Offshore wind power operation and maintenance personnel safety closed-loop control method, system and electronic equipment

Abstract

The present application belongs to the field of offshore wind power operation and maintenance personnel safety monitoring, and specifically relates to a kind of offshore wind power operation and maintenance personnel safety closed-loop control method, system and electronic equipment, comprising the following steps: S1, operation and maintenance personnel safety multi-source data acquisition and preprocessing;S2, operation and maintenance personnel state data real-time reconstruction based on communication state perception;S3, operation and maintenance personnel safety risk assessment based on fuzzy comprehensive evaluation;S4, operation and maintenance personnel safety control strategy generation;S5, operation and maintenance personnel safety closed-loop control based on man-machine collaborative feedback.The present application solves the problem of unreliable operation and maintenance personnel state data and inaccurate feedback information under unstable communication conditions of offshore wind power operation by real-time reconstruction of operation and maintenance personnel state data and man-machine collaborative feedback mechanism, resulting in poor closed-loop control effect.The present application can solve the problem of unreliable operation and maintenance personnel state data and inaccurate feedback information under unstable communication conditions of offshore wind power operation, resulting in poor closed-loop control effect.

Description

Technical Field

[0001] This invention belongs to the field of safety monitoring for offshore wind power operation and maintenance personnel, and specifically relates to a closed-loop management method, system and electronic equipment for the safety of offshore wind power operation and maintenance personnel. Background Technology

[0002] With the continuous expansion of offshore wind power installed capacity, offshore wind farms are increasingly developing towards deeper waters and larger capacities. Operation and maintenance (O&M) operations are characterized by complex environments, wide work areas, high personnel mobility, and high risks. O&M personnel are easily affected by a combination of factors, including environmental conditions, turbine operating status, and their own physiological and behavioral states, during offshore boarding, equipment inspections, and hoisting operations, resulting in significant safety risks. Therefore, establishing an advanced safety management system for O&M personnel is of great importance for improving the safety management level of offshore wind power O&M personnel.

[0003] Currently, offshore wind power operation and maintenance personnel generally adopt an open-loop safety management model. This involves first monitoring personnel status and the working environment using wearable devices, video surveillance, and environmental sensors. Then, risk assessments are conducted based on the monitoring data. When abnormal conditions or risk events are detected, corresponding safety management strategies, such as early warnings, are implemented. This model lacks a closed-loop feedback mechanism, particularly reliable feedback based on on-site confirmation by operation and maintenance personnel. It cannot fully incorporate the personnel's own judgment for correction, making reliable iterative optimization of safety management strategies difficult. Furthermore, due to the distance between offshore wind farms and land, and insufficient base station coverage, unstable communication issues such as data transmission latency and packet loss exist, making it impossible to obtain real-time data on the status of operation and maintenance personnel in a timely manner, further reducing the effectiveness of safety management.

[0004] In summary, existing methods lack reliable feedback information, making it difficult to meet the safety closed-loop management needs of offshore wind power operation and maintenance personnel. Therefore, there is an urgent need to provide a method, system, and electronic equipment for safety closed-loop management of offshore wind power operation and maintenance personnel under unstable communication conditions. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, and electronic equipment for closed-loop safety management of offshore wind power operation and maintenance personnel, in order to solve the problems of unreliable status data of operation and maintenance personnel and inaccurate feedback information caused by unstable communication conditions in offshore wind power operations, resulting in poor closed-loop management effect.

[0006] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0007] The purpose of this invention is to provide a method and system for closed-loop safety management of maintenance personnel under unstable communication conditions in offshore wind power operations. By real-time reconstruction of maintenance personnel status data and a human-machine collaborative feedback mechanism, this invention addresses the problems of unreliable maintenance personnel status data and inaccurate feedback information leading to poor closed-loop management under unstable communication conditions in offshore wind power operations.

[0008] To achieve the above objectives, the present invention employs the following technical solution:

[0009] In a first aspect, the present invention provides a closed-loop management method for the safety of offshore wind power operation and maintenance personnel, comprising the following steps:

[0010] S1. Collect maintenance personnel status data, maintenance task data, maintenance environment data, and unit operation status data to form raw multi-source data. Perform missing value completion processing, outlier correction processing, and standardization processing on the raw multi-source data, and output standardized multi-source data.

[0011] S2. Based on the operation and maintenance personnel status data in the standardized multi-source data, monitor the real-time communication status and make a status determination. Adaptively process and reconstruct the operation and maintenance personnel status data according to the communication status, and output valid operation and maintenance personnel status data.

[0012] S3. Based on the effective maintenance personnel status data and the maintenance task data, maintenance environment data, and unit operation status data in the standardized multi-source data, perform data fusion processing at a unified time scale to generate multi-source fused data for the current moment. According to the weight vector, safety risk level set, and membership function, perform single-factor fuzzy evaluation, construct fuzzy evaluation matrix, and comprehensive calculation processing in sequence. Determine the safety risk level through the maximum membership principle and output the real-time maintenance personnel safety risk level signal.

[0013] S4. Based on the real-time security risk level signal of maintenance personnel, retrieve the security management strategy library for maintenance personnel, perform strategy matching processing, and output a security management strategy that is adapted to the current risk.

[0014] S5. Execute the security control strategy and, based on the risk feedback information after the strategy execution, iteratively correct the risk assessment-related parameters and the security control strategy library to achieve closed-loop security control throughout the entire process.

[0015] Optionally, in step S1, the original multi-source data undergoes missing value completion, outlier correction, and standardization, and standardized multi-source data is output, specifically as follows:

[0016] Missing value detection is performed on the collected raw multi-source data. When a missing value is detected, the time series data before and after the missing value is retrieved, and missing value completion processing is performed through a preset missing value imputation model to obtain the completed multi-source data.

[0017] Anomaly detection is performed on the completed multi-source data. When an anomaly is detected, the time-series data before and after the anomaly is obtained, and anomaly data correction processing is performed through a preset smoothing processing model to obtain corrected compliant data.

[0018] The corrected compliant data is standardized and normalized using the Z-score algorithm to output the standardized multi-source data.

[0019] Optionally, the specific processing flow of S2 is as follows:

[0020] S21. Based on the transmission process of the maintenance personnel status data in the communication network, calculate the data transmission delay at the current moment according to the timestamp information of the data sending end and receiving end within the preset historical time window, and calculate the packet loss rate at the current moment according to the number of packets of the data sending end and receiving end within the preset historical time window.

[0021] S22. Based on the comparison between the data transmission delay and data packet loss rate and the preset communication status threshold, determine whether there is an abnormality in the current communication status.

[0022] S23. If the communication status is normal, the operation and maintenance personnel status data is forwarded directly. If the communication status is abnormal, the operation and maintenance personnel status data at the current moment is reconstructed by combining the historical operation and maintenance personnel status data and communication status information through the long short-term memory network, and the reconstructed operation and maintenance personnel status data is output.

[0023] Optionally, the specific processing flow of S3 is as follows:

[0024] S31. Based on the effective maintenance personnel status data and the maintenance operation task data, maintenance operation environment data, and unit operation status data in the standardized multi-source data, perform data fusion processing according to a unified time scale to construct the multi-source fused data at the current moment.

[0025] S32. Determine the corresponding weight vector based on the difference in the degree of impact of each multi-source fusion data item on the security risk of operation and maintenance personnel, establish a set of security risk levels, establish the membership relationship between multi-source fusion data items and different security risk levels, and set the corresponding membership function.

[0026] S33. The multi-source fusion data and membership functions are used to perform single-factor fuzzy evaluation on each multi-source fusion data to form a single-factor fuzzy evaluation result; the single-factor fuzzy evaluation results of each multi-source fusion data are combined to construct a fuzzy evaluation matrix; the weight vector is used to perform comprehensive calculation on the fuzzy evaluation matrix to obtain the final fuzzy comprehensive evaluation result of the security risk of operation and maintenance personnel; combined with the security risk level set, the maximum membership principle is used to determine the security risk level at the current moment, and the real-time security risk level signal of operation and maintenance personnel is output.

[0027] Optionally, S4 specifically includes:

[0028] S41. Based on the early warning and control requirements corresponding to different security risk levels, pre-establish the mapping relationship between security risks and security control strategies for operation and maintenance personnel, and build a security control strategy library for operation and maintenance personnel.

[0029] S42. Based on the real-time security risk level signal of the operation and maintenance personnel, perform mapping retrieval in the security management strategy library for the operation and maintenance personnel, and output a security management strategy that is adapted to the current risk.

[0030] Optionally, the step of iteratively correcting the risk assessment-related parameters specifically includes:

[0031] Repeat the S1 to S3 process to reassess and obtain the secondary security risk level after the implementation of the control strategy;

[0032] Collect actual safety risk levels, compare the deviation between actual safety risk levels and secondary safety risk levels, and analyze the impact of various multi-source fusion data on risk assessment results;

[0033] Based on the analysis results, the weight vector, security risk level set, and membership relationship between multi-source fusion data items and different security risk levels adopted by S3 were adjusted and corrected.

[0034] Optionally, the step of iteratively revising the security control strategy library specifically includes:

[0035] Obtain information on changes in security risk levels before and after the implementation of control strategies, and determine the degree of matching between the current security control strategy and the corresponding security risk level in conjunction with the implemented security control strategies;

[0036] Based on the results of the policy matching, the mapping relationship between the content of the security management policy, the security risk level and the security management policy is adjusted and corrected to achieve iterative updates of the security management policy library for operation and maintenance personnel.

[0037] Optionally, in S1, the status data of the maintenance personnel includes body temperature, heart rate, blood pressure, personnel location, and movement status;

[0038] The operation and maintenance task data includes task type, task duration, task area, and task intensity;

[0039] The operation and maintenance environment data includes wind speed, wind direction, temperature, and humidity;

[0040] The unit operating status data includes unit shutdown information.

[0041] Secondly, the offshore wind power operation and maintenance personnel safety closed-loop management system of the present invention includes:

[0042] The data acquisition and preprocessing module is used to collect raw multi-source data, including status data of operation and maintenance personnel, operation and maintenance task data, operation and maintenance environment data, and unit operation status data. The module then performs missing value completion, outlier correction, and standardization processing on the raw multi-source data in sequence, and outputs standardized multi-source data.

[0043] The real-time data reconstruction module is used to receive the status data of operation and maintenance personnel from the standardized multi-source data output by the data acquisition and preprocessing module, monitor the communication transmission latency and packet loss rate within the preset historical time window, determine the real-time communication status, adaptively process and reconstruct the status data of operation and maintenance personnel based on the communication status, and output valid status data of operation and maintenance personnel.

[0044] The security risk assessment module receives valid operation and maintenance personnel status data output by the real-time data reconstruction module and standardized multi-source data output by the data acquisition and preprocessing module. It completes data fusion at a unified time scale to generate multi-source fused data. Based on the weight vector, security risk level set, and membership function, it completes single-factor fuzzy evaluation, fuzzy evaluation matrix construction, and comprehensive calculation processing, and outputs real-time operation and maintenance personnel security risk level signals.

[0045] The management strategy generation module is used to pre-build a security management strategy library for operations and maintenance personnel that includes a mapping relationship between security risk levels and security management strategies. It retrieves and matches the security risk level signals of operations and maintenance personnel in real time and outputs security management strategies that are adapted to the current risk.

[0046] The security closed-loop management module is used to execute security management strategies. It reuses the previous data processing, data reconstruction and risk assessment processes to obtain a secondary security risk level. It iteratively corrects the relevant parameters of risk assessment based on the deviation of the actual security risk level, and adaptively updates the security management strategy library according to the strategy execution effect to achieve full-process closed-loop security management.

[0047] Thirdly, the present invention provides an electronic device for executing the offshore wind power operation and maintenance personnel safety closed-loop management method of the present invention, wherein the electronic device includes a wearable device and a cloud server device; the wearable device and the cloud server device are connected through a wireless communication network.

[0048] The wearable device is used to collect and preprocess the status data of operation and maintenance personnel in real time, collect the actual security risk level, and send the collected data to the cloud server device.

[0049] The cloud server device is used to perform real-time reconstruction of communication status data of operation and maintenance personnel, fuzzy comprehensive evaluation of security risk assessment, generation of security control strategies, and closed-loop security control of human-machine collaborative feedback, and to store business-related data.

[0050] The beneficial effects of this invention are:

[0051] 1. This invention achieves accurate judgment of communication status by monitoring maritime communication transmission latency and packet loss rate. Relying on long short-term memory networks combined with historical time-series data and communication status information, it reconstructs the status data of maintenance personnel under abnormal communication scenarios. This effectively solves problems such as insufficient coverage of maritime base stations, data loss, data distortion, and insufficient real-time performance caused by long-distance transmission. It avoids interference from invalid and delayed abnormal data under unstable communication conditions on risk assessment, significantly improves the accuracy and completeness of maintenance personnel status data collection, and provides a reliable data foundation for subsequent security risk assessment.

[0052] 2. This invention abandons the traditional open-loop assessment mode of single-sensor monitoring and constructs a human-machine collaborative correction mechanism that combines machine intelligent assessment with on-site manual feedback. By comparing the secondary safety risk level assessed by the algorithm with the actual on-site risk level of the operation and maintenance personnel, it reverse-engineers the risk correlation patterns of multi-source fusion data, and realizes adaptive correction of risk assessment parameters such as weight vector, safety risk level set, and membership relationship. This solves the defects of existing technologies, such as lack of manual feedback correction, rigid risk assessment model, and poor fit of assessment results with actual on-site working conditions, and greatly improves the accuracy and environmental adaptability of safety risk assessment.

[0053] 3. This invention establishes a two-way iterative optimization mechanism for safety risk levels and control strategies. It determines the degree of strategy matching based on the risk changes before and after strategy execution, dynamically corrects the content of control strategies and the risk-strategy mapping relationship, and realizes continuous iterative updates of the control strategy library. This completely solves the problems of traditional open-loop control strategies being fixed, unable to be dynamically optimized according to on-site operating conditions, and having poor control adaptability, effectively improving the pertinence and effectiveness of safety control strategies.

[0054] 4. This invention forms a closed-loop management and control system covering the entire process, including data acquisition, adaptive communication reconstruction, intelligent risk assessment, strategy matching and execution, and human-machine feedback iterative optimization. It achieves comprehensive optimization from three dimensions: the reliability of the underlying data, the accuracy of the mid-level risk assessment, and the adaptability of the upper-level management and control strategy. It overcomes the technical shortcomings of existing offshore wind power operation and maintenance safety management and control, such as inaccurate feedback information, lack of closed-loop iteration capability, and poor overall management and control effect. It can adapt to the complex and ever-changing offshore wind power operation environment and significantly improve the intelligent level of safety management and control and the on-site implementation effect of offshore operation and maintenance.

[0055] In summary, this invention solves the problem of unreliable status data of maintenance personnel and inaccurate feedback information leading to poor closed-loop management under unstable communication conditions in offshore wind power operations. Attached Figure Description

[0056] This application can be further illustrated by the non-limiting embodiments given in the accompanying drawings.

[0057] Figure 1 This is a flowchart of the closed-loop safety management method for offshore wind power operation and maintenance personnel described in the embodiments of this application;

[0058] Figure 2 This is a flowchart of the multi-source data collection and preprocessing process for security by operation and maintenance personnel, as described in this application embodiment.

[0059] Figure 3 This is a flowchart illustrating the real-time reconstruction of maintenance personnel status data based on communication status awareness, according to an embodiment of this application.

[0060] Figure 4 This is a schematic diagram of a closed-loop safety management system for operation and maintenance personnel based on human-machine collaborative feedback, according to an embodiment of this application.

[0061] Figure 5 This is a block diagram of the offshore wind power operation and maintenance personnel safety closed-loop management system according to an embodiment of this application;

[0062] Figure 6 This is a schematic diagram of the electronic device composition according to an embodiment of this application. Detailed Implementation

[0063] The embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are used to illustrate the principles of this application by way of example, but should not be used to limit the scope of this application, that is, this application is not limited to the described embodiments.

[0064] like Figure 1 As shown in the embodiments of this application, a closed-loop safety management method for offshore wind power operation and maintenance personnel includes the following steps:

[0065] S1. Safety multi-source data collection and preprocessing for operation and maintenance personnel: Collect operation and maintenance personnel status data, operation and maintenance task data, operation and maintenance environment data, and unit operation status data to form raw multi-source data. Perform missing value completion processing, outlier correction processing, and standardization processing on the raw multi-source data, and output standardized multi-source data.

[0066] S2. Real-time reconstruction of maintenance personnel status data based on communication status awareness: Based on maintenance personnel status data in standardized multi-source data, monitor real-time communication status and make status judgments. Adaptively process and reconstruct maintenance personnel status data according to communication status, and output effective maintenance personnel status data.

[0067] S3. Safety Risk Assessment of Operation and Maintenance Personnel Based on Fuzzy Comprehensive Evaluation: Based on effective operation and maintenance personnel status data and standardized multi-source data including operation and maintenance task data, operation and maintenance environment data, and unit operation status data, data fusion processing is performed at a unified time scale to generate multi-source fused data at the current moment. According to the weight vector, safety risk level set, and membership function, single-factor fuzzy evaluation, fuzzy evaluation matrix construction, and comprehensive calculation processing are performed in sequence. The safety risk level is determined by the maximum membership principle, and the real-time operation and maintenance personnel safety risk level signal is output.

[0068] S4. Generation of security management policies for operations and maintenance personnel: Based on the real-time security risk level signal of operations and maintenance personnel, the system retrieves the security management policy library for operations and maintenance personnel, performs policy matching processing, and outputs security management policies that are adapted to the current risk.

[0069] S5. Safety closed-loop management of operation and maintenance personnel based on human-machine collaborative feedback: Execute safety management policies and iteratively revise relevant parameters of risk assessment and safety management policy library based on risk feedback information after policy execution to achieve closed-loop safety management throughout the entire process.

[0070] The following is a detailed explanation of each step:

[0071] like Figure 2 As shown in the embodiment of this application, S1, the security multi-source data collection by operation and maintenance personnel, specifically includes the following process:

[0072] S11. Security multi-source data collection by maintenance personnel:

[0073] The multi-source safety data for operations and maintenance personnel includes personnel status data, operations and maintenance task data, operations and maintenance environment data, and unit operating status data. Personnel status data includes, but is not limited to, body temperature, heart rate, blood pressure, personnel location, and movement status; operations and maintenance task data includes, but is not limited to, task type, task duration, task area, and task intensity; operations and maintenance environment data includes, but is not limited to, wind speed, wind direction, temperature, and humidity; and unit operating status data includes, but is not limited to, unit shutdown information.

[0074] S12. Security preprocessing of multi-source data by maintenance personnel:

[0075] First, missing value processing is performed on the security multi-source data collected from operations and maintenance personnel. Specifically, this includes: detecting missing values ​​in the original multi-source data; when missing values ​​are detected, obtaining the time-series data before and after the missing value; and performing missing value completion processing through a preset missing value imputation model to obtain the completed multi-source data.

[0076] Next, outlier detection is performed on the completed multi-source data. When an outlier is detected, the time series data before and after the outlier is obtained, and the outlier data correction is performed through a preset smoothing model to obtain the corrected compliant data.

[0077] Finally, the corrected compliant data is standardized and normalized using the Z-score algorithm to output standardized multi-source data.

[0078] like Figure 3 As shown in the embodiment of this application, S2, real-time reconstruction of the status data of operation and maintenance personnel based on communication status awareness, specifically includes the following process:

[0079] S21. Communication status information collection:

[0080] Based on the transmission process of maintenance personnel status data in the communication network, and according to the timestamp information of the data sender and receiver within a preset historical time window, the data transmission delay at the current moment is calculated and expressed as:

[0081]

[0082] In the formula, Representing the current moment Data transmission latency, This indicates the number of packets within the preset historical time window. Indicates the first Data packet sending time, Indicates the first Data packet reception time.

[0083] Based on the number of packets sent and received by the data sender and receiver within a preset historical time window, the packet loss rate at the current moment is calculated and expressed as:

[0084]

[0085] In the formula, Indicates the current time Data packet loss rate This indicates the number of packets sent by the data sender within the preset historical time window. This indicates the number of packets received by the data receiver within the preset historical time window.

[0086] S22. Communication Status Determination:

[0087] The system compares data transmission latency and packet loss rate with a preset communication status threshold to determine if there is an anomaly in the current communication status, as shown below:

[0088]

[0089] In the formula, Indicates the current time The communication status determination result, when Indicates the current time The communication status is normal. Indicates the current time The communication status is abnormal. Indicates the data transmission delay threshold. This represents the data packet loss rate threshold.

[0090] S23. Real-time reconstruction of maintenance personnel status data:

[0091] Based on the communication status determination result obtained in S22, the status data of the maintenance personnel received at the current moment is reconstructed in real time. First, if the communication status determination result... When the current communication status is normal, the received status data from the operations and maintenance personnel will be directly output, as follows:

[0092]

[0093] In the formula, This represents the status data of operations and maintenance personnel after reconstruction at the current moment (i.e., the status data of valid operations and maintenance personnel). Indicates the current time Received status data from operations and maintenance personnel.

[0094] If the communication status determination result If the current communication status is abnormal, then the long short-term memory network is used to reconstruct the current maintenance personnel status data by combining historical maintenance personnel status data and communication status information, as follows:

[0095]

[0096] In the formula, Represents the Long Short-Term Memory network. Indicates the backtracking time step.

[0097] S3. Security risk assessment for operations and maintenance personnel based on fuzzy comprehensive evaluation, specifically including the following process:

[0098] S31. Security data fusion for operations and maintenance personnel:

[0099] Based on effective maintenance personnel status data, combined with preprocessed maintenance task data, maintenance environment data, and unit operating status data, data fusion processing is performed according to a unified time scale to construct multi-source fused data for the current moment, represented as follows:

[0100]

[0101] In the formula, Indicates time Multi-source fusion data, , and These represent the time intervals after preprocessing. The operation and maintenance task data, operation and maintenance environment data, and unit operating status data.

[0102] S32. Construction of a fuzzy evaluation model for the safety of maintenance personnel:

[0103] Multi-source fusion data and membership functions are used to perform single-factor fuzzy evaluation on each multi-source fusion data point, forming single-factor fuzzy evaluation results. A fuzzy evaluation matrix is ​​constructed by combining the single-factor fuzzy evaluation results of each multi-source fusion data point. Weight vectors are used to perform comprehensive calculations on the fuzzy evaluation matrix to obtain the final fuzzy comprehensive evaluation result of the security risk for operations and maintenance personnel. Based on the set of security risk levels, the maximum membership principle is used to determine the security risk level at the current moment, outputting a real-time security risk level signal for operations and maintenance personnel. The specific process includes:

[0104] Based on the differences in the degree of impact of each multi-source fusion data item on the security risks to operations and maintenance personnel, a weight vector is determined through expert experience, expressed as follows:

[0105]

[0106] In the formula, This represents the preset weight vector. Indicates the first The weights corresponding to the multi-source fused data items, and satisfying , This indicates the total number of items in the multi-source fusion data.

[0107] Establish a set of security risk levels, represented as:

[0108]

[0109] In the formula, Represents a set of security risk levels. Indicates the first Class of security risk levels.

[0110] For various multi-source data, establish the membership relationship between multi-source fusion data items and different security risk levels, and set the corresponding membership functions.

[0111] S33. Security Risk Assessment for Operation and Maintenance Personnel:

[0112] Based on the multi-source fusion data obtained in S31, the membership function established in S32 is used to perform single-factor fuzzy evaluation on each multi-source fusion data point, forming the single-factor fuzzy evaluation result, which is expressed as:

[0113]

[0114] In the formula, Indicates the current At this moment The results of single-factor fuzzy evaluation of multi-source fusion data. Indicates the current At this moment The multi-source fusion data belongs to the first Membership degree of risk class level.

[0115] By combining the single-factor fuzzy evaluation results of various multi-source fusion data, a fuzzy evaluation matrix is ​​constructed, represented as follows:

[0116]

[0117] In the formula, Indicates the current time The fuzzy evaluation matrix.

[0118] Using the weight vector established by S32, the fuzzy evaluation matrix is ​​comprehensively processed to obtain the final fuzzy comprehensive evaluation result of the safety risks for operation and maintenance personnel, expressed as:

[0119]

[0120] In the formula, Indicates the current time The results of the fuzzy comprehensive evaluation of safety risks for maintenance personnel. This represents the weighted average operator.

[0121] Based on the set of security risk levels established by S31, the security risk level at the current moment is determined using the principle of maximum membership. Output real-time safety risk level signals for maintenance personnel;

[0122] S4. Generation of security management policies for operations and maintenance personnel, specifically including the following processes:

[0123] S41. Construction of a security management strategy library for operations and maintenance personnel:

[0124] Based on the early warning and control requirements corresponding to different security risk levels, a mapping relationship between security risks and security control strategies for operations and maintenance personnel is pre-established, and a security control strategy library for operations and maintenance personnel is constructed, represented as follows:

[0125]

[0126] In the formula, This represents the security management strategy library for operations and maintenance personnel. Indicates the first Security management strategies for operations and maintenance personnel corresponding to different security risk levels.

[0127] S42. Generation of security management policies for operations and maintenance personnel:

[0128] Based on the real-time security risk level signal of maintenance personnel output by S3, a mapping and retrieval is performed in the maintenance personnel security management policy database established by S41 to generate the security management policy for the current moment. .

[0129] like Figure 4 As shown in the embodiment of this application, S5, the closed-loop management of maintenance personnel safety based on human-machine collaborative feedback, specifically includes:

[0130] S51. Human-machine collaborative feedback information collection and fusion analysis:

[0131] After the security control strategy for operations and maintenance personnel is implemented, the security risks of operations and maintenance personnel are reassessed in sequence according to the methods from S1 to S3 to obtain the security risk level of operations and maintenance personnel after the implementation of the control strategy, i.e., the secondary security risk level.

[0132] The actual security risk level is collected, specifically by collecting manually confirmed feedback information, including whether the security risk assessment by the operations and maintenance personnel is accurate, whether the security control strategies of the operations and maintenance personnel are effective, and the actual security risk level confirmed by the personnel.

[0133] The actual security risk level is integrated with the security risk level of operations and maintenance personnel after the implementation of the control strategy to determine the current feedback result type. Specifically, when the manually confirmed feedback information indicates that the security risk level is inconsistent with the actual level, it is determined that the fuzzy security evaluation model for operations and maintenance personnel (including weight vectors, security risk level sets, and the membership relationships between multi-source fusion data items and different security risk levels) has insufficient adaptability, and a closed-loop correction of the fuzzy security evaluation model for operations and maintenance personnel is performed; when the manually confirmed feedback information indicates that the security control strategy has not achieved the expected results, it is determined that the current control strategy is not well matched with the security risk level, and a closed-loop correction of the security control strategy library for operations and maintenance personnel is performed.

[0134] S52. Closed-loop correction of the fuzzy evaluation model for the safety of maintenance personnel:

[0135] Based on the deviation between the actual security risk level confirmed by manual verification and the security risk level of maintenance personnel after the implementation of control strategies, this paper analyzes the degree of influence of various multi-source fusion data in the current fuzzy evaluation model of maintenance personnel security on the risk judgment results.

[0136] Based on the above analysis results, the weight vector, the set of security risk levels, and the membership relationship between multi-source fusion data items and different security risk levels in the security fuzzy evaluation model for operation and maintenance personnel are adjusted and corrected.

[0137] S53, Closed-loop revision of the security management strategy library for operations and maintenance personnel:

[0138] Based on the changes in the security risk level of operations and maintenance personnel before and after the implementation of the control strategy, and in conjunction with the security control strategy already implemented, the degree of matching between the current security control strategy and its corresponding security risk level is determined, and the security control strategy and the mapping and matching relationship between the security risk level and the security control strategy are adjusted and corrected accordingly.

[0139] like Figure 5As shown in this embodiment, a closed-loop safety management system for offshore wind power operation and maintenance personnel includes a data acquisition and preprocessing module, a real-time data reconstruction module, a safety risk assessment module, a management strategy generation module, and a safety closed-loop management module. The data acquisition and preprocessing module collects raw multi-source data, including operation and maintenance personnel status data, operation and maintenance task data, operation and maintenance environment data, and turbine operating status data. It then sequentially performs missing value completion, outlier correction, and standardization processing on the raw multi-source data, outputting standardized multi-source data. The real-time data reconstruction module receives the operation and maintenance personnel status data from the standardized multi-source data output by the data acquisition and preprocessing module, monitors communication transmission latency and packet loss rate within a preset historical time window, determines the real-time communication status, and adaptively processes and reconstructs the operation and maintenance personnel status data based on the communication status, outputting valid operation and maintenance personnel status data. The security risk assessment module receives valid maintenance personnel status data output from the real-time data reconstruction module and standardized multi-source data output from the data acquisition and preprocessing module. It then fuses this data at a unified time scale to generate multi-source fused data. Based on weight vectors, a set of security risk levels, and membership functions, it performs single-factor fuzzy evaluation, fuzzy evaluation matrix construction, and comprehensive calculations, outputting real-time maintenance personnel security risk level signals. The control strategy generation module pre-builds a security control strategy library for maintenance personnel, containing a mapping relationship between security risk levels and security control strategies. It retrieves and matches strategies based on real-time maintenance personnel security risk level signals, outputting security control strategies adapted to the current risk level. The security closed-loop control module executes the security control strategies, reusing the pre-processing, data reconstruction, and risk assessment steps to obtain secondary security risk levels. It iteratively corrects relevant risk assessment parameters based on actual security risk level deviations and adaptively updates the security control strategy library based on strategy execution results, achieving full-process closed-loop security control.

[0140] like Figure 6 As shown in the embodiments of this application, an electronic device is used to execute the offshore wind power operation and maintenance personnel safety closed-loop management method of this application. The electronic device includes a wearable device and a cloud server device; the wearable device and the cloud server device are connected via a wireless communication network. The wearable device is used to collect and preprocess operation and maintenance personnel status data in real time, collect the actual safety risk level, and send the collected data to the cloud server device. The cloud server device is used to perform real-time reconstruction of operation and maintenance personnel status data with communication status awareness, fuzzy comprehensive evaluation of safety risk assessment, generation of safety control strategies, human-machine collaborative feedback of safety closed-loop management, and storage of business-related data.

[0141] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A closed-loop safety management method for offshore wind power operation and maintenance personnel, characterized in that, Includes the following steps: S1. Collect maintenance personnel status data, maintenance task data, maintenance environment data, and unit operation status data to form raw multi-source data. Perform missing value completion processing, outlier correction processing, and standardization processing on the raw multi-source data, and output standardized multi-source data. S2. Based on the operation and maintenance personnel status data in the standardized multi-source data, monitor the real-time communication status and make a status determination. Adaptively process and reconstruct the operation and maintenance personnel status data according to the communication status, and output valid operation and maintenance personnel status data. S3. Based on the effective maintenance personnel status data and the maintenance task data, maintenance environment data, and unit operation status data in the standardized multi-source data, perform data fusion processing at a unified time scale to generate multi-source fused data for the current moment. According to the weight vector, safety risk level set, and membership function, perform single-factor fuzzy evaluation, construct fuzzy evaluation matrix, and comprehensive calculation processing in sequence. Determine the safety risk level through the maximum membership principle and output the real-time maintenance personnel safety risk level signal. S4. Based on the real-time security risk level signal of maintenance personnel, retrieve the security management strategy library for maintenance personnel, perform strategy matching processing, and output a security management strategy that is adapted to the current risk. S5. Execute the security control strategy and, based on the risk feedback information after the strategy execution, iteratively correct the risk assessment-related parameters and the security control strategy library to achieve closed-loop security control throughout the entire process.

2. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, In step S1, the original multi-source data undergoes missing value completion, outlier correction, and standardization, and standardized multi-source data is output. Specifically: Missing value detection is performed on the collected raw multi-source data. When a missing value is detected, the time series data before and after the missing value is retrieved, and missing value completion processing is performed through a preset missing value imputation model to obtain the completed multi-source data. Anomaly detection is performed on the completed multi-source data. When an anomaly is detected, the time-series data before and after the anomaly is obtained, and anomaly data correction processing is performed through a preset smoothing processing model to obtain corrected compliant data. The corrected compliant data is standardized and normalized using the Z-score algorithm to output the standardized multi-source data.

3. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, The specific processing flow of S2 is as follows: S21. Based on the transmission process of the maintenance personnel status data in the communication network, calculate the data transmission delay at the current moment according to the timestamp information of the data sending end and receiving end within the preset historical time window, and calculate the packet loss rate at the current moment according to the number of packets of the data sending end and receiving end within the preset historical time window. S22. Based on the comparison between the data transmission delay and data packet loss rate and the preset communication status threshold, determine whether there is an abnormality in the current communication status. S23. If the communication status is normal, the operation and maintenance personnel status data is forwarded directly. If the communication status is abnormal, the operation and maintenance personnel status data at the current moment is reconstructed by combining the historical operation and maintenance personnel status data and communication status information through the long short-term memory network, and the reconstructed operation and maintenance personnel status data is output.

4. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, The specific processing flow of S3 is as follows: S31. Based on the effective maintenance personnel status data and the maintenance operation task data, maintenance operation environment data, and unit operation status data in the standardized multi-source data, perform data fusion processing according to a unified time scale to construct the multi-source fused data at the current moment. S32. Determine the corresponding weight vector based on the difference in the degree of impact of each multi-source fusion data item on the security risk of operation and maintenance personnel, establish a set of security risk levels, establish the membership relationship between multi-source fusion data items and different security risk levels, and set the corresponding membership function. S33. The multi-source fusion data and membership functions are used to perform single-factor fuzzy evaluation on each multi-source fusion data to form a single-factor fuzzy evaluation result; and a fuzzy evaluation matrix is ​​constructed by combining the single-factor fuzzy evaluation results of each multi-source fusion data. The weight vector is used to perform comprehensive calculations on the fuzzy evaluation matrix to obtain the final fuzzy comprehensive evaluation result of the safety risks of operation and maintenance personnel. Based on the aforementioned set of security risk levels, the security risk level at the current moment is determined using the maximum membership principle, and a real-time security risk level signal for maintenance personnel is output.

5. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, Specifically, S4 is: S41. Based on the early warning and control requirements corresponding to different security risk levels, pre-establish the mapping relationship between security risks and security control strategies for operation and maintenance personnel, and build a security control strategy library for operation and maintenance personnel. S42. Based on the real-time security risk level signal of the operation and maintenance personnel, perform mapping retrieval in the security management strategy library for the operation and maintenance personnel, and output a security management strategy that is adapted to the current risk.

6. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, The steps for iteratively correcting the risk assessment-related parameters specifically include: Repeat the S1 to S3 process to reassess and obtain the secondary security risk level after the implementation of the control strategy; Collect actual safety risk levels, compare the deviation between actual safety risk levels and secondary safety risk levels, and analyze the impact of various multi-source fusion data on risk assessment results; Based on the analysis results, the weight vector, security risk level set, and membership relationship between multi-source fusion data items and different security risk levels adopted by S3 were adjusted and corrected.

7. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 6, characterized in that, The steps for iteratively revising the security control strategy library specifically include: Obtain information on changes in security risk levels before and after the implementation of control strategies, and determine the degree of matching between the current security control strategy and the corresponding security risk level in conjunction with the implemented security control strategies; Based on the results of the policy matching, the mapping relationship between the content of the security management policy, the security risk level and the security management policy is adjusted and corrected to achieve iterative updates of the security management policy library for operation and maintenance personnel.

8. The closed-loop safety management method for offshore wind power operation and maintenance personnel according to claim 1, characterized in that, In S1, the status data of the maintenance personnel includes body temperature, heart rate, blood pressure, personnel location, and movement status; The operation and maintenance task data includes task type, task duration, task area, and task intensity; The operation and maintenance environment data includes wind speed, wind direction, and temperature and humidity; The unit operating status data includes unit shutdown information.

9. A closed-loop safety management and control system for offshore wind power operation and maintenance personnel, characterized in that, include: The data acquisition and preprocessing module is used to collect raw multi-source data, including status data of operation and maintenance personnel, operation and maintenance task data, operation and maintenance environment data, and unit operation status data. The module then performs missing value completion, outlier correction, and standardization processing on the raw multi-source data in sequence, and outputs standardized multi-source data. The real-time data reconstruction module is used to receive the status data of operation and maintenance personnel from the standardized multi-source data output by the data acquisition and preprocessing module, monitor the communication transmission latency and packet loss rate within the preset historical time window, determine the real-time communication status, adaptively process and reconstruct the status data of operation and maintenance personnel based on the communication status, and output valid status data of operation and maintenance personnel. The security risk assessment module receives valid operation and maintenance personnel status data output by the real-time data reconstruction module and standardized multi-source data output by the data acquisition and preprocessing module. It completes data fusion at a unified time scale to generate multi-source fused data. Based on the weight vector, security risk level set, and membership function, it completes single-factor fuzzy evaluation, fuzzy evaluation matrix construction, and comprehensive calculation processing, and outputs real-time operation and maintenance personnel security risk level signals. The management strategy generation module is used to pre-build a security management strategy library for operations and maintenance personnel that includes a mapping relationship between security risk levels and security management strategies. It retrieves and matches the security risk level signals of operations and maintenance personnel in real time and outputs security management strategies that are adapted to the current risk. The security closed-loop management module is used to execute security management strategies. It reuses the previous data processing, data reconstruction and risk assessment processes to obtain a secondary security risk level. It iteratively corrects the relevant parameters of risk assessment based on the deviation of the actual security risk level, and adaptively updates the security management strategy library according to the strategy execution effect to achieve full-process closed-loop security management.

10. An electronic device, characterized in that, The electronic device is used to implement the closed-loop safety management method for offshore wind power operation and maintenance personnel according to any one of claims 1-8, wherein the electronic device includes a wearable device and a cloud server device; the wearable device and the cloud server device are connected through a wireless communication network. The wearable device is used to collect and preprocess the status data of operation and maintenance personnel in real time, collect the actual security risk level, and send the collected data to the cloud server device. The cloud server device is used to perform real-time reconstruction of communication status data of operation and maintenance personnel, fuzzy comprehensive evaluation of security risk assessment, generation of security control strategies, and closed-loop security control of human-machine collaborative feedback, and to store business-related data.

Images