Logistics transportation vehicle real-time positioning and monitoring intelligent sensing system

Abstract

The invention discloses a logistics transportation vehicle real-time positioning and monitoring intelligent sensing system, and belongs to the technical field of logistics transportation positioning. The system comprises a transportation supervision platform. The transportation supervision platform is in communication connection with a transportation multi-source data acquisition module, a shallow transportation data analysis module and a deep risk analysis module. Three types of core data sets are constructed through the transportation multi-source data acquisition module, and in combination with sliding window analysis, fixed rule verification and fluctuation threshold analysis of the shallow layer module, multi-sensor time sequence asynchronous hidden danger full-dimension capture and three-level early warning are realized, and data failure and risk missed judgment are avoided; the deep layer module carries out full-link root cause tracing on early warning, three-dimensional evaluation quantitative grading is combined, the defect that only surface abnormity can be found traditionally is overcome, and an accurate decision basis is provided for positioning sensing linkage analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics transportation positioning, in particular to a logistics transportation vehicle real-time positioning monitoring intelligent sensing system. BACKGROUND

[0002] The logistics transportation vehicle real-time positioning monitoring intelligent sensing system is an intelligent management tool integrating positioning technology, sensing technology, communication technology and data processing technology. The core architecture comprises a vehicle-mounted sensing layer, a data transmission layer and a management platform layer. The vehicle-mounted sensing layer integrates a GPS / Beidou dual-mode positioning module, speed, temperature, humidity, vibration sensors and an OBD device, and is responsible for collecting vehicle position, running state and cargo environment data. The transmission layer realizes low-delay data transmission through 4G / 5G, satellite communication or Wi-Fi, etc. The management platform layer stores, analyzes and visualizes the data to provide decision support for logistics enterprises.

[0003] The conventional logistics transportation vehicle real-time positioning monitoring intelligent sensing system has a single data dimension for monitoring the time sequence asynchronous hidden troubles of multiple sensors, and only focuses on the positioning data or the timestamp information of a single sensor, lacking comprehensive capture of clock synchronization quality, cross-sensor logical association and transmission link stability, which leads to inability to early identify batch time sequence misplacement and transmission link fluctuations and other hidden troubles, and the early warning threshold is mostly a fixed value, which is difficult to adapt to the needs of different transportation scenarios and cargo types, resulting in early warning lag or high false alarm rate.

[0004] In view of the above technical defects, a solution is proposed. SUMMARY

[0005] The purpose of the present application is to provide a logistics transportation vehicle real-time positioning monitoring intelligent sensing system to solve the problems.

[0006] To achieve the above purpose, the present application provides the following technical scheme: a logistics transportation vehicle real-time positioning monitoring intelligent sensing system, comprising a transportation supervision platform, the transportation supervision platform being communicatively connected with a transportation multi-source data acquisition module, a shallow transportation data analysis module and a deep risk analysis module; The transportation multi-source data acquisition module acquires time synchronization quality data sets, cross-sensor logical association consistency data sets and data transmission link time sequence stability data sets according to the supervision cycle of the target transportation vehicle load, and analyzes and processes the obtained data in sequence; The shallow transportation data analysis module analyzes the collected data according to the sub-period to generate corresponding multi-level early warning signals and the judgment reasons for generating multi-level early warning signals; The deep risk analysis module carries out root cause backtracking analysis combined with a plurality of groups of data collected by a plurality of levels of early warning signals, obtains corresponding judgment basis, and constructs a comprehensive score rule combined with a duration dimension to generate a risk level and a corresponding processing signal.

[0007] Further, the data acquisition and processing process of the transportation multi-source data acquisition module is as follows: According to the starting moment of the target transportation vehicle carrying goods to the current data acquisition cutoff moment, it is marked as a supervision period, and the supervision period is divided into i sub-periods on average, combined with a plurality of types of sensors deployed on the target transportation vehicle, corresponding related data is classified and multi-sourced, time synchronization quality data set, cross-sensor logical correlation consistency data set and data transmission link time sequence stability data set are obtained.

[0008] Further, a plurality of sub-periods are randomly selected in a reverse manner, and the local clock data value of the continuous acquisition sensor in the sub-period is compared with the time reference. The absolute value of the obtained value is marked as the clock drift cumulative value, and the time difference from receiving the transportation supervision platform synchronization signal to completing the clock calibration is recorded. After each synchronization, it is collected in real time and marked as synchronization signal response delay data; analyze the data proportion of the sensor data timestamp and the platform synchronization time deviation within 10 milliseconds in a single sub-period, and mark it as timestamp consistency proportion data. It is marked as time synchronization quality data set; The matching of the vehicle state recorded by the recording positioning module and the speed data output by the OBD device and the vibration data output by the vibration sensor of the target transportation vehicle in the supervision period is analyzed, and the state correlation abnormal number is constructed according to the time of data acquisition; collect the time difference between the change of the environmental sensor data deployed on the target transportation vehicle and the change of the position recorded in the positioning module, and mark it as environment-position response delay data. It is marked as cross-sensor logical correlation consistency data set.

[0009] Further, the time difference between the data collected from the sensor and successfully received by the transportation supervision platform is analyzed, the fluctuation range of all transmission delays in a plurality of sub-periods is randomly intercepted, the average value is taken, and it is marked as transmission link delay fluctuation rate; for large data volume sensors, collect the deviation number of the reordering sequence after packet transmission and the original collection sequence, and mark it as data packet reordering time deviation; compare the cache data storage length of the vehicle-mounted transformation computing node with the preset length, mark the comparison result as edge node cache backlog length, and mark it as data transmission link time sequence stability data set.

[0010] Further, the analysis process of the shallow transportation data analysis module on the plurality of levels of early warning signals is as follows: According to the preset value of the type of sensor deployed in the target transport vehicle during use, the preset deviation threshold is obtained and compared with the accumulated deviation value in the sub-period. If the accumulated deviation value exceeds the deviation threshold, it is judged that the clock synchronization is abnormal, and a first-level warning signal is generated. If the selected sub-period has continuous multiple synchronization response delays that exceed the preset upper limit, which is a floating range constructed according to historical data combined with the time difference in the current monitoring period, it is determined that the synchronization response is timed out, and a second-level warning signal is generated. When the selected sub-period has data proportion lower than the lower limit of the platform synchronization time deviation, it is judged that the batch data has time sequence dislocation, and there is a timestamp consistency anomaly, generating a first-level warning signal.

[0011] Further, if the number of state correlation anomalies in a sub-period exceeds the preset upper limit of anomaly number, it is determined that the sensor state time sequence is contradictory, and a second-level warning signal is generated. The environment-position response delay analysis data is processed. If the response delay exceeds the preset time, which is a preset standard according to the target transport vehicle loading goods for advance replacement, it is determined that the environment-position time sequence is lagging, and a first-level warning signal is generated. The transmission link delay fluctuation rate is compared with the retrieved preset delay fluctuation range. If the delay fluctuation range exceeds the preset fluctuation threshold, it is determined that the transmission time sequence is unstable, and a first-level warning signal is generated. If there is a fluctuation range exceeding the preset deviation number in a cycle node, it is judged that the packet recombination time sequence is abnormal, and a second-level warning signal is generated. If there is an edge node cache backlog duration exceeding the duration fluctuation range, it is determined that there is a cache time sequence superposition risk, and a first-level warning signal is generated.

[0012] Further, the analysis process of the deep risk analysis module is as follows: The root cause tracing of time synchronization quality related warning is as follows: when the warning signals of clock synchronization anomaly, synchronization response timeout and timestamp consistency anomaly are determined, the complete record of time synchronization quality data set in the corresponding sub-period is extracted, and the root cause is disassembled in combination with sensor hardware archives and vehicle-mounted environment monitoring data. The root cause tracing of cross-sensor logic correlation related warning is as follows: when the warning signals of sensor state time sequence contradiction and environment-position time sequence lag are generated, the cross-sensor logic correlation consistency data set in the corresponding sub-period and the time sequence synchronization record of each sensor are linked to carry out correlation root cause analysis. The root cause tracing of transmission link time sequence stability related warning is as follows: when the warning signals of transmission time sequence instability, packet recombination time sequence anomaly and cache time sequence superposition risk are generated, the complete record of time sequence stability data set of data transmission link is combined with network operator signal log and edge node running state data to carry out root cause positioning.

[0013] Further, the risk quantitative evaluation of time asynchronous hidden danger is based on the root cause tracing result, a three-dimensional risk evaluation system is constructed, the hidden danger is quantitatively scored from three dimensions of influence range, duration and data validity influence degree, and finally a low, medium and high three-level risk judgment signal is generated.

[0014] Further, according to the corresponding risk level, a monitoring type processing signal, an intervention type processing signal and an emergency processing signal are generated, and a comprehensive score rule is constructed: if the root cause is a local slight problem and only a small amount of non-core data is affected, it is determined as low risk and a monitoring type processing signal is generated; if the root cause is a local persistent problem and has caused the validity of part of core data to decrease, it is determined as medium risk and an intervention type processing signal is generated; if the root cause is a global systematic problem and has caused a large amount of core data to be invalid or there is a transportation safety hidden danger, it is determined as high risk and an emergency processing signal is generated.

[0015] The beneficial effects of the present application are: 1. The present application constructs three types of core data sets of time synchronization quality, cross-sensor logical association consistency and data transmission link time sequence stability through the transportation multi-source data acquisition module, combines the sliding window analysis, fixed rule verification and fluctuation threshold analysis method of the shallow transportation data analysis module, realizes full-dimensional capture and three-level accurate early warning of multi-sensor time asynchronous hidden danger, identifies slight hidden danger to serious failure from the multi-dimensional of clock synchronization, sensor association and transmission link, the early warning response is more timely and comprehensive, and the data invalidation or transportation risk missed judgment caused by time asynchronous is effectively avoided.

[0016] 2. The present application carries out full-link root cause tracing on the early warning signal through the deep risk analysis module, realizes hidden danger quantitative grading combined with the three-dimensional risk evaluation system, provides accurate decision basis for positioning sensor linkage analysis, and solves the defects that the traditional method can only find surface abnormalities and cannot locate the root cause. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 The system flow chart of the present application is shown in the figure. Figure 2 The flow chart of the shallow transportation data analysis module of the present application is shown in the figure. DETAILED DESCRIPTION

[0019] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0020] Embodiment one: please refer to Figure 1 - Figure 2 As shown in the figure, the embodiment is a logistics transport vehicle real-time positioning monitoring intelligent sensing system, comprising a transport supervision platform, the transport supervision platform being communicatively connected with a transport multi-source data acquisition module, a shallow transport data analysis module and a deep risk analysis module: The transport supervision platform obtains high-precision standard time in a cloud database based on Beidou / GPS dual mode, combines a timer internally provided by a target transport vehicle to jointly construct a time reference, periodically sends a synchronization signal to all sensors disposed in the target transport vehicle, the sensors automatically calibrate local clocks after receiving the synchronization signal, and ensures that the collection time of all devices is based on the same time axis, thereby reducing clock drift caused time sequence asynchrony from the source; The transport multi-source data acquisition module, according to the starting time of the target transport vehicle to the current data acquisition cutoff time, marks it as a supervision period, divides the supervision period into i sub-periods on average, i is a natural number greater than zero, and combines a plurality of types of sensors disposed on the target transport vehicle to classify and collect corresponding related data, to obtain a time synchronization quality data set, a cross-sensor logical association consistency data set and a data transmission link time sequence stability data set; The time synchronization quality data set includes clock drift cumulative value, synchronization signal response delay data and timestamp consistency proportion data, and specifically as follows: Randomly frame multiple groups of sub-periods in a reverse way, and compare the local clock data value of the continuously collected sensors in the sub-periods with the time reference, and mark the absolute value of the obtained value as the clock drift cumulative value; Record the time difference from when the sensor receives the synchronization signal of the transport supervision platform to when the clock calibration is completed, collect in real time after each synchronization, mark it as synchronization signal response delay data, and analyze the data proportion of the sensor data timestamp and the platform synchronization time deviation within 10 milliseconds in a single sub-period, and mark it as timestamp consistency proportion data; It should be noted that the clock drift cumulative value represents the clock offset of the monitoring sensor due to hardware aging and electromagnetic interference. Excessive deviation will directly cause timing asynchrony. The synchronization signal response delay data represents whether the sensor is out of synchronization due to communication failure or excessive load. The timestamp consistency ratio data represents the overall effectiveness of the sensor timing synchronization. A low ratio indicates a risk of batch timing misplacement. The data collection sources are the transport supervision platform time reference log, sensor synchronization response feedback information, and vehicle-mounted edge computing node data preprocessing records. The collection frequency is consistent with the sensor synchronization period.

[0021] The cross-sensor logical association consistency data set includes state association abnormality times and environment-location response delay data, as follows: The matching of the vehicle state output by the recording positioning module and the speed data output by the OBD device and the vibration data output by the vibration sensor for the target transport vehicle within the supervision period is recorded. The state association abnormality times are constructed according to the data acquisition time. The time difference between the change of the environmental sensor data deployed on the target transport vehicle and the change of the location recorded in the positioning module is collected and marked as environment-location response delay data. The data collection sources are real-time output data from each sensor and positioning module geographic fence trigger records. The collection frequency is every several milliseconds. The association results are generated by real-time comparison through the edge computing node; The data transmission link timing stability data set includes transmission link delay fluctuation rate and data packet recombination timing deviation, as follows: The time difference between the completion of data collection from the sensor and the successful reception by the transport supervision platform is analyzed. The fluctuation range of all transmission delays in multiple sub-periods is randomly intercepted, and the average value is taken as the transmission link delay fluctuation rate. For large data volume sensors, the deviation times of the recombination order after packet transmission and the original collection order are collected, and marked as data packet recombination timing deviation. The comparison between the cache data storage duration of the vehicle-mounted transformation computing node and the preset duration is made, and the comparison result is marked as edge node cache backlog duration. It should be noted that excessive fluctuation of the transmission link delay fluctuation rate indicates that the transmission link is unstable, which may cause the order of data arriving at the platform to be inconsistent with the collection order. Data packet recombination timing deviation may cause the data collected later to arrive at the platform first, causing recombination timing misplacement, if network congestion occurs during packet transmission. If data cannot be uploaded in time due to bandwidth limitations, excessive cache backlog will cause subsequent data timing to be superimposed, forming an asynchronous hidden danger. The data collection sources are network transmission logs, edge node cache state monitoring records, and transport supervision platform data reception logs. The collection frequency is millisecond level, which can be adjusted according to actual needs. The timing fluctuation of the transmission link is captured in real time.

[0022] The shallow layer transportation data analysis module obtains a time synchronization quality data set, adopts a sliding window method to randomly frame a plurality of sub-periods, and jointly analyzes clock drift cumulative values, synchronization signal response delay data and timestamp consistency proportion data in all sub-periods: Taking the clock drift cumulative value as the main factor, a preset deviation threshold value is obtained according to the preset value of the sensor type deployed in the target transportation vehicle during use, and the cumulative deviation value in the sub-period is compared. If the cumulative deviation value exceeds the deviation threshold value, it is judged that the clock synchronization is abnormal, and a first-level warning signal is generated; Taking the synchronization signal response delay data as the main factor, if the response delay of synchronization exists continuously for multiple times in the selected sub-period, and the response delay exceeds the preset upper limit, the preset upper limit is a floating range constructed according to historical data and the time difference in the current supervision period, it is determined that the synchronization response is overtime, and a second-level warning signal is generated; Taking the timestamp consistency proportion data as the main factor, when the data proportion in the framed sub-period is lower than the lower limit of the platform synchronization time deviation, it is judged that the batch data exists time sequence dislocation, and the timestamp consistency is abnormal, a first-level warning signal is generated, and the lower limit of the platform synchronization time deviation is obtained according to historical data; For the cross-sensor logical association consistency data set, a preset fixed rule checking method is adopted, and under the condition of presetting 100 logical association rules: If the number of state association abnormalities in a sub-period exceeds the preset upper limit of abnormal times, it is determined that the sensor state time sequence is contradictory, and a second-level warning signal is generated; The sub-periods generating the second-level warning signals are sequentially supervised, and if there are continuous sub-periods with abnormalities, the third-level warning signal is upgraded.

[0023] The environment-position response delay analysis data is processed, if the response delay exceeds the preset time, the preset time is according to the target transportation vehicle loading goods to replace the corresponding preset standard in advance, it is determined that the environment-position time sequence is lagging, and a first-level warning signal is generated.

[0024] For the data transmission link time sequence stability data set, a fluctuation threshold analysis method is adopted, and the fluctuation threshold includes a delay fluctuation range, a deviation number fluctuation range and a time length fluctuation range: The transmission link delay fluctuation rate is compared with the preset delay fluctuation range: if the delay fluctuation range exceeds the preset fluctuation threshold, it is determined that the transmission time sequence is unstable, and a first-level warning signal is generated; When the first-level warning signal is generated, the specific value of the transmission link delay fluctuation rate still exists continuous growth, then it is replaced by the second-level warning signal.

[0025] The data packet reorganization timing deviation is compared with the preset deviation frequency fluctuation range: if the preset deviation frequency fluctuation range is exceeded within a cycle node, it is judged that the data packet reorganization timing is abnormal, and a secondary early warning signal is generated.

[0026] The edge node cache backlog duration is compared with the preset duration fluctuation range: if the edge node cache backlog duration exceeds the duration fluctuation range, it is determined that there is a cache timing superposition risk, a primary early warning signal is generated, and the primary early warning signal is continuously monitored after being generated. If it affects the next group of sub-periods, it will be upgraded to a tertiary early warning signal.

[0027] It should be noted that the early warning signal contains core information such as abnormal type, involved sensor ID, abnormal occurrence timestamp, and associated data number. The primary early warning signal indicates that there is a slight timing asynchronous hidden danger, which does not affect the validity of the core data, and only needs to be continuously monitored by the platform. The secondary early warning signal indicates that the timing asynchronous hidden danger has caused partial data logic contradiction, and needs to trigger sensor self-check or slight intervention. The tertiary early warning signal indicates that the timing asynchronous hidden danger is serious, which may cause data invalidation or transportation risk, and needs to trigger immediate active intervention measures.

[0028] Embodiment two: The deep risk analysis module takes the primary, secondary, and tertiary early warning signals generated by the shallow transportation data analysis module as input, combines the original collection data and preprocessing records of the time synchronization quality data set, cross-sensor logic correlation consistency data set, and data transmission link timing stability data set, carries out full-link root cause tracing and risk quantification evaluation, and finally generates accurate judgment signals to provide decision basis for the positioning sensor linkage analysis module; Root cause tracing of time synchronization quality related early warning: when the early warning signals of clock synchronization exception, synchronization response timeout, and timestamp consistency exception are determined, the complete records of the time synchronization quality data set in the corresponding sub-period are extracted, and the root cause is disassembled in combination with the sensor hardware archives and vehicle-mounted environment monitoring data: When the clock drift cumulative value presents the characteristics of continuous increase and stable increase rate, the use duration and maintenance records of the sensor are called, and if the use duration exceeds 18 months and no crystal oscillator calibration has been performed, it is determined that the root cause is sensor hardware aging and crystal oscillator performance degradation. If the clock drift cumulative value suddenly increases at vehicle start and engine high load period, combined with vehicle-mounted electromagnetic interference monitoring data, it is determined that the root cause is vehicle-mounted electromagnetic interference leading to clock count deviation. For the case of continuous over-standard synchronous response delay data, first check the transport supervision platform synchronization signal sending log, if the platform synchronization signal is normally sent, further analyze the communication link state of the sensor receiving end; if the signal strength is lower than the standard preset measurement value or the bus load exceeds 80%, determine that the reason is that the communication link is congested or the signal attenuation causes the synchronization instruction receiving delay; if the communication link state is normal, call the sensor running log, if there is a record of high CPU load of the sensor, determine that the reason is that the sensor itself processing capacity is insufficient, which causes the synchronization calibration lag; When the timestamp consistency proportion data is lower than the lower limit and the batch data appears timing dislocation, compare the timestamp consistency proportion of other sensors in the same batch, if only a single sensor has this problem, combined with its synchronization response feedback information, determine that the reason is that the single sensor clock calibration is invalid; if all sensors have low proportion, check the transport supervision platform time reference log; if there is a record of satellite time service interruption of the reference server and standard deviation of crystal oscillator time service exceeding the standard, determine that the reason is that the global time reference synchronization fails.

[0029] The root cause tracing of the cross-sensor logical association related warning, when generating the sensor state timing contradiction, environment-position timing lag warning signal, the cross-sensor logical association consistency data set in the corresponding sub-period and the timing synchronization record of each sensor are linked to carry out association root cause analysis: For the case of over-standard number of state association abnormalities, extract the positioning module timestamp, OBD device timestamp, and vibration sensor timestamp at the time of abnormality, if the timestamp deviation of the three exceeds 15 milliseconds, and the positioning module timestamp and the platform reference time deviation is the largest, determine that the reason is that the positioning module timing dislocation causes the state judgment contradiction; if the timestamp of the three is consistent but the data logic is still contradictory, further check the sensor installation position, exclude the hardware installation problem, and determine that the reason is that the multi-sensor data acquisition timing is out of synchronization, which causes the state association failure; For the warning of environment-position response delay exceeding the standard, call the positioning module geographic fence trigger timestamp, environment sensor data change timestamp and transmission link log, if the timestamp deviation of the two is within 50 milliseconds but the platform receiving time deviation exceeds the preset threshold, combined with the transmission link delay fluctuation rate data, determine that the reason is that the environment data transmission delay causes the response lag; if the timestamp deviation of the two itself exceeds 50 milliseconds, and the timestamp consistency proportion of the environment sensor is lower than 90%, determine that the reason is that the environment sensor clock synchronization failure causes the data acquisition timing to lag behind the position change.

[0030] The root cause tracing of the transmission link timing stability related warning, when generating the transmission timing instability, packet recombination timing anomaly and cache timing superposition risk warning signal, combined with the complete record of the data transmission link timing stability data set, the network operator signal log and edge node running state data are linked to carry out root cause positioning: When the transmission link delay fluctuation rate exceeds the threshold and continues to grow, compare the transmission link data of other transport vehicles in the same region during the same period; if there is a general delay fluctuation, determine that the root cause is weak regional network signal coverage or operator network congestion; if only the target vehicle has this problem, check the signal reception strength of the on-board communication module and the antenna connection state; if the signal strength is normal but the delay fluctuation is significant, determine that the root cause is hardware failure of the on-board communication module causing transmission timing disorder; For the case where the data packet reassembly timing deviation exceeds the standard, extract the original data packet number, sending timestamp, and receiving timestamp of the packet transmission; if the data packet sent later arrives at the platform first and the transmission link delay fluctuation rate during this period exceeds 20 milliseconds, determine that the root cause is network congestion causing data packet transmission sequence disorder; if the data packet sending and receiving timestamp sequence is consistent but the reassembled timing is still abnormal, check the edge node packet reassembly algorithm log; if there are records of algorithm version not updated and parameter configuration error, determine that the root cause is logical defect of the reassembly algorithm causing timing misplacement; When the edge node cache backlog duration exceeds the threshold, compare the sensor data acquisition frequency, data volume, and network bandwidth occupancy rate during the cache backlog period; if the data volume and acquisition frequency do not exceed the upper limit of bandwidth carrying but the cache backlog occurs, determine that the root cause is inefficient edge node data upload scheduling algorithm causing transmission queuing; if the data volume exceeds the upper limit of bandwidth carrying, combined with the network coverage map of the transportation route, determine that the root cause is insufficient network bandwidth in remote sections causing data transmission blockage.

[0031] Risk quantification evaluation of timing asynchronous hidden dangers, based on root cause tracing results, build a three-dimensional risk evaluation system, quantify and score hidden dangers from three dimensions of influence range, duration, and data validity impact degree, and finally generate low, medium, and high level risk judgment signals: Statistical number of sensor types affected by timing asynchrony, 1 type of sensor affected is 1 point, 2-3 types of sensors affected are 2 points, 4 types and above sensors affected are 3 points, build the influence range dimension; 1 point for hidden danger in a single sub-period, 2-3 consecutive sub-periods for 2 points, 4 consecutive sub-periods and above for 3 points, build the duration dimension; less than 10% of invalid data caused by timing asynchrony is 1 point, 10%-30% is 2 points, more than 30% is 3 points, build the data validity impact degree dimension; total score 3-4 points is low risk, 5-7 points is medium risk, 8-9 points is high risk, corresponding to generate monitoring type processing signal, intervention type processing signal, and emergency processing signal, build comprehensive score rules: If the root cause is a local minor problem that only affects a small amount of non-core data, the signal contains hidden danger type, involved sensor ID, and recommended monitoring period, etc. information, determine as low risk, generate monitoring type processing signal; If the root cause is a local persistent problem, which has caused the validity of part of the core data to decline, the signal contains information such as root cause type, intervention measure list, and expected repair effect, and is determined as medium risk, and an intervention type processing signal is generated; If the root cause is a global systemic problem, which has caused a large amount of core data to be invalid or has hidden transportation safety hazards, the signal contains information such as root cause level, emergency treatment process, and backup scheme trigger condition, and is determined as high risk, and an emergency processing signal is generated; It should be noted that the three-dimensional dimension weight of risk assessment can be dynamically adjusted according to the type of transported goods. The weight of the influence degree of data validity is increased by 50% in cold chain transportation and dangerous goods transportation, and the weight of the influence range is increased by 30% in ordinary goods transportation. In the root cause tracing process, an AI auxiliary model trained based on historical cases is introduced, and by matching a plurality of time sequence asynchronous hidden danger case libraries, the accuracy of root cause identification is improved.

[0032] The above is only an example and description of the structure of the present application. Those skilled in the art can make various modifications or supplements or use similar ways to replace the described specific embodiments, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present claims, and should belong to the protection scope of the present application.

[0033] In combination with Embodiment One and Embodiment Two, three types of core data sets are constructed through the transportation multi-source data acquisition module, and shallow module sliding window analysis, fixed rule verification and fluctuation threshold analysis are combined to realize multi-sensor time sequence asynchronous hidden danger full-dimensional capture and three-level early warning, avoiding data invalidation and risk omission; the deep module makes full-link root cause tracing on the early warning, and combines three-dimensional evaluation quantization grading to solve the defect that the traditional method can only find surface abnormalities, and provides accurate decision basis for positioning sensor linkage analysis.

[0034] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are contained in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0035] The preferred embodiments of the application disclosed above are only to facilitate the elucidation of the application. The preferred embodiments do not describe all the details of the application and limit the application to the specific embodiments. Obviously, many modifications and variations can be made in light of the teachings above. The description is chosen and described in order to provide the best illustration of the application and its practical application to those skilled in the art and to enable those skilled in the art to best utilize the application. The application is limited only by the claims and their full scope and equivalents.

Claims

1. A real-time positioning monitoring intelligent sensing system for logistics transport vehicles, characterized in that, The transport supervision platform is communicatively connected with a transport multi-source data acquisition module, a shallow transport data analysis module and a deep risk analysis module. The transport multi-source data acquisition module acquires time synchronization quality data sets, cross-sensor logical correlation consistency data sets and data transmission link time sequence stability data sets according to a target transport vehicle cargo supervision cycle, and sequentially analyzes and processes the obtained data. The shallow transport data analysis module analyzes the collected data according to a sub-period to generate corresponding multi-level early warning signals and judgment reasons for the multi-level early warning signals. The deep risk analysis module performs root cause backtracking analysis on the multi-level early warning signals and the collected multiple groups of data to obtain corresponding judgment basis, and constructs a comprehensive score rule in combination with a duration dimension to generate a risk level and a corresponding processing signal.

2. The real-time positioning monitoring intelligent sensing system for a logistics transport vehicle according to claim 1, characterized in that, The data acquisition and processing process of the transport multi-source data acquisition module is as follows: According to the starting time of the target transport vehicle cargo and the current data acquisition cutoff time, the target transport vehicle cargo is marked as a supervision cycle, the supervision cycle is divided into i sub-periods on average, and related data corresponding to a plurality of types of sensors deployed on the target transport vehicle are classified and multi-sourced to obtain time synchronization quality data sets, cross-sensor logical correlation consistency data sets and data transmission link time sequence stability data sets.

3. The real-time positioning monitoring intelligent sensing system for a logistics transport vehicle according to claim 2, characterized in that, A plurality of sub-periods are randomly selected in a reverse manner, and the local clock data values of the continuous acquisition sensors in the sub-periods are compared with the time reference. The absolute value of the obtained value is marked as a clock drift cumulative value, the time difference from when the sensor receives the transport supervision platform synchronization signal to when the clock calibration is completed is recorded, and the value is marked as a synchronization signal response delay data after each synchronization. The data proportion of the sensor data timestamp and the platform synchronization time deviation within 10 milliseconds in a single sub-period is analyzed, and the data is marked as a timestamp consistency proportion data. The time synchronization quality data sets are obtained by aggregating the data. The matching of the vehicle state output by the recording positioning module, the speed data output by the OBD device and the vibration data output by the vibration sensor of the target transport vehicle in the supervision cycle is analyzed, and the state correlation abnormal number is constructed according to the data acquisition time. The time difference between the data change of the environmental sensor deployed on the target transport vehicle and the position change recorded in the positioning module is collected, and the time difference is marked as an environmental-position response delay data. The cross-sensor logical correlation consistency data sets are obtained by aggregating the data.

4. The real-time positioning monitoring intelligent sensing system of a logistics transport vehicle according to claim 3, characterized in that, The time difference between the analysis data collected from the sensor and successfully received by the transport supervision platform is analyzed, the fluctuation range of all transmission delays in a plurality of sub-periods is randomly intercepted, the average value is taken, and the average value is marked as a transmission link delay fluctuation rate. For a large amount of sensor data, the deviation number of the reordering sequence after packet transmission and the original collection sequence is collected, and the deviation number is marked as a data packet reordering time deviation. The cache data storage duration of the vehicle-mounted transformation calculation node is compared with the preset duration, the comparison result is marked as an edge node cache backlog duration, and the data transmission link time sequence stability data sets are obtained by aggregating the data.

5. The real-time positioning monitoring intelligent sensing system for a logistics transport vehicle according to claim 1, characterized in that, The analysis process of the shallow transport data analysis module on the multi-level early warning signals is as follows: According to the preset value of the type of sensor deployed in the target transport vehicle during use, the preset deviation threshold is obtained and compared with the accumulated deviation value in the sub-period. If the accumulated deviation value exceeds the deviation threshold, it is judged that the clock synchronization is abnormal, and a first-level warning signal is generated; If the selected sub-period exists continuously multiple times, the response delay exceeds the preset upper limit, and the preset upper limit is a floating range constructed according to historical data combined with the time difference in the current monitoring period, it is determined that the synchronous response is timed out, and a second-level warning signal is generated; When the selected sub-period exists, the data proportion is lower than the lower limit of the platform synchronization time deviation, it is judged that the batch data exists time sequence dislocation, there is timestamp consistency anomaly, and a first-level warning signal is generated.

6. The real-time positioning monitoring intelligent sensing system of a logistics transport vehicle according to claim 5, characterized in that, If the number of state correlation anomalies in a sub-period exceeds the preset upper limit of the number of anomalies, it is determined that the sensor state time sequence is contradictory, and a second-level warning signal is generated; The response delay analysis data of the environment-position is processed. If the response delay exceeds the preset time, the preset time is according to the target transport vehicle loading goods to replace the corresponding preset standard in advance, it is determined that the environment-position time sequence lags, and a first-level warning signal is generated; The transmission link delay fluctuation rate is compared with the preset delay fluctuation range: if there is a delay fluctuation range exceeding the preset fluctuation threshold, it is determined that the transmission time sequence is unstable, and a first-level warning signal is generated; When there is a cycle node exceeding the preset deviation number of fluctuation range, it is judged that the package recombination time sequence is abnormal, and a second-level warning signal is generated; If there is an edge node cache backlog duration exceeding the duration fluctuation range, it is determined that there is a cache time sequence superposition risk, and a first-level warning signal is generated.

7. The real-time positioning monitoring intelligent sensing system for a logistics transport vehicle according to claim 1, characterized in that, The analysis process of the deep risk analysis module is as follows: Root cause tracing of time synchronization quality related warning: when the warning signals of clock synchronization abnormality, synchronous response timeout and timestamp consistency anomaly are determined, the complete record of time synchronization quality data set in the corresponding sub-period is extracted, and the root cause is analyzed combined with sensor hardware archives and vehicle-mounted environment monitoring data; Root cause tracing of cross-sensor logic correlation related warning: when the sensor state time sequence contradiction, environment-position time sequence lag warning signals are generated, the cross-sensor logic correlation consistency data set in the corresponding sub-period is linked with the time sequence synchronization record of each sensor, and the correlation root cause analysis is carried out; Root cause tracing of transmission link time sequence stability related warning: when the transmission time sequence is unstable, the package recombination time sequence is abnormal, and the cache time sequence superposition risk warning signal is generated, combined with the complete record of time sequence stability data set of data transmission link, the root cause positioning is carried out combined with network operator signal log and edge node running state data.

8. The real-time positioning monitoring intelligent sensing system of a logistics transport vehicle according to claim 7, characterized in that, Risk quantitative evaluation of time sequence asynchronous hidden danger: based on the root cause tracing result, a three-dimensional risk evaluation system is constructed, and the hidden danger is quantitatively scored from three dimensions of influence range, duration and data validity influence degree, and finally a low, medium and high level risk judgment signal is generated.

9. The real-time positioning monitoring intelligent sensing system of a logistics transport vehicle according to claim 8, characterized in that, According to the corresponding risk level, monitoring type processing signal, intervention type processing signal and emergency processing signal are generated, and a comprehensive score rule is constructed: if the root cause is a local minor problem, only a small amount of non-core data is affected, it is determined as low risk, and a monitoring type processing signal is generated; If the root cause is local persistent problem, which has caused partial core data validity to decline, it is determined as medium risk, and an intervention type processing signal is generated; If the root cause is global systemic problem, which has caused a large number of core data to be invalid or there is a transportation safety hazard, it is determined as high risk, and an emergency processing signal is generated.

Images