Safety monitoring method and system based on electric power working environment

Abstract

The invention discloses a safety monitoring method and system based on an electric power operation environment, and relates to the technical field of electric power operation safety management, and the method comprises the steps: collecting first data and second data, and carrying out the real-time analysis through combining the first data and the second data; identifying the first state through a first model, and performing evaluation to generate a first evaluation score; and predicting a first behavior mode through a first method, and adjusting a first threshold to guide an operator to take safety measures. According to the method provided by the invention, behavior identification is carried out by combining the ConvLSTM model, and non-standard operation behaviors can be found in time, so that potential safety hazards are effectively avoided; by constructing a safety risk assessment model, the risk of accidents is effectively reduced; by dynamically adjusting the monitoring threshold value, the safety monitoring system is more sensitive and accurate, the change of the working environment can be responded in time, and the safety of operators is ensured; through behavior trend prediction, potential non-standard behaviors can be recognized in advance, and accidents are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power operation safety management, and in particular to a safety monitoring method and system based on a power operation environment. Background Art

[0002] With the development of the power industry and the continuous increase in electricity demand, power workers face increasingly complex and dangerous working environments when performing tasks such as maintenance, inspection and installation of high-voltage power facilities. Traditional power operation safety management mainly relies on manual supervision and traditional safety protection measures. However, with the increasing complexity of the operating environment and technological advancement, traditional safety management methods have gradually shown their limitations. Especially in complex cross-operation environments, manual supervision cannot achieve real-time, comprehensive and accurate risk warnings, resulting in many potential risks not being identified and handled in a timely manner.

[0003] In order to meet the challenges in power operation safety management, intelligent monitoring and early warning mechanisms based on data collection and analysis have gradually emerged in recent years. These mechanisms use advanced sensor technology, machine learning and deep learning technology to identify and analyze the behavior of power workers, as well as real-time monitoring and analysis of the working environment, thereby achieving early warning of potential safety hazards.

[0004] Although the existing intelligent monitoring and early warning mechanism has made certain progress, there are still some shortcomings. First, the existing behavior recognition technology mainly relies on single behavioral data, and ignores the impact of environmental factors on behavior. For example, in high temperature and high humidity environments, the actions of operators may be affected, resulting in irregular behavior. Second, the existing environmental monitoring method mainly relies on single environmental data, and ignores the impact of other environmental factors such as noise and light on work safety. In addition, the existing safety monitoring system generally lacks the ability to comprehensively judge and dynamically adjust multiple factors. The risk assessment method usually adopts a fixed threshold judgment standard and lacks adaptability to changes in operator behavior and environment, resulting in low early warning accuracy. In order to address this problem, how to combine real-time collected behavioral data, environmental data and operation risk assessment results, dynamically adjust the monitoring threshold, and promptly guide operators to take corresponding safety measures has become a technical problem that needs to be solved urgently in the power operation safety monitoring system. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: the existing power operation safety management method has the problems of insufficient monitoring of operator behavior and environmental factors, inaccurate risk assessment, and insufficient timely warning, as well as the problem of how to achieve intelligent, safe and efficient operation management.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a safety monitoring method based on an electric power operation environment, comprising collecting first data and second data, and performing real-time analysis in combination with the first data and the second data; identifying a first state through a first model, and performing an evaluation to generate a first evaluation score; predicting a first behavior pattern through a first method, and adjusting a first threshold to guide operators to take safety measures.

[0008] As a preferred solution of the safety monitoring method based on the electric power operation environment described in the present invention, the collecting of the first data and the second data includes collecting the first data and the second data in real time through smart devices and sensor devices.

[0009] As a preferred solution of the safety monitoring method based on the electric power operation environment described in the present invention, the real-time analysis includes preprocessing the collected first data and second data, and extracting the first feature based on the preprocessed data.

[0010] As a preferred solution of the safety monitoring method based on the electric power operation environment described in the present invention, the identifying the first state through the first model includes taking the first feature as the input of the first model and outputting the first state through the first model.

[0011] As a preferred solution of the safety monitoring method based on the electric power operation environment described in the present invention, the generating of the first evaluation score includes defining an evaluation index according to the first state in combination with the second data, and obtaining the first evaluation score through the evaluation model.

[0012] As a preferred solution of the safety monitoring method based on the power operation environment described in the present invention, wherein: predicting the first behavior pattern by the first method includes establishing a first database using historical data; and predicting the first behavior pattern by the first method based on the first database.

[0013] As a preferred solution of the safety monitoring method based on the electric power operation environment described in the present invention, the adjusting the first threshold includes dynamically adjusting the first threshold in combination with the first state, the first behavior pattern and the first evaluation score.

[0014] Another object of the present invention is to provide a safety monitoring system based on the power operation environment, which can dynamically adjust the monitoring threshold by combining real-time collected behavioral data, environmental data and operation risk assessment results, thereby solving the technical problem that the current power operation safety monitoring system lacks the ability to comprehensively judge and dynamically adjust multiple factors.

[0015] As a preferred solution of the safety monitoring system based on the electric power operation environment described in the present invention, it includes: a data module, an evaluation module, and an adjustment module; the data module is used to collect first data and second data, and perform real-time analysis in combination with the first data and the second data; the evaluation module is used to identify the first state through a first model, and perform an evaluation to generate a first evaluation score; the adjustment module is used to predict the first behavior pattern through a first method, and adjust the first threshold to guide the operator to take safety measures.

[0016] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a safety monitoring method based on an electric power operation environment.

[0017] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a safety monitoring method based on an electric power operation environment.

[0018] Beneficial effects of the present invention: The safety monitoring method based on the electric power operation environment provided by the present invention collects the behavioral data of the operators in real time and combines the ConvLSTM model for behavior recognition, so as to timely discover irregular operating behaviors, thereby effectively avoiding potential safety hazards; by comprehensively considering the behavioral data and environmental data, a safety risk assessment model is constructed, the risk assessment score is calculated in real time, and corresponding safety measures are taken according to the risk level, effectively reducing the risk of accidents; by dynamically adjusting the monitoring threshold, the safety monitoring system is made more sensitive and accurate, and can respond to changes in the operating environment in a timely manner to ensure the safety of operators; through behavioral trend prediction, potential irregular behaviors can be identified in advance to avoid the occurrence of accidents; the system triggers an alarm system when the risk assessment score exceeds the threshold, which can avoid the expansion of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 This is an overall flow chart of a safety monitoring method based on an electric power operation environment provided by the first embodiment of the present invention.

[0021] Figure 2 This is an overall flow chart of a safety monitoring system based on an electric power operation environment provided by the third embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0023] Example 1, with reference to Figure 1 , as an embodiment of the present invention, provides a safety monitoring method based on an electric power operation environment, comprising:

[0024] S1: Collect first data and second data, and perform real-time analysis based on the first data and the second data.

[0025] Furthermore, collecting the first data and the second data includes collecting the first data and the second data in real time through a smart device and a sensor device.

[0026] The first data includes but is not limited to behavioral data such as the operator's posture and movement data, tool usage, protective equipment wearing, work position and safety distance, behavioral anomaly detection, heart rate monitoring, movement frequency, and voice and environment interaction collected through the smart wearable device worn by the operator.

[0027] The second data includes but is not limited to environmental data such as temperature, humidity, wind speed and direction, atmospheric pressure, light intensity, noise level, air quality, current, voltage, equipment vibration, equipment temperature and physical obstacles in the environment collected by sensor devices worn by operators.

[0028] It should be noted that the real-time analysis includes preprocessing the collected first data and second data, and extracting the first feature based on the preprocessed data.

[0029] The first features include but are not limited to behavioral features such as action and posture features, behavior timing features, and environment and background features.

[0030] In the embodiment of the present application, the motion and posture characteristics include acceleration characteristics, gyroscope characteristics and operator posture characteristics.

[0031] In an embodiment of the present application, the environment and background features include environmental data and operating scenes of the operating environment.

[0032] It should also be noted that in the embodiments of the present application, through the smart wearable devices and various sensors worn by the operators, the behavioral data and working environment data of the operators are collected in real time, providing a comprehensive and accurate data basis for subsequent risk assessment and safety monitoring. The behavioral data can reflect the operating status and safety behavior of the operators, identify possible irregular operations or physical abnormalities, and the environmental data can fully perceive the external risk factors and equipment operating status of the work site, evaluate the environment's dynamic perception of the operators and environmental status, and provide data support for subsequent behavior identification, risk assessment, trend prediction and intelligent early warning, thereby improving the overall safety and management level of the power operation environment.

[0033] S2: Identify a first state through a first model and perform an evaluation to generate a first evaluation score.

[0034] Furthermore, identifying the first state through the first model includes taking the first feature as an input of the first model and outputting the first state through the first model.

[0035] The first model can be a ConvLSTM model, a Transformer model, or other deep learning models suitable for recognition and prediction tasks.

[0036] The first state includes but is not limited to the current behavior state of the operator.

[0037] In an embodiment of the present application, the first model uses a ConvLSTM model to output a first state. Specifically, the first data collected from the operator is manually labeled to construct a labeled data set; the labeling types include standardized operations and non-standard operations; 70% of the labeled data set is set as a training set, 15% of the labeled data set is set as a validation set, and 15% of the labeled data set is set as a test set; the ConvLSTM model is trained using the training set, and the weights are gradually adjusted to optimize the cross-entropy loss function; during the training process, the performance of the ConvLSTM model is monitored using the validation set; the test set is used to evaluate the calculation accuracy, recall rate, and F1-score of the ConvLSTM model; the first feature is used as the input of the trained ConvLSTM model for real-time analysis. The output of the ConvLSTM model is the current behavior state of the operator, which includes standardized operations and non-standard operations.

[0038] In an optional embodiment, the first model uses a Transformer model to output a first state. Specifically, the preprocessed first data is input into the Transformer model, and the spatiotemporal features are automatically extracted using the self-attention mechanism; the collected behavioral data is labeled, and the cross-entropy loss function is used as the optimization target. The Adam optimizer is used to set the learning rate and learning decay strategy, and the Transformer model is trained on the training set. The model parameters are gradually adjusted, and the model performance is monitored through the validation set. The early stopping strategy is used to stop training when the validation set loss no longer decreases. The Transformer model outputs the probability distribution of the behavioral state at each time step. According to the output probability distribution, the type with the highest probability is selected as the current behavioral state of the operator.

[0039] It should be noted that generating the first evaluation score includes defining an evaluation indicator based on the first state in combination with the second data, and obtaining the first evaluation score through an evaluation model.

[0040] The evaluation indicators include but are not limited to behavioral risk indicators and environmental risk indicators.

[0041] In the embodiment of the present application, the formula of the evaluation model is expressed as:

[0042]

[0043] Among them, S ′ (R) represents the optimized comprehensive risk score, w Bi represents the weight of the i-th behavioral risk indicator, S Bi represents the score of the i-th behavioral risk indicator, n represents the number of behavioral risk indicators, S Ej Represents the j-th environmental risk indicator score.

[0044] S ′ The value range of (R) is from 0 to positive infinity.

[0045] The behavioral risk indicators include 1 point for standard work and 5 points for non-standard work.

[0046] The environmental risk indicators include but are not limited to temperature between 15℃ and 30℃, which is 1 point; temperature above 30℃ or below 10℃, which is 3 points; temperature above 35℃ or below 0℃, which is 5 points; humidity between 50% and 70%, which is 1 point; humidity below 30% or above 80%, which is 3 points; humidity above 90% or below 10%, which is 5 points; wind speed below 5m / s, which is 1 point; wind speed between 5 and 10m / s, which is 3 points; wind speed above 10m / s, which is 5 points; light between 500 and 1000Lux, which is 1 point; light between 100 and 500Lux, which is 3 points; light below 100Lux or light above 1000Lux, which is 5 points.

[0047] It should also be noted that in the embodiment of the present application, by combining the behavioral data of the operators and the real-time data of the working environment, an intelligent safety risk assessment mechanism is established to achieve comprehensive monitoring and accurate assessment of the working status. By extracting motion and posture features, behavioral timing features, and environmental and background features, the system can deeply analyze the relationship between the operating behavior of the operators and environmental factors, identify the current behavioral state, and use the ConvLSTM model for training and optimization based on these data to ensure the accuracy and robustness of real-time identification of behavioral states. Combined with environmental risk indicators and behavioral risk indicators, a safety risk assessment model is further constructed to achieve quantitative and dynamic evaluation of risk scores. It can timely discover potential safety hazards, generate accurate risk assessment scores, and provide targeted safety warnings and measures guidance for operators, ultimately improving the safety management level of power operations and reducing the risk of personal accidents.

[0048] S3: Predict the first behavior pattern using the first method, adjust the first threshold and guide the operator to take safety measures.

[0049] Furthermore, predicting the first behavior pattern by the first method includes establishing a first database using historical data; and predicting the first behavior pattern by the first method based on the first database.

[0050] The establishing of the first database includes utilizing historical data to analyze the behavior patterns and common operation processes of operators and establishing a behavior trajectory library.

[0051] The first method may be to predict the first behavior pattern through a DTW dynamic time warping algorithm, or through a first model, or other suitable methods.

[0052] The first behavior pattern includes but is not limited to an irregular behavior pattern.

[0053] In an embodiment of the present application, the first method is to use a ConvLSTM model. By adding a future step length prediction mechanism to the ConvLSTM model, the behavior trend of the operator in the next few seconds is predicted, and whether the operator is likely to perform irregular operations in the next few seconds is determined.

[0054] In an optional embodiment, the first method is to use a Transformer model to predict whether the operator is likely to perform irregular operations in the next few seconds.

[0055] In an optional embodiment, the first method uses the DTW dynamic time warping algorithm to compare real-time behavior trajectories with historical trajectories to identify potential irregular behavior patterns.

[0056] It should be noted that adjusting the first threshold includes dynamically adjusting the first threshold in combination with the first state, the first behavior pattern, and the first evaluation score.

[0057] In the embodiment of the present application, the first threshold is dynamically adjusted. Specifically, a comprehensive feature vector is constructed by combining the behavior recognition state, the behavior prediction result, and the current evaluation score, and a dynamic threshold is set to guide the operator to take corresponding safety measures. The formula of the comprehensive feature vector is expressed as:

[0058]

[0059] Among them, F(t) is the feature vector, B(t) is the current behavior state, To predict the future behavior trend, S ′ (R) is the current comprehensive risk score.

[0060] The formula for the dynamic monitoring threshold is expressed as:

[0061]

[0062] Among them, Θ(t) represents the dynamic monitoring threshold, w i represents the weight coefficient of the i-th feature, Φ i (F i (t)) represents the i-th characteristic function, F i (t) represents the i-th eigenvalue at time t; according to the dynamic monitoring threshold, the comprehensive risk score is mapped to different risk levels, and the formula is expressed as:

[0063]

[0064] Among them, Θ(t) represents the dynamic monitoring threshold, S ′ (R) indicates the current comprehensive risk score. When the risk level is low, normal operations continue and the existing monitoring status is maintained. When the risk level is medium, safety reminders are sent to operators, who are advised to wear protective equipment and adjust their working postures. Risk warnings are also sent to managers, who are advised to strengthen on-site monitoring. When the risk level is high, the alarm system is immediately triggered, notifying operators to stop current operations and initiate emergency plans.

[0065] It should also be noted that in the embodiment of the present application, by combining the behavior recognition results, behavior trend prediction results and the current comprehensive risk score, an accurate assessment of the safety risk of the operating personnel and an adaptive adjustment of the dynamic threshold are achieved, thereby improving the sensitivity and practicality of the safety monitoring system. The DTW dynamic time warping algorithm is used to compare the real-time behavior trajectory with the historical trajectory, which can identify potential irregular behaviors; by introducing the future step prediction mechanism in the ConvLSTM model, the behavior trend of the operating personnel in the next few seconds is predicted in advance, further enhancing the system's foresight and risk prediction capabilities. By combining the above data to construct a feature vector and set a dynamic monitoring threshold, the safety risks can be graded according to the real-time changes in the operating environment, and safety warnings and guidance measures can be provided to operating personnel and managers in a timely manner, ensuring the accuracy, real-time and adaptability of risk management, effectively reducing the risk of safety accidents caused by irregular behaviors during power operations, and ensuring the safety and stability of the operation site.

[0066] Example 2 is an embodiment of the present invention, which provides a safety monitoring method based on an electric power operation environment. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0067] First, 10 power workers with different levels of experience were selected to wear smart wearable devices, including heart rate sensors, gyroscopes, accelerometers, and voice recognition devices. At the same time, various sensors such as temperature and humidity sensors, light intensity meters, and noise monitoring equipment were deployed in the working environment.

[0068] Collect operator posture data (acceleration, angular velocity), tool usage, protective equipment wearing, heart rate monitoring, work position and safety distance, voice interaction data, temperature, humidity, wind speed, light intensity, noise level, air quality, etc.

[0069] The collected behavioral data and environmental data are preprocessed, including denoising, data normalization and time alignment.

[0070] The data is input into the ConvLSTM model for training, and the job behavior status is labeled as "standard job" and "non-standard job", with 70% of the training set, 15% of the validation set, and 15% of the test set.

[0071] Using the DTW (Dynamic Time Warping) algorithm, real-time behavioral trajectories are compared with a database of historical trajectories to predict behavioral trends over the next five seconds. A comprehensive risk score is calculated by combining behavioral status with environmental risk indicators. Monitoring thresholds are dynamically adjusted, and the results are mapped to risk levels. Different safety measures are implemented for workers based on the risk level. The experimental data is shown in Table 1.

[0072] Table 1 Experimental data table

[0073]

[0074] As shown in Table 1, during the behavior recognition process, the ConvLSTM model was optimized using training and validation sets. Test results show an accuracy of 95.3%, a recall of 92.6%, and an F1-score of 93.8%, outperforming a traditional convolutional neural network (CNN) model (which only achieved an accuracy of 88.2%). Using the DTW dynamic time warping algorithm, the match between real-time and historical behavior trajectories reached over 90%. In terms of behavioral trend prediction, the accuracy of predicting irregular operator operations within the next five seconds reached 87%. Risk scoring demonstrated significant differentiation across different environmental conditions and behavioral states. For example, Operator 6's risk score reached 7.0 (high risk), indicating irregular behavior and a high environmental risk, triggering a high-risk warning. Operators 3 and 7 received scores of 2.1 and 2.3 (low risk), respectively, indicating standard behavior and a safe environment, and no alarm was triggered.

[0075] The system calculates dynamic thresholds in real time and adaptively adjusts monitoring standards to ensure accurate risk warnings under different operating personnel and environmental conditions. For example, in an environment with high temperature, high humidity and high noise, the system automatically lowers the monitoring threshold to provide early warning of potential risks.

[0076] Example 3, reference Figure 2 , as an embodiment of the present invention, provides a safety monitoring system based on an electric power operation environment, including a data module, an evaluation module, and an adjustment module.

[0077] The data module is used to collect first data and second data, and perform real-time analysis in combination with the first data and the second data; the evaluation module is used to identify the first state through the first model, and perform evaluation to generate a first evaluation score; the adjustment module is used to predict the first behavior pattern through the first method, and adjust the first threshold to guide the operator to take safety measures.

[0078] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0079] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0080] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0081] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc. It should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention and are not intended to be limiting. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced with equivalents without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications should be encompassed by the claims of the present invention.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A safety monitoring method based on an electric power operation environment, characterized in that: include: Collecting first data and second data, and performing real-time analysis on the first data and the second data; Identifying a first state through a first model and performing an evaluation to generate a first evaluation score; The first behavior pattern is predicted by the first method, and the first threshold is adjusted to guide the operator to take safety measures.

2. The safety monitoring method based on the power operation environment according to claim 1, characterized in that: The collecting of the first data and the second data includes collecting the first data and the second data in real time through a smart device and a sensor device.

3. The safety monitoring method based on the power operation environment according to claim 2, characterized in that: The real-time analysis includes preprocessing the collected first data and second data, and extracting a first feature based on the preprocessed data.

4. The safety monitoring method based on the power operation environment according to claim 3, characterized in that: The identifying the first state through the first model includes taking the first feature as an input of the first model and outputting the first state through the first model.

5. The safety monitoring method based on the power operation environment according to claim 4, characterized in that: Generating the first evaluation score includes defining an evaluation indicator according to the first state in combination with the second data, and obtaining the first evaluation score through an evaluation model.

6. The safety monitoring method based on the power operation environment according to claim 5, characterized in that: The predicting of the first behavior pattern by the first method includes establishing a first database using historical data; A first behavior pattern is predicted by a first method based on the first database.

7. The safety monitoring method based on the power operation environment according to claim 6, characterized in that: The adjusting the first threshold includes dynamically adjusting the first threshold in combination with the first state, the first behavior pattern, and the first evaluation score.

8. A system using the safety monitoring method based on the power operation environment according to any one of claims 1 to 7, characterized in that: Including data module, evaluation module and adjustment module; The data module is used to collect the first data and the second data, and perform real-time analysis on the first data and the second data; The evaluation module is used to identify the first state through the first model, and perform evaluation to generate a first evaluation score; The adjustment module is used to predict the first behavior pattern through a first method, and adjust the first threshold to guide the operator to take safety measures.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the safety monitoring method based on the electric power operation environment described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the safety monitoring method based on the electric power operation environment according to any one of claims 1 to 7 are implemented.

Images