Hazardous chemicals transportation risk assessment method and system based on neural network

Abstract

The present invention discloses a method and system for assessing the risk of hazardous chemicals transportation based on a neural network, which belongs to the technical field of safe transportation assessment of hazardous chemicals, and includes an information collection module, an objective risk assessment module, a subjective risk weighting module, a result determination module, and an optimization solution generation module; the information collection module collects information on hazardous chemicals related to transportation, including types of hazardous chemicals, hazardous chemical transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies. The method and system for assessing the risk of hazardous chemicals transportation based on a neural network of the present invention can realize data acquisition, objective risk assessment, subjective risk assessment, subjective and objective risk weighting assessment, result determination, and optimization solution generation, which is helpful to comprehensively assess the relevant hazardous chemicals transportation risks from multiple dimensions, and provide risk optimization suggestions for relevant enterprises, thereby enhancing the accuracy and reliability of hazardous chemicals risk assessment.

Description

Technical Field

[0001] The present invention belongs to the technical field of hazardous chemicals safe transportation assessment, and specifically is a hazardous chemicals transportation risk assessment method and system based on neural network. Background Art

[0002] Hazardous chemicals (hereinafter referred to as hazardous chemicals) refer to chemical substances with dangerous properties, such as flammability, explosiveness, corrosion, toxicity, reactivity, etc., which may cause harm to human health, environment and property safety. Therefore, it is necessary to collect relevant information before the transportation of hazardous chemicals to assess the risks of the transportation of hazardous chemicals;

[0003] Most of the existing dangerous goods transportation risk assessment systems conduct risk assessment on the dangerous goods transportation based on objective factors between the two transportation locations, but fail to take into account relevant subjective factors of the dangerous goods transportation, such as the accident rate of the transportation company, the violation of the transportation company, the operating time of the transportation company, the supporting services of the transportation company, etc. These subjective factors cannot be incorporated into the dangerous goods risk transportation assessment process, resulting in the risk assessment results having a weak reference value in actual situations, and having problems of low practicality and functionality;

[0004] In response to the above, this case proposes a hazardous chemicals transportation risk assessment method and system based on neural network, which solves the above technical problems through subjective risk weighting module, objective risk assessment module and optimization solution generation module. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method and system for assessing the risk of hazardous chemicals transportation based on a neural network, which solves the above technical problems by improving the detection method and the processing method.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A risk assessment method and system for hazardous chemicals transportation based on neural networks, including an information collection module, an objective risk assessment module, a subjective risk weighting module, a result determination module, and an optimization solution generation module;

[0008] The information collection module collects information on hazardous chemicals related to transportation, including types of hazardous chemicals, hazardous chemical transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies, which includes company names, accident records, violations, years of operation, and related transportation route supporting services;

[0009] The objective risk assessment module, based on the transportation-related hazardous chemicals information collected by the information collection module, collects the road condition information and weather information of the two transportation locations, collects historical data and establishes an assessment model through a convolutional neural network, conducts a first round of risk assessment on the relevant hazardous chemicals transportation, and obtains an objective risk assessment value;

[0010] The subjective risk weighting module calculates the subjective risk weighting value based on the hazardous chemicals transportation company database and the transportation route of the hazardous chemicals, and outputs the subjective risk weighting value to the objective risk assessment module, performs subjective risk assessment on the objective risk assessment value and obtains the final risk assessment value;

[0011] The result determination module is based on the final risk assessment value and risk threshold , the risk of hazardous chemicals transportation is determined, and when the determination result is unqualified, the output signal enters the optimization solution generation module;

[0012] The optimization scheme generation module combines the relevant data of the subjective risk weighting module and the objective risk assessment module to optimize the risk of this hazardous chemicals transportation plan, conduct risk assessment on the optimization scheme, and output the results to the result judgment module.

[0013] Furthermore, the objective risk assessment module, based on the transportation-related hazardous chemicals information collected by the information collection module, collects the road condition information and weather information of the two transportation locations, collects historical data, establishes an objective risk assessment model through a convolutional neural network, conducts a first round of risk assessment on the relevant hazardous chemicals transportation, and obtains an objective risk assessment value. The specific steps are as follows:

[0014] Collect historical hazardous chemicals information, road conditions, weather information and risk values, extract relevant features from the original data, and build an objective risk assessment model based on a convolutional neural network. The specific steps are as follows:

[0015] Hazardous chemicals data include types, total transport volume, transport times, transport methods, routes; road condition data include low traffic volume, medium traffic volume, and high traffic volume; weather data include low, medium, and high visibility, as well as weather conditions such as sunny, light rain, moderate rain, and heavy rain. Visibility, weather conditions, and traffic volume data are encoded using unique hot coding;

[0016] Divide historical data into training set, validation set and test set, build an objective risk assessment model based on the convolutional neural network through the training set, and verify the constructed objective risk assessment model based on the validation set and test set;

[0017] choose As an activation function, the output value of the risk assessment objective model is compressed into the range of 0 to 1 and converted into the range of 0 to 100. The algorithm formula is: ;

[0018] in, represent The output value of the activation function, Represents the output value of the objective risk assessment model constructed by the convolutional neural network;

[0019] Combined with the hazardous chemicals transportation information collected by the information collection module, the model risk value is calculated through the risk assessment objective model to obtain the objective risk assessment value.

[0020] Furthermore, the subjective risk weighting module calculates the subjective risk weighting value based on the hazardous chemicals transportation company database and the transportation route of the hazardous chemicals, and outputs the subjective risk weighting value to the objective risk assessment module, performs subjective risk assessment on the objective risk assessment value and obtains the final risk assessment value, and the specific steps are as follows:

[0021] Through the hazardous chemicals transportation company database in the information collection module, obtain the violation information and accident records, operating years, and related transportation route supporting services of the relevant hazardous chemicals transportation companies, and conduct a comprehensive assessment of each item based on the above information;

[0022] Based on the results of the comprehensive assessment, a weight is set for each influencing project to obtain a subjective risk weighted value, and the subjective risk weighted value is output to the objective risk assessment module. A subjective risk assessment is performed on the objective risk assessment value to obtain the final risk assessment value.

[0023] Furthermore, the hazardous chemicals transportation company database in the information collection module is used to obtain the violation information, accident records, operating years, and related transportation route supporting services of the relevant hazardous chemicals transportation companies, and a comprehensive assessment is conducted on each item based on the above information. The specific steps are as follows:

[0024] Obtain the total number of transport dispatches of relevant hazardous chemicals transport companies through the information collection module And the number of points deducted for transport violations , calculate the violation rate of the relevant hazardous chemicals transportation companies: ;

[0025] in, Represents the violation rate of the relevant hazardous chemicals transportation company, and obtains the violation rate score based on the violation rate of the vehicle. ,in ;

[0026] Obtain the total number of hazardous chemicals transported by the relevant hazardous chemicals transport company through the information collection module and the number of hazardous chemicals transportation accidents , calculate the accident rate of relevant hazardous chemicals transportation companies: ;

[0027] in, Represents the accident rate of the relevant hazardous chemicals transportation company, and obtains the accident rate score based on the accident rate calculation ,in ;

[0028] Obtain the operating years of relevant hazardous chemicals transportation companies through the information collection module , based on the operating years value, obtain the operating years score:

[0029] when When ;

[0030] when When ;

[0031] when When ;

[0032] The information collection module is used to obtain the supporting services of the relevant operating routes of the relevant hazardous chemicals transportation companies, including the number of emergency stations, maintenance service points, and fire emergency points, and the total coverage mileage of the security points of the relevant sections. , and based on the total mileage of the relevant road section , calculate the coverage rate of the guarantee points : ;

[0033] Based on coverage of guarantee points Calculate the score of supporting services ;

[0034] Collect the number of times drivers travel the relevant road segments , calculate the route familiarity score, the steps are:

[0035] when When ;

[0036] when When ;

[0037] when When .

[0038] Furthermore, based on the result of the comprehensive assessment, a weight is set for each influencing project to obtain a subjective risk weighted value, and the specific algorithm formula is: ;

[0039] ;

[0040] in, Represent the weight proportion of the corresponding scores respectively. Represents the subjective risk weighted value. Based on the objective risk assessment value and the subjective risk weighted value, the final risk assessment value is obtained. , and its algorithm formula is: ;

[0041] ;

[0042] in, Represent the weight ratio of objective risk assessment value and subjective risk weighted value respectively. Represents the objective risk assessment value obtained based on the objective risk assessment model.

[0043] Furthermore, the optimization scheme generation module combines the relevant data of the subjective risk weighting module and the objective risk assessment module to optimize the risk of this hazardous chemicals transportation scheme, conduct risk assessment on the optimization scheme, and output the results to the result determination module. The specific steps are as follows:

[0044] Based on the relevant data of the objective risk assessment module, it generates optimization suggestions for the total transport volume, number of transports, and transport time. Based on the relevant data of the subjective risk weighting module, it generates optimization adjustment opinions for the drivers and sets the optimization adjustment priority.

[0045] According to the objective parameter adjustment value and the subjective parameter adjustment value, the relevant data is transmitted to the objective risk assessment module and the subjective risk weighting module respectively to calculate the optimized risk assessment value. , transmission optimization risk assessment value Enter the result determination module and the risk threshold Make a judgment:

[0046] when When the optimization risk assessment fails, the hazardous chemicals transportation plan is transmitted for risk marking;

[0047] when When the optimization risk assessment is qualified, an optimization plan is generated and transmitted to the relevant hazardous chemicals transportation company, waiting for confirmation by the hazardous chemicals transportation company.

[0048] Furthermore, based on the relevant data of the objective risk assessment module, the adjustment optimization suggestions are generated for the total transportation volume, transportation times, and transportation time. Based on the relevant data of the subjective risk weighting module, the optimization adjustment opinions are generated for the drivers, and the optimization adjustment priority is set. The specific steps are:

[0049] Set a single transport volume threshold for the total transport volume and number of transports , according to the total transport volume value The number of transports , calculate the single transport volume of the original plan : ;

[0050] when ≤ No adjustment is made when

[0051] when > When the transport times are adjusted, ≤ ;

[0052] With regard to the transportation time, the original transportation time is postponed in combination with the weather and road conditions during the relevant time period, and the weather and road conditions during the relevant time period are re-obtained;

[0053] For drivers who have traveled the relevant route less than 40 times, optimization adjustment suggestions are generated, that is, drivers who have traveled the relevant route more than 40 times are selected;

[0054] Set optimization adjustment priorities and calculate the pending optimization risk assessment value after each priority optimization adjustment and risk threshold Perform a pre-alignment:

[0055] when < When the optimization adjustment signal is transmitted to the optimization scheme generation module for optimization, when both the subjective parameter and objective parameter related optimization strategies are used, Substituting values ,transmission Enter the result determination module;

[0056] When any < When the optimization adjustment signal is transmitted to the optimization scheme generation module for optimization, when both the subjective parameter and objective parameter related optimization strategies are used, Substituting values ,transmission Enter the result determination module;

[0057] When any ≥ When Substituting values ,transmission Enter the result determination module;

[0058] Among them, the priority of optimization adjustment is driver adjustment > single transport volume adjustment > transport time adjustment.

[0059] Furthermore, the information collection module collects information on hazardous chemicals related to transportation, including types of hazardous chemicals, hazardous chemical transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies, which includes company names, accident records, violations, years of operation, and related transportation route supporting services;

[0060] The supporting services of the relevant transport routes include the number of driving times along the routes and the number of supporting stations along the routes, including the number of first aid stations, maintenance service points, and fire emergency points;

[0061] The result determination module is based on the final risk assessment value and risk threshold , the risk of hazardous chemicals transportation is determined. When the determination result is unqualified, the output signal enters the optimization solution generation module. The specific steps are:

[0062] when When , it means that the final risk assessment is unqualified, and the transmission signal enters the optimization solution generation module;

[0063] when , it means the final risk assessment is qualified.

[0064] The risk assessment method for hazardous chemicals transportation based on neural network includes the following steps:

[0065] S1. Based on this hazardous chemicals transportation plan, collect relevant hazardous chemicals information, including types of hazardous chemicals, hazardous chemicals transportation volume, transportation methods, and transportation routes. At the same time, establish a hazardous chemicals transportation company database to collect and record the names of relevant hazardous chemicals transportation companies, accident records, violations, years of operation, and related transportation route supporting services;

[0066] S2. Based on the relevant hazardous chemicals information of this transportation, collect the road condition information and weather information of the two transportation locations, and establish an objective risk assessment model through a convolutional neural network to conduct an objective risk assessment on this hazardous chemicals transportation plan and obtain an objective risk assessment value;

[0067] S3. Based on the relevant company data in the hazardous chemicals transportation company database, including the hazardous chemicals transportation company's violation and accident records, operating years, and related transportation route supporting services, conduct a comprehensive assessment of each item and set a weight value to obtain a subjective risk weighted value;

[0068] S4. Obtain the final risk assessment value by weighted calculation of the objective risk assessment value and the subjective risk weighted value, and compare it with the risk threshold. If the result is unqualified, optimize and adjust the transportation plan based on the priority of the objective parameters and the subjective parameters.

[0069] S5. Compare and judge the optimized risk assessment value after optimization and adjustment with the risk threshold. If it is qualified, transmit the optimized adjustment plan to the relevant hazardous chemicals transportation company for confirmation. If it is unqualified, mark the risk for this hazardous chemicals transportation plan.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. In the present invention, by setting a subjective risk weighting module, based on the relevant subjective influencing factors of the hazardous chemicals transportation company, the violation situation of the relevant hazardous chemicals transportation company, accident record, operating years, and related transportation route supporting services are evaluated, and the above parameters are weighted to generate a subjective risk weighted value, which can more comprehensively evaluate the actual risk of the transportation company and make a more accurate prediction of the risk level of the transportation company;

[0072] 2. In the present invention, the final risk assessment value of the transportation of hazardous chemicals is obtained by weighted addition of the objective risk assessment value and the subjective risk weighted value, so as to achieve a comprehensive assessment of the transportation risk of hazardous chemicals. The accuracy of the assessment of the transportation of hazardous chemicals is improved through a comprehensive assessment method, and more comprehensive risk information can be captured;

[0073] 3. In the present invention, by adjusting the weight value based on the actual situation, the influence proportion of each influencing parameter can be flexibly adjusted according to different types of hazardous chemicals. By adjusting the weight value, the specific risk characteristics of each hazardous chemical can be more accurately reflected, the accuracy of the assessment can be improved, and the practicality and functionality can be enhanced;

[0074] 4. In the present invention, by setting an optimization scheme generation module to cooperate with a subjective risk weighting module and an objective risk assessment module, the transportation plan that fails the final risk assessment is optimized and adjusted, and risk optimization suggestions are provided to relevant transportation companies for reference. The transportation companies can make flexible choices based on the actual adjusted costs and impacts, helping them find the best balance between safety, cost and efficiency;

[0075] The entire neural network-based hazardous chemicals transportation risk assessment method and system can realize data acquisition, objective risk assessment, subjective risk assessment, subjective and objective risk weighted assessment, result judgment, and optimization plan generation. It helps to comprehensively evaluate relevant hazardous chemicals transportation risks from multiple dimensions and provide risk optimization suggestions for related companies, thereby enhancing the accuracy and reliability of hazardous chemicals risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a neural network-based hazardous chemicals transportation risk assessment method and system block diagram of the present invention. DETAILED DESCRIPTION

[0077] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0078] like Figure 1 As shown, the risk assessment system for hazardous chemicals transportation based on neural network includes information collection module, objective risk assessment module, subjective risk weighting module, result determination module and optimization scheme generation module;

[0079] The information collection module collects information related to the transportation of hazardous chemicals, including types of hazardous chemicals, transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies, which includes company names, accident records, violations, years of operation, and related transportation route supporting services;

[0080] The supporting services of relevant transport routes include the number of route driving times and the number of supporting stations along the route, including the number of first aid stations, maintenance service points, and fire emergency points.

[0081] It should be noted that the types of hazardous chemicals collected and transported, including different categories of products such as flammable, explosive, corrosive, and toxic, help identify their specific transportation requirements and potential risks. The transportation routes include the starting point, end point, places along the way, and distance.

[0082] The objective risk assessment module, based on the transportation-related hazardous chemicals information collected by the information collection module, collects the road conditions and weather information of the two transportation locations, collects historical data and establishes an assessment model through a convolutional neural network, conducts a first round of risk assessment on the relevant hazardous chemicals transportation, and obtains an objective risk assessment value. The specific steps are as follows:

[0083] Collect historical hazardous chemicals information, road conditions, weather information and risk values, extract relevant features from the original data, and build an objective risk assessment model based on a convolutional neural network. The specific steps are as follows:

[0084] The hazardous chemicals data include type, total transportation volume, number of transportation times, transportation method, and route. The road condition data includes low traffic volume, medium traffic volume, and high traffic volume. The weather data includes low, medium, and high visibility, as well as weather conditions such as sunny, light rain, moderate rain, and heavy rain. The visibility, weather conditions, and traffic volume data are encoded using unique hot encoding.

[0085] It should be noted that the risk values ​​in the historical data need to be set in advance by professionals in related fields, and the value range is 0-100. The unique hot encoding takes the weather status as an example, and creates a new binary feature column for "sunny", "light rain", "moderate rain" and "heavy rain". Each category is mapped to a unique binary feature column. At the same time, in each data, only one feature column has a value of 1, and the rest are 0;

[0086] Original weather conditions sunny light rain moderate rain rainstorm sunny 1 0 0 0 light rain 0 1 0 0 moderate rain 0 0 1 0 rainstorm 0 0 0 1

[0087] Divide historical data into training set, validation set and test set, build an objective risk assessment model based on the convolutional neural network through the training set, and verify the constructed objective risk assessment model based on the validation set and test set;

[0088] choose As an activation function, the output value of the risk assessment objective model is compressed into the range of 0 to 1 and converted into the range of 0 to 100. The algorithm formula is: ;

[0089] in, represent The output value of the activation function, Represents the output value of the objective risk assessment model constructed by the convolutional neural network;

[0090] Combined with the hazardous chemicals transportation information collected by the information collection module, the model risk value is calculated through the risk assessment objective model to obtain the objective risk assessment value.

[0091] It should be noted that the output value of the risk assessment objective model For example: ;

[0092] That is, the objective risk assessment value this time is 76.8.

[0093] The subjective risk weighting module calculates the subjective risk weighting value based on the hazardous chemicals transportation company database and the transportation route of the hazardous chemicals, and outputs the subjective risk weighting value to the objective risk assessment module, performs subjective risk assessment on the objective risk assessment value and obtains the final risk assessment value. The specific steps are as follows:

[0094] Through the hazardous chemicals transportation company database in the information collection module, obtain the violation information and accident records, operating years, and related transportation route supporting services of the relevant hazardous chemicals transportation companies. Based on the above information, conduct a comprehensive assessment of each item. The specific steps are as follows:

[0095] Obtain the total number of transport dispatches of relevant hazardous chemicals transport companies through the information collection module And the number of points deducted for transport violations , calculate the violation rate of the relevant hazardous chemicals transportation companies: ;

[0096] in, Represents the violation rate of the relevant hazardous chemicals transportation company, and obtains the violation rate score based on the violation rate of the vehicle. ,in ;

[0097] Obtain the total number of hazardous chemicals transported by the relevant hazardous chemicals transport company through the information collection module and the number of hazardous chemicals transportation accidents , calculate the accident rate of relevant hazardous chemicals transportation companies: ;

[0098] in, Represents the accident rate of the relevant hazardous chemicals transportation company, and obtains the accident rate score based on the accident rate calculation ,in ;

[0099] Obtain the operating years of relevant hazardous chemicals transportation companies through the information collection module , based on the operating years value, obtain the operating years score:

[0100] when When ;

[0101] when When ;

[0102] when When ;

[0103] The information collection module is used to obtain the supporting services of the relevant operating routes of the relevant hazardous chemicals transportation companies, including the number of emergency stations, maintenance service points, and fire emergency points, and the total coverage mileage of the security points of the relevant sections. , and based on the total mileage of the relevant road section , calculate the coverage rate of the guarantee points : ;

[0104] Based on coverage of guarantee points ;

[0105] Collect the number of times drivers travel the relevant road segments , calculate the route familiarity score, the steps are:

[0106] when When ;

[0107] when When ;

[0108] when When .

[0109] Based on the results of the comprehensive assessment, a weight is set for each influencing project to obtain a subjective risk weighted value, and the subjective risk weighted value is output to the objective risk assessment module. A subjective risk assessment is performed on the objective risk assessment value to obtain the final risk assessment value. The specific algorithm formula is: ;

[0110] ;

[0111] in, Represent the weight proportion of the corresponding scores respectively. Represents the subjective risk weighted value. Based on the objective risk assessment value and the subjective risk weighted value, the final risk assessment value is obtained. , and its algorithm formula is: ; ;

[0112] in, Represent the weight ratio of objective risk assessment value and subjective risk weighted value respectively. Represents the objective risk assessment value obtained based on the objective risk assessment model.

[0113] project Weight Fraction Weighted value Objective risk assessment value γ 0.6 88 88×0.6=52.8 Subjective risk weighting value ϑ 0.4 75 75×0.4=30 Final risk assessment value μ 52.8+30=82.8

[0114] The optimization scheme generation module combines the relevant data of the subjective risk weighting module and the objective risk assessment module to optimize the risk of this hazardous chemicals transportation plan, conduct risk assessment on the optimization scheme, and output the results to the result judgment module. The specific steps are as follows:

[0115] Based on the relevant data of the objective risk assessment module, optimization suggestions are generated for the total transport volume, number of transports, and transport time. Based on the relevant data of the subjective risk weighting module, optimization adjustment opinions are generated for the drivers, and the optimization adjustment priority is set. The specific steps are as follows:

[0116] Set a single transport volume threshold for the total transport volume and number of transports , according to the total transport volume value The number of transports , calculate the single transport volume of the original plan : ;

[0117] when ≤ No adjustment is made when

[0118] when > When the transport times are adjusted, ≤ ;

[0119] With regard to the transportation time, the original transportation time is postponed in combination with the weather and road conditions during the relevant time period, and the weather and road conditions during the relevant time period are re-obtained;

[0120] For drivers who have traveled the relevant route less than 40 times, optimization adjustment suggestions are generated, that is, drivers who have traveled the relevant route more than 40 times are selected;

[0121] Set optimization adjustment priorities and calculate the pending optimization risk assessment value after each priority optimization adjustment and risk threshold Perform a pre-alignment:

[0122] when < When the optimization adjustment signal is transmitted to the optimization scheme generation module for optimization, when both the subjective parameter and objective parameter related optimization strategies are used, Substituting values ,transmission Enter the result determination module;

[0123] When any ≥ When Substituting values ,transmission Enter the result determination module;

[0124] Among them, the priority of optimization adjustment is driver adjustment > single transport volume adjustment > transport time adjustment.

[0125] It should be noted that by setting the priority of optimization and adjustment, the part with the least cost impact in the relevant transportation plan can be adjusted first. When the priority is adjusted in sequence and still does not meet the risk assessment threshold, it means that there are major risks in this hazardous chemicals transportation plan.

[0126] According to the objective parameter adjustment value and the subjective parameter adjustment value, the relevant data is transmitted to the objective risk assessment module and the subjective risk weighting module respectively to calculate the optimized risk assessment value. , transmission optimization risk assessment value Enter the result determination module and the risk threshold Make a judgment:

[0127] when When the optimization risk assessment fails, the hazardous chemicals transportation plan is transmitted for risk marking;

[0128] when When the optimization risk assessment is qualified, an optimization plan is generated and transmitted to the relevant hazardous chemicals transportation company, waiting for confirmation by the hazardous chemicals transportation company.

[0129] It should be noted that after the optimization plan is generated and transmitted to the relevant hazardous chemicals transportation company, it is necessary to wait for the hazardous chemicals transportation company to confirm the relevant optimization plan and adjust the transportation plan. After the adjustment is completed, it needs to be re-entered into the system for risk assessment.

[0130] Result judgment module, based on the final risk assessment value and risk threshold , the risk of hazardous chemicals transportation is determined. When the determination result is unqualified, the output signal enters the optimization solution generation module. The specific steps are:

[0131] when When , it means that the final risk assessment is unqualified, and the transmission signal enters the optimization solution generation module;

[0132] when , it means the final risk assessment is qualified.

[0133] The method for assessing the risk of hazardous chemicals transportation based on a neural network of the present invention comprises the following steps:

[0134] S1. Based on this hazardous chemicals transportation plan, collect relevant hazardous chemicals information, including types of hazardous chemicals, hazardous chemicals transportation volume, transportation methods, and transportation routes. At the same time, establish a hazardous chemicals transportation company database to collect and record the names of relevant hazardous chemicals transportation companies, accident records, violations, years of operation, and related transportation route supporting services;

[0135] S2. Based on the relevant hazardous chemicals information of this transportation, collect the road condition information and weather information of the two transportation locations, and establish an objective risk assessment model through a convolutional neural network to conduct an objective risk assessment on this hazardous chemicals transportation plan and obtain an objective risk assessment value;

[0136] S3. Based on the relevant company data in the hazardous chemicals transportation company database, including the hazardous chemicals transportation company's violation and accident records, operating years, and related transportation route supporting services, conduct a comprehensive assessment of each item and set a weight value to obtain a subjective risk weighted value;

[0137] S4. Obtain the final risk assessment value by weighted calculation of the objective risk assessment value and the subjective risk weighted value, and compare it with the risk threshold. If the result is unqualified, optimize and adjust the transportation plan based on the priority of the objective parameters and the subjective parameters.

[0138] S5. Compare and judge the optimized risk assessment value after optimization and adjustment with the risk threshold. If it is qualified, transmit the optimized adjustment plan to the relevant hazardous chemicals transportation company for confirmation. If it is unqualified, mark the risk for this hazardous chemicals transportation plan.

[0139] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0140] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. Hazardous chemicals transportation risk assessment system based on neural network, characterized by: It includes information collection module, objective risk assessment module, subjective risk weighting module, result determination module and optimization solution generation module; The information collection module collects information on hazardous chemicals related to transportation, including types of hazardous chemicals, hazardous chemical transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies, which includes company names, accident records, violations, years of operation, and related transportation route supporting services; The objective risk assessment module, based on the transportation-related hazardous chemicals information collected by the information collection module, collects the road condition information and weather information of the two transportation locations, collects historical data and establishes an assessment model through a convolutional neural network, conducts a first round of risk assessment on the relevant hazardous chemicals transportation, and obtains an objective risk assessment value; The subjective risk weighting module calculates the subjective risk weighting value based on the hazardous chemicals transportation company database and the transportation route of the hazardous chemicals, and outputs the subjective risk weighting value to the objective risk assessment module, performs subjective risk assessment on the objective risk assessment value and obtains the final risk assessment value. The specific steps are as follows: Through the hazardous chemicals transportation company database in the information collection module, obtain the violation information and accident records, operating years, and related transportation route supporting services of the relevant hazardous chemicals transportation companies. Based on the above information, conduct a comprehensive assessment of each item. The specific steps are as follows: Obtain the total number of transport dispatches of relevant hazardous chemicals transport companies through the information collection module And the number of points deducted for transport violations , calculate the violation rate of the relevant hazardous chemicals transportation companies: ; in, Represents the violation rate of the relevant hazardous chemicals transportation company, and obtains the violation rate score based on the violation rate of the vehicle. ,in ; Obtain the total number of hazardous chemicals transported by the relevant hazardous chemicals transport company through the information collection module and the number of hazardous chemicals transportation accidents , calculate the accident rate of relevant hazardous chemicals transportation companies: ; in, Represents the accident rate of the relevant hazardous chemicals transportation company, and obtains the accident rate score based on the accident rate calculation ,in ; Obtain the operating years of relevant hazardous chemicals transportation companies through the information collection module , based on the operating years value, obtain the operating years score: when When ; when When ; when When ; The information collection module is used to obtain the supporting services of the relevant operating routes of the relevant hazardous chemicals transportation companies, including the number of emergency stations, maintenance service points, and fire emergency points, and the total coverage mileage of the security points of the relevant sections. , and based on the total mileage of the relevant road section , calculate the coverage rate of the guarantee points : ; Based on coverage of guarantee points Calculate the score of supporting services ; Collect the number of times drivers travel the relevant road segments , calculate the route familiarity score, the steps are: when When ; when When ; when When ; Based on the results of the comprehensive assessment, a weight is set for each influencing project to obtain a subjective risk weighted value, and the subjective risk weighted value is output to the objective risk assessment module, and a subjective risk assessment is performed on the objective risk assessment value to obtain the final risk assessment value; The result determination module is based on the final risk assessment value and risk threshold , the risk of hazardous chemicals transportation is determined, and when the determination result is unqualified, the output signal enters the optimization solution generation module; The optimization scheme generation module combines the relevant data of the subjective risk weighting module and the objective risk assessment module to optimize the risk of this hazardous chemicals transportation plan, conduct risk assessment on the optimization scheme, and output the results to the result judgment module.

2. The hazardous chemicals transportation risk assessment system based on neural network according to claim 1 is characterized by: The objective risk assessment module, based on the transportation-related hazardous chemicals information collected by the information collection module, collects the road condition information and weather information of the two transportation locations, collects historical data, establishes an objective risk assessment model through a convolutional neural network, conducts a first round of risk assessment on the relevant hazardous chemicals transportation, and obtains an objective risk assessment value. The specific steps are as follows: Collect historical hazardous chemicals information, road conditions, weather information and risk values, extract relevant features from the original data, and build an objective risk assessment model based on a convolutional neural network. The specific steps are as follows: Hazardous chemicals data include types, total transport volume, transport times, transport methods, routes; road condition data include low traffic volume, medium traffic volume, and high traffic volume; weather data include low, medium, and high visibility, as well as weather conditions such as sunny, light rain, moderate rain, and heavy rain. Visibility, weather conditions, and traffic volume data are encoded using unique hot coding; Divide historical data into training set, validation set and test set, build an objective risk assessment model based on the convolutional neural network through the training set, and verify the constructed objective risk assessment model based on the validation set and test set; choose As an activation function, the output value of the risk assessment objective model is compressed into the range of 0 to 1 and converted into the range of 0 to 100. The algorithm formula is: ; in, represent The output value of the activation function, Represents the output value of the objective risk assessment model constructed by the convolutional neural network; Combined with the hazardous chemicals transportation information collected by the information collection module, the model risk value is calculated through the risk assessment objective model to obtain the objective risk assessment value.

3. The hazardous chemicals transportation risk assessment system based on neural network according to claim 1 is characterized by: Based on the results of the comprehensive assessment, a weight is set for each influencing project to obtain a subjective risk weighted value. The specific algorithm formula is: ; ； in, Represent the weight proportion of the corresponding scores respectively. Represents the subjective risk weighted value. Based on the objective risk assessment value and the subjective risk weighted value, the final risk assessment value is obtained. , and its algorithm formula is: ; ; in, Represent the weight ratio of objective risk assessment value and subjective risk weighted value respectively. Represents the objective risk assessment value obtained based on the objective risk assessment model.

4. The hazardous chemicals transportation risk assessment system based on neural network according to claim 1 is characterized by: The optimization scheme generation module combines the relevant data of the subjective risk weighting module and the objective risk assessment module to optimize the risk of this hazardous chemicals transportation plan, conduct risk assessment on the optimization scheme, and output the results to the result determination module. The specific steps are as follows: Based on the relevant data of the objective risk assessment module, it generates optimization suggestions for the total transport volume, number of transports, and transport time. Based on the relevant data of the subjective risk weighting module, it generates optimization adjustment opinions for the drivers and sets the optimization adjustment priority. According to the objective parameter adjustment value and the subjective parameter adjustment value, the relevant data is transmitted to the objective risk assessment module and the subjective risk weighting module respectively to calculate the optimized risk assessment value. , transmission optimization risk assessment value Enter the result determination module and the risk threshold Make a judgment: when When the optimization risk assessment fails, the hazardous chemicals transportation plan is transmitted for risk marking; when When the optimization risk assessment is qualified, an optimization plan is generated and transmitted to the relevant hazardous chemicals transportation company, waiting for confirmation by the hazardous chemicals transportation company.

5. The hazardous chemicals transportation risk assessment system based on neural network according to claim 4 is characterized by: Based on the relevant data of the objective risk assessment module, the adjustment optimization suggestions are generated for the total transportation volume, transportation times, and transportation time. Based on the relevant data of the subjective risk weighting module, the optimization adjustment opinions are generated for the drivers, and the optimization adjustment priority is set. The specific steps are: Set a single transport volume threshold for the total transport volume and number of transports , according to the total transport volume value The number of transports , calculate the original plan single transport volume : ; when ≤ No adjustment is made when when > When the transport times are adjusted, ≤ ; With regard to the transportation time, the original transportation time is postponed in combination with the weather and road conditions during the relevant time period, and the weather and road conditions during the relevant time period are re-obtained; For drivers who have traveled the relevant route less than 40 times, optimization adjustment suggestions are generated, that is, drivers who have traveled the relevant route more than 40 times are selected; Set optimization adjustment priorities and calculate the pending optimization risk assessment value after each priority optimization adjustment and risk threshold Perform a pre-alignment: when < When the optimization adjustment signal is transmitted to the optimization scheme generation module for optimization, when both the subjective parameter and objective parameter related optimization strategies are used, Substituting values ,transmission Enter the result determination module; When any ≥ When Substituting values ,transmission Enter the result determination module; Among them, the priority of optimization adjustment is driver adjustment > single transport volume adjustment > transport time adjustment.

6. The hazardous chemicals transportation risk assessment system based on neural network according to claim 1 is characterized by: The information collection module collects information on hazardous chemicals related to transportation, including types of hazardous chemicals, hazardous chemical transportation volume, transportation methods, and transportation routes, and establishes a database of hazardous chemical transportation companies, which includes company names, accident records, violations, years of operation, and related transportation route supporting services; The supporting services of the relevant transport routes include the number of driving times along the routes and the number of supporting stations along the routes, including the number of first aid stations, maintenance service points, and fire emergency points; The result determination module is based on the final risk assessment value and risk threshold , the risk of hazardous chemicals transportation is determined. When the determination result is unqualified, the output signal enters the optimization solution generation module. The specific steps are: when When , it means that the final risk assessment is unqualified, and the transmission signal enters the optimization solution generation module; when , it means the final risk assessment is qualified.

7. A risk assessment method for hazardous chemicals transportation based on neural network, characterized in that: The method adopts the hazardous chemicals transportation risk assessment system based on a neural network as described in any one of claims 1 to 6, and comprises the following steps: S1. Based on this hazardous chemicals transportation plan, collect relevant hazardous chemicals information, including types of hazardous chemicals, hazardous chemicals transportation volume, transportation methods, and transportation routes. At the same time, establish a hazardous chemicals transportation company database to collect and record the names of relevant hazardous chemicals transportation companies, accident records, violations, years of operation, and related transportation route supporting services; S2. Based on the relevant hazardous chemicals information of this transportation, collect the road condition information and weather information of the two transportation locations, and establish an objective risk assessment model through a convolutional neural network to conduct an objective risk assessment on this hazardous chemicals transportation plan and obtain an objective risk assessment value; S3. Based on the relevant company data in the hazardous chemicals transportation company database, including the hazardous chemicals transportation company's violation and accident records, operating years, and related transportation route supporting services, conduct a comprehensive assessment of each item and set a weight value to obtain a subjective risk weighted value; S4. Obtain the final risk assessment value by weighted calculation of the objective risk assessment value and the subjective risk weighted value, and compare it with the risk threshold. If the result is unqualified, optimize and adjust the transportation plan based on the priority of the objective parameters and the subjective parameters. S5. Compare and judge the optimized risk assessment value after optimization and adjustment with the risk threshold. If it is qualified, transmit the optimized adjustment plan to the relevant hazardous chemicals transportation company for confirmation. If it is unqualified, mark the risk for this hazardous chemicals transportation plan.

Images