CN-122019673-A - Dynamic construction method and system of data relation model for different traffic scenes of highway

CN122019673ACN 122019673 ACN122019673 ACN 122019673ACN-122019673-A

Abstract

The invention discloses a dynamic construction method and a system of a data relation model for different highway service scenes, which relate to the technical field of data management and comprise the steps of obtaining a highway service scene set and generating an output data format matrix; grouping business scenes based on output data items to obtain business scene subgroups, determining a target output data format matrix in the subgroups, reversely pushing an input data item set and the input data format matrix by combining a calculation model, searching corresponding data items on a data platform, establishing a data association relation table, judging format consistency, executing data processing when the data items are not satisfied, generating a data acquisition request or a user input request when the data items are not satisfied, constructing a data association model fetch, carrying out input compliance check, and iteratively updating the matrix, the association table and the relation model according to user evaluation feedback. By the technical scheme, the cross-scene data caliber unification and multiplexing are realized, the maintenance cost is reduced, and the fetch stability and the input data quality are improved.

Inventors

Huang Yeran
ZHU JIERUI
ZHAO SHUO
CAO XUAN
LI MINGRUI
GUO YUQI
NIU SHUYUN
GAO JIAN

Assignees

交通运输部公路科学研究所

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (10)

1. A dynamic construction method of a data relation model for different traffic scenes of a highway is characterized by comprising the following steps: Acquiring a highway service scene set to be supported, determining output data items of the service scenes and corresponding data format requirements for each service scene in the highway service scene set, and generating an output data format matrix corresponding to each service scene; Grouping the highway service scene sets based on the output data items of each service scene to obtain a service scene subgroup and the same output data item set corresponding to the service scene subgroup; Determining a target output data format matrix of the service scene subgroup based on the data format requirements of the same output data item set in the service scene subgroup aiming at the service scene subgroup; determining an input data item set and an input data format matrix for meeting the output requirement of the calculation model based on the target output data format matrix of the service scene subgroup and the calculation model corresponding to the service scene subgroup; For the input data item set, searching data item of the data platform corresponding to each input data item in the data platform, and establishing a data association relation table, wherein the data association relation table at least comprises output data item and output data format, input data item and input data format, data item identification of the data platform and data format requirement of the data platform; Based on the data association relation table, carrying out compliance judgment on the data format requirement of the data platform data item and the input data format matrix; when the data format requirement of the data item of the data platform is judged to be not met by the input data format matrix, executing data processing on the data item of the corresponding data platform; Constructing a data relation model corresponding to the business scene subgroup based on the data association relation table and the consistency judging result, and acquiring input data for the calculation model based on the data relation model; Updating the output data item and the corresponding data format requirement based on evaluation feedback used by a user, and iteratively updating the output data format matrix, the data association relation table and the data relation model according to the output data format matrix and the data association relation table.
2. The dynamic construction method of data relation models for different traffic scenes of a highway according to claim 1, wherein the traffic scene set of the highway at least comprises traffic scenes in the fields of highway construction, maintenance, operation and road politics; Aiming at the business scene, an application scene demand data item is formed by questionnaire investigation of highway management personnel and trip personnel; The output data format matrix includes at least output data items, field descriptions, data types, data value fields, data precision, and units of measure.
3. The method for dynamically constructing data relation models for different road traffic scenes according to claim 1, wherein the specific process of grouping the road traffic scene sets comprises: Comparing the output data items of the service scenes to be grouped with the same output data item set of the established service scene subgroup, and establishing a new service scene subgroup when the number of the same output data items is 0; And when the number of the same output data items is not smaller than a preset threshold value, adding the to-be-grouped business scenes into a corresponding business scene subgroup, and updating the same output data item set of the corresponding business scene subgroup.
4. The method for dynamically constructing data relation models for different traffic scenes on a highway according to claim 1, wherein the specific process of determining the target output data format matrix comprises the following steps: and aiming at the same output data item set, comparing data format requirements corresponding to the same output data item, and calculating based on a highest requirement function to obtain the target output data format matrix, wherein the highest requirement function aiming at data precision outputs the data precision requirement with smaller precision value.
5. The method for dynamically constructing data relation models for different traffic scenarios on a highway according to claim 1, wherein the specific process of determining the set of input data items and the input data format matrix based on the target output data format matrix and the calculation model comprises: and mapping the calculation model and the target output data format matrix into an input data item set and an input data format matrix for meeting the output requirement of the calculation model based on a preset mapping function.
6. The dynamic construction method of data relation model for different traffic scenes according to claim 1, wherein the data association table further comprises a support application identifier, and the data platform data item identifier is represented by a manner of "thematic database identifier-data item identifier".
7. The dynamic construction method of data relation model for different traffic scenes according to claim 1, wherein the data processing is used for supplementing data items of data platform which do not meet the input data format matrix to meet the input data format matrix, and the missing data item list at least comprises missing input data item identifiers and corresponding input data format requirements.
8. The method for dynamically constructing the data relationship model for different traffic scenes on the highway according to claim 1, wherein the established data relationship model is searched for non-identical output data items in the application scene to determine corresponding input data items and input data formats, and when the search result is empty, the data relationship table is updated and the data relationship model corresponding to the non-identical output data items is re-established.
9. The method for dynamically constructing data relation models for different traffic scenes on a highway according to claim 1, wherein when the traffic scene subgroup comprises daily travel path selection, differentiated charging path selection, travel path planning, emergency path planning and guidance traffic scenes, the input data item set at least comprises service area queuing information, toll station queuing information, maintenance planning and construction information, traffic running state information, traffic congestion early warning information, weather disaster early warning information, traffic event road control measures, emergency command control information and historical ETC traffic records.
10. A system for dynamically constructing a data relationship model for different traffic scenes on a highway, which is used for realizing the method for dynamically constructing the data relationship model for different traffic scenes on a highway according to any one of claims 1 to 9, comprising: The system comprises a service scene management module, a data format matrix generation module and a data format management module, wherein the service scene management module is used for acquiring a highway service scene set to be supported, determining output data items of the service scene and corresponding data format requirements aiming at each service scene in the highway service scene set, and generating an output data format matrix corresponding to each service scene; the service scene grouping module is used for grouping the road service scene sets based on the output data items of each service scene to obtain a service scene subgroup and the same output data item set corresponding to the service scene subgroup; A target format determining module, configured to determine, for the service scene subgroup, a target output data format matrix of the service scene subgroup based on data format requirements of the same set of output data items in the service scene subgroup; The input demand determining module is used for determining an input data item set and an input data format matrix for meeting the output demand of the computing model based on the target output data format matrix of the service scene subgroup and the computing model corresponding to the service scene subgroup; the data association construction module is used for searching data items of the data platform corresponding to each input data item in the data platform aiming at the input data item set, and establishing a data association relation table, wherein the data association relation table at least comprises output data items and output data formats, input data items and input data formats, data item identifiers of the data platform and data format requirements of the data platform; The compliance judging module is used for carrying out compliance judgment on the data format requirement of the data platform data item and the input data format matrix based on the data association relation table; The data supplementing and collecting module is used for executing data processing on the data items of the corresponding data platform when the data format requirements of the data items of the data platform are judged to be not met by the input data format matrix by the coincidence judgment; The model construction checking module is used for constructing a data relation model corresponding to the business scene subgroup based on the data association relation table and the consistency judging result, and acquiring input data for the calculation model based on the data relation model; and the feedback iteration updating module is used for updating the output data item and the corresponding data format requirement based on evaluation feedback used by a user, and iteratively updating the output data format matrix, the data association relation table and the data relation model according to the output data item and the corresponding data format requirement.

Description

Dynamic construction method and system of data relation model for different traffic scenes of highway Technical Field The invention relates to the technical field of data management, in particular to a dynamic construction method of a data relationship model for different traffic scenes of a highway and a dynamic construction system of the data relationship model for different traffic scenes of the highway. Background In different business fields of construction, maintenance, operation, administration of roads and the like, application requirements for multi-class business scenes, such as daily travel path selection, differentiated charging path selection, travel path planning, emergency induction and the like, exist commonly in the road industry. The above-mentioned business scenes usually need to output specific data results to the upper layer application, and have differentiation requirements on output data items and data formats (such as field types, value fields, precision, measurement units and the like) thereof, and meanwhile, each business scene depends on multi-source data provided by a data platform as input data of a calculation model. In the prior art, mapping relations among output data items, input data items and data items of a data platform required by a service scene are established in a manual combing and pre-curing configuration mode. Because different business scenes have inconsistent format requirements on the same type of data and the business scenes are iterated frequently, the mapping relation needs to be repeatedly maintained, cross-scene multiplexing is difficult to achieve, when the data format requirements of existing data items of a data platform are not matched with the input format of a business model, additional manual investigation and processing are often needed, the problems of inconsistent data caliber, unsatisfied format, non-compliance of model input and the like are easily introduced, and when the data items required by the data platform are absent, a missing item list and an acquisition/input request mechanism facing the input format of the business model are absent, so that the business model is difficult to stably operate. Meanwhile, the prior art generally lacks a mechanism for continuously iterating output requirements, association relations and relation models based on evaluation feedback by users, and is difficult to form a data relation model construction flow capable of realizing closed-loop optimization. Disclosure of Invention Aiming at the problems, the invention provides a dynamic construction method and a system for a data relation model of different traffic scenes of a highway, which are characterized in that an output data format matrix is firstly generated and grouped according to the same output data item through a dynamic construction mechanism of the data relation model driven by traffic scene output requirements, format calibers of different scenes are unified in subgroups by utilizing a highest requirement function, an input data item set and an input data format matrix are reversely determined by combining a calculation model, manual carding configuration and cross-scene repeated maintenance are reduced, further, data processing and supplementing are carried out on unsatisfied items by establishing an association relation table of the output/input and data items of a data platform, data acquisition or user input is triggered on the missing items to form a data supply closed loop, the consistency and the usability of the input of the calculation model are improved, the input compliance inspection is carried out based on the data relation model, the abnormal data input module risk is reduced, the iteration updating matrix, the association table and the relation model are combined with user evaluation feedback iteration, and the continuous adaptation capability to traffic change is improved. In order to achieve the above purpose, the present invention provides a dynamic construction method of a data relationship model for different traffic scenes of a highway, comprising: Acquiring a highway service scene set to be supported, determining output data items of the service scenes and corresponding data format requirements for each service scene in the highway service scene set, and generating an output data format matrix corresponding to each service scene; Grouping the highway service scene sets based on the output data items of each service scene to obtain a service scene subgroup and the same output data item set corresponding to the service scene subgroup; Determining a target output data format matrix of the service scene subgroup based on the data format requirements of the same output data item set in the service scene subgroup aiming at the service scene subgroup; determining an input data item set and an input data format matrix for meeting the output requirement of the calculation model based on the target output data format matrix of the ser