An electronic waybill submission system and method applied to dangerous goods road transport

Abstract

The application discloses an electronic waybill reporting system and method applied to dangerous goods road transportation, relates to the technical field of data security, and comprises the following steps: collecting structured reporting data and performing field renaming, type conversion and validity detection; performing normalized coding to obtain an idempotent check key; performing sampling and smoothing processing based on the idempotent check key, calculating an event commitment value to form a space-time commitment chain; calculating a track quality base score and a space-time consistency base score according to the space-time commitment chain to generate a quality threshold passing token; screening a minimum disclosure field set according to the token and performing hashing and group encryption to generate a signed message; and completing repeated retrieval and verification based on the signed message to generate an audit log. Through the linkage mechanism of the space-time commitment chain and the quality evaluation module, the quality grading of vehicle track data and the space-time consistency quantization checking are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data security, and particularly relates to an electronic waybill reporting system and method applied to dangerous goods road transportation. BACKGROUND

[0002] With the continuous improvement of the safety supervision system of dangerous goods road transportation, electronic waybill reporting has gradually become the core means for information interconnection and dynamic monitoring between regulatory agencies and transportation enterprises. The existing electronic waybill system generally uses a structured data format to realize real-time reporting and sharing of waybill information through the Internet or a government affairs special network. The regulatory platforms in various places have successively established interfaces with vehicle satellite positioning, emergency response and enterprise safety management systems, realizing visual tracking and in-process supervision of the whole transportation process.

[0003] The current electronic waybill reporting technology mainly relies on static field matching and one-way encryption transmission. Firstly, it lacks a dynamic trajectory consistency verification mechanism, and cannot evaluate the quality and authenticity of the transportation space-time data during the reporting process, making it difficult for regulatory departments to determine the reliability of the data. Secondly, the processing of privacy protection and minimum disclosure is rough, the encryption granularity of sensitive information is not fine, and there are cases of excessive encryption or insufficient disclosure, which not only affects the efficiency of regulatory review, but also increases the risk of data leakage. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides an electronic waybill reporting method applied to dangerous goods road transportation to solve the problems of difficult verification of trajectory data authenticity and difficult balance between privacy disclosure and regulatory needs in the prior art.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides an electronic waybill reporting method applied to dangerous goods road transportation, which comprises collecting structured reporting data for field renaming, type conversion and legality detection, and obtaining a modeling context.

[0008] Normalization encoding is performed through the modeling context to obtain an idempotent check key.

[0009] Sampling and smoothing are performed through the idempotent check key, an event commitment value is calculated, and a space-time commitment chain is obtained.

[0010] Based on the space-time commitment chain, a trajectory quality base score and a space-time consistency base score are calculated, and a quality threshold passing token is obtained.

[0011] The token is weighted and screened according to a quality threshold, a minimum disclosure field set is obtained, the minimum disclosure field set is subjected to hash processing and group encryption, and a signed message is obtained;

[0012] Based on the signed message, repeated search and verification are performed to obtain an audit log.

[0013] As a preferred scheme of the electronic transport bill submission method for dangerous goods road transportation, the structured submission data is collected, field renaming, type conversion and legality detection are performed, and a modeling context is obtained.

[0014] The structured submission data is collected, a standard field mapping table and a regulatory rule file are set, field renaming, type conversion and legality detection are performed on the structured submission data, and a standardized data set is obtained.

[0015] The rule version number and security configuration parameters are extracted from the regulatory rule file to obtain a rule security binding block.

[0016] For each field in the standardized data set, a field hash digest is calculated by a hash algorithm, the hash digests of all fields are arranged into a field hash digest matrix, and the field hash digest matrix and the rule security binding block are encapsulated to obtain a modeling context.

[0017] As a preferred scheme of the electronic transport bill submission method for dangerous goods road transportation, the structured submission data is collected, field renaming, type conversion and legality detection are performed, and a modeling context is obtained.

[0018] Through the modeling context, fields related to unique identification are classified into a unique identification field set, information reflecting transportation time is classified into a time field set, and information describing the geographical range of transportation is classified into a geographic fence field set.

[0019] Character standardization is performed on the unique identification field set, time zone normalization and timestamp conversion are performed on the time field set, and geographic encoding and precision truncation processing are performed on the geographic fence field set to obtain normalized field data.

[0020] Based on the normalized field data, an idempotent check key is calculated by a security digest algorithm.

[0021] As a preferred scheme of the electronic transport bill submission method for dangerous goods road transportation, the structured submission data is collected, field renaming, type conversion and legality detection are performed, and a modeling context is obtained.

[0022] Trajectory data, operation time data, and environment and network evidence data are collected and synchronously fused according to timestamps to obtain a trajectory stream.

[0023] Based on the trajectory flow, sampling and smoothing processing is performed, abnormal positioning points are removed, and a standardized trajectory point sequence is obtained;

[0024] Through the standardized trajectory point sequence, combined with the modeling context and the idempotent check key, event recognition is performed, event records are obtained, event commitment values are sequentially calculated according to time sequence, and a space-time commitment chain is formed.

[0025] As a preferred scheme of the electronic transport document submission method for dangerous goods road transportation provided by the application, wherein, based on the space-time commitment chain, the trajectory quality base score and the space-time consistency base score are calculated, and the specific steps are as follows,

[0026] Based on the standardized trajectory point sequence, the trajectory sampling integrity rate, the positioning reliability average value and the time continuity ratio are calculated, weighted and summarized, and the trajectory quality base score is obtained.

[0027] According to the space-time commitment chain, the event time monotonicity rate, the speed rationality rate and the path coherence rate are calculated, weighted and summarized, and the space-time consistency base score is obtained.

[0028] As a preferred scheme of the electronic transport document submission method for dangerous goods road transportation provided by the application, wherein, the quality threshold passing token is obtained by the following steps,

[0029] The trajectory quality base score and the space-time consistency base score are fused and calculated to obtain a comprehensive trigger criterion score value, and when the comprehensive trigger criterion score value is greater than or equal to a preset quality threshold, a quality threshold passing token is generated.

[0030] As a preferred scheme of the electronic transport document submission method for dangerous goods road transportation provided by the application, wherein, according to the quality threshold passing token, the minimum disclosure field set is obtained by weighted screening, and the specific steps are as follows,

[0031] According to the quality threshold passing token, the corresponding instance record is retrieved in the modeling context, and a submission field set is obtained.

[0032] The privacy disclosure strategy configuration file is set, the submission field set is weighted and screened, the disclosure priority score value is calculated, the disclosure priority score value is sorted and filtered, and the minimum disclosure field set is obtained.

[0033] As a preferred scheme of the electronic transport document submission method for dangerous goods road transportation provided by the application, wherein, the minimum disclosure field set is hashed and grouped and encrypted to obtain a signature message, and the specific steps are as follows,

[0034] The minimum disclosure field set is hashed, a Merkle tree is constructed, and a Merkle path proof of each field is generated, and a disclosure root digest is obtained.

[0035] Extract a sensitive field subset from the minimum disclosure field set, group encryption according to the field category, obtain the encrypted field load, uniformly package the Merkle path proof, the disclosure root summary and the encrypted field load, and obtain the signed message.

[0036] As a preferred scheme of the application applied to the electronic waybill submission method for dangerous goods road transportation, wherein: based on the signed message, repeated search and verification are performed to obtain an audit log, and the specific steps are as follows,

[0037] Based on the signed message and the idempotent check key, repeated search and verification are performed to obtain a receipt.

[0038] When the verification fails or the receipt is not obtained, the short message channel message load is reorganized through the minimum disclosure field set, a short receipt number is obtained, and the receipt or the short receipt number is used as an index to obtain an audit log.

[0039] In a second aspect, the application provides an electronic waybill submission system applied to dangerous goods road transportation, which comprises a data processing module, a structured submission data acquisition module, a field renaming module, a type conversion module, a legality detection module, a modeling context acquisition module,

[0040] A normalization coding module acquires an idempotent check key through the modeling context.

[0041] A commitment generation module calculates an event commitment value through the idempotent check key, and acquires a space-time commitment chain.

[0042] A quality evaluation module calculates a track quality score and a space-time consistency score based on the space-time commitment chain, and acquires a quality threshold passing token.

[0043] A minimum disclosure module performs weighted screening according to the quality threshold passing token, acquires a minimum disclosure field set, and performs hash processing and group encryption through the minimum disclosure field set to obtain a signed message.

[0044] An audit verification module performs repeated search and verification based on the signed message to obtain an audit log.

[0045] The application has the following beneficial effects: through the cooperative processing mechanism of structured data analysis and normalization coding, the standardization, semantic unification and unique identification generation of the multi-source field of the electronic waybill of dangerous goods are realized; through the linkage mechanism of the space-time commitment chain and the quality evaluation module, the quality grading and space-time consistency quantization checking of the vehicle track data are realized; through the combination mechanism of the minimum disclosure and audit verification module, the minimum disclosure and verifiable encrypted submission of sensitive data are realized. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0047] Fig. 1 A flow chart of the electronic waybill reporting method applied to dangerous goods road transport.

[0048] Fig. 2 A schematic diagram of the electronic waybill reporting system applied to dangerous goods road transport.

[0049] Fig. 3 A flow chart of trajectory quality assessment and token generation.

[0050] Fig. 4 A flow chart of field disclosure and signature reporting. DETAILED DESCRIPTION

[0051] In order to make the above objectives, features and advantages of the present application more apparent and comprehensible, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0052] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0053] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an embodiment that is independent or alternative to other embodiments.

[0054] REFERENCE Figs. 1-4 For one embodiment of the present application, the embodiment provides an electronic waybill reporting method applied to dangerous goods road transport, comprising the following steps:

[0055] S1: Analyzing the electronic waybill data modeling base set to establish a modeling context.

[0056] The structured submission data of the electronic dangerous goods waybill is collected, including the information of the shipper, the information of the carrying vehicle, the type of the dangerous goods, the transportation route, the time limit node and the safety constraint field, the consistency of the electronic dangerous goods waybill is pre-checked and completed through the vehicle state, the qualification state of the practitioner in the historical waybill record, the waybill inspection record and the record information, and the structured submission data for submission is formed. The standard field mapping table and the supervision rule file issued by the supervision platform are obtained, the standard field mapping table is used to indicate the corresponding relationship and the data type between the data field and the supervision standard field, and the supervision rule file contains the version number of the current submission specification, the requirement of the compulsory field, the checking rule and the safety strategy parameter. The structured submission data is renamed, type-converted and legally detected through the standard field mapping table and the supervision rule file, and the standardized data set is obtained.

[0057] For each field in the standardized data set, the field name, the field value and the constraint number thereof in the supervision rule file are combined in turn, the hash digest corresponding to the field is obtained through the hash algorithm calculation, and the hash digests of all the fields are arranged into a field hash digest matrix.

[0058] The current rule version number and the related safety configuration parameters are extracted from the supervision rule file, the safety configuration parameters include the transmission safety level, the encryption algorithm selection, the security certificate fingerprint of the submission interface and the compliance detection strategy. The rule version number, the safety parameter set, the certificate digest and the compliance enable flag are bound to obtain a rule safety binding block. The field hash digest matrix and the rule safety binding block are uniformly encapsulated to obtain a modeling context, and the compact binary coding format is used for storage. According to the coding result of the modeling context, the current generation timestamp and the enterprise institution code, combination is carried out, the hash algorithm is calculated, and a model identifier is obtained.

[0059] S2: The idempotent check key is obtained by normalizing the coding through the modeling context.

[0060] Through the field mapping information contained in the modeling context, the fields related to the unique identifier (for example, the waybill number, the vehicle license plate number and the carrying unit code) are classified into a unique identifier field set; the information reflecting the transportation time (for example, the departure time, the expected arrival time and the return time) is classified into a time field set; and the information describing the transportation geographical range (such as the loading location, the unloading location and the driving track coordinates) is classified into a geographical fence field set.

[0061] For the unique identification field set, a character standardization algorithm is used to remove redundant symbols (such as spaces and hyphens) and convert them to a uniform uppercase format to eliminate the uniqueness deviation caused by input differences; for the time field set, the time zone normalization and timestamp conversion are performed according to the time format standard (for example, ISO8601) specified in the regulatory rule file, so that all time information is represented in a unified time reference (for example, UTC+8); for the geo-fence field set, the coordinate points or address text are converted to latitude and longitude numerical values according to the geographic coding standard (for example, WGS-84 coordinate system), and the latitude and longitude precision is truncated, so that the same fence area has comparability in different reporting channels and use scenarios.

[0062] In the unique identification field set, the time field set and the geo-fence field set, the field group that best reflects the uniqueness of the waybill is extracted, combined in a fixed order, and the fixed order is specified by the regulatory rule file (for example, waybill number → departure time → loading latitude and longitude → unloading latitude and longitude), and the combined field values are calculated by a secure digest algorithm (for example, SM3) to obtain a fixed-length idempotent check key, which can be used as an association index between electronic waybill records, waybill inspection records and out-of-province waybill management records, to identify the repeated reporting of the same transportation task at different reporting channels and different times. All normalization rules, field categories and field sequences involved in the calculation are packaged as field normalization templates, and the normalization templates are bound with model identifiers, and the binding method is to add a reference item in the modeling context object, recording the number, generation time and associated model identifier of the template. Through binding, any reporting instance can quickly retrieve its corresponding normalization template and idempotent check key through the model identifier.

[0063] S3: According to the idempotent check key, the vehicle trajectory data, operation time and geographic location are sampled to establish a space-time commitment chain.

[0064] The three types of data are continuously accessed from the vehicle terminal, including trajectory data, operation time data, and environmental and network evidence. The trajectory data includes latitude and longitude, collection time, speed, and positioning accuracy. The operation time data includes loading start / end, departure, entry / exit of key road sections, unloading completion, and return to the factory. The environmental and network evidence includes cell ID, satellite signal-to-noise ratio, and optional on-site forensic photo hash, which is used to ensure that the loading start, departure, unloading completion, and return to the factory time nodes in the operation time data are consistent with the departure time, arrival time, return time, and vehicle state change records in the record information in the electronic dangerous goods transport bill. The environmental and network evidence can be matched with the photos, cell switching information, and satellite signal quality information formed in the on-site inspection records and law enforcement inspection records. Based on the timestamp, the trajectory data, operation time data, and environmental and network evidence data are synchronized and fused to form a trajectory stream containing spatial position, task state, and confidence index.

[0065] A sampling strategy combining fixed time intervals and minimum spatial displacement is used for the trajectory stream. Specifically, when the time interval between adjacent samples exceeds the preset time threshold or the spatial displacement exceeds the preset distance, the current point is included in the effective sampling sequence. Light smoothing is performed on the continuous sampling points (e.g., sliding average of position and speed with a limited window) to suppress transient jitter. If there is a significant positioning jump (e.g., crossing an impossible distance in a short time), it is marked as abnormal and excluded in subsequent event recognition. A standardized trajectory point sequence is obtained in chronological order.

[0066] It should be noted that the preset time threshold and the preset distance are based on the sampling frequency of the vehicle positioning terminal and the statistical results of the average speed of historical running trajectories, combined with the shipping record and driving trajectory sample of dangerous goods transportation tasks in a certain period of time. The time interval and displacement distribution of adjacent trajectory points are statistically analyzed for different vehicle types and route categories, so that the selected preset time threshold and preset distance can cover normal transportation conditions and be sensitive to abnormal stop, detour, and other situations.

[0067] By standardizing the sequence of track points, combining the geofence parameters and operation time windows recorded in the modeling context, event recognition is performed on the vehicle operation state. When a track point enters or stays in a specific fence (e.g., loading / unloading point, provincial border, dangerous section) for a minimum stay or travel condition, the corresponding event is triggered. When the time of a track point approaches the planned time window and meets the location condition, a planned event (e.g., departure, appointment window approach) is triggered. When the speed, direction, or stay duration meets the state change judgment threshold (e.g., long time stationary or high speed cruising), a state event (e.g., mid-stop, entry into high speed) is triggered. An event record is generated for each triggered event, including event type, occurrence time, occurrence location, hit fence identifier, sampling point index range, idempotent check key reference, and auxiliary evidence related to the event (e.g., summary of cellular / satellite quality indicators). All triggered events are combined to obtain an event trigger set.

[0068] It should be noted that the state change judgment threshold is a threshold determined by offline training and empirical analysis based on the statistical results of the speed change rate, direction change rate, and stay time distribution in the vehicle historical track samples. Referring to the track features corresponding to previous abnormal events or illegal and irregular transportation cases, the threshold is adjusted by type to have a low false positive rate in normal transportation records corresponding to conventional electronic waybills, and to be preferentially identified in transportation records with obvious abnormal stay or abnormal detour features.

[0069] According to the time sequence of event occurrence, the event core summary (e.g., event type, occurrence time, occurrence location, idempotent check key, and necessary quality indicator summary) and corresponding timestamp of each event are concatenated with the commitment value of the previous event, and a secure summary calculation is performed again using the national standard SM3 to obtain the commitment value of the current event. The first event uses a fixed initial value as the previous commitment value, and the event core summary and corresponding timestamp of each subsequent event reference the commitment value of the previous event to obtain the commitment value of the current event, forming a chain-like dependent structure. When any event content is tampered with, all subsequent commitment values will change, thereby achieving overall proof of sequence and integrity. The start time, end time, number of events, chain head commitment value (commitment value of the last event), and idempotent check key of the entire chain form a spatiotemporal commitment chain, which can be saved together with the corresponding dangerous goods electronic waybill number, out-of-province waybill number, and historical inspection record, used to establish a one-to-one correspondence between the actual operation track and the reported record of a single or cross-regional transportation task, and quickly locate the corresponding commitment chain and waybill reporting record through the idempotent check key during subsequent verification, review, or statistical analysis.

[0070] After the spatiotemporal commitment chain is formed, structure checking and timing checking are performed. The structure checking is to replay the calculation to confirm that each commitment value can be uniquely derived from its event summary, timestamp receipt and previous commitment value. The timing checking is to check whether the event time is strictly monotonic (allowing the same second but not allowing backtracking) and whether the timestamp receipt time is not too late from the event entry chain time (to prevent post-forgery). After the checking, the commitment chain is marked as valid; if it fails, the specific event is located and the commitment of the event is rolled back, waiting for data correction or re-evidence. When the commitment chain structure is found to be abnormal or the time sequence is found to be abnormal, the related transport order can be marked as a key check object to support the risk judgment of dangerous goods transportation.

[0071] S4: Based on the spatiotemporal commitment chain, the trajectory data quality score and the spatiotemporal consistency score are calculated, and the quality threshold passing token is obtained by weighted combination.

[0072] The standardized trajectory point sequence is extracted from the valid time window of the spatiotemporal commitment chain, and the trajectory quality score and the spatiotemporal consistency score are calculated. The trajectory quality score is composed of the trajectory sampling completeness rate, the average position accuracy and the time continuity proportion. Specifically, the trajectory sampling completeness rate is obtained by the ratio of the actual sampling trajectory point number to the theoretical sampling point number (calculated according to the set sampling frequency and driving time); the average position accuracy is obtained by the average value of the position accuracy field of all trajectory points, which is used to reflect the overall positioning accuracy; the time continuity proportion is obtained by the proportion of the sample whose adjacent trajectory point time interval does not exceed the upper threshold of the sampling period, which is used to measure the continuity of the trajectory in the time dimension. The trajectory quality score is obtained by weighted summary of the trajectory sampling completeness rate, the average position accuracy and the time continuity proportion. The spatiotemporal consistency score is composed of the event time monotonicity rate, the speed reasonability rate and the path continuity rate. Specifically, the event time monotonicity rate is calculated by checking the incrementality of the adjacent event timestamps in the spatiotemporal commitment chain, i.e. the proportion of the number of times that the time does not appear to be reversed in the total number of events; the speed reasonability rate is calculated according to the average speed of the spatial distance and the time difference between adjacent events, and the proportion of events falling within the normal operating speed range of the vehicle (determined by the vehicle type and road speed limit) is counted; the path continuity rate is calculated by analyzing the continuity of the trajectory points in the map topology structure, and the proportion of the events whose spatial path does not appear abnormal jump, fracture or cross inaccessible area between adjacent events is counted. The event time monotonicity rate, the speed reasonability rate and the path continuity rate are normalized and weighted to obtain the spatiotemporal consistency score. The spatiotemporal consistency score, after being associated with the inspection records, spot check results and historical violation handling information of the corresponding electronic transport order, can be used to distinguish normal transportation tasks from transportation tasks with abnormal bypass, long-time detention or overspeed driving, etc. risk characteristics, and provide quantitative basis for risk stratification and key object screening in transport order inspection statistics and annual transport order analysis.

[0073] It should be noted that the upper limit of the sampling period is an empirical parameter determined based on the upload frequency of the vehicle positioning terminal, communication delay characteristics, and the 95th percentile of the time interval between adjacent points in historical trajectory samples. This parameter limits the maximum allowable range of time intervals between trajectory points. The weighting of the trajectory sampling completeness rate, the mean of location reliability, and the proportion of time continuity in the weighted summation is obtained through multiple regression analysis of the correlation between the trajectory sampling completeness rate, the mean of location reliability, and the proportion of time continuity in historical transportation task samples and the overall trajectory quality level determined manually. This ensures that each indicator contributes to the final evaluation result in the comprehensive quality score. The contribution level is matched with its actual impact. The weights of the event time monotonicity, speed reasonableness, and path coherence rate are weighted by collecting trajectory data samples with manually labeled quality levels, calculating the event time monotonicity, speed reasonableness, and path coherence rate for each sample, and using the manual quality level as the real label. Then, a combination of multiple linear regression and sensitivity analysis is used to evaluate the contribution of the three indicators to the overall spatiotemporal consistency. The standardized regression coefficients are averaged and normalized to their respective weights. The manual quality level is formed through daily waybill spot checks, on-site inspection records, and historical accident reviews.

[0074] The comprehensive trigger criterion score is obtained by fusing the trajectory quality baseline and the spatiotemporal consistency baseline. The expression is as follows:

[0075] ;

[0076] in, This represents the comprehensive trigger criterion score. This represents the Sigmoid function. This represents the adjustment coefficient. Indicates the trajectory quality baseline. Indicates the spatiotemporal consistency basis. Indicates the time dispersion. Indicates spatial dispersion. This represents the time-discrete tolerance constant. Represents the spatial discrete tolerance constant. Indicates the severity coefficient of the abnormal penalty. Indicates abnormal distance. This indicates the upper limit of the abnormal distance.

[0077] It should be noted that, It uses a large number of labeled samples to calculate different The mean squared error (MSE) between the comprehensive trigger criterion score and the actual pass rate under the given values ​​is selected to minimize the error. As the initial value, and adjusted using Bayesian optimization or adaptive gradient adjustment... Perform dynamic fine-tuning to ensure always reflects the real impact strength of trajectory quality and spatiotemporal consistency on the overall triggering decision, The value range of [1, 5], when Less than 1, the impact of trajectory quality and spatiotemporal consistency on the comprehensive triggering criterion is significantly weakened. It is not sensitive to quality differences, resulting in low-quality data still passing the criterion, when Greater than 5, the exponential amplification effect is too strong, which can amplify slight fluctuations into a large fluctuation in the criterion, resulting in decreased stability and increased false triggering rate. is calculated by the standard deviation of the timestamps of each event in the spatiotemporal commitment chain; is determined by analyzing the 95th percentile of the distribution of the standard deviation of event timestamps in a large number of historical trajectory samples; is obtained by calculating the standard deviation of the Euclidean distance of all valid trajectory points relative to the centroid (average latitude and longitude) within a set time window; is determined by analyzing the 95th percentile of the distribution of the spatial dispersion in historical trajectory samples; is obtained by fitting the error distribution between the comprehensive triggering criterion score and the manual quality assessment results in historical trajectory samples. The grid search combined with cross-validation method is used to gradually adjust the value of to find the parameter point that achieves the optimal balance between the pass rate and the abnormal sample suppression rate, The value range of [0.5, 3], when Less than 0.5, the abnormal penalty is too weak, and significant abnormalities (such as drift, jump) in the trajectory may still pass the quality threshold, when Greater than 3, the penalty is too strong, and slight noise or short-term signal fluctuations will cause the criterion to drop sharply, resulting in excessive rejection. is obtained by calculating the Mahalanobis distance between the multi-dimensional features (including average speed, acceleration variance, turning angle change rate, signal switching frequency, etc.) of each trajectory window in the spatiotemporal commitment chain and the mean value of the historical normal sample features; is obtained by statistical analysis of the distribution of abnormal distances in historical trajectory samples, and the 99th percentile of the distribution is taken as the upper limit of the abnormal distance.

[0078] A quality threshold is set, and when the comprehensive trigger criterion score is greater than or equal to the quality threshold, a quality threshold passing token is generated, the quality threshold passing token including a model identifier, an idempotent check key, a trigger time, a comprehensive trigger criterion score, a track quality base score, a space-time consistency base score, and a rule version number. After the quality threshold passing token is associated with a corresponding dangerous goods electronic waybill number, an acceptance result, and an inspection record, the quality threshold passing token serves as a quality control basis for waybill management, random inspection screening, and statistical analysis, is used to quickly identify waybills with low track quality or potential risks in a large number of submission records, and supports comparison and evaluation of the quality levels of waybills of different enterprises, different routes, or different goods types.

[0079] It should be noted that the quality threshold is determined by collecting a large number of historical submission samples, calculating the comprehensive trigger criterion score of each sample, using the qualified / unqualified results determined by a supervisor as true labels, and then determining the score threshold point that optimally balances the identification accuracy and false rejection rate through ROC curve analysis as the quality threshold.

[0080] S5: Weighted screening is performed according to the quality threshold passing token, a minimum disclosure field set is obtained, the minimum disclosure field set is subjected to hash processing and group encryption, and a signed message is obtained.

[0081] A privacy disclosure policy configuration file is set, the privacy disclosure policy configuration file including a disclosure level (for example, public field, regulatory field, sensitive field), a weight coefficient (reflecting the balance between the regulatory value and the privacy risk of the field), and an encryption method (plain text, digest, group encryption); after the quality threshold passing token is obtained, the corresponding instance record is retrieved in the modeling context according to the model identifier and the idempotent check key in the quality threshold passing token, the standardized field structure of the current instance record is restored through the field normalization template, and the standardized field structure is used as a submission field set.

[0082] The submission field set is subjected to weighted screening through the privacy disclosure policy configuration file, the regulatory value weight and the privacy risk weight are assigned to each field, and the disclosure priority score of each field is obtained by comprehensive scoring combined with a privacy constraint coefficient; and then the disclosure priority scores are sorted and filtered according to a disclosure threshold, only the fields necessary for regulatory verification are retained, and a minimum disclosure field set is formed.

[0083] It should be noted that the regulatory value weight is determined by extracting the use frequency and regulatory call times of each field from the historical transportation task and reporting records as the main indicators of regulatory value, combined with the field importance label of the regulatory department and the policy provisions, and using multi-dimensional feature regression analysis to determine the influence degree of each field on the regulatory target, thereby generating the regulatory value weight. The historical transportation task and reporting records include the archived dangerous goods electronic waybill records, spot check and audit records, and out-of-province city waybill check records; the privacy risk weight is determined by identifying and classifying the sensitive information such as enterprise operation information, personnel identity information and geographic location in the field content through the enterprise internal data classification and grading system, and counting the leakage risk and access frequency of these information in historical data, thereby determining the privacy risk weight of the field. The privacy constraint coefficient is determined by analyzing the balance point of the occurrence rate of privacy leakage risk events and the regulatory verification demand in the history reporting; the disclosure threshold is determined by analyzing the relationship between the field disclosure score and the regulatory audit pass rate in the historical reporting samples in the offline stage, using ROC curve analysis to determine the critical value that optimally balances the compliance pass rate and privacy risk, and usually taking the 80-90 percentile of the score distribution as the default disclosure threshold. In specific application, the disclosure threshold can be adjusted periodically by combining the dangerous goods electronic waybill spot check return records and problem rectification records in recent years, so that the field disclosure strategy can meet the pass rate requirements of regulatory audit, and sensitive data beyond the necessary range will not be exposed in regular waybill query and report statistics.

[0084] The minimum disclosure field set is hashed, the SM3 algorithm is used to calculate the hash value of each field content, a binary Merkle tree is constructed by arranging all field hash values in a fixed order, a corresponding Merkle path proof (hash chain from field node to tree root) is generated for each field, and the generated Merkle tree root hash value is used as the disclosure root digest.

[0085] Sensitive field subsets are selected from the minimum disclosure field set, grouped and encrypted according to field categories, independent symmetric keys are generated for each group using the national SM4, encryption initial vectors are generated using model identifiers and rule version numbers in the quality threshold pass token, and symmetric encryption algorithms are used to perform grouped encryption operations on all data in the field group using independent symmetric keys and encryption initial vectors to obtain encrypted field payloads. After encryption, the symmetric key of the group is non-symmetrically encrypted and packaged using the public key provided by the regulatory agency, so that the regulatory agency can restore it by private key decryption when needed.

[0086] The plaintext field (non-sensitive part), encrypted field payload, disclosure root abstract, and corresponding Merkle path proof are packaged into a standardized JSON report payload, and the report payload is signed using the enterprise private key, signed using the enterprise private key (SM2), to obtain a signature value, and the signature value is packaged with the report payload to obtain a signed message.

[0087] S6: Based on the signed message, repeated search and verification are performed to obtain an audit log.

[0088] The signed message is sent to the supervision platform, which verifies the signed message.

[0089] Further, the SM3 is used to recalculate the abstract value of the message payload, the signature algorithm (SM2) is called to verify the consistency of the signature and the abstract value, if the signature verification is passed, it is proved that the signed message has not been tampered with in the transmission process, and the data source is reliable; the idempotent check key is extracted from the message, and it is queried whether there is a record with the same key, if there is, it is determined that the message is a repeated report, and the previous acceptance receipt is returned directly to realize idempotent guarantee, at this time the idempotent check key is consistent with the same transportation task in the dangerous goods electronic waybill record, waybill inspection record and waybill record of other provinces and cities generated before, and a new acceptance record is not established again, only the access and verification time of this time is appended in the original record; if not, a new acceptance record is created and processing continues; a unique acceptance number is generated by using timestamp, enterprise identifier and abstract value, and the acceptance number and processing time are packaged into a receipt message. The receipt message is returned to the enterprise end through the original encryption channel, after receiving the receipt message, the acceptance number is bound with the quality threshold token and archived to form a complete supervision traceability record.

[0090] When it is detected that no receipt message is received within a preset exceeding time (e.g., 60 seconds) or the acceptance number check fails, the SMS channel emergency reporting is triggered. Specifically, the key data items (including model identifier, idempotent check key, disclosure root abstract, quality threshold token abstract and signature value) in the minimum disclosure field set are reassembled into an SMS channel message payload. In order to adapt to the length limit of SMS transmission, the SMS channel message payload is fragmented and encoded, and the size of each fragment is not more than the standard SMS data frame (e.g., 140 bytes). After splitting the message content into several SMS fragments, in order to avoid incomplete information caused by fragment loss during transmission, based on the error correction coding principle, K redundant fragments are generated based on the original N data fragments, so that the supervision end can recover the complete message content even if some fragments are missing, thereby improving the reliability of the SMS channel transmission. The parity redundancy coding algorithm is used to check and enhance the SMS fragments to generate redundant fragments, and the expression is:

[0091] ;

[0092] wherein, represents the thredundant segment, represents the total number of original segments, represents the thoriginal segment, bit of data, represents the bitwise XOR operator.

[0093] The redundant segments are grouped together with the original segments to form a fragmented set. The generated fragmented set is sequentially numbered, and the task number (corresponding to the model identifier), the segment number and the total number of segments, and the supervision check mark (derived from the idempotent check key) are attached to the header of each segment. All segments are sent to the supervision end through the SMS gateway. After receiving the segments, the supervision end reorganizes them according to the task number and the sequence number. If some segments are missing, they are reconstructed through the redundant segments to recover the original SMS channel message payload. The signature value carried in the SMS channel message payload is matched with the enterprise public key certificate to verify the source and integrity of the submission content. Then, the disclosure root digest and quality threshold token digest in the message are compared with the last successfully submitted version stored in the supervision end to confirm that the SMS submission content indeed comes from the same submission task of the main channel and the fields have not been tampered with. After verification, the minimum field consistency check is performed. The disclosure field name, number, and disclosure root digest are compared to confirm that the content is completely consistent with the minimum disclosure set, so that this SMS submission is marked as an emergency submission completion status in the dangerous goods electronic waybill record, the out-of-province / city waybill management record, and the waybill inspection record. According to the signature digest of the SMS submission payload, the SMS reception time, and the supervision access identifier, a short receipt number is generated and returned to the enterprise end through the SMS gateway. After receiving the short receipt number, it is bound with the quality threshold token and recorded in the audit log as an emergency closed-loop voucher in case of main channel submission failure. The short receipt number can be used as a marker field in historical waybill queries, waybill inspection statistics, and annual waybill statistics to identify which waybills are completed through the emergency channel.

[0094] Read the success and failure report records in the past period of time (for example, the last 30 times), extract the quality threshold value, regulatory value weight, privacy risk weight and comprehensive trigger criterion score, calculate the mean difference and variance difference between the success samples and the failure samples, dynamically adjust the parameters including the quality threshold value, field disclosure weight, regulatory value weight, privacy risk weight and adjustment coefficient by the least variance principle, so that the high-quality samples are more stable and the abnormal samples are more sensitive, the success and failure report records can be extracted from the dangerous goods electronic waybill record, waybill inspection record, out-of-province and city waybill inspection record, and annual waybill statistics, monthly waybill abnormal rate, monthly transportation analysis and other statistical results, and by comparing the performances of different enterprises, different routes and different goods categories in the success rate and abnormal rate, the dynamically adjusted parameters are more suitable for the actual supervision scene.

[0095] The embodiment also provides a dangerous goods road transportation electronic waybill reporting system, comprising: a data processing module, which collects structured reporting data, performs field renaming, type conversion and legality detection, and obtains a modeling context;

[0096] A normalization coding module performs normalization coding through the modeling context and obtains an idempotent check key;

[0097] A commitment generation module performs sampling and smoothing processing through the idempotent check key, calculates an event commitment value, and obtains a space-time commitment chain;

[0098] A quality evaluation module calculates a track quality base score and a space-time consistency base score based on the space-time commitment chain, and obtains a quality threshold passing token;

[0099] A minimum disclosure module performs weighted screening according to the quality threshold passing token, obtains a minimum disclosure field set, performs hash processing and group encryption through the minimum disclosure field set, and obtains a signed message;

[0100] An audit verification module performs repeated search and verification based on the signed message, and obtains an audit log. The embodiment also provides a computer device suitable for the dangerous goods road transportation electronic waybill reporting method, comprising: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the dangerous goods road transportation electronic waybill reporting method proposed in the above embodiment.

[0101] In summary, the application realizes the standardization, semantic unification and unique identification generation of the multi-source fields of the electronic transport document of dangerous goods through the cooperative processing mechanism of structured data analysis and normalized coding; realizes the quality grading and space-time consistency quantitative checking of the vehicle trajectory data through the linkage mechanism of the space-time commitment chain and the quality evaluation module; realizes the minimum disclosure and verifiable encryption submission of sensitive data through the combination mechanism of the minimum disclosure and audit verification module.

[0102] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for submitting electronic waybills for road transport of dangerous goods, characterized in that: include, Collect structured reporting data, perform field renaming, type conversion, and validity checks to obtain the modeling context; The idempotency check key is obtained by normalizing the encoding through the modeling context. Sampling and smoothing are performed using idempotent check keys to calculate event commitment values ​​and obtain the spatiotemporal commitment chain. Based on the spatiotemporal commitment chain, calculate the trajectory quality basis score and the spatiotemporal consistency basis score, and obtain the quality threshold pass token; Based on the quality threshold, the token is used for weighted filtering to obtain the minimum set of disclosure fields. The minimum set of disclosure fields is then hashed and grouped for encryption to obtain the signed message. Based on the signed message, perform repeated retrieval and verification to obtain the audit log; The quality threshold is obtained via a token, and the specific steps are as follows: The trajectory quality baseline and the spatiotemporal consistency baseline are fused and calculated to obtain a comprehensive trigger criterion score. When the comprehensive trigger criterion score is greater than or equal to the preset quality threshold, a quality threshold pass token is generated. The specific steps for obtaining the minimum set of disclosure fields by weighted filtering using tokens based on quality thresholds are as follows: Based on the quality threshold, the corresponding instance record is retrieved in the modeling context using a token to obtain the set of reporting fields; Configure a privacy disclosure policy configuration file, perform weighted filtering on the set of reported fields, calculate the disclosure priority score, sort and filter the disclosure priority scores, and obtain the minimum set of disclosure fields.

2. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 1, characterized in that: The process involves collecting structured reporting data, renaming fields, converting types, and performing validity checks to obtain the modeling context. The specific steps are as follows: Collect structured reporting data, set up standard field mapping tables and regulatory rule files, rename fields, convert types and perform legality checks on structured reporting data, and obtain its standardized dataset; Extract the rule version number and security configuration parameters from the regulatory rule file to obtain the rule security binding block; For each field in the standardized dataset, a field hash digest is calculated using a hash algorithm. The hash digests of all fields are then organized into a field hash digest matrix. This field hash digest matrix is ​​then encapsulated with a rule security binding block to obtain the modeling context.

3. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 2, characterized in that: The steps for obtaining the idempotent check key by modeling the context and performing normalized encoding are as follows: By modeling the context, fields related to unique identifiers are grouped into a unique identifier field set, information reflecting transportation time is grouped into a time field set, and information describing the geographical scope of transportation is grouped into a geofence field set. Perform character standardization on the unique identifier field set, perform time zone normalization and timestamp conversion on the time field set, perform geocoding and precision truncation on the geofence field set, and obtain normalized field data; Based on normalized field data, an idempotent check key is calculated using a secure digest algorithm.

4. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 3, characterized in that: The specific steps for calculating the event commitment value and obtaining the spatiotemporal commitment chain using an idempotent verification key are as follows: Collect trajectory data, operation time data, and environmental and network supporting data, and synchronize and fuse them according to timestamps to obtain trajectory stream; Based on trajectory flow, sampling and smoothing processes are performed to remove abnormal location points and obtain a standardized trajectory point sequence. By standardizing trajectory point sequences, combining modeling context and idempotent verification keys, event identification is performed, event records are obtained, and the event commitment values ​​are calculated sequentially according to the event records to form a spatiotemporal commitment chain.

5. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 4, characterized in that: The specific steps for calculating the trajectory quality basis score and spatiotemporal consistency basis score based on the spatiotemporal commitment chain are as follows: Based on the standardized trajectory point sequence, the trajectory sampling completeness rate, the mean of location reliability and the proportion of time continuity are calculated, and a weighted summary is performed to obtain the trajectory quality baseline score; Based on the spatiotemporal commitment chain, the event time monotonicity, velocity rationality, and path coherence rate are calculated, weighted and summarized to obtain the spatiotemporal consistency basis score.

6. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 5, characterized in that: The specific steps for obtaining the signed message by hashing and block encryption using the minimum set of disclosure fields are as follows: Hash the minimum set of disclosure fields, construct a Merkle tree and generate Merkle path proofs for each field to obtain the disclosure root digest; Extract a subset of sensitive fields from the minimum set of disclosure fields, group and encrypt them according to field category, obtain the encrypted field payload, and package the Merkle path proof, disclosure root digest and encrypted field payload into a unified package to obtain the signature message.

7. The method for submitting electronic waybills for road transport of dangerous goods as described in claim 6, characterized in that: The specific steps for performing repeated retrieval and verification based on the signed message to obtain the audit log are as follows: Perform repeated retrieval and verification based on the signature message and idempotent check key to obtain the acceptance receipt; When verification fails or no acceptance receipt is obtained, the SMS channel message payload is reassembled using the minimum disclosure field set to obtain the short receipt number. The acceptance receipt or short receipt number is then used as an index to retrieve the audit log.

8. A system for submitting electronic waybills for road transport of dangerous goods, based on the method for submitting electronic waybills for road transport of dangerous goods as described in any one of claims 1 to 7, characterized in that: This includes a data processing module, which collects structured reporting data, performs field renaming, type conversion, and validity checks, and obtains the modeling context; The normalization encoding module performs normalization encoding by modeling the context to obtain the idempotency check key; The commitment generation module performs sampling and smoothing processing through an idempotent verification key, calculates the event commitment value, and obtains the spatiotemporal commitment chain; The quality assessment module, based on the spatiotemporal commitment chain, calculates the trajectory quality basis score and the spatiotemporal consistency basis score, and obtains the quality threshold pass token; The minimum disclosure module uses tokens to perform weighted filtering based on quality thresholds to obtain a minimum disclosure field set. Then, it performs hashing and block encryption on the minimum disclosure field set to obtain a signed message. The audit verification module performs repeated retrieval and verification based on the signed message to obtain the audit log.

Images