Security risk determination method, device and equipment
By inputting data into the LS model and LEC model and weighting processing, and adjusting the weighting coefficients in combination with big data, the problem of inaccurate security risk judgment methods is solved, and more efficient and accurate risk level determination is achieved.
Patent Information
- Application Number
- CN202310988658.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-07
- Publication Date
- 2025-02-14
AI Technical Summary
The existing security risk judgment methods rely on natural person participation and simple calculation methods, resulting in inaccurate risk judgment and failure to effectively comprehensively consider different risk judgment methods and big data information.
A security risk level determination method that comprehensively considers big data processing and different risk judgment methods is adopted. By inputting the collected data into the LS model and the LEC model, weighting is carried out to form the final risk judgment result, and using big data pairs The weighting coefficient is adjusted to improve judgment accuracy.
The accuracy of security risk judgment is improved, and the influence of human factors is reduced through data screening and weighting processing, and the risk judgment model is continuously optimized through feedback adjustment.
Smart Images
Figure CN119443765A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of computer technology, and in particular, relates to a security risk determination method, device and equipment. Background Art
[0002] In actual production and operation activities, in order to strengthen safety management and accurately identify the risk points and harmful factors in production and operation activities, various entities have determined the severity of the risks through the acquisition of some data, and carried out risk classification, so as to determine the risk control of possible risks, clarify the priority and risk control measures of risk control, and achieve the purpose of prior prevention, hazard reduction, and risk level reduction. Many entities have formulated safety management measures.
[0003] There are many studies on the judgment of safety risks. For example, existing methods include the risk assessment method of the operating condition hazard analysis and evaluation method LCE (L (likelihood, the possibility of an accident), E (exposure, the frequency of personnel exposure to dangerous environments), C (consequence, the possible consequences of an accident)) and the risk matrix method (LS). In the prior art, there are many discussions on the judgment of these safety risks and standards have been formed.
[0004] The LCE risk assessment method mainly takes the danger of the working conditions as the dependent variable D, the possibility of accidents or dangerous events L, the frequency of people's exposure to dangerous environments E, and the losses C caused by accidents as independent variables, adopts a semi-quantitative valuation method, and determines the scores of the three independent variables respectively, and finally calculates the risk size through D = L × E × C. The risk matrix method (LS) mainly identifies the hazards that may exist in each operating unit, and determines the possible consequences of such hazards and the possibility of such consequences. The two are multiplied to obtain the risk of the determined hazards.
[0005] These safety risk assessment methods are relatively traditional. In the process of determining the variables in various risk assessment methods, there are many human factors due to the inevitable participation of natural persons. In addition, since the calculation is simply multiplication, and different risk assessment methods are not comprehensively considered, and big data is not combined, the risk assessment is inaccurate.
[0006] Based on this, a method for determining the security risk level is needed that utilizes big data and comprehensively considers different risk judgment methods. Summary of the invention
[0007] To solve the above problems, the present disclosure provides a security risk determination method, device and equipment, which comprehensively consider the processing of big data and different risk judgment methods, and can make feedback adjustments based on the judgment results.
[0008] The present invention provides a method for determining a security risk level, the method comprising:
[0009] Inputting the collected data into a risk model, wherein the risk model includes an LS model and a LEC model;
[0010] The LS model and the LEC model process the received data respectively, and form a processing result R and a processing result D respectively;
[0011] The processing result R and the processing result D are weighted to form the risk judgment result Y.
[0012] Furthermore, the step of inputting the collected data into the risk model includes:
[0013] Collect risk data, screen the collected risk data, and form screened data;
[0014] The screened data is input into the risk model.
[0015] Further, the risk judgment result Y is formed in the following manner:
[0016] Y = α × r1 × R + β × r2 × D;
[0017] Among them, α and β are weighting coefficients, r1 and r2 are normalization factors;
[0018] r1=1 / 25;
[0019] α+β=1, and 0<α, β<1.
[0020] Furthermore, α = [0.4-0.49], β = [0.51-0.6].
[0021] Furthermore, the method further comprises:
[0022] The risk model is trained using big data to determine the weighted coefficients of the weighting.
[0023] The present disclosure also provides a security risk judgment device, the device comprising:
[0024] A collection unit configured to input the collected data into a risk model, wherein the risk model includes an LS model and a LEC model;
[0025] The processing unit is configured to process the received data using the LS model and the LEC model respectively to form a processing result R and a processing result D respectively; and weight the processing result R and the processing result D to form a risk judgment result Y.
[0026] Further, the risk judgment result Y is formed in the following manner:
[0027] Y = α × r1 × R + β × r2 × D;
[0028] Among them, α and β are weighting coefficients, r1 and r2 are normalization factors;
[0029] r1=1 / 25;
[0030] α+β=1, and 0<α, β<1.
[0031] Furthermore, α = [0.4-0.49], β = [0.51-0.6].
[0032] Furthermore, the device also includes:
[0033] The weighting coefficient determination unit is configured to train the risk model using big data to determine the weighted weighting coefficient.
[0034] The present disclosure also provides a risk judgment device, the risk judgment device comprising at least one processor and at least one memory, the memory being electrically connected to the processor, wherein:
[0035] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the security risk determination method described in the present disclosure.
[0036] Compared with the prior art, the present invention has the following advantages:
[0037] In this disclosure, the collected data is screened to make the data before risk judgment more accurate, providing a basic guarantee for the accuracy of judgment;
[0038] The present disclosure comprehensively considers the two risk judgment methods of LEC and LS. Instead of simply using the existing LEC and LS, it fully utilizes the advantages of the two risk judgments to further improve the accuracy of the judgment;
[0039] In the present disclosure, during the risk judgment process, the risk judgment model is adjusted in real time according to the judgment result, that is, the risk judgment model is adjusted by providing feedback through the judgment result. Therefore, the inaccuracy of risk judgment can be further corrected.
[0040] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A schematic diagram of a security risk level determination structure according to an embodiment of the present disclosure is shown;
[0043] Figure 2 A schematic diagram of a security risk level process according to an embodiment of the present disclosure is shown;
[0044] Figure 3 A block diagram of a device for determining a security risk level according to an embodiment of the present disclosure is shown;
[0045] Figure 4 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.
[0047] The security risk determination method of the embodiment of the present disclosure mainly includes risk data collection, screening of the collected risk data, inputting the screened risk data into the LS risk judgment model and the LEC risk judgment model respectively, and the LS risk judgment model and the LEC risk judgment model process the received risk data respectively. Finally, the processing results of the LS risk judgment model and the LEC risk judgment model are weighted to output the final risk judgment result. Considering that the output results of the LS risk judgment model and the LEC risk judgment model may have a large deviation, the present disclosure also uses the judgment result to adjust the weighting coefficient in the above-mentioned weighted processing.
[0048] In the embodiments of the present disclosure, the safety production of petrochemical industry is taken as an example for illustration, but it should be noted that it is not limited to the existing petrochemical production. Figure 1 A schematic diagram of a security risk level determination structure according to an embodiment of the present disclosure is shown. Figure 1As shown in Figure 1, it mainly includes data collection, data screening and risk judgment. In the risk judgment part, the LS model and LEC model are used in combination. Figure 2 FIG. 4 shows a schematic diagram of the security risk level process of an embodiment of the present disclosure. Figure 2 As shown in the figure, it mainly includes collecting risk data, screening and checking the collected risk data; inputting the screened risk data into the risk model, making risk judgments, and finally outputting the risk judgment results. Figure 2 The step of judging whether the result is accurate in the dotted line is to adjust the relevant parameters (such as weight coefficient) in the risk model. This step can be used to adjust the relevant parameters in the risk model in a timely manner before the risk judgment of the actual project or when the risk judgment deviates.
[0049] Exemplarily, a computer collects data such as the probability of accidents occurring in the petrochemical production process (e.g., the score of an anti-static grounding device that is damaged but not repaired), the severity of the consequences of the incident (e.g., the score of causing injuries to petrochemical workers), etc.; the collected data is screened to improve the accuracy of risk judgment; the screened data is input into the risk model, i.e., the data is respectively entered into the LS model and the LEC model for calculation, and the results of the LS model and the LEC model are weightedly added to form a judgment result, and the weighting coefficient in the weighted calculation can be adjusted accordingly according to the judgment result to further improve the accuracy of the final data judgment result.
[0050] There are various complex on-site conditions in petrochemical engineering sites, and there are many interference factors, so the measured data may be wrong, invalid or repeated due to interference problems. Based on this, in the embodiment of the present disclosure, a data screening step is set to exclude data that is obviously inconsistent with the actual situation.
[0051] In the data screening step, the collected data can be excluded by comparing it with the set threshold. If the collected data exceeds the set threshold, it means that the data has exceeded the normal range and is an abnormal data, so the data can be deleted. It can also be excluded by comparing the previous and next data and using the gradient calculation method. If the difference between the later data and the previous one or two or the data within a reasonably short time range exceeds the set threshold, if the difference between the previous and next data is large, it means that a very large change has occurred in a short period of time, which is not in line with the actual situation. Therefore, it can be considered that the data of the receipt is an abnormal data, so the data can be deleted.
[0052] In order to further improve the accuracy of data screening, when the collected data is identified as abnormal data, an alarm message or prompt message can be issued to prompt the on-site personnel to make a manual judgment to determine whether the collected data is abnormal data, and finally determine whether to delete the data based on the manual judgment result. In the embodiment of the present disclosure, when screening data, the data can be labeled, for example, exclusion tags can be set for low-importance and repeated data to prevent these data from entering the risk model and affecting the judgment of risk.
[0053] The collected data are screened through the data screening step to generate screened data Data. The screened data are respectively input into the LS model and the LEC model in the risk model.
[0054] In the embodiment of the present disclosure, the LS model can be defined as follows:
[0055] Risk matrix analysis method (LS), R (risk value) = L × S.
[0056] According to relevant safety requirements, the criteria for determining the probability of an accident (L) are shown in the following table:
[0057]
[0058] The criteria for determining the severity of the consequences of an incident (S) are shown in the following table:
[0059] Points Security Requirements Personal injury Economic losses 5 Violation of laws, regulations and standards die More than one million 4 Potential violations of regulations and standards Loss of ability to work More than 500,000 3 Does not comply with parent company or industry standards fracture More than 10,000 2 Not in compliance with company operating procedures Minor injuries Less than 10,000 yuan 1 Fully compliant No casualties No loss
[0060] After the above scores pass through the LS model, the risk value R is calculated using the calculation method of the model: R (risk value) = L×S.
[0061] The safety risk level determination criteria (R) and control measures are shown in the following table:
[0062]
[0063]
[0064] The risk matrix can be formed by using the criteria of the possibility of accident (L) and the severity of the consequences (S), as shown in the following table:
[0065] Consequence level 1 2 3 4 5 5 Slightly dangerous Significant danger Highly dangerous Extremely dangerous Extremely dangerous 4 Slightly dangerous Slightly dangerous Significant danger Highly dangerous Extremely dangerous 3 Slightly dangerous Slightly dangerous Significant danger Significant danger Highly dangerous 2 Slightly dangerous Slightly dangerous Slightly dangerous Slightly dangerous Significant danger 1 Slightly dangerous Slightly dangerous Slightly dangerous Slightly dangerous Slightly dangerous
[0066] It can be seen from the above table that the consequence level values are all 5, that is, when the calculation result is 25, the corresponding level is extremely dangerous, and this consequence has the highest consequence level; and the consequence level values are all 1, that is, when the calculation result is 1, the corresponding level is slightly dangerous, and this consequence has the lowest consequence level.
[0067] It should be noted that the values and definitions in the above tables are an exemplary description of the present technical solution and can be set according to actual conditions in actual application projects, and are not limited to the above values and definitions.
[0068] In the embodiment of the present disclosure, the LEC model can be defined as follows:
[0069] The risk assessment method for operating conditions (LEC) is as follows:
[0070] D=L×E×C
[0071] The criteria for determining the probability of an accident (L) are shown in the following table:
[0072]
[0073] The criteria for determining the frequency of exposure to hazardous environments (E) are shown in the following table:
[0074] Points Frequency 10 Continuous exposure 6 Exposure during daily working hours 3 Weekly or occasional exposure 2 Exposure once a month 1 Exposure several times a year 0.5 Very rare exposure
[0075] The criteria for determining the severity of the consequences (C) of the deviation in an accident event are shown in the following table:
[0076]
[0077]
[0078] Through the calculation of D = L × E × C, the risk level determination criteria (D) and control measures are shown in the following table:
[0079]
[0080] As can be seen from the above table, a risk value less than 20 belongs to a slightly dangerous risk level, and a risk value greater than 320 belongs to an extremely dangerous risk level.
[0081] Similarly, the values and definitions in the above tables are an exemplary description of the present technical solution, and can be set according to actual conditions in actual application projects, and are not limited to the above values and definitions.
[0082] In the embodiment of the present disclosure, in order to further improve the accuracy of risk judgment, two risk judgment models are comprehensively considered, and the results R and D output by the LS model and the LEC model are weighted:
[0083] The final output result Y=α×r1×R+β×r2×D, where α and β are weighting coefficients, and r1 and r2 are normalization factors.
[0084] In the embodiment of the present disclosure, the maximum values of the LS model and the LEC model and the risk levels corresponding to the values are considered respectively, and r1 and r2 are set as follows:
[0085] r1=1 / 25; in Indicates rounding up.
[0086] The above two normalization factors are set with full consideration that the output value range of the LS model is 1-25, and each risk level range is evenly distributed. The output value of the LEC model is between greater than 0 and less than or equal to 10000, and the length of the value range of each risk level is different. And the two models are set in consideration of unified calculation.
[0087] α and β are weighting coefficients of α+β=1, and 0<α, β<1. In the embodiment of the present disclosure, considering that the LEC model takes into account more factors and is more conducive to the judgment of the risk level, its weight is set to be larger, and it is preferred to adopt α=[0.4-0.49], β=[0.51-0.6], that is, α values between 0.4-0.49 and β values between 0.51-0.6 are the better way. In actual use, the above two weighting coefficients can be appropriately adjusted within the above range. In the embodiment of the present disclosure, the two risk judgment methods of LEC and LS are comprehensively considered. Instead of simply using the existing LEC and LS, the advantages of the two risk judgments are fully utilized, and the specific methods of the two risk judgments are fully considered. Corresponding parameters such as normalization and weighting coefficients are set to further improve the accuracy of the judgment.
[0088] Through the above calculation of Y=α×r1×R+β×r2×D, the judgment value is output. In the embodiment of the present disclosure, different risk levels corresponding to the Y value can be set according to the actual engineering situation, as shown in the following table:
[0089] Value at Risk Risk Level >0.8 Highest risk 0.6-0.8 Higher risk 0.4-0.6 High risk 0.2-0.4 Low risk <0.2 Lower risk
[0090] In order to improve the accuracy of the model, in the implementation of this disclosure, Figure 1 The risk model in the figure can be trained in advance using big data to determine the two weighting coefficient parameters α and β.
[0091] The training process can be: select some determined risk data from the historical big data collected and calculated over a long period of time, input the selected risk data into the risk model, compare the judgment result output by the risk model with the accurate judgment result, and if the two judgment results are quite different, such as being at different risk levels, adjust the two values of α and β respectively until all specified risk data are verified, and the output judgment result is the same as the accurate judgment result or within the set error range. In the risk judgment process of the disclosed embodiment, the risk judgment model is adjusted or trained in real time according to the judgment result, that is, the risk judgment model is adjusted by feedback of the judgment result, so that the inaccuracy of the risk judgment can be further corrected.
[0092] The final α and β parameters formed by training are used for actual risk judgment. In the actual judgment process, due to the large number of parameters, the input data may be inaccurate or repeated, which may lead to deviations in the output results. At this time, the α and β parameters can be adjusted according to the output judgment results.
[0093] Based on the above method, the embodiment of the present disclosure also provides a risk determination device corresponding to the above method. Figure 3 A block diagram of a security risk determination device according to an embodiment of the present disclosure is shown. Figure 3 As shown, the risk determination device includes an electrically connected collection unit, a processing unit and a weighting coefficient determination unit, wherein the collection unit is configured to input the collected data into a risk model, wherein the risk model includes an LS model and a LEC model; the processing unit is configured to use the LS model and the LEC model to process the received data respectively to form a processing result R and a processing result D respectively; the processing result R and the processing result D are weighted to form a risk judgment result Y; the weighting coefficient determination unit is configured to use big data to train the risk model to determine the weighting coefficient of the weighting.
[0094] Based on the above disclosed content, the present disclosure also provides an electronic device. Figure 4 As shown, the electronic device of an embodiment of the present disclosure includes at least one electrically connected processor and at least one memory, wherein the memory is electrically connected to the processor, wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method as described above.
[0095] It should be noted that the electrical connection between the above-mentioned units does not necessarily mean the connection between lines. An indirect connection method can be applied to the embodiments of the present disclosure as long as the purpose of the present disclosure is achieved.
[0096] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for determining security risks, the method comprising: Inputting the collected data into a risk model, wherein the risk model includes an LS model and a LEC model; The LS model and the LEC model process the received data respectively, and form a processing result R and a processing result D respectively; The processing result R and the processing result D are weighted to form the risk judgment result Y.
2. The method according to claim 1, wherein: The step of inputting the collected data into the risk model includes: Collect risk data, screen the collected risk data, and form screened data; The screened data is input into the risk model.
3. The method according to claim 1, wherein: The risk judgment result Y is formed in the following way: Y = α × r1 × R + β × r2 × D; Among them, α and β are weighting coefficients, r1 and r2 are normalization factors; r1＝1 / 25； α+β=1, and 0<α, β<1.
4. The method according to claim 3, wherein: α＝[0.4-0.49]，β＝[0.51-0.6]。 5. The method according to any one of claims 1 to 4, further comprising: The risk model is trained using big data to determine the weighted coefficients of the weighting.
6. A security risk determination device, the device comprising: A collection unit configured to input the collected data into a risk model, wherein the risk model includes an LS model and a LEC model; A processing unit configured to process the received data using the LS model and the LEC model to form a processing result R and a processing result D respectively; The processing result R and the processing result D are weighted to form the risk judgment result Y.
7. The device according to claim 6, wherein: The risk judgment result Y is formed in the following way: Y = α × r1 × R + β × r2 × D; Among them, α and β are weighting coefficients, r1 and r2 are normalization factors; r1＝1 / 25； α+β=1, and 0<α, β<1.
8. The device according to claim 7, wherein: α＝[0.4-0.49]，β＝[0.51-0.6]。 9. The device according to any one of claims 6 to 8, further comprising: The weighting coefficient determination unit is configured to train the risk model using big data to determine the weighted weighting coefficient.
10. A risk determination device, comprising at least one processor and at least one memory, wherein the memory is electrically connected to the processor, wherein: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.