Intelligent supervision method and system for operation safety information

Abstract

The invention discloses an intelligent supervision method and system for operation safety information, and belongs to the technical field of safety monitoring. The method comprises the following steps: collecting environmental data in a historical operation site and records of dangerous accidents during historical operation by using a sensor, and searching feature data reflecting the dangerous accidents; comparing and screening environmental data during safety early warning in different working sites in history to obtain potential data influencing the feature data; constructing a scatter diagram between each type of potential data and feature data, and building a function influence relationship between each type of potential data and feature data; calculating a feature threshold value for safety early warning; acquiring potential data in the operation site in real time, and optimizing the feature threshold value in real time by utilizing a function influence relationship between each type of potential data and the feature data; and collecting feature data in the operation site in real time, and performing real-time early warning on the safety during operation by using the optimized feature threshold.

Description

Technical Field

[0001] The present invention relates to the technical field of safety supervision, and in particular to an intelligent supervision method and system for operation safety information. Background Art

[0002] With the development of technologies such as artificial intelligence, big data analysis and the Internet of Things, intelligent supervision technology plays an increasingly important role in job safety management. Mainly by collecting, integrating and analyzing a large amount of data related to job safety, potential safety risks and trends are discovered, providing data support for accident prevention. Big data analysis helps to improve the accuracy and pertinence of supervision and better guide safety management work. Job safety information supervision can be applied in many fields, such as: construction, medical, urban planning, etc.; with the widespread attention of the social population to safety issues, technologies for achieving safety supervision using big data and information technology have emerged one after another, but there are still many problems in many safety monitoring technologies. Due to the complexity of the operating environment, the amount of data that needs to be processed for accurate prediction is relatively large, which greatly consumes the computing power of the system; and because the application field of intelligent supervision of job safety information is relatively wide, the difference in environmental data in different fields leads to large differences in the standards of safety warnings. Therefore, some safety supervision systems have great limitations when used, which undoubtedly greatly reduces the work efficiency and cost of the system; Based on the above problems, the present invention proposes an intelligent supervision method and system for job safety information. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent monitoring method and system for operation safety information to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] An intelligent supervision method for operation safety information, the method comprising the following steps:

[0006] S100, using sensors to collect environmental data in historical work sites and records of dangerous accidents that occurred during historical operations, and searching for characteristic data that reflects dangerous accidents;

[0007] Furthermore, the specific steps for finding characteristic data reflecting dangerous accidents are as follows:

[0008] S101. Collect the records of dangerous accidents that occurred during operation in history, extract the environmental data before and after the dangerous accident, and assume that the environmental data before the dangerous accident is {H1, H2, H3, ..., H n}, H1, H2, H3, ..., H nrepresents the 1st, 2nd, 3rd, ..., nth environmental data before the dangerous accident occurred in the history, where n is a positive integer; suppose the environmental data after the dangerous accident is {H'1, H'2, H'3, ..., H'n}, H'1, H'2, H'3, ..., H' n Indicates the 1st, 2nd, 3rd, ..., nth environmental data after a dangerous accident occurred during operation in history; extracts the environmental data from the mth record of dangerous accidents that occurred during operation in history;

[0009] S102. Calculate the difference between each environmental data before and after a dangerous accident occurred during operation in history. The formula is:

[0010]

[0011] In the formula, H_c represents the average difference of each environmental data before and after the dangerous accident occurred during the operation in history, H represents each environmental data before the dangerous accident occurred during the operation in history, H' represents each environmental data after the dangerous accident occurred during the operation in history, and m represents the number of records of dangerous accidents during the operation in history collected; after calculation, the difference of environmental data before and after the dangerous accident occurred during the operation in history is {H_c1, H_c2, H_c3, ..., H_c n}, H_c1, H_c2, H_c3,..., H_c n Indicates the differences in the 1st, 2nd, 3rd, ..., nth environmental data before and after a dangerous accident occurred during operation in history;

[0012] S103, screening n types of environmental data differences before and after dangerous accidents occurred during operation in history, selecting the type of environmental data corresponding to the maximum difference as characteristic data reflecting the dangerous accident, and setting the obtained characteristic data as T.

[0013] Using the environmental data of dangerous accidents in the past, we can find the environmental data with the biggest difference before and after the dangerous accident as the characteristic data. When making real-time safety warnings during operations, we only need to judge the characteristic data to get the real-time safety situation in the operation site, which greatly reduces the amount of calculation and data processing of the system and increases the work efficiency of the system.

[0014] S200, compare and screen the environmental data of safety warnings in different work sites in history to obtain potential data that affects the characteristic data;

[0015] Furthermore, the specific steps of obtaining potential data that affects the characteristic data through comparison and screening are as follows:

[0016] S201, collect the environmental data of safety warnings in different working sites in history, and assume that the collected environmental data of safety warnings are {S1, S2, S3, ..., Sn}, S1, S2, S3, ..., S n represents the 1st, 2nd, 3rd, ..., nth environmental data during the safety warning in history, and collects the characteristic data of G safety warnings at different sites in history as {T1, T2, T3, ..., T G}, T1, T2, T3, ..., T G Indicates the characteristic data of the 1st, 2nd, 3rd, ..., Gth safety warnings at different sites in history, each characteristic data corresponds to a set of environmental data, and a set of environmental data includes n types of environmental data;

[0017] S202, sorting the characteristic data of safety warnings at different sites collected in the G history from small to large, and calculating the difference between the characteristic data and the difference between each corresponding environmental data according to the arrangement order, the formula is:

[0018] T_c=T i+1 -T i

[0019]

[0020] In the formula, T_c represents the difference between the last feature data and the previous one according to the arrangement order, T i+1 Indicates the next feature data in the sorting, T i represents the previous feature data in the sorting, i∈(1, G), S_c represents the difference between the next and previous features of each environment data in the sorting, Indicates the next environmental data value of the same environmental data. Represents the previous environmental data value of the same environmental data, r∈(1,n);

[0021] S203, after calculation, the G-1 feature data differences are {T_c1, T_c2, T_c3, ..., T_c G-1}, T_c1, T_c2, T_c3,..., T_c G-1 Indicates the 1st, 2nd, 3rd, ..., G-1th characteristic data values ​​calculated, G-1 is a positive integer; correspondingly, G-1 differences of n kinds of environmental data are calculated; the difference between the characteristic data and the environmental data is judged, specifically:

[0022] When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c≠0, the characteristic data difference T_c=0, the corresponding environmental data difference S_c=0, it is judged that the corresponding environmental data type has an impact risk;

[0023] When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c=0, and the characteristic data difference T_c=0, the corresponding environmental data difference S_c≠0, it is judged that the corresponding environmental data type has no impact risk;

[0024] S204, judge G-1 feature data differences and environmental data differences. When the environmental data difference has an impact risk in G-1 judgments, determine the corresponding environmental data type as potential data that affects the feature data; when the environmental data difference has no impact risk in one of G-1 judgments, determine the corresponding environmental data type is not potential data that affects the feature data; the potential data that affects the feature data is obtained through judgment as {Q1, Q2, Q3, ..., Q p}, Q1, Q2, Q3, ..., Q p It represents the 1st, 2nd, 3rd, ..., pth potential data affecting the characteristic data obtained by judgment, where p is a positive integer.

[0025] After obtaining the characteristic data that reflects the dangerous accident, different environments may cause changes in the characteristic data, resulting in large errors in the judgment of the safety situation of the work site and leading to misjudgment. Therefore, it is more important to find potential environmental data that affects the characteristic data. The potential data can be analyzed to reduce the error of the characteristic data and avoid system misjudgment.

[0026] S300, constructing a scatter plot between each potential data and feature data, and performing linear regression calculation on the scatter plot to obtain a functional influence relationship between each potential data and feature data;

[0027] Furthermore, the specific steps for obtaining the functional influence relationship between each potential data and feature data are as follows:

[0028] S301, extract different feature data when performing operations in the history

[0029] {T_z1, T_z2, T_z3,...,T_z g}, T_z1, T_z2, T_z3,..., T_z g Indicates the 1st, 2nd, 3rd, ..., feature data in the collected history, g is a positive integer; according to the p kinds of potential data obtained by judgment, correspondingly collect g potential data values ​​of each potential data; and establish a scatter plot of each potential data and feature data, with the potential data as the independent variable and the feature data as the variable to construct the scatter plot;

[0030] S302. After establishing a scatter plot of each potential data and characteristic data, calculate the regression curve in each scatter plot, and obtain the regression curve equation as T_z=a×Q+b, where T_z represents the variable characteristic data, Q represents the independent variable potential data, a represents the slope of the regression curve, and b represents the intercept of the regression curve; the slope a and intercept b in the regression curve are calculated based on the existing points in the scatter plot, and the formula is:

[0031]

[0032] b=T_z pi -a×Q pi

[0033] In the formula, T_z pi represents the average value of g variable characteristic data, Q pi represents the average value of the potential data of g independent variables, T_z u Represents the characteristic data of each variable, Q u Represents the potential data of each independent variable, u∈(1,g);

[0034] S303, after calculating, the regression curve equation of each scatter plot is obtained, and the regression curve equation in each scatter plot is used as the functional influence relationship between the independent variable potential data and the variable characteristic data in the scatter plot, and finally the functional influence relationship between the p kinds of potential data and the characteristic data is calculated as {f(Q1), f(Q2), f(Q3), ..., f(Q p )}, f(Q1), f(Q2), f(Q3),..., f(Q p ) represents the calculated functional influence relationship of the 1st, 2nd, 3rd, ..., kinds of potential data.

[0035] Draw a scatter plot based on the values ​​of potential data and feature data in history; use the scatter plot to calculate the regression curve, and clearly calculate the impact of potential data on feature data; when judging feature data based on potential data, accurately calculate the feature data after being affected by potential data;

[0036] S400, collecting characteristic data of dangerous accidents that occurred during operation in the past, and calculating characteristic thresholds for safety warning;

[0037] Furthermore, the specific steps for calculating the feature threshold for security warning are:

[0038] S401. Collect characteristic data of the work site when dangerous accidents occurred in history.

[0039] {T_d1, T_d2, T_d3,..., T_d h}, T_d1, T_d2, T_d3,..., T_dh It represents the characteristic data collected in the work site when the 1st, 2nd, 3rd, ..., hth dangerous accidents occurred in the history, where h is a positive integer;

[0040] S402. Calculate the characteristic threshold for safety warning based on the characteristic data of the work site when dangerous accidents occurred in the history collected. The formula is:

[0041]

[0042] In the formula, T_y represents the characteristic threshold used for safety warning, T_d represents the characteristic data of the work site when dangerous accidents occurred in the history collected in real time, T_st represents the standard deviation of the characteristic data, and h represents the number of characteristic data collected in the work site when dangerous accidents occurred in the history.

[0043] S500, collecting potential data in the work site in real time, and optimizing the feature threshold in real time by using the functional influence relationship between each potential data and feature data;

[0044] Furthermore, the specific steps for real-time optimization of feature thresholds are as follows:

[0045] S501, real-time collection of potential data in the work site {Q_s1, Q_s2, Q_s3, ..., Q_s p}, Q_s1, Q_s2, Q_s3,..., Q_s p represents the 1st, 2nd, 3rd, ..., pth potential data values ​​collected in real time;

[0046] S502, extracting the functional influence relationship between each potential data and feature data in S303, and calculating the optimized value of the feature threshold, the formula is:

[0047]

[0048] In the formula, T_op represents the optimized value of the feature threshold, (a×Q_s+b) t represents the functional influence relationship between each potential data and feature data, t∈(1,p), Q_s represents the potential data collected in real time, T_y represents the feature threshold before optimization; p represents the number of potential data types;

[0049] S503. Optimize the real-time feature threshold according to the calculated optimized value of the feature threshold. The optimization formula is: T_after=T_y+T_op, where T_after represents the optimized feature threshold, T_y represents the feature threshold before optimization, and T_op represents the optimized value of the feature threshold. Finally, the optimized feature threshold is T_after.

[0050] The optimized value of the feature threshold is calculated using the functional influence relationship between potential data and feature data. This is because the different potential data in different work sites may cause the feature threshold to change after affecting the feature data. If no optimization is performed, a large feature error will occur, leading to system misjudgment. After optimization using the optimized value, the feature error is removed, which will not affect the system safety warning and can avoid system misjudgment.

[0051] S600, collecting feature data in the work site in real time, and using the optimized feature threshold to issue a real-time warning for safety during the work.

[0052] Furthermore, the specific steps of using the optimized feature threshold to provide real-time warning for safety during operation are as follows:

[0053] S601. Collect the characteristic data of the work site in real time as T_s, and use the optimized characteristic threshold to issue a safety warning for the work site. When T_s≥T_after, it is determined that there is a safety risk at the work site and a warning is issued. When T_s<T_after, it is determined that there is no safety risk at the work site and normal operations can continue.

[0054] An intelligent monitoring system for operation safety information, the intelligent monitoring system comprising a data collection module, a potential data analysis module, a threshold calculation module, a threshold optimization module and an early warning module;

[0055] The data collection module is used to collect environmental data when dangerous accidents occurred in history, and find out characteristic data reflecting dangerous accidents;

[0056] The potential data analysis module is used to search for potential data that affects characteristic data in the environmental data according to the records of safety warnings that occurred in the history, and calculate the functional influence relationship between each potential data and characteristic data;

[0057] The threshold calculation module is used to calculate the characteristic threshold value for determining whether there is a safety risk at the work site based on the characteristic data of dangerous accidents that occurred in the past;

[0058] The threshold optimization module is used to calculate the optimized value of the feature threshold by using the functional influence relationship between each potential data and feature data, and optimize the feature threshold according to the optimized value;

[0059] The early warning module is used to collect feature data in the work site in real time, and use the optimized feature threshold to issue safety warnings to the work site.

[0060] The data collection module includes a historical data collection unit and a characteristic data search unit;

[0061] The historical data collection unit is used to collect environmental data when dangerous accidents occurred in the past;

[0062] The characteristic data search unit is used to analyze the environmental data of dangerous accidents that have occurred in the history collected, calculate the difference and search for characteristic data that reflects the dangerous accidents.

[0063] The potential data analysis module includes a potential data search unit and a function impact relationship calculation unit;

[0064] The potential data search unit is used to perform difference calculation on the environmental data and the characteristic data according to the records of the safety warnings in history, and search for the potential data that affects the characteristic data;

[0065] The function influence relationship calculation unit is used to establish a scatter plot of potential data and feature data, calculate the regression curve equation in the scatter plot, and determine the regression curve equation as the function influence relationship between the potential data and feature data.

[0066] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0067] 1. The present invention utilizes the environmental data of dangerous accidents that occurred during operations in history, calculates the difference before and after the dangerous accident, and finds out the characteristic data that reflects the dangerous accident. When the work site is monitored for safety, only the characteristic data needs to be collected, processed and analyzed, which greatly reduces the amount of calculation and data processing of the system and increases the work efficiency of the system.

[0068] 2. The present invention utilizes the difference between the environmental data and the characteristic data during the safety warning during the historical operation, finds out the potential data that affects the characteristic data through judgment, and calculates the functional influence relationship between each potential data and the characteristic data, so as to clearly obtain the characteristic data value after the potential data affects the characteristic data, thereby reducing the error when analyzing the characteristic data.

[0069] 3. The present invention utilizes the functional influence relationship between potential data and feature data to calculate the optimized value of the feature threshold, and then utilizes the optimized value to optimize the feature threshold. This allows the system to flexibly change the feature threshold in different work sites and accurately complete the judgment of the safety risks of the work sites; it avoids the system from misjudging the safety risks and causing operational losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0071] Figure 1 It is a module distribution diagram of an intelligent monitoring system for operation safety information of the present invention;

[0072] Figure 2 It is a schematic diagram of the steps of an intelligent monitoring method for operation safety information of the present invention;

[0073] Figure 3 It is a temperature and humidity scatter diagram of an intelligent monitoring method for operation safety information of the present invention. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0075] See also Figure 1-Figure 3 , the present invention provides a technical solution:

[0076] An intelligent supervision method for operation safety information, the method comprising the following steps:

[0077] S100, using sensors to collect environmental data in historical work sites and records of dangerous accidents that occurred during historical operations, and searching for characteristic data that reflects dangerous accidents;

[0078] The specific steps for finding characteristic data that reflect dangerous accidents are:

[0079] S101. Collect the records of dangerous accidents that occurred during operation in history, extract the environmental data before and after the dangerous accident, and assume that the environmental data before the dangerous accident is {H1, H2, H3, ..., H n}, H1, H2, H3, ..., H n represents the 1st, 2nd, 3rd, ..., nth environmental data before the dangerous accident occurred in the history, where n is a positive integer; suppose the environmental data after the dangerous accident is {H'1, H'2, H'3, ..., H'n}, H'1, H'2, H'3, ..., H' n Indicates the 1st, 2nd, 3rd, ..., nth environmental data after a dangerous accident occurred during operation in history; extracts the environmental data from the mth record of dangerous accidents that occurred during operation in history;

[0080] S102. Calculate the difference between each environmental data before and after a dangerous accident occurred during operation in history. The formula is:

[0081]

[0082] In the formula, H_c represents the average difference of each environmental data before and after the dangerous accident occurred during the operation in history, H represents each environmental data before the dangerous accident occurred during the operation in history, H' represents each environmental data after the dangerous accident occurred during the operation in history, and m represents the number of records of dangerous accidents during the operation in history collected; after calculation, the difference of environmental data before and after the dangerous accident occurred during the operation in history is {H_c1, H_c2, H_c3, ..., H_c n}, H_c1, H_c2, H_c3,..., H_c n Indicates the differences in the 1st, 2nd, 3rd, ..., nth environmental data before and after a dangerous accident occurred during operation in history;

[0083] S103, screening n types of environmental data differences before and after dangerous accidents occurred during operation in history, selecting the type of environmental data corresponding to the maximum difference as characteristic data reflecting the dangerous accident, and setting the obtained characteristic data as T.

[0084] Using the environmental data of dangerous accidents in the past, we can find the environmental data with the biggest difference before and after the dangerous accident as the characteristic data. When making real-time safety warnings during operations, we only need to judge the characteristic data to get the real-time safety situation in the operation site, which greatly reduces the amount of calculation and data processing of the system and increases the work efficiency of the system.

[0085] S200, compare and screen the environmental data of safety warnings in different work sites in history to obtain potential data that affects the characteristic data;

[0086] The specific steps to obtain potential data that affects feature data through comparison and screening are:

[0087] S201, collect the environmental data of safety warnings in different working sites in history, and assume that the collected environmental data of safety warnings are {S1, S2, S3, ..., S n}, S1, S2, S3, ..., S n represents the 1st, 2nd, 3rd, ..., nth environmental data during the safety warning in history, and collects the characteristic data of G safety warnings at different sites in history as {T1, T2, T3, ..., T G}, T1, T2, T3, ..., T G Indicates the characteristic data of the 1st, 2nd, 3rd, ..., Gth safety warnings at different sites in history, each characteristic data corresponds to a set of environmental data, and a set of environmental data includes n types of environmental data;

[0088] S202, sorting the characteristic data of safety warnings at different sites collected in the G history from small to large, and calculating the difference between the characteristic data and the difference between each corresponding environmental data according to the arrangement order, the formula is:

[0089] T_c=T i+1 -T i

[0090]

[0091] In the formula, T_c represents the difference between the last feature data and the previous one according to the arrangement order, T i+1 Indicates the next feature data in the sorting, T i represents the previous feature data in the sorting, i∈(1, G), S_c represents the difference between the next and previous features of each environment data in the sorting, Indicates the next environmental data value of the same environmental data. Represents the previous environmental data value of the same environmental data, r∈(1,n);

[0092] S203, after calculation, the G-1 feature data differences are {T_c1, T_c2, T_c3, ..., T_c G-1}, T_c1, T_c2, T_c3,..., T_c G-1 Indicates the 1st, 2nd, 3rd, ..., G-1th characteristic data values ​​calculated, G-1 is a positive integer; correspondingly, G-1 differences of n kinds of environmental data are calculated; the difference between the characteristic data and the environmental data is judged, specifically:

[0093] When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c≠0, the characteristic data difference T_c=0, the corresponding environmental data difference S_c=0, it is judged that the corresponding environmental data type has an impact risk;

[0094] When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c=0, and the characteristic data difference T_c=0, the corresponding environmental data difference S_c≠0, it is judged that the corresponding environmental data type has no impact risk;

[0095] S204, judge G-1 feature data differences and environmental data differences. When the environmental data difference has an impact risk in G-1 judgments, determine the corresponding environmental data type as potential data that affects the feature data; when the environmental data difference has no impact risk in one of G-1 judgments, determine the corresponding environmental data type is not potential data that affects the feature data; the potential data that affects the feature data is obtained through judgment as {Q1, Q2, Q3, ..., Q p}, Q1, Q2, Q3, ..., Q p It represents the 1st, 2nd, 3rd, ..., pth potential data affecting the characteristic data obtained by judgment, where p is a positive integer.

[0096] After obtaining the characteristic data that reflects the dangerous accident, different environments may cause changes in the characteristic data, resulting in large errors in the judgment of the safety situation of the work site and leading to misjudgment. Therefore, it is more important to find potential environmental data that affects the characteristic data. The potential data can be analyzed to reduce the error of the characteristic data and avoid system misjudgment.

[0097] S300, constructing a scatter plot between each potential data and feature data, and performing linear regression calculation on the scatter plot to obtain a functional influence relationship between each potential data and feature data;

[0098] The specific steps to obtain the functional influence relationship between each potential data and feature data are:

[0099] S301, extract different feature data when performing operations in the history

[0100] {T_z1, T_z2, T_z3,...,T_z g}, T_z1, T_z2, T_z3,..., T_z g Indicates the 1st, 2nd, 3rd, ..., feature data in the collected history, g is a positive integer; according to the p kinds of potential data obtained by judgment, correspondingly collect g potential data values ​​of each potential data; and establish a scatter plot of each potential data and feature data, with the potential data as the independent variable and the feature data as the variable to construct the scatter plot;

[0101] S302. After establishing a scatter plot of each potential data and characteristic data, calculate the regression curve in each scatter plot, and obtain the regression curve equation as T_z=a×Q+b, where T_z represents the variable characteristic data, Q represents the independent variable potential data, a represents the slope of the regression curve, and b represents the intercept of the regression curve; the slope a and intercept b in the regression curve are calculated based on the existing points in the scatter plot, and the formula is:

[0102]

[0103] b=T_z pi -a×Q pi

[0104] In the formula, T_z pi represents the average value of g variable characteristic data, Q pi represents the average value of the potential data of g independent variables, T_z u Represents the characteristic data of each variable, Q uRepresents the potential data of each independent variable, u∈(1,g);

[0105] S303, after calculating, the regression curve equation of each scatter plot is obtained, and the regression curve equation in each scatter plot is used as the functional influence relationship between the independent variable potential data and the variable characteristic data in the scatter plot, and finally the functional influence relationship between the p kinds of potential data and the characteristic data is calculated as {f(Q1), f(Q2), f(Q3), ..., f(Q p )}, f(Q1), f(Q2), f(Q3),..., f(Q p ) represents the calculated functional influence relationship of the 1st, 2nd, 3rd, ..., kinds of potential data.

[0106] Draw a scatter plot based on the values ​​of potential data and feature data in history; use the scatter plot to calculate the regression curve, and clearly calculate the impact of potential data on feature data; when judging feature data based on potential data, accurately calculate the feature data after being affected by potential data;

[0107] S400, collecting characteristic data of dangerous accidents that occurred during operation in the past, and calculating characteristic thresholds for safety warning;

[0108] The specific steps for calculating the feature threshold for security warning are:

[0109] S401. Collect characteristic data of the work site when dangerous accidents occurred in history.

[0110] {T_d1, T_d2, T_d3,..., T_d h}, T_d1, T_d2, T_d3,..., T_d h It represents the characteristic data collected in the work site when the 1st, 2nd, 3rd, ..., hth dangerous accidents occurred in the history, where h is a positive integer;

[0111] S402. Calculate the characteristic threshold for safety warning based on the characteristic data of the work site when dangerous accidents occurred in the history collected. The formula is:

[0112]

[0113] In the formula, T_y represents the characteristic threshold used for safety warning, T_d represents the characteristic data of the work site when dangerous accidents occurred in the history collected in real time, T_st represents the standard deviation of the characteristic data, and h represents the number of characteristic data collected in the work site when dangerous accidents occurred in the history.

[0114] S500, collecting potential data in the work site in real time, and optimizing the feature threshold in real time by using the functional influence relationship between each potential data and feature data;

[0115] The specific steps for real-time optimization of feature thresholds are:

[0116] S501, real-time collection of potential data in the work site {Q_s1, Q_s2, Q_s3, ..., Q_s p}, Q_s1, Q_s2, Q_s3,..., Q_s p represents the 1st, 2nd, 3rd, ..., pth potential data values ​​collected in real time;

[0117] S502, extracting the functional influence relationship between each potential data and feature data in S303, and calculating the optimized value of the feature threshold, the formula is:

[0118]

[0119] In the formula, T_op represents the optimized value of the feature threshold, (a×Q_s+b) t represents the functional influence relationship between each potential data and feature data, t∈(1,p), Q_s represents the potential data collected in real time, T_y represents the feature threshold before optimization; p represents the number of potential data types;

[0120] S503. Optimize the real-time feature threshold according to the calculated optimized value of the feature threshold. The optimization formula is: T_after=T_y+T_op, where T_after represents the optimized feature threshold, T_y represents the feature threshold before optimization, and T_op represents the optimized value of the feature threshold. Finally, the optimized feature threshold is T_after.

[0121] The optimized value of the feature threshold is calculated using the functional influence relationship between potential data and feature data. This is because the different potential data in different work sites may cause the feature threshold to change after affecting the feature data. If no optimization is performed, a large feature error will occur, leading to system misjudgment. After optimization using the optimized value, the feature error is removed, which will not affect the system safety warning and can avoid system misjudgment.

[0122] S600, collecting feature data in the work site in real time, and using the optimized feature threshold to issue a real-time warning for safety during the work.

[0123] The specific steps of using the optimized feature threshold to provide real-time warning for safety during operation are as follows:

[0124] S601. Collect the characteristic data of the work site in real time as T_s, and use the optimized characteristic threshold to issue a safety warning for the work site. When T_s≥T_after, it is determined that there is a safety risk at the work site and a warning is issued. When T_s<T_after, it is determined that there is no safety risk at the work site and normal operations can continue.

[0125] An intelligent monitoring system for operation safety information, the intelligent monitoring system comprising a data collection module, a potential data analysis module, a threshold calculation module, a threshold optimization module and an early warning module;

[0126] The data collection module is used to collect environmental data when dangerous accidents occurred in history, and find out characteristic data reflecting dangerous accidents;

[0127] The potential data analysis module is used to search for potential data that affects characteristic data in the environmental data according to the records of safety warnings that occurred in the history, and calculate the functional influence relationship between each potential data and characteristic data;

[0128] The threshold calculation module is used to calculate the characteristic threshold value for determining whether there is a safety risk at the work site based on the characteristic data of dangerous accidents that occurred in the past;

[0129] The threshold optimization module is used to calculate the optimized value of the feature threshold by using the functional influence relationship between each potential data and feature data, and optimize the feature threshold according to the optimized value;

[0130] The early warning module is used to collect feature data in the work site in real time, and use the optimized feature threshold to issue safety warnings to the work site.

[0131] The data collection module includes a historical data collection unit and a characteristic data search unit;

[0132] The historical data collection unit is used to collect environmental data when dangerous accidents occurred in the past;

[0133] The characteristic data search unit is used to analyze the environmental data of dangerous accidents that have occurred in the history collected, calculate the difference and search for characteristic data that reflects the dangerous accidents.

[0134] The potential data analysis module includes a potential data search unit and a function impact relationship calculation unit;

[0135] The potential data search unit is used to perform difference calculation on the environmental data and the characteristic data according to the records of the safety warnings in history, and search for the potential data that affects the characteristic data;

[0136] The function influence relationship calculation unit is used to establish a scatter plot of potential data and feature data, calculate the regression curve equation in the scatter plot, and determine the regression curve equation as the function influence relationship between the potential data and feature data.

[0137] Example 1

[0138] Collect environmental data of dangerous accidents that occurred during construction work in history, calculate the difference of each environmental data, and select the maximum difference as the characteristic data, then the characteristic data is temperature;

[0139] Collect the records of safety warnings issued during construction work in the past, extract the environmental data from the three records, the temperature is 20, 30, 40 respectively; the other environmental data include humidity and oxygen content; the humidity record values ​​are 10, 20, 30; the oxygen content is 34, 19, 19; calculate the difference of the potential data of the characteristic data, the temperature difference is 10, 10; the humidity difference is 10, 20; the oxygen content difference is 15, 0; after judgment, the potential data is humidity;

[0140] Collect the temperature and humidity in history and construct a temperature and humidity scatter plot such as Figure 3 As shown, the regression curve equation is T_z=1.0354×Q+8.8771;

[0141] The temperature of dangerous accidents in history is collected, and the temperature threshold is calculated to be 30; the humidity value in real-time construction operation is 19, and the optimal value of temperature is calculated. The formula is:

[0142]

[0143] Because the potential data at this time is only humidity, we only need to calculate the functional influence of humidity, and get the optimized value of temperature to be about 1.5 degrees. We use the optimized value to optimize the temperature threshold, and the optimized temperature threshold is 28.5.

[0144] The real-time temperature value collected during construction operations is 29 degrees. Based on the optimized temperature threshold, it is determined that there are safety risks in real-time construction operations and a safety warning is issued.

[0145] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0146] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent supervision method for operation safety information, characterized in that: The method comprises the following steps: S100, using sensors to collect environmental data in historical work sites and records of dangerous accidents that occurred during historical operations, and searching for characteristic data that reflects dangerous accidents; S200, compare and screen the environmental data of safety warnings in different work sites in history to obtain potential data that affects the characteristic data; S300, constructing a scatter plot between each potential data and feature data, and performing linear regression calculation on the scatter plot to obtain a functional influence relationship between each potential data and feature data; S400, collecting characteristic data of dangerous accidents that occurred during operation in the past, and calculating characteristic thresholds for safety warning; S500, collecting potential data in the work site in real time, and optimizing the feature threshold in real time by using the functional influence relationship between each potential data and feature data; S600: Collect feature data in the work site in real time, and use the optimized feature threshold to provide real-time warning for safety during work.

2. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of searching for characteristic data reflecting dangerous accidents in S100 are: S101. Collect the records of dangerous accidents that occurred during operation in history, extract the environmental data before and after the dangerous accident, and assume that the environmental data before the dangerous accident is {H1, H2, H3, ..., H n }, H1, H2, H3, ..., H n represents the 1st, 2nd, 3rd, ..., nth environmental data before the dangerous accident occurred in the history, where n is a positive integer; suppose the environmental data after the dangerous accident is {H'1, H'2, H'3, ..., H' n }, H'1, H'2, H'3,..., H' n Indicates the 1st, 2nd, 3rd, ..., nth environmental data after a dangerous accident occurred during operation in history; extracts the environmental data from the mth record of dangerous accidents that occurred during operation in history; S102. Calculate the difference between each environmental data before and after a dangerous accident occurred during operation in history. The formula is: In the formula, H_c represents the average difference of each environmental data before and after the dangerous accident occurred during the operation in history, H represents each environmental data before the dangerous accident occurred during the operation in history, H' represents each environmental data after the dangerous accident occurred during the operation in history, and m represents the number of records of dangerous accidents during the operation in history collected; after calculation, the difference of environmental data before and after the dangerous accident occurred during the operation in history is {H_c1, H_c2, H_c3, ..., H_c n }, H_c1, H_c2, H_c3,..., H_c n Indicates the differences in the 1st, 2nd, 3rd, ..., nth environmental data before and after a dangerous accident occurred during operation in history; S103, screening n types of environmental data differences before and after dangerous accidents occurred during operation in history, selecting the type of environmental data corresponding to the maximum difference as characteristic data reflecting the dangerous accident, and setting the obtained characteristic data as T.

3. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of obtaining the potential data affecting the characteristic data through comparison and screening in 200 are: S201, collect the environmental data of safety warnings in different working sites in history, and assume that the collected environmental data of safety warnings are {S1, S2, S3, ..., S n }, S1, S2, S3, ..., S n represents the 1st, 2nd, 3rd, ..., nth environmental data during the safety warning in history, and collects the characteristic data of G safety warnings at different sites in history as {T1, T2, T3, ..., T G }, T1, T2, T3, ..., T G Indicates the characteristic data of the 1st, 2nd, 3rd, ..., Gth safety warnings at different sites in history, each characteristic data corresponds to a set of environmental data, and a set of environmental data includes n types of environmental data; S202, sorting the characteristic data of safety warnings at different sites collected in the G history from small to large, and calculating the difference between the characteristic data and the difference between each corresponding environmental data according to the arrangement order, the formula is: T_c=T i+1 -T i In the formula, T_c represents the difference between the last feature data and the previous one according to the arrangement order, T i+1 Indicates the next feature data in the sort, T i represents the previous feature data in the sorting, i∈(1, G), S_c represents the difference between the next and previous features of each environment data in the sorting, Indicates the next environmental data value of the same environmental data. Represents the previous environmental data value of the same environmental data, r∈(1,n); S203, after calculation, the G-1 feature data differences are {T_c1, T_c2, T_c3, ..., T_c G-1 }, T_c1, T_c2, T_c3,..., T_c G-1 Indicates the 1st, 2nd, 3rd, ..., G-1th characteristic data values ​​calculated, G-1 is a positive integer; correspondingly, G-1 differences of n kinds of environmental data are calculated; the difference between the characteristic data and the environmental data is judged, specifically: When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c≠0, the characteristic data difference T_c=0, the corresponding environmental data difference S_c=0, it is judged that the corresponding environmental data type has an impact risk; When the characteristic data difference T_c≠0, the corresponding environmental data difference S_c=0, and the characteristic data difference T_c=0, the corresponding environmental data difference S_c≠0, it is judged that the corresponding environmental data type has no impact risk; S204, judge G-1 feature data differences and environmental data differences. When the environmental data difference has an impact risk in G-1 judgments, determine the corresponding environmental data type as potential data that affects the feature data; when the environmental data difference has no impact risk in one of G-1 judgments, determine the corresponding environmental data type is not potential data that affects the feature data; the potential data that affects the feature data is obtained through judgment as {Q1, Q2, Q3, ..., Q p }, Q1, Q2, Q3, ..., Q p It represents the 1st, 2nd, 3rd, ..., pth potential data affecting the characteristic data obtained by judgment, where p is a positive integer.

4. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of obtaining the functional influence relationship between each potential data and characteristic data in S300 are: S301, extract different feature data when performing operations in the history as {T_z1, T_z2, T_z3, ..., T_z g }, T_z1, T_z2, T_z3,..., T_z g Indicates the 1st, 2nd, 3rd, ..., feature data in the collected history, g is a positive integer; according to the p kinds of potential data obtained by judgment, correspondingly collect g potential data values ​​of each potential data; and establish a scatter plot of each potential data and feature data, with the potential data as the independent variable and the feature data as the variable to construct the scatter plot; S302. After establishing a scatter plot of each potential data and characteristic data, calculate the regression curve in each scatter plot, and obtain the regression curve equation as T_z=a×Q+b, where T_z represents the variable characteristic data, Q represents the independent variable potential data, a represents the slope of the regression curve, and b represents the intercept of the regression curve; the slope a and intercept b in the regression curve are calculated based on the existing points in the scatter plot, and the formula is: b=T_z pi -a×Q pi In the formula, T_z pi represents the average value of g variable characteristic data, Q pi represents the average value of the potential data of g independent variables, T_z u Represents the characteristic data of each variable, Q u Represents the potential data of each independent variable, u∈(1,g); S303, after calculating, the regression curve equation of each scatter plot is obtained, and the regression curve equation in each scatter plot is used as the functional influence relationship between the independent variable potential data and the variable characteristic data in the scatter plot, and finally the functional influence relationship between the p kinds of potential data and the characteristic data is calculated as {f(Q1), f(Q2), f(Q3), ..., f(Q p )}, f(Q1), f(Q2), f(Q3),..., f(Q p ) represents the calculated functional influence relationship of the 1st, 2nd, 3rd, ..., kinds of potential data.

5. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of calculating the characteristic threshold for safety warning in S400 are: S401, collect the characteristic data of the work site when dangerous accidents occurred in history {T_d1, T_d2, T_d3, ..., T_d h }, T_d1, T_d2, T_d3,..., T_d h It represents the characteristic data collected in the work site when the 1st, 2nd, 3rd, ..., hth dangerous accidents occurred in the history, where h is a positive integer; S402. Calculate the characteristic threshold for safety warning based on the characteristic data of the work site when dangerous accidents occurred in the history collected. The formula is: In the formula, T_y represents the characteristic threshold used for safety warning, T_d represents the characteristic data of the work site when dangerous accidents occurred in the history collected in real time, T_st represents the standard deviation of the characteristic data, and h represents the number of characteristic data collected in the work site when dangerous accidents occurred in the history.

6. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of performing real-time optimization on the feature threshold in S500 are: S501, real-time collection of potential data in the work site {Q_s1, Q_s2, Q_s3, ..., Q_s p }, Q_s1, Q_s2, Q_s3,..., Q_s p represents the 1st, 2nd, 3rd, ..., pth potential data values ​​collected in real time; S502, extracting the functional influence relationship between each potential data and feature data in S303, and calculating the optimized value of the feature threshold, the formula is: In the formula, T_op represents the optimized value of the feature threshold, (a×Q_s+b) t represents the functional influence relationship between each potential data and feature data, t∈(1, p), Q_s represents the potential data collected in real time, T_y represents the feature threshold before optimization; p represents the number of potential data types; S503. Optimize the real-time feature threshold according to the calculated optimized value of the feature threshold. The optimization formula is: T_after=T_y+T_op, where T_after represents the optimized feature threshold, T_y represents the feature threshold before optimization, and T_op represents the optimized value of the feature threshold. Finally, the optimized feature threshold is T_after.

7. The intelligent monitoring method for operation safety information according to claim 1 is characterized in that: The specific steps of using the optimized characteristic threshold in S600 to provide a real-time warning for safety during operation are: S601. Collect the characteristic data of the work site in real time as T_s, and use the optimized characteristic threshold to issue a safety warning for the work site. When T_s≥T_after, it is determined that there is a safety risk at the work site and a warning is issued. When T_s<T_after, it is determined that there is no safety risk at the work site and normal operations can continue.

8. An intelligent monitoring system for operation safety information, characterized in that: The intelligent supervision system includes a data collection module, a potential data analysis module, a threshold calculation module, a threshold optimization module and an early warning module; The data collection module is used to collect environmental data when dangerous accidents occurred in history, and find out characteristic data reflecting dangerous accidents; The potential data analysis module is used to search for potential data that affects characteristic data in the environmental data according to the records of safety warnings that occurred in the history, and calculate the functional influence relationship between each potential data and characteristic data; The threshold calculation module is used to calculate the characteristic threshold value for determining whether there is a safety risk at the work site based on the characteristic data of dangerous accidents that occurred in the past; The threshold optimization module is used to calculate the optimized value of the feature threshold by using the functional influence relationship between each potential data and feature data, and optimize the feature threshold according to the optimized value; The early warning module is used to collect feature data in the work site in real time, and use the optimized feature threshold to issue safety warnings to the work site.

9. The intelligent monitoring system for operation safety information according to claim 8 is characterized in that: The data collection module includes a historical data collection unit and a feature data search unit; The historical data collection unit is used to collect environmental data when dangerous accidents occurred in the past; The characteristic data search unit is used to analyze the environmental data of dangerous accidents that have occurred in the history collected, calculate the difference and search for characteristic data that reflects the dangerous accidents.

10. The intelligent monitoring system for operation safety information according to claim 8, characterized in that: The potential data analysis module includes a potential data search unit and a function impact relationship calculation unit; The potential data search unit is used to perform difference calculation on the environmental data and the characteristic data according to the records of the safety warnings in history, and search for the potential data that affects the characteristic data; The function influence relationship calculation unit is used to establish a scatter plot of potential data and feature data, calculate the regression curve equation in the scatter plot, and determine the regression curve equation as the function influence relationship between the potential data and feature data.

Images