's identity by comparing the operator's identity information with the authorization data stored in the database, checks whether the data is consistent, and generates an identity verification result;

[0074] Permission comparison submodule: Based on the identity authentication result, the permission management data of the current device is extracted, and by comparing the operator's permission information and the device permission requirements, the permission level of the personnel is checked to see if it meets the operating standards, the permission compliance is confirmed, and the permission verification status is generated;

[0075] Operation confirmation submodule: based on the permission verification status, extract the current operation status data of the device, determine whether the operation behavior is allowed by comparing the permission and the device operation status, verify the match between the operation permission and the device status, and generate the device operation permission result;

[0076] Identity verification submodule: Based on the operator's identity information, the operator's identity data is encrypted using the SHA-256 hash algorithm, a hash value of the operator's identity data is generated by calling a hash function, and the hash value is compared with the authorization data stored in the database. During the comparison process, the two are confirmed to be consistent by bit-by-bit matching. If the match is successful, the corresponding authorization information in the database is extracted, the integrity and consistency of the identity data are checked, the comparison status during the identity verification process is recorded, and the identity verification result is generated;

[0077] Permission comparison submodule: Based on the identity authentication result, the permission management module is called, and the permission management data of the current device is extracted by using the role-based access control model. The extracted operator permission information is gradually compared with the device permission requirements. First, the permission level function is called to calculate the permission level of the current operator, and the permission mapping table is used to confirm whether the permission meets the requirements of the current operating device. The permission level parameters are compared to check whether the permission level of the personnel is lower than or equal to the device permission requirements. The comparison result status is recorded by calling the verification function to generate the permission verification status;

[0078] Operation confirmation submodule: Based on the permission verification status, call the data interface of the device status module, extract the operating status data of the current device, use the logical judgment algorithm to perform a secondary match between the permission verification status and the current status of the device, check whether the permission and device status meet the operation conditions, and judge through Boolean values. If the conditions are met, call the permission confirmation function to record the operation permission, generate the device operation permission identifier, and generate the device operation permission result.

[0079] See also Figure 2 The real-time operation monitoring module includes a status monitoring submodule, an abnormal screening submodule, and an operation recording submodule, wherein:

[0080] Status monitoring submodule: extracts the current operating parameters of the device based on the device operating status information, monitors the device status changes by comparing the parameters with the normal operating standards of the device, checks whether the parameters of the device remain within the normal range, and generates real-time status records;

[0081] Abnormal screening submodule: Based on the real-time status record and the equipment operation authority result, the isolation forest algorithm is used to screen the operation parameters of the equipment, check the abnormal parameters of the equipment through the set threshold, extract the status data beyond the normal range, screen and mark the abnormal status, and generate abnormal screening results;

[0082] Operation record submodule: based on the abnormal screening results, record the operation behavior of the equipment, mark the time of occurrence of the abnormal operation by timestamp, record the information of related operation behavior, and generate abnormal handling record;

[0083] Status monitoring submodule: Based on the equipment operation status information, the PID control algorithm is used to monitor the operation parameters of the current equipment in real time, extract the key variables in the equipment operation parameters, and then call the proportional, integral, and differential coefficients in the PID controller, which are Kp, Ki, and Kd respectively, to calculate the error between the current parameters and the normal operation standard of the equipment, and adjust the state through the PID controller, call the recording function to synchronize the difference between the equipment parameters and the set point, and generate real-time status records;

[0084] Abnormal screening submodule: Based on the real-time status record and the equipment operation authority result, the isolation forest algorithm is used to screen the abnormal operation parameters of the equipment. First, the number of decision trees of the isolation forest is set to 100, and the normal state data in the training set is used to build an isolation forest model. By randomly selecting samples and performing spatial division, the equipment parameters in the abnormal state are compared with the operation parameters in the normal state. The isolation forest algorithm calculates the abnormal score of the equipment operation parameters, screens and marks the parameters that exceed the normal range by setting thresholds, and calls the marking function to record the category and time of the abnormal state to generate abnormal screening results;

[0085] Operation record submodule: Based on the exception screening results, the device operation behavior log module is called to mark the occurrence time of the abnormal operation. First, the event ID, operator information and exception category in the abnormal operation behavior are extracted, and the insert command in the SQL database is called to write the above data into the operation log table. The database transaction processing function is called to perform data consistency check and submission, and finally an exception handling record is generated.

[0086] See also Figure 2 , the isolation forest algorithm, according to the formula:

[0087]

[0088] Where: S′(x) is the improved anomaly score, h(x) is the path length from data point x to the leaf node in the isolated tree, C(n) is the average path length adjustment factor of sample size n, P(k) is the equipment operation authority weight, D(t) is the equipment operation time dimension, M(v) is the fluctuation range of equipment operation parameters, w1 and w2 are weight coefficients;

[0089] Execution process: Calculate the path length h(x) of the data point x in the isolated tree to measure whether the equipment is operating normally. Calculate the average path length adjustment factor C(n) based on the sample size n to normalize the path length under different sample sizes to make the scores comparable. Introduce the equipment operation authority weight P(k). By analyzing the authority level of the equipment, ensure that it has a higher sensitivity for abnormal detection in important equipment operations. The equipment operation time dimension D(t) is used to reflect the impact of equipment operation time on abnormal detection. Equipment that has been running stably for a long time has a lower false alarm rate in detection. It is dynamically adjusted based on the changes in the equipment's operating cycle. Calculate the fluctuation range M(v) of the equipment operation parameters. By analyzing the fluctuation of equipment parameters, determine whether the equipment is running stably. The greater the fluctuation, the higher the possibility of equipment abnormality. Combined with the weight coefficients w1 and w2, adjust the impact of path length related parameters and fluctuation range parameters on the final abnormal score respectively. The weight coefficient is obtained through historical data optimization to ensure the rationality and accuracy of the score. Score the equipment's operating parameters and generate an abnormal status mark value.

[0090] See also Figure 2 The alarm control module includes an alarm triggering submodule, a signal sending submodule, and an equipment control submodule, wherein:

[0091] Alarm trigger submodule: based on the abnormal handling record, extract abnormal data, check whether the alarm trigger conditions are met by comparing the equipment operation status with the preset alarm conditions item by item, determine the alarm level according to the equipment status information, and generate alarm trigger information;

[0092] Signal sending submodule: based on the alarm trigger information, by activating the sound and light alarm device, selecting the corresponding alarm form, sending the sound and light signal corresponding to the alarm level, and locking the device operation interface, generating alarm signal data;

[0093] Equipment control submodule: Based on the alarm signal data, the deep Q network algorithm is used to call the equipment control system, lock the equipment operation authority, and generate a safety control signal through the control operation of shutting down or switching to a safe mode;

[0094] Alarm trigger submodule: Based on the abnormal handling record, the abnormal data is extracted, and the Boolean logic judgment algorithm is used to compare the equipment operation status with the preset alarm conditions item by item. First, the key operation parameters of the equipment and the preset alarm threshold are extracted, and each parameter is checked to see whether it exceeds the alarm trigger condition, and matched with the preset value range one by one. For parameters exceeding the threshold, the mapping function is used to call the abnormal type and alarm level mapping table. By comparing the threshold exceeding conditions corresponding to different parameters and the overall status of the equipment, the severity of the alarm is determined. Subsequently, the result marking function is called according to the judgment result to record the alarm category and level, and the alarm trigger information is generated according to the preset alarm response rules, and the alarm trigger information is input into the system database for subsequent handling reference;

[0095] Signal sending submodule: Based on the alarm trigger information, by calling the sound and light alarm device interface, the activation state of the alarm device is controlled by the GPIO port. The alarm level information is first extracted. According to the alarm level and the sound and light signal activation rules, the form of activating the sound and light device is determined by the logic judgment algorithm, including the color of the signal light. Red indicates a major warning, yellow indicates a general warning, and green indicates a safe state. A continuous beep indicates an emergency state, and an intermittent beep indicates a secondary warning. The sound and light signal type is set through the controller interface function, and the intensity of the sound and light signal is adjusted according to the settings of different alarm levels, including the buzzer volume and the flashing light frequency. The LCD control interface is called to lock the device operation interface, and the corresponding alarm information and disposal suggestions are displayed on the screen to ensure that the operator obtains complete abnormal information signals at the first time;

[0096] Equipment control submodule: Based on the alarm signal data, a deep Q network algorithm is used to analyze the equipment status and make control decisions. By calling the value function estimation module in the deep Q network, the value of different operation instructions is calculated according to the current equipment status and alarm information, and the optimal action is selected according to the principle of maximizing value. After determining to shut down or switch to safe mode, the corresponding operation is executed by calling the equipment control system interface function, and the operation authority is locked to prevent external interference during the operation. At the same time, a variety of safety control signals are generated according to the level and category of the alarm signal, and synchronized with the system database to ensure the integrity of the operation records in the subsequent disposal process.

[0097] See also Figure 2 , the deep Q network algorithm, according to the formula:

[0098] Q(s,a)=r+γ·max a′ [Q(s′,a′)·α+β·η(s,a)+θ·T(s,a)]

[0099] Where: Q(s,a) is the expected return value, r is the immediate return signal, γ is the discount factor, α is the weight coefficient, β is the equipment operation risk coefficient, η(s,a) is the environmental interference coefficient, θ is the execution time weight factor, and T(s,a) is the operation time consumption;

[0100] Execution process: Enter the current state s and receive the immediate return signal r, which is generated by the equipment alarm data and represents the direct feedback of the current operation on the equipment safety. The weight of the future return is adjusted by the discount factor γ to ensure that both immediate safety and long-term system stability are considered. By searching for the optimal operation a′ and the next state s′, the maximum expected return obtained by executing action a′ in state s′ is calculated. The weight coefficient α is combined to measure the contribution of the current expected return. The equipment operation risk coefficient β is calculated by evaluating the equipment's historical data and failure rate and other factors to reflect the risk level of the current operation and adjust the return value. The environmental interference coefficient η(s,a) is introduced to represent the impact of the external environment on the equipment operation. The external environmental conditions are monitored by sensors to ensure that external factors are considered during the equipment operation process. The time consumption T(s,a) of the equipment operation is included in the formula. The execution time weight factor θ is used to measure the impact of the operation time on the system safety signal. The time consumption is calculated through the real-time feedback of the equipment and added to the formula to generate a comprehensive safety control signal.

[0101] See also Figure 2 The state recovery module includes a fault self-detection submodule, a state judgment submodule, and an operation recovery submodule, wherein:

[0102] Fault self-check submodule: Based on the safety control signal, the historical operation data and current status data of the equipment are extracted, and the difference between the current equipment status and the historical data is analyzed by comparing each equipment parameter item by item, so as to check whether there is any hidden fault in the equipment and generate the fault self-check result;

[0103] Status judgment submodule: Based on the fault self-check result, analyze the current device status, check whether the device has been restored to the standard operating state by judging each key parameter of the device item by item, and generate a status recovery judgment result;

[0104] Operation recovery submodule: based on the state recovery judgment result, gradually recover the operation function of the device, by gradually unlocking the locked state of the device, reactivating the operation authority and operation interface of the device, and generating device recovery operation path information;

[0105] Fault self-check submodule: Based on the safety control signal, the historical operation data and current state data of the equipment are extracted, and the dynamic time warping algorithm is used to compare each equipment parameter item by item. By calculating the minimum alignment path between the current equipment parameter sequence and the historical operation data sequence, the distance of each parameter point is calculated in turn and cumulative minimization processing is performed to determine the difference between the current state and the historical operation data, and the state analysis function is called to mark the parameters with differences, and the fault self-check results are generated;

[0106] State judgment submodule: Based on the fault self-check results, the device state is judged by using a logistic regression algorithm. First, the key parameters of the device are extracted, and the model is trained through the known normal and fault state data in the training set. The binary classifier is called to judge the device parameter state item by item, and check whether each device parameter has recovered to the standard operating range. The state marking function is called to record the judgment result, and the state recovery judgment result is generated;

[0107] Operation recovery submodule: Based on the status recovery judgment result, call the device control module to gradually release the locked state of the device. First, call the operation permission management function to restore the device's access permission, activate the operation interface, unlock key operation functions through the device operation control interface, call the status synchronization function to synchronize the current state of the device, record the steps of restoring each function, and generate device recovery operation path information.

[0108] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. Maintenance and independent safety information system, characterized by: The system comprises: Condition monitoring module: Based on the operating parameters of the internal sensors of the equipment, the data of environmental and operating parameters are called, and the standard operating range of the equipment is compared with the real-time data, the difference value is calculated, and the equipment status is judged whether it exceeds the normal range. The abnormal parameters are marked and the equipment operating status information is generated; Personnel identification module: Based on the camera capturing the facial information of the current operator, the facial features of the operator are matched with the data stored in the database, the identity is confirmed through the comparison results, and the conformity between the identity of the operator and the system authority is judged in combination with the current operating environment to generate the identity information of the operator; Device authority verification module: Based on the operator's identity information, the device's authority management data is called to compare the current operation authority. Combined with the current device status, it is determined whether the operator is allowed to operate, whether the operation behavior meets the authority requirements, and the device operation authority result is generated; Real-time operation monitoring module: Based on the equipment operation status information and the equipment operation authority results, the isolation forest algorithm is used to perform real-time comparison of equipment operation data, determine the stability of various parameters, screen abnormal states by setting thresholds, mark abnormal situations and record operation behaviors, and generate abnormal handling records; Alarm control module: Based on the abnormal handling record, the deep Q network algorithm is used to call the alarm signal control system, determine the alarm triggering conditions, issue sound and light prompts for different alarm levels, control the operation interface of the equipment to lock, and execute equipment shutdown or safety mode to generate a safety control signal; Status recovery module: Based on the safety control signal, the historical operation data and current status data of the equipment are called to perform fault self-check of the equipment, determine whether the equipment has returned to normal state, gradually restore the operation function of the equipment, and generate equipment recovery operation path information.

2. The maintenance and self-safety information system according to claim 1, characterized in that: The operator identity information includes personnel identity, authority level, and authorization validity period; the device operation authority result includes allowed operation status, operation authority level, and operation restriction conditions; the device recovery operation path information includes fault recovery steps, device restart sequence, and status check records.

3. The maintenance and self-safety information system according to claim 1, characterized in that: The state monitoring module includes a parameter acquisition submodule, a state calculation submodule, and an abnormal marking submodule, wherein: Parameter collection submodule: Based on the operating parameters of the internal sensors of the equipment, the data of the environment and operating parameters are called. By acquiring real-time data from each sensor, the operating data of the equipment in the current working state is collected, and the data is synchronized to generate an operating parameter set. State calculation submodule: based on the operating parameter set, compare each parameter with the standard operating range of the equipment, determine whether each parameter exceeds the allowable range through difference calculation, and classify and sort out the parameters exceeding the threshold to generate a state difference result; Abnormal marking submodule: Based on the status difference results, the abnormal parameters are screened, the parameters are associated with the equipment status through the marking mechanism, and the category and occurrence time of the abnormal parameters are recorded to generate equipment operation status information.

4. The maintenance and self-safety information system according to claim 1, characterized in that: The personnel identification module includes a face capture submodule, an identity comparison submodule, and an authority verification submodule, wherein: Facial capture submodule: Based on the device camera, the operator's facial features are captured, the facial key point information is recorded in real time through the camera, the operator's facial feature set is extracted, the integrity of all feature points is ensured and saved, and facial feature data is generated; Identity comparison submodule: Based on the facial feature data, compare it with the facial information stored in the database, confirm the identity of the person by calculating the similarity of each feature point, mark the comparison result status, and generate the identity comparison result; Permission verification submodule: Based on the identity comparison result, call the personnel permission data to perform permission verification under the current operating environment, judge the legitimacy of the operating permission by comparing the current permission level and operation requirements, and generate the operator identity information.

5. The maintenance and self-safety information system according to claim 1, characterized in that: The device authority verification module includes an identity authentication submodule, an authority comparison submodule, and an operation confirmation submodule, wherein: Identity verification submodule: extracts the identity data of the operator based on the operator's identity information, confirms the validity of the current operator's identity by comparing the operator's identity information with the authorization data stored in the database, checks whether the data is consistent, and generates an identity verification result; Permission comparison submodule: Based on the identity authentication result, the permission management data of the current device is extracted, and by comparing the operator's permission information and the device permission requirements, the permission level of the personnel is checked to see if it meets the operating standards, the permission compliance is confirmed, and the permission verification status is generated; Operation confirmation submodule: Based on the permission verification status, extract the current operation status data of the device, determine whether the operation behavior is allowed by comparing the permission and the device operation status, verify the match between the operation permission and the device status, and generate the device operation permission result.

6. The maintenance and self-safety information system according to claim 1, characterized in that: The real-time operation monitoring module includes a status monitoring submodule, an abnormal screening submodule, and an operation recording submodule, wherein: Status monitoring submodule: extracts the current operating parameters of the device based on the device operating status information, monitors the device status changes by comparing the parameters with the normal operating standards of the device, checks whether the parameters of the device remain within the normal range, and generates real-time status records; Abnormal screening submodule: Based on the real-time status record and the equipment operation authority result, the isolation forest algorithm is used to screen the operation parameters of the equipment, check the abnormal parameters of the equipment through the set threshold, extract the status data beyond the normal range, screen and mark the abnormal status, and generate abnormal screening results; Operation record submodule: based on the abnormal screening results, record the operation behavior of the equipment, mark the time of occurrence of the abnormal operation by timestamp, record the information of related operation behavior, and generate abnormal handling record.

7. The maintenance and self-safety information system according to claim 1, characterized in that: The isolation forest algorithm is based on the formula: Where: S′(x) is the improved anomaly score, h(x) is the path length from data point x to the leaf node in the isolated tree, C(n) is the average path length adjustment factor of sample size n, P(k) is the equipment operation authority weight, D(t) is the equipment operation time dimension, M(v) is the fluctuation range of equipment operation parameters, and w1 and w2 are weight coefficients.

8. The maintenance and self-safety information system according to claim 1, characterized in that: The alarm control module includes an alarm triggering submodule, a signal sending submodule, and an equipment control submodule, wherein: Alarm trigger submodule: based on the abnormal handling record, extract abnormal data, check whether the alarm trigger conditions are met by comparing the equipment operation status with the preset alarm conditions item by item, determine the alarm level according to the equipment status information, and generate alarm trigger information; Signal sending submodule: based on the alarm trigger information, by activating the sound and light alarm device, selecting the corresponding alarm form, sending the sound and light signal corresponding to the alarm level, and locking the device operation interface, generating alarm signal data; Equipment control submodule: Based on the alarm signal data, the deep Q network algorithm is used to call the equipment control system, lock the equipment operation authority, and generate a safety control signal through the control operation of shutting down or switching to a safe mode.

9. The maintenance and self-safety information system according to claim 1, characterized in that: The deep Q network algorithm is based on the formula: Q(s,a)=r+γ·max a′ [Q(s′,a′)·α+β·η(s,a)+θ·T(s,a)] Where: Q(s,a) is the expected return value, r is the immediate return signal, γ is the discount factor, α is the weight coefficient, β is the equipment operation risk coefficient, η(s,a) is the environmental interference coefficient, θ is the execution time weight factor, and T(s,a) is the operation time consumption.

10. The maintenance and self-safety information system according to claim 1, characterized in that: The state recovery module includes a fault self-detection submodule, a state judgment submodule, and an operation recovery submodule, wherein: Fault self-check submodule: Based on the safety control signal, the historical operation data and current status data of the equipment are extracted, and the difference between the current equipment status and the historical data is analyzed by comparing each equipment parameter item by item, so as to check whether there is any hidden fault in the equipment and generate the fault self-check result; Status judgment submodule: Based on the fault self-check result, analyze the current device status, check whether the device has been restored to the standard operating state by judging each key parameter of the device item by item, and generate a status recovery judgment result; Operation recovery submodule: based on the state recovery judgment result, gradually recover the operation function of the device, by gradually releasing the locked state of the device, reactivating the operation authority and operation interface of the device, and generating device recovery operation path information.

Images