Power safety production method, system, equipment and medium combined with power organization

Abstract

The invention relates to the technical field of information, and discloses an electric power safety production method, system and equipment combined with an electric power organization and a medium, and the method comprises the steps: obtaining the real-time operation data of electric power equipment; performing anomaly detection on the real-time operation data through a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm combining chaotic initialization, dynamic inertia weight and fuzzy logic; determining a coping scheme and a hidden danger level of the power equipment according to the abnormal operation data; and based on the hidden danger level, matching a target person from a power organization database through an artificial intelligence matching algorithm, and executing the coping scheme, so as to realize the safe production of the power equipment, thereby greatly improving the safe production efficiency of the power equipment, and reducing the social influence and economic loss caused by a power accident.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method, system, equipment and medium for safe production of electric power combined with electric power organization. Background Art

[0002] In the process of power production safety, abnormal detection of power equipment, formulation of response measures and matching of implementation personnel are becoming increasingly important.

[0003] At present, most companies usually use sensors to monitor the working status of power equipment in real time, such as changes in parameters such as pressure, temperature, and humidity, to determine whether the equipment is in normal working condition, but this method is inefficient; when formulating response measures and assigning implementation personnel, they usually rely on the experience of technical personnel. However, the training cycle of technical personnel is long and the number is small, resulting in the existing power production safety efficiency becoming lower and lower.

[0004] It can be seen that how to solve the problem of low efficiency in existing power production safety has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the invention

[0005] The present invention provides a method, system, equipment and medium for safe production of electric power combined with electric power organization, so as to solve the problem of how to improve the efficiency of existing safe production of electric power.

[0006] In order to solve the above technical problems, the first aspect of the present invention provides a method for safe production of electric power combined with electric power organization, comprising: Obtain real-time operating data of power equipment; Anomaly detection is performed on the real-time operation data by using a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; Determine a response plan and a hidden danger level of the power equipment according to the abnormal operation data; Based on the hidden danger level, the target personnel are matched from the power organization database through the artificial intelligence matching algorithm and the response plan is executed to achieve safe production of the power equipment.

[0007] As one of the preferred solutions, the method of performing anomaly detection on the real-time operation data by using a hybrid multi-objective optimization algorithm to obtain abnormal operation data includes: Generate the initial position of the particle group through chaotic mapping, and initialize the speed of each particle; each particle is a set of detection thresholds for the real-time operation data of the power equipment; A multi-objective function is constructed based on the false alarm rate, response speed and computing resource consumption of the power equipment, and the weight coefficient of each objective is adjusted through the fuzzy logic and its corresponding fuzzy rule base according to preset business requirements to obtain a multi-objective optimization function; The multi-objective optimization function is used as a fitness function to calculate the fitness of each particle, and the individual optimal position and the global optimal position of the particle swarm are updated based on the fitness calculation result, so as to adjust the speed and position of each particle by introducing the particle swarm speed and position update formula of the dynamic inertia weight; Iteratively perform fitness calculation and particle speed and position adjustment processes according to the adjusted particle swarm to obtain several global optimal positions to construct a Pareto solution set, and screen the Pareto solution set through a non-dominated sorting method to obtain an optimal detection threshold; The real-time operation data is subjected to abnormality detection using the optimal detection threshold to obtain abnormal operation data.

[0008] As one preferred solution, the dynamic inertia weight is expressed by the following formula: In the formula, is the dynamic inertia weight; , are the minimum and maximum values ​​of the dynamic inertia weight respectively; is the attenuation coefficient; t is the number of iterations; is the aggregation degree of the particle group; is the adjustment factor for the effect of aggregation; is the total number of particle groups; is the position of the ith particle; is the particle group center; is the Euclidean distance.

[0009] As one of the preferred solutions, determining the response plan and hidden danger level of the power equipment according to the abnormal operation data includes: The abnormal operation data is processed by a risk assessment model to obtain a comprehensive risk score, and a response plan for the power equipment is determined based on the comprehensive risk score; wherein the risk assessment model is represented by the following formula: In the formula, To generate a comprehensive risk score; is the importance factor of the nth power equipment; is the state correction factor of the nth power equipment; is the risk adjustment coefficient; M is the total number of power equipment.

[0010] As one of the preferred solutions, determining the response plan and hidden danger level of the power equipment according to the abnormal operation data further includes: Determine the hidden danger type of the power equipment according to the abnormal operation data, quantify the hidden danger level of the power equipment through a hidden danger assessment model, and calculate the severity of each hidden danger type to obtain a hidden danger severity score; wherein the hidden danger assessment model is represented by the following formula: In the formula, is the level of hidden danger; is the importance weight of the ath hidden danger type; is the processing weight of the ath hidden danger type; A is the total number of hidden danger types.

[0011] As one of the preferred solutions, based on the hidden danger level, the target personnel are matched from the power organization database through an artificial intelligence matching algorithm and the response plan is executed to achieve the safe production of the power equipment, including: Acquire the real-time environmental data of the electric power equipment and the basic data of each person in the electric power organization database, and construct the hidden danger level, the real-time environmental data and each basic data as input features; The input features are processed by a deep learning network model constructed by a multi-layer perceptron to obtain the skill matching degree of each person, and the skill matching degree of each person is optimized by using a machine learning algorithm and a reinforcement learning algorithm to obtain the skill matching optimization result of each person; A multi-objective matching function model is constructed based on the skill matching optimization result and the hidden danger severity score, and a particle swarm optimization algorithm is used to perform a global search on the multi-objective matching function model to obtain target personnel and execute the response plan to achieve safe production of the power equipment.

[0012] As one of the preferred solutions, the machine learning algorithm and the reinforcement learning algorithm are used to optimize the skill matching of each person to obtain the skill matching optimization result of each person, including: The hidden danger severity score is processed by an isolation forest algorithm to obtain an optimized score, and the skill matching degree of each person is adjusted according to the optimized score to obtain a comprehensive matching degree of each person; The hidden danger level and the comprehensive matching degree of each person are used as state information, and a candidate person is determined from the power organization database as an action, an agent is assigned to each person, and a reward function is constructed based on the time, completion degree, and resource waste required for the agent to execute the response plan. With the goal of maximizing the cumulative reward, the skill matching of each person is optimized through the deep Q network to obtain the skill matching optimization results of each person.

[0013] A second aspect of the present invention provides a power safety production system combined with a power organization, comprising: A data acquisition module is used to obtain real-time operation data of power equipment; An anomaly detection module, used to perform anomaly detection on the real-time operation data through a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; A level determination module, used to determine the response plan and hidden danger level of the power equipment according to the abnormal operation data; The personnel matching module is used to match the target personnel from the power organization database based on the hidden danger level through an artificial intelligence matching algorithm and execute the response plan to achieve safe production of the power equipment.

[0014] A third aspect of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the electric power safety production method combined with the electric power organization as described above.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the power production safety method combined with the power organization as described above is implemented.

[0016] Compared with the prior art, the embodiments of the present invention have the following advantages: (1) Through real-time data acquisition and the application of hybrid multi-objective optimization algorithms, it is possible to achieve continuous and efficient monitoring of the operating status of power equipment, detect and handle abnormal situations in a timely manner, and thus improve monitoring efficiency and accuracy; (2) Based on the response plan and hidden danger level determined by abnormal operation data, targeted treatment measures can be formulated to avoid waste of resources and ineffective investment. At the same time, accurate matching of personnel can be achieved through artificial intelligence matching algorithms to ensure the effective implementation of response plans; (3) Through the comprehensive use of advanced technologies and management methods, it is possible to timely discover and deal with safety hazards of power equipment and effectively prevent accidents, thereby enhancing the power company's ability to ensure safe production. The power organization database is combined with the safe production of power equipment, which not only improves the management level of power equipment, but also promotes the innovation and development of power organization work. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of a method for safe production of electric power combined with electric power organization provided by a certain embodiment of the present invention; Figure 2 It is a structural diagram of a power safety production system combined with a power organization provided by a certain embodiment of the present invention; Figure 3 It is a structural diagram of an electronic device provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0021] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0022] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0023] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides a method for safe production of electric power combined with electric power organization, comprising: S1. Obtain real-time operation data of power equipment; S2. Performing anomaly detection on the real-time operation data by using a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; S3. Determine the response plan and hidden danger level of the power equipment according to the abnormal operation data; S4. Based on the hidden danger level, the target personnel are matched from the power organization database through the artificial intelligence matching algorithm and the response plan is executed to achieve safe production of the power equipment.

[0024] Specifically, the present invention uses advanced sensor technology and data acquisition systems to obtain real-time operating data of power equipment. These data include but are not limited to key parameters such as current, voltage, temperature, vibration, etc., and power equipment includes transformers, generators, motors, switchgear, power supplies, etc. Then the hidden danger level and response plan correspond to the number of power equipment, that is, each power equipment has its corresponding hidden danger level and response plan.

[0025] There are multiple challenges in anomaly detection of power equipment: the equipment sensor data is diverse and dynamically changing, and the optimization strategy needs to be adjusted in real time to maintain the detection accuracy and response speed. In this environment, the traditional particle swarm optimization algorithm may fail due to local optimal problems, and although the genetic algorithm has strong global search capabilities, its convergence speed is relatively slow. In order to overcome the shortcomings of the two methods and process various complex and dynamic monitoring data in power equipment, the present invention proposes an innovative solution based on a hybrid multi-objective optimization algorithm, which combines adaptive chaos initialization, dynamic inertia weight and fuzzy logic decision-making, can effectively cope with various complex scenarios in the operation of power equipment, enhance the efficiency and stability of the algorithm, and deal with conflicts and trade-offs between multiple objectives at the same time, thereby detecting anomalies in the real-time operation data of power equipment; wherein, chaos initialization: the traditional particle swarm initialization method may cause the particle swarm to fall into a local optimum, and the chaotic initialization generated by the introduction of chaotic mapping increases the randomness and diversity of the initialized particles, and improves the breadth of the search; dynamic inertia weight: by adaptively adjusting the dynamic inertia weight, it is possible to flexibly switch between global search and local search, especially in the monitoring process of power equipment, the temperature, current, voltage and other data fluctuate greatly, and a flexible search strategy is required to quickly capture abnormal changes; fuzzy logic decision: by introducing fuzzy logic and its corresponding fuzzy rule base, dynamic adjustment is performed for the conflicts between multiple objectives, thereby improving the adaptive ability of the solution, ensuring that the power equipment control system can flexibly cope with different working scenarios (such as abnormality detection under high load and high temperature environment). The present invention utilizes the powerful search capability and optimization performance of the hybrid multi-objective optimization algorithm to accurately and quickly locate equipment failures or potential problems to improve the efficiency of power safety production.

[0026] Then, based on the detected abnormal operation data, combined with the historical data of the power equipment and expert knowledge, the response plan and hidden danger level are determined; among them, the hidden danger level is usually divided into different levels such as emergency, important, general, etc., and can also be expressed by numerical values ​​​​to facilitate the subsequent adoption of corresponding measures.

[0027] Finally, based on the determined hidden danger level, the artificial intelligence matching algorithm is used to screen out personnel matching the target task from the power organization database and implement the response plan, making full use of the personnel information in the power organization database, including professional skills, work experience, emergency handling capabilities, etc., to ensure that the most suitable personnel are assigned to where they are most needed. The present invention combines the power organization database with the safe production of power equipment, which not only improves the management level of power equipment, but also promotes the innovation and development of power organization work.

[0028] In one embodiment, step S2 includes: Generate the initial position of the particle group through chaotic mapping, and initialize the speed of each particle; each particle is a set of detection thresholds for the real-time operation data of the power equipment; A multi-objective function is constructed based on the false alarm rate, response speed and computing resource consumption of the power equipment, and the weight coefficient of each objective is adjusted through the fuzzy logic and its corresponding fuzzy rule base according to preset business requirements to obtain a multi-objective optimization function; The multi-objective optimization function is used as a fitness function to calculate the fitness of each particle, and the individual optimal position and the global optimal position of the particle swarm are updated based on the fitness calculation result, so as to adjust the speed and position of each particle by introducing the particle swarm speed and position update formula of the dynamic inertia weight; Iteratively perform fitness calculation and particle speed and position adjustment processes according to the adjusted particle swarm to obtain several global optimal positions to construct a Pareto solution set, and screen the Pareto solution set through a non-dominated sorting method to obtain an optimal detection threshold; The real-time operation data is subjected to abnormality detection using the optimal detection threshold to obtain abnormal operation data.

[0029] Specifically, the present invention uses a chaotic initialization method to initialize the particle swarm, that is, in order to increase the diversity of the particle swarm and avoid the initial local optimal solution, the initial position of the particle is generated by using Logistic mapping, which is expressed by the following formula: In the formula, is the control parameter of Logistic mapping; is the initial position of the particle, and the value of each position is in the interval (0, 1). For power equipment monitoring, sensor data (such as temperature, vibration, and load) fluctuate greatly. The use of chaotic initialization method can ensure the wide distribution of particles during initialization and avoid over-concentration of the search space.

[0030] The speed of each particle can be randomly generated or initialized according to a certain strategy, usually set to a random value or zero; each particle represents a set of detection thresholds for the real-time operation data of the power equipment to determine whether the data is abnormal. For example, for a transformer, its real-time operation data includes vibration data and temperature data, then the particle corresponding to the transformer is composed of the transformer vibration detection threshold and the transformer temperature detection threshold; similarly, a particle group formed by particles corresponding to the real-time operation data of all power equipment can be obtained.

[0031] When constructing the multi-objective optimization function, the false alarm rate, response speed and computing resource consumption of the power equipment are considered. The function is expressed by the following formula: In the formula, , , They are weight coefficients for false alarm rate, response delay and computing resource consumption. These weight coefficients are dynamically adjusted by fuzzy logic to balance different goals based on the real-time operation data of power equipment (such as equipment load, temperature change, vibration, etc.) and preset business needs. Among them, the adjustment mechanism of weight coefficients includes: false alarm rate refers to the ratio of the power equipment control system misjudging the normal operation equipment as abnormal. When the equipment load is low or the equipment state is relatively stable, the power equipment control system tends to reduce the weight of the false alarm rate to avoid excessive intervention; when the equipment load is heavy (such as high temperature, overload operation, etc.), it is necessary to increase the weight of the false alarm rate to reduce the impact of false alarms on the power equipment control system; response delay refers to the time from the occurrence of equipment failure to the response of the power equipment control system. In the monitoring of power equipment, the shorter the response delay, the more conducive it is to reduce the loss caused by equipment failure. Therefore, when the equipment shows abnormal signs (such as excessive temperature, overload, etc.), it is necessary to optimize the response delay first; the monitoring of power equipment requires processing a large amount of sensor data and real-time calculation, which may cause a large burden on the resources of the power equipment control system, especially when the equipment is running stably, the power equipment control system should minimize the occupation of computing resources to avoid waste.

[0032] The present invention takes the tolerance for false alarm rate, the requirement for response speed, the limitation on computing resource consumption, etc. as business demand indicators, takes the current false alarm rate, current response speed, current computing resource consumption, etc. as status indicators, and converts these two indicators into fuzzy variables through fuzzification processing, and takes the weight coefficient of each target as the output scalar; defines fuzzy sets and membership functions for each input and output variable, such as, for the tolerance for false alarm rate, the fuzzy set can be defined as "low", "medium", "high"; for the requirement for response speed, the fuzzy set can be defined as "slow", "medium", "fast"; for the limitation on computing resource consumption, the fuzzy set can be defined as "loose", "medium", "strict"; the fuzzy set of output variables (weight coefficients) can be defined as "small", "medium", "large"; the membership function is used The invention is to map the specific input value to the membership degree of the fuzzy set, such as triangular function, trapezoidal function and Gaussian function; then by introducing fuzzy logic rules, the power equipment control system can dynamically adjust the weight of each target according to real-time data (such as equipment load, temperature change, etc.). The rule is an "IF-THEN" rule set designed based on expert experience and business needs, such as IF the temperature is too high AND the load is high THEN increase the weight of response delay and reduce the weight of computing resource consumption; IF the equipment is running normally AND the vibration is normal THEN increase the weight of false alarm rate and reduce the weight of response delay and computing resource consumption; finally, according to the current value of the input variable and the fuzzy rule base, the Mamdani method and Sugeno method are used to calculate the fuzzy value of the output variable and defuzzify it to obtain the weight coefficient of each target. Through this dynamic adjustment mechanism, the present invention can adaptively balance the three optimization goals and ensure the accuracy and efficiency of power equipment monitoring.

[0033] The present invention uses a multi-objective optimization function as a fitness function to calculate the fitness value of each particle, then updates the individual optimal position (pbest) and the global optimal position (gbest) of the particle swarm according to the fitness calculation results, and introduces a dynamic inertia weight to adjust the speed and position update formula of the particle swarm; wherein the dynamic inertia weight is expressed by the following formula: In the formula, is the dynamic inertia weight; , are the minimum and maximum values ​​of the dynamic inertia weight respectively; is the attenuation coefficient; t is the number of iterations; is the aggregation degree of the particle group; is the adjustment factor for the effect of aggregation; is the total number of particle groups; is the position of the ith particle; is the particle group center; is the Euclidean distance.

[0034] Since the inertia weight controlling the search behavior of particles may lead to an overly extensive search and fall into the local optimum, the present invention uses dynamic inertia weight to balance the global and local searches to improve the convergence speed and accuracy of the algorithm; wherein the speed and position update formulas introducing the dynamic inertia weight are as follows: In the formula, v i is the velocity of the ith particle; c1 and c2 are learning factors; r1 and r2 are random numbers.

[0035] Then, according to the adjusted particle swarm, the fitness calculation and the particle speed and position adjustment process are iteratively performed; during the iteration process, the global optimal position is continuously updated until the stopping condition is met (such as reaching the maximum number of iterations or the fitness change is less than the threshold); through multiple iterations, several global optimal positions are obtained and constructed into a Pareto solution set. Each solution in the Pareto solution set is a non-dominated solution to the multi-objective optimization problem, that is, it is better than other solutions in a certain goal, but not inferior to other solutions in other goals; the Pareto solution set is screened using the non-dominated sorting method to obtain the optimal detection threshold to retain the solutions on the Pareto frontier and ensure the diversity and optimality of the solution set; finally, the optimal detection threshold is used to perform anomaly detection on the real-time operation data to screen out abnormal operation data.

[0036] Taking transformer abnormality monitoring as an example, assuming that the monitoring objectives are: false alarm rate ≤ 5%, response delay ≤ 2 seconds, and CPU occupancy rate ≤ 30%, a hybrid multi-objective optimization algorithm is applied in transformer monitoring. The weights of each optimization objective are dynamically adjusted according to the actual operating status of the power equipment (such as equipment vibration, temperature change, etc.). Finally, due to frequent abnormal transformer vibration, the vibration weight is increased from 0.6 to 0.7, and the detection threshold sensitivity is reduced from 1.5 to 1.2, in order to reduce false alarms of minor abnormalities and ensure timely response to potential faults.

[0037] The present invention initializes a particle swarm through chaotic mapping, combines multi-objective optimization and fuzzy logic to dynamically adjust weights, and uses a particle swarm algorithm for iterative optimization to ultimately obtain an optimal detection threshold, which not only effectively improves the accuracy, real-time performance, and resource efficiency of abnormal detection of power equipment, but also optimizes the utilization of computing resources, and has high practical value; at the same time, combined with the actual needs of power equipment monitoring, the power equipment control system can adaptively adjust the weights of the optimization targets in real time, ensuring good comprehensive performance under different operating conditions.

[0038] In one embodiment, step S3 includes: The abnormal operation data is processed by a risk assessment model to obtain a comprehensive risk score, and a response plan for the power equipment is determined based on the comprehensive risk score; wherein the risk assessment model is represented by the following formula: In the formula, To generate a comprehensive risk score; is the importance factor of the nth power equipment, evaluated according to the criticality of the equipment and its influence on the power equipment control system; is the state correction factor of the nth power equipment, and adjusts the risk factor of the power equipment based on the real-time monitoring data; is the risk adjustment coefficient, which usually takes into account factors such as the historical failure record of power equipment and the service life of the equipment; M is the total number of power equipment.

[0039] Specifically, after detecting an abnormality, the present invention generates processing suggestions based on the comprehensive risk score calculated by the risk assessment model, such as immediate shutdown inspection, replacement of damaged parts, adjustment of operating procedures (adjustment of illegal or non-compliant behaviors), etc., to ensure equipment safety. Assuming that a certain power production equipment (such as a transformer) is undergoing risk assessment, the system monitors the following factors: the importance factor of the transformer is set to 5 according to the key role of the transformer in the power system; the fault risk of the transformer is set to 0.8 according to the real-time monitoring data (the equipment temperature and vibration indicators are both beyond the normal range), then through the formula calculation, the comprehensive risk score of the transformer is: 5*0.8=4, this result shows that the risk score of the transformer is high, suggesting that immediate measures need to be taken for inspection and repair; among them, various response plans correspond to a comprehensive risk score range. When the calculated comprehensive risk score falls into a certain range, the response plan to which the range belongs is executed. The specific range division can be determined based on historical data, which is not detailed as the focus here. In addition, once an abnormal warning is triggered, the power equipment control system will automatically generate a hidden danger report and push it to relevant management personnel in real time through the linkage platform; the warning report will include detailed information such as the type, severity, time of occurrence, location, etc. of the hidden danger. The system will automatically match appropriate personnel for processing based on the severity and type of the warning information.

[0040] The risk assessment model of the present invention comprehensively considers the severity and probability of hidden dangers, provides a scientific comprehensive risk score, and helps decision makers fully understand equipment risks; based on risk levels and hidden danger types, formulates reasonable response plans to ensure the effectiveness and pertinence of response measures.

[0041] In one embodiment, step S3 further includes: Determine the hidden danger type of the power equipment according to the abnormal operation data, quantify the hidden danger level of the power equipment through a hidden danger assessment model, and calculate the severity of each hidden danger type to obtain a hidden danger severity score; wherein the hidden danger assessment model is represented by the following formula: In the formula, is the level of hidden danger; is the importance weight of the ath hidden danger type; is the processing weight of the ath hidden danger type; A is the total number of hidden danger types.

[0042] Specifically, the present invention divides the hidden danger types into several categories according to the characteristics of abnormal operation data, such as electrical hidden dangers: such as short circuit, leakage, overload, etc.; mechanical hidden dangers: such as equipment wear, abnormal vibration, etc.; environmental hidden dangers: such as excessive temperature, excessive humidity, etc.; control hidden dangers: such as signal loss, control failure, etc.; the hidden danger type (such as equipment failure, operating error, safety violation, etc.) can also be determined based on the response plan, and then the hidden danger level of the power equipment can be quantified through the hidden danger assessment model, and a severity score (S) is assigned to each hidden danger according to the impact of the hidden danger. The calculation formula is as follows: In the formula, They are preset weights, which respectively reflect the importance of power equipment, the scope of faults and the impact of production interruptions. Their weights and respective adjustment objects can be determined based on historical fault data.

[0043] The present invention can accurately identify potential hidden dangers of power equipment through analysis of abnormal operation data and classification of hidden danger types. The hidden danger assessment model can quantify the hidden danger level and provide clear scoring standards to facilitate decision makers' understanding and management.

[0044] In one embodiment, step S4 includes: Acquire the real-time environmental data of the electric power equipment and the basic data of each person in the electric power organization database, and construct the hidden danger level, the real-time environmental data and each basic data as input features; The input features are processed by a deep learning network model constructed by a multi-layer perceptron to obtain the skill matching degree of each person, and the skill matching degree of each person is optimized by using a machine learning algorithm and a reinforcement learning algorithm to obtain the skill matching optimization result of each person; A multi-objective matching function model is constructed based on the skill matching optimization result and the hidden danger severity score, and a particle swarm optimization algorithm is used to perform a global search on the multi-objective matching function model to obtain the target personnel and execute the response plan to achieve safe production of the power equipment.

[0045] Specifically, for the matching of tasks and processing personnel, the present invention designs a multi-layer nested artificial intelligence matching framework, adopts deep learning network (multi-layer perceptron, MLP) scoring, combines the anomaly detection algorithm based on isolation forest, and reinforcement learning (RL) optimization allocation strategy, and matches the response plan tasks with each person in the power organization database to improve the adaptability of the matching model; wherein, the power organization database stores each person's skill label, team, historical tasks, qualification level, task feedback index, responsibility fulfillment index, etc.

[0046] First, the skill labels of each person are retrieved from the power organization database, that is, basic data, such as maintenance capabilities: mechanical maintenance, electrical maintenance, structural fault repair, electrical expertise: circuit design analysis, sensor debugging, etc., hidden danger analysis experience: hidden danger warning analysis, historical processing records, etc.; combined with the hidden danger level: the emergency response level required by the task (such as high risk, medium risk, low risk) and the qualification level implied in the response plan: the task's requirements for professional certificates and work experience, and contextual features: the current team structure (such as the skill composition of other members of the team) and the real-time environmental data of the power equipment: such as temperature, humidity, air pressure, vibration, etc., and constructed as input features, the input features On the basis of basic features, more dimensions are introduced to increase the learning ability of the model.

[0047] The input features are then processed by a deep learning network model constructed using multi-layer perceptrons to predict the skill match of the handler. ; Among them, the deep learning network model includes an input layer (accepting feature vectors ), hidden layer (using ReLU activation function: in are the weights and biases of the i-th layer) and the output layer (output matching score ), and the model is trained in combination with historical data (including hidden danger levels, environmental data, basic personnel data and actual matching results). The loss function is defined as the mean square error between the predicted value and the true score. The weight parameters are updated through the back propagation algorithm, and the skill matching degree of each person is predicted by the trained deep learning network model; then the skill matching degree is preliminarily optimized using the machine learning algorithm to improve the accuracy of the matching degree through feature selection and hyperparameter tuning, and the skill matching degree is further optimized using the reinforcement learning algorithm. The goal of reinforcement learning is to maximize the matching effect, and the skill matching optimization results of each person are obtained by continuously interacting with the environment (simulating matching scenarios) and adjusting the matching strategy.

[0048] Finally, a multi-objective optimization algorithm is introduced in the matching and optimization process to make the system more flexible and efficient. The present invention comprehensively considers factors such as skill matching, task priority, and team workload to define a multi-objective matching function model. : In the formula, , , , They are weight coefficients of their corresponding targets, which can be determined according to business needs and historical data; Score the completion of historical tasks; For current staff workload.

[0049] The particle swarm optimization algorithm is used to perform a global search on the multi-objective matching function model to find the best allocation solution. When the improvement is lower than the threshold or the number of iterations reaches the set value, the algorithm is stopped, the target personnel corresponding to each response plan are obtained and the response plan is executed to achieve safe production of power equipment.

[0050] Assume that the task queue contains three high-risk tasks and the power organization database contains five people, whose skill scores, workloads, and historical completion scores are as follows: Task characteristics: Task 1: Hidden danger level 9, abnormal score 0.8; Task 2: Hidden danger level 7, abnormal score 0.6; Task 3: Hidden danger level 6, abnormal score 0.4; Personnel characteristics: , ; After calculation by the algorithm described in the present invention, the final matching result between tasks and personnel is: Task 1 is assigned to Personnel 4; Task 2 is assigned to Personnel 1; Task 3 is assigned to Personnel 2.

[0051] The present invention uses a deep learning network model and an optimization algorithm to accurately match personnel skills with hidden danger handling requirements and improve handling efficiency; a multi-objective matching function model is constructed by comprehensively considering skill matching and hidden danger severity scores to ensure the comprehensiveness and scientificity of the matching solution; the particle swarm optimization algorithm used can search for the optimal matching solution in a global range to avoid falling into a local optimal solution; the reinforcement learning algorithm used can dynamically adjust the matching strategy according to the actual matching effect to improve the adaptability and robustness of the system; through accurate matching and rapid response, the hidden danger risk can be effectively reduced to ensure the safe production of power equipment; the automated matching and optimization process reduces manual intervention and improves the efficiency and accuracy of hidden danger handling; the scheme achieves accurate matching and efficient response to hidden danger handling of power equipment through data acquisition, feature construction, deep learning models, multi-objective matching functions and particle swarm optimization algorithms, is scientific, practical and efficient, and can effectively improve the safe production level of power equipment.

[0052] In one embodiment, the skill matching degree of each person is optimized by using a machine learning algorithm and a reinforcement learning algorithm to obtain the skill matching optimization result of each person, including: The hidden danger severity score is processed by an isolation forest algorithm to obtain an optimized score, and the skill matching degree of each person is adjusted according to the optimized score to obtain a comprehensive matching degree of each person; The hidden danger level and the comprehensive matching degree of each person are used as state information, and a candidate person is determined from the power organization database as an action, an agent is assigned to each person, and a reward function is constructed based on the time, completion degree, and resource waste required for the agent to execute the response plan. With the goal of maximizing the cumulative reward, the skill matching of each person is optimized through the deep Q network to obtain the skill matching optimization results of each person.

[0053] Specifically, the present invention uses the isolation forest algorithm to perform anomaly auxiliary detection on the task hidden danger score to ensure that high-risk tasks are matched first. By constructing a random tree structure, the optimization score of each data point is calculated (optimization score = path length / sample depth). The higher the score, the more likely the task is a hidden danger hotspot. The hidden danger severity score is used as input, and the isolation forest algorithm is used to calculate the optimization score of each score. The hidden danger severity score is adjusted according to the optimization score to calculate the comprehensive matching degree. ;in, is the adjustment coefficient.

[0054] Then, reinforcement learning is used to dynamically optimize the matching results to improve the allocation efficiency in complex task scenarios. The time limit corresponding to the hidden danger level and the response plan is used as the task characteristics, the comprehensive matching degree and workload of each person are used as the person status, and the task characteristics and the person status are used as the state (State, S); the optimal processing personnel are selected from the power organization database based on the skill matching degree of each person. , indicating that the i-th person is assigned to the task as an action (Action, ), such as plan 1 selects person A, plan 2 selects person B, plan 3: selects person C, etc., and assigns an agent to each person. Each agent is responsible for selecting and executing the response plan, such as agent A: responsible for matching and task execution of person A, agent B: responsible for matching and task execution of person B, agent C: responsible for matching and task execution of person C, etc.; and calculates the reward value based on the effect of the allocation result, the reward function as follows: Finally, the deep Q learning algorithm is used to optimize the skill matching of each person with the goal of maximizing the cumulative reward, and the skill matching optimization result of each person is obtained. The optimization process is expressed by the following formula: In the formula, is the discount factor.

[0055] When using the deep Q network to optimize the strategy of each agent, its input is state information, and its output is the Q value of each action (i.e., expected cumulative reward). By continuously interacting with the environment, the Q value is updated and the strategy is gradually optimized. The deep Q network training process includes: initializing the network parameters and experience replay buffer of DQN; for each agent: selecting an action according to the current state (using the ε-greedy strategy; executing the action, observing the reward and the next state; storing the experience (state, action, reward, next state) in the experience replay buffer; randomly sampling a batch of experience from the experience replay buffer and updating the network parameters of DQN; repeating the above process until DQN converges, and finally optimizing the skill matching of each person through DQN to obtain the skill matching optimization result of each person.

[0056] The present invention optimizes hidden danger severity scores through the isolation forest algorithm, reduces scoring bias, and improves scoring reliability; adjusts skill matching according to the optimized hidden danger scores to ensure that high-risk hidden dangers are handled by highly skilled personnel; through the design of intelligent agents and reward functions, it is possible to comprehensively consider task completion, time and resource waste, and optimize matching strategies; uses a deep Q network to optimize matching strategies, and is able to dynamically adjust strategies to adapt to different hidden danger handling scenarios; through deep learning and reinforcement learning algorithms, it is possible to accurately match personnel skills with hidden danger handling requirements and improve handling efficiency; and by optimizing matching strategies, it is possible to reduce resource waste and improve resource utilization.

[0057] In the embodiment of the present application, based on the problem of how to improve the existing power safety production efficiency, a power safety production method combined with a power organization is designed, which uses a hybrid multi-objective optimization algorithm formed by a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic to perform anomaly detection on the real-time operating data of each power equipment, thereby improving the detection accuracy and response speed, and optimizing the utilization of computing resources, ensuring that the power equipment can maintain good comprehensive performance under different operating conditions; at the same time, a multi-layer nested artificial intelligence matching framework is designed, which uses deep learning network scoring, combined with an anomaly detection algorithm based on isolation forest, and a reinforcement learning optimization allocation strategy, and matches the response plan tasks with the personnel in the power organization database to enhance the adaptive ability of the matching model, thereby greatly improving the safety production efficiency of the power equipment and reducing the social impact and economic losses caused by power accidents.

[0058] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0059] In another embodiment, if Figure 2 As shown, the second aspect of the present invention provides a power safety production system combined with a power organization, comprising: The data acquisition module 10 is used to acquire real-time operation data of the power equipment; An anomaly detection module 20 is used to perform anomaly detection on the real-time operation data through a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; A level determination module 30, used to determine a response plan and a hidden danger level of the power equipment according to the abnormal operation data; The personnel matching module 40 is used to match the target personnel from the power organization database based on the hidden danger level through an artificial intelligence matching algorithm and execute the response plan to achieve safe production of the power equipment.

[0060] It should be noted that each module in the above-mentioned power safety production system combined with a power organization can be fully or partially implemented by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of a power safety production system combined with a power organization, please refer to the definition of a power safety production method combined with a power organization above. The two have the same functions and effects, which will not be repeated here.

[0061] A third aspect of the present invention provides an electronic device, the electronic device comprising: processor, memory, and bus; The bus is used to connect the processor and the memory; The memory is used to store operation instructions; The processor is used to call the operation instruction, and the executable instruction enables the processor to perform operations corresponding to a method for safe power production combined with power organization as shown in the first aspect of the present application.

[0062] In an alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3 The electronic device 5000 shown includes: a processor 5001 and a memory 5003. The processor 5001 and the memory 5003 are connected, such as through a bus 5002. Optionally, the electronic device 5000 may also include a transceiver 5004. It should be noted that in actual applications, the transceiver 5004 is not limited to one, and the structure of the electronic device 5000 does not constitute a limitation on the embodiments of the present application.

[0063] Processor 5001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 5001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0064] The bus 5002 may include a path to transmit information between the above components. The bus 5002 may be a PCI bus or an EISA bus, etc. The bus 5002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0065] The memory 5003 may be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, a CD-ROM or other optical disk storage, an optical disk storage (including a compressed optical disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0066] The memory 5003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 5001. The processor 5001 is used to execute the application code stored in the memory 5003 to implement the content shown in any of the above method embodiments.

[0067] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc.

[0068] The fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements a method for safe production of electric power combined with an electric power organization as shown in the first aspect of the present application.

[0069] Another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0070] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented.

[0071] In summary, the present invention relates to the field of information technology, and discloses a method, system, equipment and medium for safe production of electric power combined with an electric power organization, the method comprising acquiring real-time operation data of electric power equipment; performing anomaly detection on the real-time operation data through a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; determining a response plan and a hidden danger level of the electric power equipment according to the abnormal operation data; based on the hidden danger level, matching target personnel from the electric power organization database through an artificial intelligence matching algorithm and executing the response plan to achieve safe production of the electric power equipment, thereby greatly improving the safe production efficiency of the electric power equipment and reducing the social impact and economic losses caused by power accidents.

[0072] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for safe production of electric power combined with electric power organization, characterized in that: include: Obtain real-time operating data of power equipment; Performing anomaly detection on the real-time operation data by using a hybrid multi-objective optimization algorithm to obtain abnormal operation data; The hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; Determine a response plan and a hidden danger level of the power equipment according to the abnormal operation data; Based on the hidden danger level, the target personnel are matched from the power organization database through the artificial intelligence matching algorithm and the response plan is executed to achieve safe production of the power equipment.

2. A method for safe production of electric power combined with electric power organization according to claim 1, characterized in that: The performing anomaly detection on the real-time operation data by using a hybrid multi-objective optimization algorithm to obtain abnormal operation data includes: Generate the initial position of the particle group through chaotic mapping, and initialize the speed of each particle; each particle is a set of detection thresholds for the real-time operation data of the power equipment; A multi-objective function is constructed based on the false alarm rate, response speed and computing resource consumption of the power equipment, and the weight coefficient of each objective is adjusted through the fuzzy logic and its corresponding fuzzy rule base according to preset business requirements to obtain a multi-objective optimization function; The multi-objective optimization function is used as a fitness function to calculate the fitness of each particle, and the individual optimal position and the global optimal position of the particle swarm are updated based on the fitness calculation result, so as to adjust the speed and position of each particle by introducing the particle swarm speed and position update formula of the dynamic inertia weight; Iteratively perform fitness calculation and particle speed and position adjustment processes according to the adjusted particle swarm to obtain several global optimal positions to construct a Pareto solution set, and screen the Pareto solution set through a non-dominated sorting method to obtain an optimal detection threshold; The real-time operation data is subjected to abnormality detection using the optimal detection threshold to obtain abnormal operation data.

3. A method for safe production of electric power combined with electric power organization according to claim 2, characterized in that: The dynamic inertia weight is expressed by the following formula: In the formula, is the dynamic inertia weight; , are the minimum and maximum values ​​of the dynamic inertia weight respectively; is the attenuation coefficient; t is the number of iterations; is the aggregation degree of the particle group; is the adjustment factor for the effect of aggregation; is the total number of particle groups; is the position of the ith particle; is the particle group center; is the Euclidean distance.

4. A method for safe production of electric power combined with electric power organization according to claim 1, characterized in that: The determining of the response plan and hidden danger level of the electric power equipment according to the abnormal operation data includes: The abnormal operation data is processed by a risk assessment model to obtain a comprehensive risk score, and a response plan for the power equipment is determined based on the comprehensive risk score; wherein the risk assessment model is represented by the following formula: In the formula, To generate a comprehensive risk score; is the importance factor of the nth power equipment; is the state correction factor of the nth power equipment; is the risk adjustment coefficient; M is the total number of power equipment.

5. The method for safe production of electric power combined with electric power organization according to claim 1, characterized in that: The determining of the response plan and hidden danger level of the electric power equipment according to the abnormal operation data also includes: Determine the hidden danger type of the power equipment according to the abnormal operation data, quantify the hidden danger level of the power equipment through a hidden danger assessment model, and calculate the severity of each hidden danger type to obtain a hidden danger severity score; wherein the hidden danger assessment model is represented by the following formula: In the formula, is the level of hidden danger; is the importance weight of the ath hidden danger type; is the processing weight of the ath hidden danger type; A is the total number of hidden danger types.

6. A method for safe production of electric power combined with electric power organization according to claim 5, characterized in that: Based on the hidden danger level, the target personnel are matched from the power organization database through an artificial intelligence matching algorithm and the response plan is executed to achieve safe production of the power equipment, including: Acquire the real-time environmental data of the electric power equipment and the basic data of each person in the electric power organization database, and construct the hidden danger level, the real-time environmental data and each basic data as input features; The input features are processed by a deep learning network model constructed by a multi-layer perceptron to obtain the skill matching degree of each person, and the skill matching degree of each person is optimized by using a machine learning algorithm and a reinforcement learning algorithm to obtain the skill matching optimization result of each person; A multi-objective matching function model is constructed based on the skill matching optimization result and the hidden danger severity score, and a particle swarm optimization algorithm is used to perform a global search on the multi-objective matching function model to obtain target personnel and execute the response plan to achieve safe production of the power equipment.

7. A method for safe production of electric power combined with electric power organization according to claim 6, characterized in that: The machine learning algorithm and the reinforcement learning algorithm are used to optimize the skill matching of each person, and the skill matching optimization results of each person are obtained, including: The hidden danger severity score is processed by an isolation forest algorithm to obtain an optimized score, and the skill matching degree of each person is adjusted according to the optimized score to obtain a comprehensive matching degree of each person; The hidden danger level and the comprehensive matching degree of each person are used as state information, and a candidate person is determined from the power organization database as an action, an agent is assigned to each person, and a reward function is constructed based on the time, completion degree, and resource waste required for the agent to execute the response plan. With the goal of maximizing the cumulative reward, the skill matching of each person is optimized through the deep Q network to obtain the skill matching optimization results of each person.

8. A power safety production system combined with a power organization, characterized in that: include: A data acquisition module is used to obtain real-time operation data of power equipment; An anomaly detection module, used to perform anomaly detection on the real-time operation data through a hybrid multi-objective optimization algorithm to obtain abnormal operation data; the hybrid multi-objective optimization algorithm is a particle swarm algorithm that combines chaos initialization, dynamic inertia weight and fuzzy logic; A level determination module, used to determine the response plan and hidden danger level of the power equipment according to the abnormal operation data; The personnel matching module is used to match the target personnel from the power organization database based on the hidden danger level through an artificial intelligence matching algorithm and execute the response plan to achieve safe production of the power equipment.

9. An electronic device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the power safety production method combined with power organization as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the power safety production method combined with power organization as described in any one of claims 1 to 7 is implemented.

Images