A work area safety early warning method

Abstract

The application discloses a kind of working area safety early warning method, it is related to safety early warning technical field.The method includes: collecting worker operation trajectory and interaction record in working area, relevant coefficient is calculated by behavior correlation analysis, extract risk preference coefficient, potential synergy coefficient, and construct behavior characteristic mapping library;Real-time interaction of worker is split into mimic and collaborative behavior, and the total value and concentration coefficient of group mimic and collaboration are calculated, and collaborative mimic individual is screened, and group synchronization model is constructed based on mean field theory, and the rule violation synchronization rate of mimic track and collaborative track is split;Dynamic game model is constructed by compliance hedging coefficient, and the advantage of rule violation or compliance game is judged;When compliance is superior, screening early warning model is constructed based on decision tree, and invalid early warning under the support of compliance is filtered;The application is favorable for positioning group rule violation transmission source, reduces invalid early warning rate, shortens early warning response time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety warning, in particular to a working area safety warning method. BACKGROUND

[0002] The safety warning of working area (especially the area involving high-risk operation) is the key of safety management. Although the existing technology has realized automatic monitoring through video monitoring identification and sensor collection, and can obtain worker operation trajectory, illegal behavior data and give a warning based on static threshold, there are still significant limitations.

[0003] The existing technology only focuses on individual illegal identification and cannot deal with illegal transmission in group interaction. In the interoperation, workers have imitation and cooperative behavior. The existing technology cannot separate these two types of behavior, cannot identify the core individual leading to illegal transmission, and it is more difficult to quantify the influence of double tracks on group illegal synchronization. The existing technology often gives a warning after the group illegal behavior forms a scale, which is obviously lagging.

[0004] The existing technology does not consider the hedging effect of compliance behavior. The teaching and reminding of compliant workers can inhibit the spread of illegal behavior, but the existing technology only triggers a warning based on illegal data. When compliance can offset the risk, the existing technology still gives a warning, resulting in many invalid warnings, wasting management resources and reducing the response efficiency of real high-risk warnings.

[0005] Therefore, the present application provides a working area safety warning method. SUMMARY

[0006] The present application aims to provide a working area safety warning method to solve at least one of the above-mentioned problems of the prior art.

[0007] A working area safety warning method, comprising the following steps:

[0008] Collecting worker operation trajectory and interaction record data of the working area, mining the implicit features of the operation trajectory and the interaction record through behavior association analysis, and constructing a behavior feature mapping library;

[0009] Based on the behavior feature mapping library, the real-time interaction of the workers is separated into imitation and cooperative behavior, and the cooperative imitation individuals are screened. Based on the cooperative imitation individuals, a group synchronization model is constructed to extract the illegal synchronization rate. The illegal synchronization rate is double-tracked and separated to obtain the illegal synchronization rate of the imitation track and the cooperative track;

[0010] Based on the illegal synchronization rate of the imitation track and the cooperative track, the compliance behavior hedging analysis is performed to obtain the compliance hedging coefficient of the imitation track and the cooperative track and construct a dynamic game model to determine the game advantage type;

[0011] If compliance hedging is dominant, a screening early warning model is constructed, violation synchronous judgment is identified and optimized, and invalid early warning under compliance support is filtered.

[0012] As a further technical solution of the application, the behavior correlation analysis is used as follows:

[0013] The operation trajectory data and interaction record data of workers in high-risk areas are collected, the high-risk stay ratio and violation correction rate in the operation trajectory data are extracted, and the correlation coefficients of the high-risk stay ratio and the violation correction rate are calculated ;

[0014] The imitated rate and the collaborative dominance in the interaction record data are extracted, the non-imitated rate is obtained by negative conversion of the imitated rate, and the correlation coefficients of the non-imitated rate and the collaborative dominance are calculated ;

[0015] Based on the correlation coefficient and the correlation coefficient , a verification process is performed;

[0016] If the correlation coefficient and the correlation coefficient are verified, the gradient weight hidden feature extraction is performed on the worker operation trajectory data and the interaction record data, and the risk preference coefficient and the potential collaboration coefficient are obtained;

[0017] Based on the risk preference coefficient and the potential collaboration coefficient, the operation trajectory data and the interaction record data are field-split to construct a feature mapping library.

[0018] As a further technical solution of the application, the double-track splitting is performed as follows:

[0019] Based on the constructed group synchronization model, the violation synchronization rate is extracted and it is judged whether the violation is out of limit, if it is out of limit, the imitation track splitting equation and the collaborative track splitting equation are constructed;

[0020] The imitation field strength and the collaborative field strength of the group synchronization model are obtained, and the corresponding imitation track splitting equation and collaborative track splitting equation are inputted combined with the violation synchronization rate, to obtain the violation synchronization rate of the imitation track and the violation synchronization rate of the collaborative track.

[0021] As a further technical solution of the application, the group synchronization model is constructed as follows:

[0022] The individual violation state and the average field variable of all workers corresponding to the collaborative imitation individual are obtained;

[0023] Based on the average field variable, the imitation field strength and the collaborative field strength are extracted;

[0024] The mean field equation is established to input the imitated field strength and the cooperative field strength, and a violation synchronization rate of comprehensive violation is obtained.

[0025] As a further technical solution of the present application, the way to obtain the cooperative imitated individual is:

[0026] According to a fixed monitoring period, time slices are divided, and the interaction records of workers in the time slices are split into imitated behaviors and cooperative behaviors;

[0027] Based on the imitated behaviors and the cooperative behaviors, and combined with a behavior feature mapping library, imitated track transmission strength and cooperative track transmission strength of the workers are extracted;

[0028] Based on the imitated track transmission strength and the cooperative track transmission strength, centralized trend analysis is performed to obtain imitated centralized coefficients and cooperative centralized coefficients;

[0029] Based on the imitated centralized coefficients, imitated core individuals among the workers are screened, and based on the cooperative centralized coefficients, cooperative core individuals among the workers are screened;

[0030] Workers that simultaneously satisfy the cooperative core individuals and the imitated core individuals are obtained as cooperative imitated individuals.

[0031] As a further technical solution of the present application, the way to extract the imitated field strength and the cooperative field strength is:

[0032] An imitated field strength equation is constructed to obtain imitated chain synchronization rates and group total value parameters, and combined with mean field variable, the imitated field strength equation is input to obtain the imitated field strength;

[0033] A cooperative field strength equation is constructed to obtain cooperative circle synchronization rates and group total value parameters, and combined with mean field variable, the cooperative field strength equation is input to obtain the cooperative field strength.

[0034] As a further technical solution of the present application, the way to determine the game advantage type is:

[0035] Based on the violation synchronization rates of the imitated track and the cooperative track, compliance behavior hedging analysis is performed to obtain compliance hedging coefficients of the imitated track and the cooperative track;

[0036] Combined with the imitated track compliance hedging coefficient and the cooperative track compliance hedging coefficient, game characteristics of both parties of a game model are established;

[0037] Based on the game characteristics of both parties of the game, a game potential equation is established to output game potential, and the game potential is compared and analyzed to determine the game advantage type.

[0038] As a further technical solution of the present application, the way to perform the compliance behavior hedging analysis is:

[0039] Establish a compliance behavior hedging equation, and obtain the compliance hedging coefficients for the imitation track and the compliance hedging coefficients for the collaborative track.

[0040] As a further technical solution of the present invention, the method for constructing the screening reduction early warning model is as follows:

[0041] Based on the game potential energy output from the game potential energy equation, the advantage potential energy difference is decomposed to locate the weak trajectory.

[0042] Develop strategies to strengthen weak links and identify whether compliance scenarios are subject to excessive warnings;

[0043] If there are excessive warnings, a screening and reduction warning model will be built to optimize the simultaneous judgment of violations and filter out invalid warnings supported by compliance.

[0044] As a further technical solution of the present invention, the method for locating the weak track is as follows:

[0045] Obtain the difference between the compliance potential energy and violation potential energy of the imitation track and the compliance potential energy and violation potential energy of the collaborative track, conduct comparative analysis, and identify the weak track.

[0046] The beneficial effects of this invention are:

[0047] 1. Collect worker operation trajectory data and interaction record data in the work area. Through behavioral correlation analysis, uncover implicit features between the two types of data. Based on the verified correlations, extract risk preference coefficients and potential collaboration coefficients. Finally, integrate basic fields, related fields, and feature fields to form a behavioral feature mapping library. This provides comprehensive and logically connected data support for subsequent group violation analysis and early warning, reducing subsequent analytical biases caused by incomplete or isolated data.

[0048] 2. Worker real-time interactions are broken down into two core behaviors: imitation and collaboration. Collaborative imitation individuals possessing both core imitation and collaborative attributes are selected, and a group synchronization model is constructed. The overall violation synchronization rate is then broken down into the violation synchronization rates of the imitation track and the collaboration track. This approach helps identify key individuals playing a crucial role in transmitting group violations, while also clarifying the contributions of the imitation and collaboration paths to violation synchronization. This provides direction for subsequent targeted analysis of violation risks along different paths and reduces the overall ambiguity in judging group violations.

[0049] 3、Based on the double-track violation synchronization rate, first calculate the compliance hedging coefficient of the simulation track and the collaborative track through the compliance hedging equation, then build a dynamic game model with the violation party and the compliance party as the two parties of the game, and determine the game advantage type by calculating and comparing the game potential energy of the two parties. The way of incorporating violation propagation and compliance hedging into the dynamic game framework is not only to trigger early warning directly according to violation data, but also to consider the offsetting effect of compliance behavior on violation diffusion, which is helpful to reflect the actual risk state in the work area, identify the trend of violation propagation, provide more practical judgment basis for subsequent whether to need early warning and how to handle, and reduce the risk misjudgment caused by not considering compliance hedging.

[0050] 4、Through the weak track positioning and directional reinforcement strategy, the weak link in the group compliance system can be targeted for optimization, rather than generalized to improve overall compliance, which helps to more efficiently strengthen the group compliance ability; The construction of the screening early warning model can identify and filter excessive early warning, reduce management resource waste, and through dynamic rule iteration to adapt to behavior state changes, so that the early warning result is more suitable for the real-time risk and compliance balance state, improving the accuracy and practicality of early warning. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0052] Figure 1 is a flowchart of a work area safety early warning method provided by the present application;

[0053] Figure 2 is a judgment flowchart of whether to exceed the limit provided by embodiment two of the present application;

[0054] Figure 3 is a module diagram of a work area safety early warning system provided by embodiment three of the present application. DETAILED DESCRIPTION

[0055] In order to make the personnel in the technical field better understand the present application scheme, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0056] Embodiment 1:

[0057] As Figure 1 shown, the working area safety warning method provided by the embodiment of the application specifically comprises the following steps:

[0058] S1, collecting worker operation trajectories and interaction records of the working area, mining implicit features of the operation trajectories and the interaction records through behavior correlation analysis, and constructing a behavior feature mapping library;

[0059] The way of collecting the worker operation trajectories and the interaction records of the working area is:

[0060] Preferably, through historical video monitoring images of the working area, operation trajectory data and interaction record data of workers in pre-divided high-risk areas in the working area in a historical monitoring period are recognized through an OpenCV machine learning algorithm framework;

[0061] The operation trajectory data includes a high-risk stay ratio H of a time length of a single worker in the high-risk area to a total work time, and a violation correction rate M of a violation operation;

[0062] The interaction record data includes a behavior imitation rate C and a collaborative dominance L;

[0063] It should be noted that the violation correction rate can be obtained by the ratio of the number of corrected violation operations of a worker in a monitoring period to the total number of violation operations of the worker in the monitoring period after the worker corrects the behavior after the first violation operation (which is automatically recognized by the system);

[0064] The imitation rate is obtained by comparing the operation log with video behavior analysis, identifying the matching of operation actions of other workers and a target worker, and counting the total number of times that other workers copy the operation of the target worker, and then comparing it with the total operation number of the target worker;

[0065] The collaborative dominance is obtained by judging the action initiator through action timing records (such as device operation trigger sequence and limb action start time) in a two-person collaborative operation scene, and counting the proportion of the number of actions initiated by the target worker to the total number of two-person collaborative actions;

[0066] The way of mining implicit features of the operation trajectories and the interaction records through behavior correlation analysis is:

[0067] Collecting worker operation trajectory data and interaction record data in N historical monitoring periods;

[0068] Wherein, N is the number of monitoring periods, N≥30;

[0069] The correlation coefficient of the high-risk stay ratio H and the violation correction rate M in the N monitoring periods is calculated through Formula One: ; ​

[0070] wherein, , respectively represent the covariance of high-risk stay ratio H and violation correction rate M, the standard deviation of H, the standard deviation of M;

[0071] The non-imitated rate C in N monitoring periods is calculated by formula two: and the correlation coefficient of the synergy dominance degree L; ;

[0072] wherein, , respectively represent the covariance of the non-imitated rate C and the synergy dominance degree L, the standard deviation of C, the standard deviation of L; It should be noted that the non-imitated rate is obtained by negatively transforming the imitated rate (1-C); the correlation coefficient of the non-imitated rate (1-C) and the synergy dominance degree is calculated in order to verify whether the individual with strong imitation influence (low non-imitated rate, i.e., high imitated rate C) is more likely to dominate synergy; based on the correlation coefficient and the correlation coefficient

[0073] , if the correlation coefficient is negative and the correlation coefficient is positive, and the absolute values of both are higher than 0.6, then it is strongly correlated, i.e., it is determined to be true, then the gradient-weighted implicit feature extraction is performed on the worker operation trajectory data and the interaction record data to obtain the risk preference coefficient and the potential synergy coefficient; wherein, the gradient-weighted implicit feature extraction is performed in the following manner:

[0074] The risk preference coefficient C is obtained by formula one:

[0075] ; The potential synergy coefficient L is obtained by formula two:

[0076] ; It can be understood that the high-risk stay ratio H reflects the proportion of the time length that the individual actively exposes to the risk area, the higher the value, the longer the individual stays in the high-risk area, and the greater the possibility of facing risks. The violation correction rate M reflects the situation of correcting the worker's violation operation, and 1-M quantifies the stubbornness of the violation that is not corrected, the larger 1-M is, the more difficult the violation is to correct. The negative correlation coefficient

[0077] ​​​​​​This study validated the association between prolonged risk exposure and increasing difficulty in correcting violations. Multiplying these three factors integrates the intensity of proactive risk-taking and resistance to correction, while strengthening the intrinsic link between the two through correlation coefficients. This effectively identifies individuals with high-risk exposure and high violation persistence. For example, if a worker has a high high-risk stay ratio (H) and a low violation correction rate (M), and... If the two show a strong correlation, then the calculated risk preference coefficient will be larger, meaning that the worker has a higher risk preference and is more inclined to take risks in the workplace;

[0078] The imitation rate (C) quantifies how frequently an individual's actions are imitated by others, reflecting the individual's influence in the group. A higher C indicates more frequent imitation and greater influence. Collaborative dominance (L) measures an individual's control over initiating collaboration; a higher L indicates greater dominance in collaborative operations. Positive correlation coefficient. This study validated the association between stronger imitation influence and greater ability to dominate collaboration. Multiplying these three factors integrates the ability to be widely imitated and the capacity to dominate collaboration, reinforcing the collaborative relationship through correlation coefficients to identify individuals with high imitation influence and high collaborative control. For example, if a worker has a high imitation rate (C) and high collaborative dominance (L), and... The strong correlation between the two indicates a high potential synergy coefficient, meaning that the worker, in collaborative group work, can attract others to imitate through their own operations and play a leading role in the collaborative process, thus having a significant impact on group behavior.

[0079] Risk preference coefficient and potential synergy coefficient are used as implicit features;

[0080] Based on risk preference coefficient and potential synergy coefficient, the operation trajectory data and interaction record data are split into fields to construct a feature mapping library;

[0081] The method for splitting the field is as follows:

[0082] Obtain the worker's ID (number) and integrate the feature mapping library of the worker's basic fields, related fields, and feature fields;

[0083] The basic fields include high-risk stay ratio (H), violation correction rate (M), imitation rate (C), and collaborative dominance (L).

[0084] Related fields include: positive correlation coefficient negative correlation coefficient ;

[0085] The feature fields include: risk preference coefficient and potential synergy coefficient.

[0086] S2, based on the behavior feature mapping library, split the worker real-time interaction into imitation and collaborative behavior, and screen collaborative imitation individuals, based on the collaborative imitation individuals, construct a group synchronization model to extract the violation synchronization rate, split the violation synchronization rate into two tracks, and obtain the violation synchronization rate of the imitation track and the collaborative track;

[0087] The way of splitting the worker real-time interaction into imitation and collaborative behavior and screening collaborative imitation individuals is:

[0088] Preferably, the time slices are divided according to a fixed monitoring period, and the workers in the real-time video monitoring image are obtained;

[0089] The interaction records of the workers in the time slices are split into imitation behavior and collaborative behavior;

[0090] It should be noted that the imitation behavior, for example, worker B imitates and replicates the operation of worker A, and the collaborative behavior, for example, workers A and B operate collaboratively;

[0091] Based on the imitation behavior and the collaborative behavior, and combined with the behavior feature mapping library, the imitation track transmission intensity and the collaborative track transmission intensity of the worker are extracted;

[0092] Exemplarily, the way of extracting the imitation transmission intensity and the collaborative transmission intensity is:

[0093] The risk preference coefficient of B imitating A is obtained , ;

[0094] Among them, are the risk preference coefficients of A and B, respectively;

[0095] For the behavior of B imitating A, the imitation transmission intensity is obtained by the formula: ; ;

[0096] Among them, is the number of times B imitates A in the time slice, is the total number of imitations in the slice time, is the violation correction rate change of B;

[0097] The potential collaborative coefficient of A leading B is obtained ;

[0098] For the behavior of A leading B collaboration, the imitation transmission intensity is obtained by the formula: ; ;

[0099] Among them, , are the number of times A initiates collaboration in the slice time and the total number of collaborations in the slice time, respectively; is the collaborative leading degree change of B;

[0100] It can be understood that if the worker does not exist the imitated track transmission strength or the cooperative track transmission strength, the value of the corresponding imitated track transmission strength or the cooperative track transmission strength of the worker is 0;

[0101] Based on obtaining and calculating the personal imitation total value and the personal cooperation total value of each worker, the sum of the personal imitation total values of all workers is calculated as the group imitation total value Mz, and the sum of the personal cooperation total values of all workers is calculated as the group cooperation total value Qz;

[0102] For example, for worker A, all the imitation transmission strengths I of A as the imitated are summed up, that is, the personal imitation total value (A) = the sum of all B imitation worker A I values (such as B1 imitates A I1 + B2 imitates A I2 +...).

[0103] For example: if worker A is imitated by B, C, I A、B =1.2, I A、C =0.8, then the personal imitation total value (A) =1.2+0.8=2.0; the worker who is not imitated, the personal imitation total value (A) =0;

[0104] For worker A, all the cooperation transmission strengths S of A as the cooperative leader are summed up, that is, the personal cooperation total value (A) = the sum of all A leading B S values (such as A leading B S1 + A leading C S2 +...).

[0105] For example: if worker A leads B, C cooperation, S A、B =1.5, S a c=1.0, then the personal cooperation total value (A) =1.5+1.0=2.5; the personal cooperation total value (A);

[0106] The proportion of the personal imitation total value of each worker to the group imitation total value is calculated to obtain the imitation concentration coefficient .

[0107] The proportion of the personal cooperation total value of each worker to the group cooperation total value is calculated to obtain the cooperation concentration coefficient .

[0108] Based on the imitation concentration coefficient, the imitation core individual in the worker is screened, and based on the cooperation concentration coefficient, the cooperation core individual in the worker is screened;

[0109] The worker that meets the cooperation core individual and the imitation core individual at the same time is obtained as the cooperative imitation individual;

[0110] It can be understood that the core individual screening threshold , is set, and the worker that meets is the imitation core individual, and the worker that meets The worker is a core individual for coordination, and both meet a coordination imitating individual;

[0111] The core individual screening threshold of 20% is from historical data statistics, and the work area data of the past 12 monitoring periods (7 days per period) is calculated. The imitating centralization coefficient and coordination centralization coefficient of all workers are calculated, and the top 20% quantile is taken as the core individual determination standard to ensure that the core individual covers 80% of the imitating or coordination interaction transmission in the group. If the size of the work area personnel changes (such as ± 30%), the quantile threshold can be recalculated and adjusted;

[0112] Based on the coordination imitating individual, a group synchronization model is constructed by the mean field theory algorithm to extract the violation synchronization rate;

[0113] The way to construct the group synchronization model by the mean field theory algorithm is:

[0114] S201, obtaining the individual violation state and average field variable of all workers corresponding to the coordination imitating individual;

[0115] Preferably, for all workers in the current time slice, the individual violation state is defined as : For each worker i, the severity of the violation operation is identified according to real-time monitoring (0 for complete compliance and 1 for serious violation);

[0116] For each coordination imitating individual , the severity of the violation operation is identified according to real-time monitoring (0 for complete compliance and 1 for serious violation);

[0117] Wherein, the value of θ of the coordination imitating individual needs to be superimposed with its influence weight, that is, the of the coordination imitating individual = 1 + imitating centralization coefficient + coordination centralization coefficient, which is used to highlight the driving effect of the coordination imitating individual on the group;

[0118] The average field variable m is obtained by the formula:

[0119] Wherein, are the total number of coordination imitating individuals and the total number of non-coordination imitating individuals (ordinary workers), respectively;

[0120] It should be noted that according to the θᵢ of all workers, the weight of the coordination imitating individual is the sum of its imitating centralization coefficient and coordination centralization coefficient, and the weight of the non-core individual is 1. The average field variable reflects the average violation influence intensity of the group on the individual;

[0121] S202, extracting the imitating field strength and coordination field strength based on the average field variable;

[0122] ​Preferably, the field strength equation is simulated as follows: The simulated field strength is obtained as follows:

[0123] wherein, z is the total value of group simulation and group synergy respectively;

[0124] The total value of group simulation and group synergy is collectively referred to as the group total value parameter;

[0125] The synergy field strength equation is simulated as follows: The synergy field strength is obtained as follows:

[0126] wherein, is the synchronization rate of the simulation chain, i.e., the proportion of synchronization violations in the simulation followers of the synergistic simulation individual; is the synchronization rate of the synergy circle, i.e., the proportion of synchronization violations in the synergy objects of the synergistic simulation individual;

[0127] S203, input the simulation field strength and the synergy field strength into the mean field equation to obtain the synchronization rate of the comprehensive violation;

[0128] Preferably, the mean field equation is simulated as follows: The synchronization rate R of the comprehensive violation is obtained as follows:

[0129] wherein, is the sensitivity coefficient (usually 5-10, controlling the degree of reaction of the group to the change of the field strength), is the threshold parameter (when the total field strength is equal to , the synchronization rate is 50%, which can be calibrated by historical data);

[0130] It can be understood that the mean field equation simulates the phase transition characteristics of the group from asynchronous violation to synchronous violation through nonlinear conversion, and the final output R value is a quantitative index reflecting the overall violation synchronization trend of the group;

[0131] As shown in Figure 2 , the synchronization rate of the violation is compared with the preset violation synchronization threshold , if the synchronization rate of the violation is lower than the preset violation synchronization threshold , the change of the synchronization rate of the violation is continuously monitored;

[0132] If the synchronization rate of the violation is higher than or equal to the preset violation synchronization threshold, it is determined that the violation is out of limit and the synchronization rate of the violation is double-rail split to obtain the synchronization rate of the violation of the simulation rail and the synchronization rate of the violation of the synergy rail ;

[0133] The synchronization rate of the violation of the simulation rail is obtained as follows: ​​​;

[0134] By splitting the equation of the coordination track: Obtain the rule-breaking synchronization rate of the coordination track ;

[0135] It can be understood that the role of extracting the rule-breaking synchronization rate and performing double-track splitting is:

[0136] Role one, locate the group rule-breaking transmission key path, double-track splitting can obtain the rule-breaking synchronization rate of the two types of tracks respectively through the imitation track and the coordination track splitting equation, quantify the contribution of the two paths to the rule-breaking synchronization, reduce the problem of ambiguous group rule-breaking attribution, and clearly indicate the direction for subsequent intervention;

[0137] Role two, support the dynamic game model to determine the risk, the rule-breaking synchronization rate after double-track splitting is the basis for calculating the compliance hedging coefficients of the imitation track and the coordination track. If not split, a single coefficient will mask the path risk difference, and after splitting, the double-track confrontation state can be described to ensure the accuracy of the determination of the game advantage type;

[0138] Role three, the double-track rule-breaking synchronization rate is the core data for locating the weak track, and the double-track potential energy difference can be calculated in combination with the transmission strength to identify weak links; at the same time, comparison with the dynamic early warning threshold can determine whether it is an over-warning, provide a basis for screening the early warning model, filter invalid early warnings, avoid resource waste, and improve the practicality of early warning.

[0139] Embodiment two:

[0140] As shown in Figure 1 , the working area safety early warning method provided by the embodiment of the application specifically includes the following steps:

[0141] S3, perform compliance behavior hedging analysis based on the rule-breaking synchronization rate of the imitation track and the coordination track, obtain the compliance hedging coefficients of the imitation track and the coordination track, and construct a dynamic game model to determine the game advantage type;

[0142] The way of performing compliance behavior hedging analysis is:

[0143] Preferably, a compliance behavior hedging equation is established: Split the double-track hedging coefficient to obtain the imitation track compliance hedging coefficient , the coordination track compliance hedging coefficient ;

[0144] Wherein, P is the proportion of the coordination imitation individual compliance operation, and the proportion of the coordination imitation individual compliance operation P is obtained by the formula: =1-(all compliance operation coefficients+corrected rule-breaking times) / total operation coefficients

[0145] Imitation track compliance hedging coefficient , Synergy track compliance hedging coefficient A dynamic game model of double-track hedging-violation propagation is established.

[0146] The way to establish the dynamic game model is:

[0147] S301, the game characteristics of the game parties of the game model are established in combination with the imitation track compliance hedging coefficient and the synergy track compliance hedging coefficient.

[0148] Preferably, the game parties include a violation party and a compliance party:

[0149] The game characteristics of the violation party include: imitation transmission intensity , imitation transmission intensity , and comprehensive violation synchronization rate R.

[0150] The game characteristics of the compliance party are: imitation track compliance hedging coefficient , synergy track compliance hedging coefficient , and synergy imitation individual compliance operation proportion.

[0151] S302, game potential energy equation is established based on the game characteristics of the game parties, and game potential energy is output. The game advantage type is determined by comparing and analyzing the game potential energy.

[0152] Preferably, the game potential energy of the violation party is obtained by formula one:

[0153] The game potential energy of the violation party is obtained by formula two:

[0154] Wherein, is the interactive object set of the synergy imitation individual A, including the imitation object and the synergy object. is the total number of synergy imitation individuals.

[0155] If , the game advantage type is violation propagation, triggering a critical early warning signal.

[0156] If , the game advantage type is compliance hedging.

[0157] S4, if compliance hedging is dominant, a screening early warning model is constructed to identify and optimize the violation synchronization judgment, and to filter invalid early warnings under the support of compliance.

[0158] If compliance hedging is dominant, the way to construct the screening early warning model is:

[0159] ​​S401, based on the game potential energy equation output game potential energy advantage potential energy difference decomposition, positioning the weak track;

[0160] S4, if the compliance hedge is dominant, build a screening early warning model, identify and optimize the violation of the synchronous judgment, and filter the invalid early warning under the support of compliance;

[0161] If the compliance hedge is dominant, the way to build a screening early warning model is:

[0162] Preferably, by formula one: Get the mimic compliance ability of A And the potential difference of the mimic violation transmission of A to B ;

[0163] By formula two: Get the synergy compliance ability of A And the potential difference of the synergy violation transmission of A to B ;

[0164] By summation equation one: Get the mimic track compliance potential-energy violation potential difference ;

[0165] By summation equation two: Get the synergy track compliance potential-energy violation potential difference ;

[0166] Establish the weak track potential criterion: Then the synergy track is the weak track;

[0167] And Then the mimic track is the weak track;

[0168] And Then the mimic track and the synergy track are both weak, triggering the highest level of early warning;

[0169] S402, formulate a weak track reinforcement strategy, and identify whether the compliance scene is over-alarmed;

[0170] Preferably, based on the determined weak track, get the compliance proportion P of the synergy mimic individual, the influence coefficient , And the interaction transmission intensity , ​​, screen the core individuals with high influence K≥0.3 and high compliance P≥0.7, and force them to interact with the weak track individuals with low compliance P<0.5 to carry out high-frequency cooperation and imitation tasks (such as compliance teaching), and verify the reinforcement effect of weak track compliance potential energy through updated transmission intensity and violation correction rate;

[0171] Through the formula: A dynamic early warning threshold model is constructed to obtain an updated violation synchronization threshold ;

[0172] If the absolute deviation of the violation synchronization rate R and the dynamic violation synchronization threshold is less than 20%, and the time , it is determined that there is over-warning under compliance support;

[0173] S403, if there is over-warning, a screening early warning model is constructed to optimize the violation synchronization judgment and filter invalid early warnings under compliance support;

[0174] Preferably, the screening early warning model is constructed based on a decision tree and a dynamic rule iteration algorithm to filter invalid early warnings under compliance support;

[0175] As understood by those skilled in the art, the way to construct the screening early warning model based on the decision tree and the dynamic rule iteration algorithm is:

[0176] Calculate the deviation of the violation synchronization rate and the dynamic violation synchronization threshold , to obtain the synchronization deviation;

[0177] Screen the samples of in the historical period, if the violation synchronization rate R does not break through the dynamic threshold in the next 2 periods, it is marked as invalid early warning (1), otherwise it is marked as valid early warning (0);

[0178] Take , , the synchronization deviation as the core features, and integrate the compliance ratio P of the cooperative and imitative individuals, the influence coefficient , of the core individuals as parameters to calculate the features and establish the model data set, train the CART decision tree, and the decision tree automatically learns the nonlinear rules of the double-track potential energy state-synchronization rate deviation-early warning effectiveness-

[0179] Every 1 monitoring period (matching S2 time slice) or when the core individual P changes by ≥15%, the decision tree is updated with new labeled data increment, and the rule threshold is automatically adjusted (such as relaxing the screening boundary when the double-track potential energy is enhanced);

[0180] ​Real-time extraction of parameter calculation features in the model data set, input into the decision tree, output whether to screen out early warning, realize intelligent filtering of invalid early warning under compliance support.

[0181] Embodiment 3:

[0182] As shown in Figure 3 A working area safety early warning system includes the following modules:

[0183] Characteristic mapping module: used for collecting worker operation trajectory and interaction record in the working area, mining implicit features of operation trajectory and interaction record through behavior association analysis, and constructing behavior characteristic mapping library;

[0184] Violation analysis module: based on the behavior characteristic mapping library, used for splitting worker real-time interaction into mimic and collaborative behaviors and screening collaborative mimic individuals, constructing group synchronization model based on collaborative mimic individuals to extract violation synchronization rate, and performing double-track splitting on the violation synchronization rate to obtain violation synchronization rates of mimic track and collaborative track;

[0185] Game analysis module: based on the violation synchronization rates of mimic track and collaborative track, performing compliance behavior hedging analysis to obtain compliance hedging coefficients of mimic track and collaborative track, constructing dynamic game model, and determining game advantage type;

[0186] Early warning optimization module: if compliance hedging is dominant, construct a screening early warning model for identifying violation synchronization judgment and optimizing the violation synchronization judgment to filter invalid early warning under compliance support.

[0187] The above describes one embodiment of the present application in detail, but the content described is only the preferred embodiment of the present application, and cannot be considered as limiting the scope of the present application. Any equivalent changes and improvements made within the scope of the present application should still belong to the patent coverage of the present application.

Claims

1. A method for early warning of workplace safety, characterized in that, Includes the following steps: Collect worker operation trajectories and interaction records in the work area, and mine the implicit features of operation trajectories and interaction records through behavioral correlation analysis to build a behavioral feature mapping library; The behavioral correlation analysis method is as follows: Worker operation trajectory data and interaction record data in high-risk areas were collected. The high-risk stay ratio and violation correction rate were extracted from the operation trajectory data, and the correlation coefficient between the high-risk stay ratio and the violation correction rate was calculated. ; Extract the imitation rate and collaborative dominance from the interaction record data. Perform a negative transformation on the imitation rate to obtain the non-imitation rate. Calculate the correlation coefficient between the non-imitation rate and collaborative dominance. ; Based on correlation coefficient and correlation coefficient Verification process will be conducted. If the correlation coefficient and correlation coefficient If all are verified, then implicit features are extracted by gradient weighting of worker operation trajectory data and interaction record data to obtain risk preference coefficient and potential synergy coefficient. Based on risk preference coefficient and potential synergy coefficient, the operation trajectory data and interaction record data are split into fields to construct a behavioral feature mapping library; Based on the behavioral feature mapping library, the real-time interaction of workers is divided into imitation and collaborative behaviors, and individuals who cooperate and imitate are selected. Based on the individuals who cooperate and imitate, a group synchronization model is constructed to extract the violation synchronization rate. The violation synchronization rate is then divided into two tracks to obtain the violation synchronization rates of the imitation track and the collaborative track. The group synchronization model is constructed as follows: Obtain the individual violation status and mean field variables of all workers corresponding to the collaborative imitation individual; Extracting the simulated field strength and the cooperative field strength based on the mean field variable; By establishing the mean field equation and inputting the imitation field strength and the cooperative field strength, the overall violation synchronization rate is obtained. The method for obtaining the aforementioned collaborative imitator is as follows: The time slices are divided according to a fixed monitoring cycle, and the interaction records of workers within the time slices are broken down into: imitation behavior and collaborative behavior; Based on imitation behavior and collaborative behavior, and combined with a behavior feature mapping library, the imitation track transmission strength and collaborative track transmission strength of workers are extracted. Central tendency analysis is performed based on the transfer intensity of the imitation track and the transfer intensity of the cooperative track to obtain the imitation concentration coefficient and the cooperative concentration coefficient. Imitative core individuals among workers are selected based on imitation concentration coefficient, and collaborative core individuals among workers are selected based on collaborative concentration coefficient. Obtain workers who simultaneously satisfy both the collaborative core individual and the imitation core individual, and use them as collaborative imitation individuals; Compliance behavior hedging analysis is conducted based on the violation synchronization rate of the imitation track and the cooperative track. The compliance hedging coefficients of the imitation track and the cooperative track are obtained, and a dynamic game model is constructed to determine the game advantage type. The method for determining the type of game advantage is as follows: Compliance behavior hedging analysis is conducted based on the violation synchronization rate of the imitation track and the collaborative track to obtain the compliance hedging coefficients of the imitation track and the collaborative track. The game characteristics of the two sides in the game model are established by combining the compliance hedging coefficient of the imitation track and the compliance hedging coefficient of the collaborative track. Based on the game characteristics of both sides, a game potential energy equation is established to output the game potential energy. The game potential energy is compared and analyzed to determine the type of game advantage. If compliance hedging is the priority, a screening and early warning model is constructed to identify violations and make simultaneous judgments on violations, and to optimize the simultaneous judgments on violations and filter out invalid early warnings supported by compliance. The method for constructing the screening and reduction early warning model is as follows: Based on the game potential energy output from the game potential energy equation, the advantage potential energy difference is decomposed to locate the weak trajectory. Develop strategies to strengthen weak links and identify whether compliance scenarios are subject to excessive warnings; If there are excessive warnings, a screening and reduction warning model will be built to optimize the simultaneous judgment of violations and filter out invalid warnings supported by compliance.

2. The work area safety early warning method according to claim 1, characterized in that, The method for performing the dual-track split is as follows: Based on the constructed group synchronization model, the violation synchronization rate is extracted and it is determined whether the violation exceeds the limit. If it exceeds the limit, the imitation track splitting equation and the cooperative track splitting equation are constructed. Obtain the imitation field strength and cooperative field strength of the group synchronization model, and combine the illegal synchronization rate with the corresponding imitation track splitting equation and cooperative track splitting equation to obtain the illegal synchronization rate of the imitation track and the illegal synchronization rate of the cooperative track.

3. The work area safety early warning method according to claim 1, characterized in that, The methods for extracting the imitation field strength and the cooperative field strength are as follows: Construct the imitation field strength equation, obtain the imitation chain synchronization rate and population total value parameters, and combine the mean field variable with the imitation field strength equation to obtain the imitation field strength; Construct a cooperative field strength equation, obtain the synchronization rate of the cooperative circle and the total value of the population parameters, and input the mean field variable into the cooperative field strength equation to obtain the cooperative field strength.

4. A work area safety early warning method according to claim 1, characterized in that, The method for conducting the aforementioned compliance behavior hedging analysis is as follows: Establish a compliance behavior hedging equation, decompose the dual-track hedging coefficients, and obtain the compliance hedging coefficients of the imitation track and the compliance hedging coefficients of the collaboration track.

5. A work area safety early warning method according to claim 1, characterized in that, The method for locating the weak track is as follows: Obtain the difference between the compliance potential energy and violation potential energy of the imitation track and the compliance potential energy and violation potential energy of the collaborative track, conduct comparative analysis, and identify the weak track.

Images