Cold chain system key index risk perception and identification system

Abstract

The invention relates to the technical field of risk perception and recognition, in particular to a cold chain system key index risk perception and recognition system which comprises the steps of collecting multi-point environment data in a refrigerator car, extracting disturbance characteristics and path trends, synchronously analyzing the temperature, humidity and wind change relation and recognizing the boundary drift and track overlapping degree. And marking a risk path level and outputting a cold chain transportation risk identification result. According to the method, maximum difference node positioning, wind speed jump extraction and humidity gradient section analysis are introduced, the disturbance recognition precision and the structure analysis capability are improved, cooperative recognition is achieved by combining the path isotropy and the synchronous change relation, a drift recognition mechanism is constructed through fluctuation ratio and boundary crossing judgment, and the disturbance recognition precision and the structure analysis capability are improved. An attention level system is established based on trajectory overlapping and direction consistency, a classification structure is formed based on continuity and coverage breadth, and the accuracy of path risk identification and the trend judgment capability are enhanced.

Description

Technical Field

[0001] This invention relates to the field of risk perception and identification technology, and in particular to a risk perception and identification system for key indicators of a cold chain system. Background Technology

[0002] The field of risk perception and identification technology encompasses the monitoring, analysis, and identification of potential risks. It aims to promptly identify potential risks within a system or environment through real-time monitoring and data analysis of key indicators. Common applications include risk assessment and early warning in logistics, transportation, and production processes. Its core components typically involve data collection, processing, and analysis. By setting specific risk indicators, it can identify and report potential risks, thereby supporting relevant decision-making. This technology is widely used in various industries such as cold chain logistics, financial risk management, and health monitoring, demonstrating significant practical value.

[0003] The cold chain system's key indicator risk perception and identification system refers to the real-time identification of risk factors in the cold chain process through the collection and analysis of key data. This includes monitoring environmental conditions such as temperature and humidity during cold chain transportation, and combining this with information such as transportation time and equipment operating status to identify potential risks in the cold chain system. Specifically, by setting up a risk identification model, key indicators are calculated and compared based on real-time collected data to analyze whether there are any anomalies and to determine their risk levels. This process typically relies on continuous monitoring of cold chain environmental parameters and data processing technology to achieve timely identification of system operating status and risk warnings.

[0004] Existing technologies for identifying anomalies in key cold chain parameters mainly rely on static threshold settings and single-indicator trend analysis. They lack dynamic combination judgment methods based on the evolution path of disturbance characteristics. In actual transportation, it is difficult to establish an effective linkage judgment logic for synchronous fluctuations of multiple parameters, and a trend screening mechanism for complex disturbance trends has not been formed. Furthermore, there is no clear division of the attention level and risk distribution level of path fluctuations. This makes it difficult to integrate and identify multi-point joint risks in a timely manner. When risks continue to spread in areas with frequent interference, the feedback capability is lagging, affecting the monitoring system's ability to comprehensively capture risk status. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a risk perception and identification system for key indicators of cold chain systems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a risk perception and identification system for key indicators of a cold chain system, the system comprising:

[0007] The environmental disturbance capture module acquires temperature and humidity data from the middle section of the refrigerated truck shelf, the air duct opening, and the rear stacking area. It extracts temperature segment difference nodes, filters humidity curve turning points, extracts wind speed jump trajectories, compares the fluctuation amplitude in each type of path, summarizes the fluctuation range, and generates a multi-point disturbance feature set.

[0008] The indicator trend screening module constructs wind speed paths and matches their changing directions according to the temperature and humidity paths in the multi-point disturbance feature set, based on the same-direction wind speed paths within the period, extracts path combinations with consistent directions within the same time period, and screens out data that meet the synchronization of each changing direction to obtain the synchronous trend change structure.

[0009] The boundary drift identification module constructs a temperature and humidity change ratio trajectory based on the synchronous trend change structure, calls the delay path of the cargo stacking area, determines whether the ratio crosses the boundary within the original boundary, identifies the offset segment and marks the boundary adjustment position, and outputs the drift trigger boundary structure.

[0010] The attention level distribution module constructs an overlap mapping map by calling surface response data based on the offset path in the drift trigger boundary structure, classifies the level labels according to the overlap between temperature and humidity fluctuations and the direction of surface reaction, and generates a change attention tilt distribution map.

[0011] As a further embodiment of the present invention, the multi-point disturbance feature set includes temperature range node distribution, humidity fluctuation gradient features, and wind speed frequency jump segments; the synchronous trend change structure includes path direction consistent segments, time overlap combination relationships, and three types of synchronous trend patterns; the drift trigger boundary structure includes fluctuation amplitude ratio intervals, boundary offset time periods, and response path drift features; and the change attention tilt distribution map includes trajectory overlap continuous segments, direction consistency levels, and high-level fluctuation path distribution.

[0012] As a further embodiment of the present invention, the environmental disturbance capture module includes a multi-point data acquisition submodule, a sequence variation extraction submodule, and a fluctuation path integration submodule;

[0013] The multi-point data acquisition submodule acquires time-series data of temperature, humidity and wind speed in the middle section of the shelf inside the refrigerated truck, the location of the air duct, and the rear stacking coverage area. It records the time label, numerical output and data density of each type of data at each acquisition point, performs sampling interval consistency judgment on the acquired data, removes abnormal null values ​​and fills in missing segments, and generates standardized continuous data groups.

[0014] The sequence variation extraction submodule constructs the difference trajectory of each segment based on the temperature sequence in the standardized continuous data group, filters the variation node with the largest difference and extracts the continuous segment, locates the turning point of the humidity sequence and filters out the continuous change time period, calculates the gradient intensity in the change segment, extracts the jump segment of the wind speed data with a sliding window and counts the frequency of occurrence, and obtains the set of sub-item fluctuation trajectories.

[0015] The fluctuation path integration submodule, based on the difference segments, turning paths and jump trajectories extracted from the sub-fluctuation trajectory set, calls the duration range of the corresponding time period in each sequence, filters out the overlapping intervals that exhibit changing behavior in both temperature, humidity and wind, marks the interval as the fluctuation concentration segment, and calculates the joint duration span of various changes to generate a multi-point disturbance feature set.

[0016] As a further embodiment of the present invention, the indicator trend screening module includes a direction consistency identification submodule, a wind speed trend superposition submodule, and a time segment screening submodule;

[0017] The direction consistency identification submodule obtains the marked temperature path and humidity path in the multi-point disturbance feature set, detects the relative position of the trend direction within each period based on the change direction sequence of the two paths, performs path pairing for temperature and humidity combinations with consistent direction within the time segment, and filters the combinations according to the trend consistency ratio to generate temperature and humidity trend matching groups.

[0018] The wind speed trend overlay submodule calls the combined path in the temperature and humidity trend matching group, collects the wind speed path change sequence within the corresponding time period, performs time period registration based on the fluctuation direction of the wind speed path and the direction sequence of the temperature and humidity combination, extracts the wind speed path whose fluctuation direction and time range are consistent with the temperature and humidity combination, and obtains a three-directional consistent structure.

[0019] The time segment filtering submodule determines whether the three types of paths have completely overlapping time segments in a continuous period based on the time labels of each path in the three-directional consistent structure. It performs segment consistency judgment on the duration and start and end times of the overlapping segments, marks the combined paths that satisfy the full synchronization trend and time consistency, and generates a synchronization trend change structure.

[0020] As a further embodiment of the present invention, the boundary drift recognition module includes an amplitude ratio construction submodule, a boundary cross-boundary judgment submodule, and a drift segment extraction submodule;

[0021] The amplitude ratio construction submodule obtains the maximum and minimum values ​​of each path within the same time period based on the temperature and humidity combination path contained in the synchronous trend change structure, constructs the fluctuation amplitude ratio based on the numerical difference and the reference duration, arranges them according to the corresponding time period to generate a spectrum sequence, and generates a time period amplitude ratio spectrum.

[0022] The boundary crossing judgment submodule calls each ratio segment in the time period amplitude ratio map, locates the upper and lower limit segments of the boundary corresponding to each time period based on the response delay path recorded by each monitoring point in the cargo stacking area, and judges whether each ratio segment crosses the upper and lower boundary segments to obtain the boundary crossing interval segment.

[0023] The drift segment extraction submodule extracts the path offset trend under the corresponding time slice based on the marked cross-boundary time period in the boundary cross-boundary segment, identifies the start and end range of the original boundary position change within the time period, compares the difference between the upper and lower limit paths before and after the change and marks the offset, and generates the drift trigger boundary structure.

[0024] As a further embodiment of the present invention, the attention level distribution module includes a trajectory overlap extraction submodule, a time direction label division submodule, and an attention level confirmation submodule;

[0025] The trajectory overlap extraction submodule, based on the temperature and humidity combination path marked as offset in the drift trigger boundary structure, calls the cargo surface response change data within the corresponding time period, obtains the overlapping sections of the two types of paths and matches their respective start and end times, constructs a persistence mapping map of the overlapping area of ​​each section based on the intersection of time periods, and generates a response overlap persistence map.

[0026] The time direction labeling submodule calculates the directional consistency ratio of the two paths in the overlapping area based on the time intersection length of each overlapping segment in the response overlap duration map. It then compares the time length and directional ratio with a unified interval threshold, sets the corresponding label level according to the ratio level, and generates the overlapping label level division result.

[0027] The attention level confirmation submodule calls the high-level label path information in the overlapping label level division result, marks and binds this part of the path with the original temperature and humidity combination sequence, and extracts the offset frequency value and surface change intensity value of each bound path to construct the attention distribution map of its corresponding position tilt direction in the time space, thus obtaining the change attention tilt distribution map.

[0028] As a further aspect of the present invention, the system further includes:

[0029] The risk status output module calls the path of interest in the change attention tilt distribution map, counts the frequency and distribution of the path in the segment, filters the path with long change duration and large fluctuation area coverage, and outputs the risk perception and identification results in the cold chain transportation process.

[0030] The risk perception and identification results include a set of high-frequency fluctuation paths, the distribution of continuous segments, and statistics on the coverage area.

[0031] As a further embodiment of the present invention, the risk status output module includes a concern path identification submodule, a path segment statistics submodule, and a risk trend classification submodule;

[0032] The path identification submodule calls the fluctuating paths marked as objects of interest in the change attention tilt distribution map, extracts the time period, number and the sequence of transportation segments involved for each path, records the start and end time positions of the path in each segment, organizes and constructs a path time span table according to the path number, and generates a path segment distribution table.

[0033] The route segment statistics submodule extracts the frequency of occurrence, time span ratio, and number of distribution areas of each route in different segments based on the distribution of each route in the continuous transportation segment in the route segment distribution table. The routes are then filtered according to the occurrence frequency threshold and the distribution range threshold to obtain a set of covered distribution routes.

[0034] The risk trend classification submodule calls the number, time span, and fluctuation direction of the paths in each segment of the coverage distribution path set, extracts similar paths based on time continuity and regional consistency, merges and classifies them according to the consistency of time series between paths and the number of segment intersections, and generates risk perception identification results.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0036] In this invention, by introducing methods such as maximum difference node positioning, wind speed jump path extraction, and humidity gradient segment analysis, the temporal accuracy and structural integrity of disturbance capture are enhanced. By utilizing path trend unidirectional screening and multi-path synchronous change matching, collaborative identification of environmental disturbance trends is achieved. A response drift identification mechanism is constructed by combining fluctuation amplitude ratio and boundary crossing path. A concern level system is established through trajectory overlap mapping and directional consistency hierarchical labeling. Furthermore, a multi-dimensional classification structure is formed based on continuity and distribution breadth characteristics, which improves the discrimination ability of fluctuation path risk identification, the accuracy of trend judgment, and the depth of path assessment. Attached Figure Description

[0037] Figure 1 This is a system flowchart of the present invention;

[0038] Figure 2 This is a flowchart illustrating the acquisition process of the environmental disturbance capture module of the present invention.

[0039] Figure 3 This is a flowchart illustrating the acquisition process of the indicator trend screening module of the present invention.

[0040] Figure 4 This is a flowchart illustrating the acquisition process of the boundary drift recognition module of the present invention.

[0041] Figure 5 This is a flowchart illustrating the acquisition process of the attention level distribution module in this invention.

[0042] Figure 6 This is a flowchart illustrating the acquisition process of the risk status output module of this invention. Detailed Implementation

[0043] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0044] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0045] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0046] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0047] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0048] Please see Figure 1 This invention provides a technical solution: a risk perception and identification system for key indicators of a cold chain system, the system comprising:

[0049] The environmental disturbance capture module acquires time-series data on temperature, humidity, and wind speed in the middle section of the shelves inside the refrigerated truck compartment, the location of the air duct opening, and the rear stacking coverage area. It extracts the nodes with the largest difference in temperature sequence segments and identifies their duration. Based on the turning points of the humidity curve, it extracts the fluctuation duration segments and gradient segments. It extracts the jump segment paths according to the fluctuation frequency of wind speed within the sliding window. It summarizes and locates the intervals with the largest fluctuations in various change sequences and generates a multi-point disturbance feature set.

[0050] The indicator trend screening module performs a periodic same-direction judgment on the directional trend of the temperature and humidity sequence with the already located fluctuation path in the multi-point disturbance feature set, extracts the combination that meets the trend, constructs the directional trajectory of the wind speed path and the combination in the same time period, judges whether there is a completely synchronous change trend in the corresponding segment of each type of trajectory, screens out the data set that meets the conditions of time overlap and consistent direction, and obtains the synchronous trend change structure.

[0051] The boundary drift identification module constructs a fluctuation amplitude ratio and time period matching map based on the temperature and humidity combination path contained in the synchronous trend change structure. By comparing the response delay path recorded by each monitoring point in the cargo stacking area, it determines whether each ratio segment crosses the original upper and lower boundary segments in the delay path, extracts the time period corresponding to the crossed segment and marks the boundary offset, and outputs the drift trigger boundary structure.

[0052] The attention level distribution module, based on the temperature and humidity combination change path marked as offset in the drift trigger boundary structure, calls the cargo surface response change data within the corresponding time period of the path, constructs a persistent mapping map of the overlapping section of the two trajectories, classifies them into graded labels according to the time overlap length and the proportion of the direction consistency, records the fluctuation path with the higher label level as the attention item, and generates a change attention tilt distribution map.

[0053] The risk status output module calls the fluctuating paths marked as concerns in the change attention tilt distribution map, performs time matching processing on the continuity span of each path in the continuous transportation segment, performs zonal statistics on the frequency and distribution range of the paths in each segment, classifies and labels fluctuating paths with continuous occurrence trends and large coverage areas, and outputs the risk perception and identification results in the cold chain transportation process.

[0054] The set of multi-point disturbance features includes the distribution of temperature range nodes, humidity fluctuation gradient features, and wind speed frequency jump segments. The synchronous trend change structure includes segments with consistent path direction, time overlap combination relationships, and three types of synchronous trend patterns. The drift trigger boundary structure includes fluctuation amplitude ratio intervals, boundary offset periods, and response path drift features. The change attention tilt distribution map includes segments with continuous trajectory overlap, directional consistency levels, and high-level fluctuation path distribution. The risk perception identification results include the set of high-frequency fluctuation paths, the distribution of continuous segments, and statistics on the coverage area.

[0055] Please see Figure 2 The environmental disturbance capture module includes a multi-point data acquisition submodule, a sequence variation extraction submodule, and a fluctuation path integration submodule.

[0056] The multi-point data acquisition submodule acquires time-series data of temperature, humidity and wind speed in the middle section of the shelf inside the refrigerated truck, the location of the air duct, and the rear stacking coverage area. It records the time label, numerical output and data density of each type of data at each acquisition point, performs sampling interval consistency judgment on the acquired data, removes abnormal null values ​​and fills in missing segments, and generates standardized continuous data groups.

[0057] The study acquires time-series data on temperature, humidity, and wind speed at the mid-section of the refrigerated truck's internal shelving, the air duct opening, and the rear stacked area. Specifically, a Pt100 platinum resistance temperature sensor with an accuracy of ±0.1℃, a capacitive humidity sensor with an accuracy of ±2%RH, and a hot-wire anemometer with an accuracy of ±0.1m / s are deployed at the mid-section of the shelving (point A), the air duct opening (point B), and the rear stacked area (point C), respectively. Data is continuously collected for 900 cycles at a fixed sampling interval of 10 seconds. Each data point is associated with and stored as a collection point identifier (A, B, or C), a data type identifier (T, H, or W), a Unix timestamp (accurate to the second), and a numerical value, forming the raw dataset. The data density within each 60-second continuous cycle is calculated, which is the actual number of collection points divided by the expected number of collection points (6). If the data density is less than 0.9, it is marked as... This period is considered unreliable. A consistency check is performed on all collected data sampling intervals. The difference between adjacent timestamps is calculated to verify if it is 10 seconds. Data points with a difference of less than 10 seconds are marked, and abnormal null values ​​(-999) or exceeding the physical range (e.g., humidity exceeding 100% RH) are removed. For missing data segments due to removal or sampling failures, linear interpolation is used to fill in the missing data. For example, if the temperature at point A is 4.2℃ at timestamp 1663905600 and 4.4℃ at timestamp 1663905620, the missing temperature value at timestamp 1663905610 is calculated and filled in as (4.2+4.4) / 2=4.3℃. This process is repeated for all data points to generate a standardized continuous dataset with no null values ​​and a strict time interval of 10 seconds.

[0058] Table 1: Examples of Standardized Continuous Data Representation

[0059] Timestamp Collection point Data types numerical values 1663905600 A temperature 4.2℃ 1663905600 B humidity 85.1%RH 1663905600 C wind speed 1.5m / s 1663905610 A temperature 4.3℃ 1663905610 B humidity 85.3%RH 1663905610 C wind speed 1.4m / s

[0060] Table 1 lists some of the standardized continuous data, showing the recording format of different collection points and data types at continuous time points.

[0061] The sequence variation extraction submodule constructs the difference trajectory of each segment based on the temperature sequence in the standardized continuous data set, filters the variation node with the largest difference and extracts the continuous segment, locates the turning point of the humidity sequence and filters out the continuous change time period, calculates the gradient intensity in the change segment, extracts the jump segment of the wind speed data with a sliding window and counts the frequency of occurrence, and obtains the set of sub-item fluctuation trajectories.

[0062] Based on the temperature series in the standardized continuous dataset, a difference trajectory for each segment is constructed. Specifically, for each temperature time series from the three sampling points A, B, and C, the difference between every two adjacent data points is calculated. Each of these forms its own difference trajectory sequence, for example, point A at... The temperature at that moment was 4.3℃. If the temperature is 4.5℃, the difference is +0.2℃. We filter out the nodes with the largest differences and extract the persistent segments. Here, we set a threshold for filtering these nodes. This threshold was determined by statistically analyzing temperature data from 100 historical batches during normal transportation, using the 99.5th percentile of the absolute value of the difference as the baseline. The specific calculation process involved collecting all... Take the historical temperature difference data points, sort their absolute values, and locate the [number]th [value]. The value at this location is 0.4℃, therefore it is set... The difference is 0.4℃. Then that point A point in time is marked as a change node. A continuous segment is defined as the time period between two change nodes where all differences (positive or negative) maintain the same sign. This process is used to locate inflection points in the humidity series and filter out continuously changing time periods. The condition for being identified as a turning point is that the product of the differences between its two preceding and following values ​​is negative. The time interval between two inflection points is selected as the continuous change period, and the gradient intensity within this interval is calculated by dividing the difference between the humidity value at the end and the beginning of the interval by the duration of the interval. For example, if the humidity rises from 85%RH to 88%RH within a 300-second interval, the gradient intensity is (88-85) / 300 = 0.01%RH / second. Simultaneously, the wind speed data is processed using a sliding window of 6 data points (60 seconds) to extract transition segments, and the standard deviation of the wind speed data within each window is calculated. And set the threshold for determining the transition segment. The threshold was set based on the sliding standard deviation calculation of wind speed data for 24 consecutive hours under historical stable airflow conditions (constant outlet fan speed), and the 98th percentile of the calculated result was taken, which is 0.25 m / s. The speed is 0.25 m / s, when a certain window... At that time, the 60-second period was recorded as a jump segment, and the frequency of all jump segments within each hour was statistically analyzed. The extracted information on continuous temperature range, continuous humidity change time period and gradient, and wind speed jump segment and frequency was integrated to obtain a set of sub-item fluctuation trajectories.

[0063] Table 2: Examples of Sub-item Fluctuation Trajectories

[0064] Trajectory type Collection point Start timestamp End timestamp Key Indicators Temperature range A 1663905800 1663905920 Difference sign: + Continuous changes in humidity B 1663905820 1663905950 Gradient intensity: 0.01%RH / second Wind speed jump segment C 1663905810 1663905870 Frequency of occurrence: 4 times / hour

[0065] Table 2 provides an example of a set of sub-item fluctuation trajectories, which includes fluctuation events and their core attributes extracted from data from different sensors.

[0066] The fluctuation path integration submodule extracts the difference segments, turning paths and jump trajectories from the set of sub-fluctuation trajectories, calls the duration range of the corresponding time period in each sequence, filters out the overlapping intervals that have changes in temperature, humidity and wind, marks the interval as the fluctuation concentration segment, and calculates the joint duration span of various changes to generate a multi-point disturbance feature set.

[0067] Based on the difference segments, turning paths, and jump trajectories extracted from the set of sub-item fluctuation trajectories, the duration range of the corresponding time period in each sequence is called. For example, a temperature duration segment at point A is extracted from the set, with a time range of [1663905800, 1663905920], a humidity continuous change period with a time range of [1663905820, 1663905950], and a wind speed jump segment with a time range of [1663905810, 1663905870]. Overlapping intervals that exhibit changes in temperature, humidity, and wind are then selected. This selection process is achieved by calculating the intersection of the three time segments, i.e., the start time of the overlapping interval is the maximum of the three start times, and the end time is the minimum of the three end times. For the above example, the start time of the overlapping interval is max(1663905800, 1663905800, 1663905820). The calculation starts at time 905820, ends at time min(1663905920, 1663905950, 1663905870) = 1663905870. If the calculated start time is later than or equal to the end time, there is no overlapping interval. Otherwise, the overlapping interval [1663905820, 1663905870] is marked as a fluctuation concentration segment, and the joint duration span of various changes is calculated, i.e., the start time is subtracted from the end time of the overlapping interval. In this example, the joint duration span is 1663905870 - 1663905820 = 50 seconds. This process is performed on all fluctuation trajectories of all collection points (A, B, C), and all the found fluctuation concentration segments and their corresponding collection points, start and end times, joint duration spans, etc. are collected to generate a multi-point disturbance feature set.

[0068] Please see Figure 3 The indicator trend screening module includes a direction consistency identification submodule, a wind speed trend overlay submodule, and a time segment screening submodule;

[0069] The direction consistency identification submodule obtains the marked temperature and humidity paths in the multi-point disturbance feature set, detects the relative position of the trend direction within each period based on the change direction sequence of the two paths, performs path pairing for temperature and humidity combinations with consistent direction within the time segment, and filters the combinations according to the trend consistency ratio to generate temperature and humidity trend matching groups.

[0070] Obtain the labeled temperature and humidity paths from the multi-point disturbance feature set. Specifically, for a concentrated fluctuation segment, such as the temperature and humidity data sequence of point A within the time interval [1663905820, 1663905870], based on the change direction sequences of the two paths, detect the relative position of the trend direction within each period, and calculate the direction label for each of the two paths at every 10-second interval. If the current value is greater than the previous value, the label is +1; if it is less than, it is -1; and if it is equal to, it is 0. Thus, two sequences composed of +1, -1, and 0 are obtained. For example, the temperature direction sequence is [+ The humidity direction sequence is [+1,+1,-1,-1,+1]. Path pairing is performed on temperature and humidity combinations with consistent directions within a time interval. That is, at the same time point, if both the temperature and humidity direction labels are +1 or both are -1, then that time point is considered a point of consistent direction. In the example above, time points 1, 2, and 5 are points of consistent direction. Combinations are selected based on the trend consistency ratio, which is calculated by dividing the number of points of consistent direction by the total number of time points within the interval. In this example, the ratio is 3 / 5 = 0.6. A trend consistency ratio threshold is also set. This threshold was set by analyzing 50 sets of historical transportation data. Of these, 25 sets showed cargo damage, and 25 sets showed no damage. Analysis revealed that the proportion of consistent temperature and humidity trends in the damaged groups was generally above 0.8, while the proportion in the normal groups was below 0.7. To ensure accuracy, 0.8 was selected as the threshold. If the trend consistency ratio of a combination is less than 0.8, the combination is discarded. If it is greater than or equal to 0.8, the temperature and humidity combination path and its related information are retained to generate a temperature and humidity trend matching group.

[0071] The wind speed trend overlay submodule calls the combined path in the temperature and humidity trend matching group, collects the wind speed path change sequence within the corresponding time period, performs time period correspondence registration based on the fluctuation direction of the wind speed path and the direction sequence of the temperature and humidity combination, extracts the wind speed path whose fluctuation direction and time range are consistent with the temperature and humidity combination, and obtains a three-directional consistent structure.

[0072] The code calls up the combined paths in the temperature and humidity trend matching group. For example, a temperature and humidity combination with a trend consistency ratio of 0.9 within the time period [1663906200, 1663906300] collects the wind speed path change sequence within the corresponding time period, that is, extracts the wind speed data within that 100-second period, and calculates its corresponding direction label sequence, resulting in a wind speed direction sequence composed of +1, -1, 0. Based on the fluctuation direction of the wind speed path and the direction sequence of the temperature and humidity combination, time period correspondence registration is performed. Specifically, in the temperature and humidity combination, the time points where the temperature and humidity directions are consistent are identified, for example, in... If at a given time, both temperature and humidity directions are +1, then the common direction at that time is set to +1, and the wind speed direction sequence is searched. The direction labels of the time are used to determine if the two are equal. For example, if... If the common direction of temperature and humidity is +1, while the direction of wind speed is -1, then the time points are inconsistent. If the direction of wind speed is also +1, then the time points are consistent. Repeat this registration process for all time points within this 100-second period, extract the wind speed path whose fluctuation direction and time range are consistent with the temperature and humidity combination, that is, filter out all time points whose temperature, humidity, and wind direction labels are exactly the same and not 0, and connect the extracted time points into one or more sub-paths. Integrate the sub-paths with the original temperature and humidity combination path to obtain a structure with consistent three directions.

[0073] The time segment filtering submodule determines whether the three types of paths have completely overlapping time segments in a continuous period based on the time labels of each path in the three-directional consistent structure. It performs segment consistency judgment on the duration and start and end times of the overlapping segments, marks the combined paths that satisfy the fully synchronous trend and are time-consistent, and generates a synchronous trend change structure.

[0074] Based on the time labels of each path in the three-directional consistent structure, it is determined whether the three types of paths have completely overlapping time segments within a continuous period. Specifically, the set of all time points in which the directions of temperature, humidity, and wind are consistent is extracted from the structure, for example, the set {1663906210,1663906220,1663906230,1663906240,1663906270,1663906280}, and consecutive time points in this set are identified. In the example above, there are two consecutive sequences, [1663906210, 1663906240] and [1663906270, 1663906280]. The time segments represented by these sequences are completely overlapping time segments. A segment consistency check is performed on the duration and start / end times of the overlapping segments. The duration of each overlapping segment is calculated; for example, the first segment is 40 seconds long, and the second is 20 seconds long. A minimum duration threshold is set. The threshold is set based on the fact that cargo thermal inertia experiments show that for the specific fruits and vegetables being transported, when the duration of external environmental disturbances is less than 30 seconds, their internal temperature is basically unaffected. Therefore, the threshold is set accordingly. The threshold is set to 30 seconds. All overlapping segments are filtered using this threshold. Segments with a duration of less than 30 seconds are removed. In the example above, the second segment with a duration of 20 seconds is removed, and only the first segment with a duration of 40 seconds is retained. The combined paths that meet the full synchronization trend and are consistent in time are marked. That is, the segments [1663906210, 1663906240] and their corresponding temperature, humidity and wind data paths are marked after filtering, and a synchronization trend change structure is generated.

[0075] Please see Figure 4The boundary drift recognition module includes an amplitude ratio construction submodule, a boundary cross-boundary judgment submodule, and a drift segment extraction submodule;

[0076] The amplitude ratio construction submodule obtains the maximum and minimum values ​​of each path within the same time period based on the temperature and humidity combination path contained in the synchronous trend change structure. It constructs the fluctuation amplitude ratio based on the numerical difference and the base duration, arranges them according to the corresponding time period to generate a spectral sequence, and generates a time period amplitude ratio spectral.

[0077] Based on the temperature and humidity combination paths contained in the synchronous trend change structure, such as a synchronous path with a time period of [1663906210, 1663906240], the maximum and minimum values ​​of each path within the same time period are obtained. Within this 40-second segment, the temperature rises from 4.8℃ to 5.3℃, and the humidity rises from 88%RH to 91%RH. Therefore, , , , The fluctuation amplitude ratio is constructed based on the numerical difference and the base duration, where the base duration is... The fluctuation range is set to 60 seconds. The calculation method for the fluctuation range ratio is to divide the numerical difference by the actual duration and then multiply by the baseline duration, thus standardizing it to a rate of change per minute. The temperature fluctuation range ratio... Similarly, the fluctuation range of humidity is greater than The graph sequence is generated by arranging the graphs according to the corresponding time periods, which means that the graphs calculated for each synchronously changing structure are... The value pairs are associated with their corresponding time periods to form a time series, generating a time period amplitude ratio graph.

[0078] The boundary crossing judgment submodule calls each ratio segment in the time period amplitude ratio map, locates the upper and lower limit segments of the boundary corresponding to each time period based on the response delay path recorded by each monitoring point in the cargo stacking area, and judges whether each ratio segment crosses the upper and lower boundary segments to obtain the boundary crossing interval segment.

[0079] Call up the ratio segments in the time period amplitude ratio graph, for example, the ratio segments corresponding to the aforementioned time period [1663906210, 1663906240]. Based on the response delay paths recorded by monitoring points within the cargo stacking area, the upper and lower boundary limits for each time period are located. Specifically, these limits are preset according to the type of cargo. For the lettuce transported in this batch, the upper limit for temperature fluctuation rate is 1.0℃ / min, and the lower limit is -1.0℃ / min; the upper limit for humidity fluctuation rate is 5.0%RH / min, and the lower limit is -5.0%RH / min. These boundary values ​​were determined using data from 10 simulated transportation experiments. Environmental parameters were changed at different rates, and cargo quality was monitored to determine the critical rate of change that would not lead to quality deterioration. It was then determined whether each ratio segment crossed the upper and lower boundary segments. The calculated fluctuation amplitude ratios were compared with the boundary values. For temperature... It did not cross the boundary; regarding humidity, It did not cross the boundary, but if calculated in another time period... , then because The temperature ratio range within a given time period is determined to have crossed the upper boundary. This time period is marked and recorded. All time periods marked as crossing the boundary are then aggregated to obtain the boundary crossing range.

[0080] The drift segment extraction submodule extracts the path offset trend under the corresponding time slice based on the marked cross-boundary time period in the boundary cross-boundary segment, identifies the start and end range of the original boundary position change within the time period, compares the difference between the upper and lower limit paths before and after the change and marks the offset, and generates the drift trigger boundary structure.

[0081] Based on the boundary crossing time period marked in the boundary crossing interval, for example, a time period within [1663907100, 1663907150]. For out-of-bounds records of 1.2℃ / minute, the path offset trend under the corresponding time slice was extracted, i.e., the original temperature data sequence within the 50-second period was retrieved. This sequence showed that the temperature linearly increased from 4.6℃ to 5.6℃. The start and end range of the original boundary position within the time period was identified. Here, the original boundary refers to the absolute value range of the safe temperature of the goods, not the rate of change. For lettuce, this range was set to [1.0℃, 5.0℃]. The experimental verification process of this range was as follows: 30 lettuce samples were placed in a constant temperature environment from -2.0℃ to 8.0℃ with a step size of 0.5℃ for 48 hours. Their cell structure, water content, and signs of decay were observed and recorded. The results showed that ice crystal damage occurred below 1.0℃ and above 5.0℃. Microbial reproduction rates accelerate significantly above 0℃, thus this range is determined. Within the aforementioned time period of exceeding the boundary, the temperature path changes from 4.6℃ (within the boundary) to 5.6℃ (outside the boundary). Therefore, the starting point of the path offset is the moment when the temperature reaches 5.0℃, assuming this moment is 1663907133, and the ending point of the offset is the end time of the segment, 1663907150. Compare the difference between the upper and lower limit paths before and after the change and mark the offset amount, that is, calculate the maximum value of the path exceeding the boundary, the offset amount is 5.6℃-5.0℃=0.6℃. Mark this offset amount of 0.6℃ with the specific time period [1663907133, 1663907150] where the offset occurred to generate the drift trigger boundary structure.

[0082] Please see Figure 5 The attention level distribution module includes a trajectory overlap extraction submodule, a time direction label division submodule, and an attention level confirmation submodule;

[0083] The trajectory overlap extraction submodule, based on the temperature and humidity combination path marked as offset in the drift trigger boundary structure, calls the cargo surface response change data within the corresponding time period, obtains the overlapping sections of the two types of paths and matches their respective start and end times, constructs the persistence mapping map of the overlapping area of ​​each section based on the intersection of time periods, and generates a response overlap persistence map.

[0084] Based on the temperature and humidity combination path marked as offset in the drift-triggered boundary structure, such as the temperature path that shifted by 0.6℃ within the aforementioned time period [1663907133, 1663907150], the surface response change data of the goods within the corresponding time period is retrieved. This data is collected by a temperature patch pasted on the surface of the goods packaging box, which shows that the surface temperature rose from 4.8℃ to 5.2℃ within the time period [1663907140, 1663907160]. The overlapping sections of the two types of paths are obtained and their start and end times are matched. The ambient temperature offset path time is [1663907133, 1663907150]. The response path time is [1663907140, 1663907160]. By calculating the time intersection, the overlapping segment is obtained as [max(1663907133, 1663907140), min(1663907150, 1663907160)] = [1663907140, 1663907150]. Based on the time intersection, a persistence mapping map of the overlapping area of ​​each segment is constructed, that is, each drift event is mapped to the duration of the corresponding cargo surface response overlapping segment (10 seconds in this example), forming a list containing all drift events and their response durations, generating a response overlap persistence map.

[0085] The time direction labeling submodule calculates the directional consistency ratio of the two paths in the overlapping area based on the time intersection length of each overlapping segment in the response overlap duration map. It then compares the time length and directional ratio with a unified interval threshold, sets the corresponding label level according to the ratio level, and generates the overlapping label level classification result.

[0086] Based on the time intersection length corresponding to each overlapping segment in the response overlap duration map, such as the aforementioned 10-second overlapping segment, the directional consistency ratio of the two paths within the overlapping area is calculated. Specifically, within the time period [1663907140, 1663907150], the directional sequence of the ambient temperature path (one point every 10 seconds) is [+1], and the directional sequence of the cargo surface temperature path is [+1]. Therefore, the directional consistency ratio is 1 / 1 = 1.0. The time length and directional ratio are compared with a unified interval threshold, where a time length threshold is set here. With direction ratio threshold The threshold was set based on historical data analysis, which statistically analyzed 30 risk events that caused level 2 or higher damage to goods. It was found that the overlap duration of these events all exceeded 8 seconds, and the directional consistency was all higher than 0.9. In contrast, over 95% of drift events that did not cause damage did not meet these two conditions. Therefore, the threshold was set... It lasts for 8 seconds. The value is 0.9. The corresponding label level is set according to the ratio level, with the following rules: if the overlap duration > And the directional ratio is greater than The label level is "high" if one of the conditions is met, the label is "medium" if one of the conditions is met, and "low" if none of the conditions are met. In this example, the overlap duration is 10 seconds > 8 seconds and the direction ratio is 1.0 > 0.9, so its label level is set to "high". Perform this operation on all overlapping segments to generate the overlapping label level classification results.

[0087] The attention level confirmation submodule calls the high-level label path information in the overlapping label level division result, marks and binds this part of the path with the original temperature and humidity combination sequence, and extracts the offset frequency value and surface change intensity value of each bound path to construct the attention distribution map of its corresponding position tilt direction in the time space, thus obtaining the change attention tilt distribution map.

[0088] The system retrieves high-level label path information from the overlapping label classification results, specifically filtering out all paths labeled "high" and their associated time and location information. These paths are then labeled and bound to the original temperature and humidity combination sequence. Specifically, in the original, complete temperature and humidity time series database, data for the corresponding time period is located, and a "high concern" flag is added. The system then extracts the offset frequency and surface change intensity values ​​for each bound path. The offset frequency value is the number of times the path is labeled "high" during the entire transportation task, and the surface change intensity value is the magnitude of change in the corresponding cargo surface response path within the overlapping section. For example, for the section [1663907140, 1663907150], the surface temperature ranges from... If the temperature changes from 4.9℃ to 5.2℃ (hypothetically), the surface change intensity value is 0.3℃. A distribution map of the attention intensity along the tilt direction corresponding to the location in the time-space is constructed. This construction process involves creating a two-dimensional coordinate system with transportation time as the X-axis and the spatial location within the vehicle (e.g., the distance from the front to the rear) as the Y-axis. Each "high-concern" event is plotted as a point on the map. The position of the point is determined by its occurrence time and physical location. The color or size of the point represents the magnitude of the surface change intensity. Furthermore, if a "high-concern" event occurs multiple times consecutively at a location, a line is drawn connecting the points. The tilt direction reflects the propagation trend of the risk in time and space, resulting in a change-attention tilt distribution map.

[0089] Please see Figure 6 The risk status output module includes a focus path identification submodule, a path segment statistics submodule, and a risk trend classification submodule.

[0090] The path identification submodule focuses on the fluctuating paths marked as objects of interest in the change-focused tilt distribution map, extracts the time period, number, and sequence of transportation segments involved for each path, records the start and end time positions of the path in each segment, organizes and constructs a path time span table according to the path number, and generates a path segment distribution table.

[0091] The system calls upon the fluctuation paths marked as items of interest in the change-focused tilt distribution map, that is, it identifies all the paths represented by the points and lines in the map, extracts the time period, number, and the sequence of transportation segments involved for each path. For example, a marked path, numbered P001, occurs during the time period [1663907133, 1663907150], and its physical location is point A (middle section of the shelf). The corresponding transportation segment is "city segment B". The system records the start and end times of the path in each segment. If the P001 path occurs entirely within "city segment B", then its start and end times in that segment are recorded. If it crosses two segments, then they are recorded separately. The system is then organized and constructed into a path time span table according to the path number.

[0092] Table 3: Distribution of Route Segments

[0093] Path number transport section Start timestamp End timestamp P001 Urban Section B 1663907133 1663907150 P002 Highway Section A 1663912400 1663912460 P002 Mountain section C 1663925010 1663925090

[0094] As shown in Table 3, this table records in detail the specific occurrence time of each high-concern route in different transportation geographical segments, and a route segment distribution table is generated based on this table.

[0095] The route segment statistics submodule extracts the frequency of occurrence, time span ratio, and number of distribution areas of each route in different segments based on the distribution of each route in the continuous transportation segment in the route segment distribution table. The routes are then filtered according to the occurrence frequency threshold and the distribution range threshold to obtain a set of covered distribution routes.

[0096] Based on the distribution of each path within a continuous transportation segment in the path segment distribution table, the frequency of occurrence, time span ratio, and number of distribution areas for each path in different segments are extracted. For path P002 in Table 3, its frequency of occurrence is 2 times, and the number of distribution areas is 2 (highway segment A and mountain segment C). The time span ratio is calculated as the ratio of the total duration of the path to the total time of the segment. Assuming the total duration of highway segment A is 3600 seconds, the time span ratio of P002 in this segment is (1663912460-1663912400) / 3600=60 / 3600≈0.0167. Paths are filtered based on the frequency of occurrence threshold and the distribution range threshold. Here, the frequency of occurrence threshold is set. Distribution range threshold These two thresholds were set based on a review of data from 200 historical transportation missions. Statistical analysis revealed that risk events leading to bulk cargo losses all exhibited warning paths at least twice in number and affected at least two different physical or geographical areas. Therefore, this combination was selected as the screening criterion. When the frequency of a path is greater than or equal to... And the number of distribution areas is greater than or equal to When the path is selected, it is retained. In the example, path P002 has a frequency of 2 and a distribution area of ​​2, which meets the condition and is therefore selected. P001 is removed because it only appears once. All paths that meet the condition are summarized to obtain the covered distribution path set.

[0097] The risk trend classification submodule calls the number, time span and fluctuation direction of the concentrated path in each segment of the coverage distribution path, extracts similar path attributes based on time continuity and regional consistency, merges and classifies them according to the consistency of time series between paths and the number of segment intersections, and generates risk perception identification results.

[0098] The process involves retrieving the path number, time span, and fluctuation direction from the distributed path set across different segments. For example, path P002 appears in both highway segment A and mountain segment C, with a positive fluctuation direction (determined by the synchronous trend change structure) (increasing temperature and humidity). Based on temporal continuity and regional consistency, similar paths are extracted. This process considers the two occurrences of P002 in different segments as events with similar attributes. Paths are then merged and categorized according to their time series consistency and the number of segment intersections. Time series consistency is calculated using Dynamic Time Warping (DTW) distance. The standardized temperature sequences at the two occurrences of P002 are calculated, yielding a DTW distance of 0.15. A consistency threshold is then set. The threshold is determined by calculating the DTW distance for 50 pairs of known associated disturbance events and 50 pairs of unassociated events, and selecting the ROC curve inflection point value that best distinguishes the two types of events. Since 0.15 < 0.3, the time series patterns of the two occurrences are considered to be highly consistent. The number of segment intersections is the total number of segments affected by the path, which is 2 here. Paths with similar attributes (same fluctuation direction, DTW distance less than the threshold) and repetitive in time and space are merged and classified into one type of risk trend. For example, the two occurrences of P002 are classified as "continuous warming and humidification risk", and the type of risk, the scope of influence (highway segment A, mountain segment C) and the occurrence time series are output to generate risk perception and identification results.

[0099] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A risk perception and identification system for key indicators of a cold chain system, characterized in that, The system includes: The environmental disturbance capture module acquires temperature and humidity data from the middle section of the refrigerated truck shelf, the air duct opening, and the rear stacking area. It extracts temperature segment difference nodes, filters humidity curve turning points, extracts wind speed jump trajectories, compares the fluctuation amplitude in each type of path, summarizes the fluctuation range, and generates a multi-point disturbance feature set. The indicator trend screening module constructs wind speed paths and matches their changing directions according to the temperature and humidity paths in the multi-point disturbance feature set, based on the same-direction wind speed paths within the period, extracts path combinations with consistent directions within the same time period, and screens out data that meet the synchronization of each changing direction to obtain the synchronous trend change structure. The boundary drift identification module constructs a temperature and humidity change ratio trajectory based on the synchronous trend change structure, calls the delay path of the cargo stacking area, determines whether the ratio crosses the boundary within the original boundary, identifies the offset segment and marks the boundary adjustment position, and outputs the drift trigger boundary structure. The attention level distribution module constructs an overlap mapping map by calling surface response data based on the offset path in the drift trigger boundary structure, classifies the level labels according to the overlap between temperature and humidity fluctuations and the direction of surface reaction, and generates a change attention tilt distribution map.

2. The risk perception and identification system for key indicators of cold chain systems according to claim 1, characterized in that: The set of multi-point disturbance features includes temperature range node distribution, humidity fluctuation gradient features, and wind speed frequency jump segments. The synchronous trend change structure includes path direction consistent segments, time overlap combination relationships, and three types of synchronous trend patterns. The drift trigger boundary structure includes fluctuation amplitude ratio intervals, boundary offset time periods, and response path drift features. The change attention tilt distribution map includes trajectory overlap continuous segments, direction consistency levels, and high-level fluctuation path distribution.

3. The risk perception and identification system for key indicators of a cold chain system according to claim 1, characterized in that, The environmental disturbance capture module includes a multi-point data acquisition submodule, a sequence variation extraction submodule, and a fluctuation path integration submodule. The multi-point data acquisition submodule acquires time-series data of temperature, humidity and wind speed in the middle section of the shelf inside the refrigerated truck, the location of the air duct, and the rear stacking coverage area. It records the time label, numerical output and data density of each type of data at each acquisition point, performs sampling interval consistency judgment on the acquired data, removes abnormal null values ​​and fills in missing segments, and generates standardized continuous data groups. The sequence variation extraction submodule constructs the difference trajectory of each segment based on the temperature sequence in the standardized continuous data group, filters the variation node with the largest difference and extracts the continuous segment, locates the turning point of the humidity sequence and filters out the continuous change time period, calculates the gradient intensity in the change segment, extracts the jump segment of the wind speed data with a sliding window and counts the frequency of occurrence, and obtains the set of sub-item fluctuation trajectories. The fluctuation path integration submodule, based on the difference segments, turning paths and jump trajectories extracted from the sub-fluctuation trajectory set, calls the duration range of the corresponding time period in each sequence, filters out the overlapping intervals that exhibit changing behavior in both temperature, humidity and wind, marks the interval as the fluctuation concentration segment, and calculates the joint duration span of various changes to generate a multi-point disturbance feature set.

4. The risk perception and identification system for key indicators of a cold chain system according to claim 1, characterized in that, The indicator trend screening module includes a direction consistency identification submodule, a wind speed trend overlay submodule, and a time segment screening submodule. The direction consistency identification submodule obtains the marked temperature path and humidity path in the multi-point disturbance feature set, detects the relative position of the trend direction within each period based on the change direction sequence of the two paths, performs path pairing for temperature and humidity combinations with consistent direction within the time segment, and filters the combinations according to the trend consistency ratio to generate temperature and humidity trend matching groups. The wind speed trend overlay submodule calls the combined path in the temperature and humidity trend matching group, collects the wind speed path change sequence within the corresponding time period, performs time period registration based on the fluctuation direction of the wind speed path and the direction sequence of the temperature and humidity combination, extracts the wind speed path whose fluctuation direction and time range are consistent with the temperature and humidity combination, and obtains a three-directional consistent structure. The time segment filtering submodule determines whether the three types of paths have completely overlapping time segments in a continuous period based on the time labels of each path in the three-directional consistent structure. It performs segment consistency judgment on the duration and start and end times of the overlapping segments, marks the combined paths that satisfy the full synchronization trend and time consistency, and generates a synchronization trend change structure.

5. The risk perception and identification system for key indicators of a cold chain system according to claim 1, characterized in that, The boundary drift recognition module includes an amplitude ratio construction submodule, a boundary cross-boundary judgment submodule, and a drift segment extraction submodule; The amplitude ratio construction submodule obtains the maximum and minimum values ​​of each path within the same time period based on the temperature and humidity combination path contained in the synchronous trend change structure, constructs the fluctuation amplitude ratio based on the numerical difference and the reference duration, arranges them according to the corresponding time period to generate a spectrum sequence, and generates a time period amplitude ratio spectrum. The boundary crossing judgment submodule calls each ratio segment in the time period amplitude ratio map, locates the upper and lower limit segments of the boundary corresponding to each time period based on the response delay path recorded by each monitoring point in the cargo stacking area, and judges whether each ratio segment crosses the upper and lower boundary segments to obtain the boundary crossing interval segment. The drift segment extraction submodule extracts the path offset trend under the corresponding time slice based on the marked cross-boundary time period in the boundary cross-boundary segment, identifies the start and end range of the original boundary position change within the time period, compares the difference between the upper and lower limit paths before and after the change and marks the offset, and generates the drift trigger boundary structure.

6. The risk perception and identification system for key indicators of a cold chain system according to claim 1, characterized in that, The attention level distribution module includes a trajectory overlap extraction submodule, a time direction label division submodule, and an attention level confirmation submodule; The trajectory overlap extraction submodule, based on the temperature and humidity combination path marked as offset in the drift trigger boundary structure, calls the cargo surface response change data within the corresponding time period, obtains the overlapping sections of the two types of paths and matches their respective start and end times, constructs a persistence mapping map of the overlapping area of ​​each section based on the intersection of time periods, and generates a response overlap persistence map. The time direction labeling submodule calculates the directional consistency ratio of the two paths in the overlapping area based on the time intersection length of each overlapping segment in the response overlap duration map. It then compares the time length and directional ratio with a unified interval threshold, sets the corresponding label level according to the ratio level, and generates the overlapping label level division result. The attention level confirmation submodule calls the high-level label path information in the overlapping label level division result, marks and binds this part of the path with the original temperature and humidity combination sequence, and extracts the offset frequency value and surface change intensity value of each bound path to construct the attention distribution map of its corresponding position tilt direction in the time space, thus obtaining the change attention tilt distribution map.

7. The risk perception and identification system for key indicators of a cold chain system according to claim 1, characterized in that, The system also includes: The risk status output module calls the path of interest in the change attention tilt distribution map, counts the frequency and distribution of the path in the segment, filters the path with long change duration and large fluctuation area coverage, and outputs the risk perception and identification results in the cold chain transportation process. The risk perception and identification results include a set of high-frequency fluctuation paths, the distribution of continuous segments, and statistics on the coverage area.

8. The risk perception and identification system for key indicators of a cold chain system according to claim 7, characterized in that, The risk status output module includes a focus path identification submodule, a path segment statistics submodule, and a risk trend classification submodule. The path identification submodule calls the fluctuating paths marked as objects of interest in the change attention tilt distribution map, extracts the time period, number and the sequence of transportation segments involved for each path, records the start and end time positions of the path in each segment, organizes and constructs a path time span table according to the path number, and generates a path segment distribution table. The route segment statistics submodule extracts the frequency of occurrence, time span ratio, and number of distribution areas of each route in different segments based on the distribution of each route in the continuous transportation segment in the route segment distribution table. The routes are then filtered according to the occurrence frequency threshold and the distribution range threshold to obtain a set of covered distribution routes. The risk trend classification submodule calls the number, time span, and fluctuation direction of the paths in each segment of the coverage distribution path set, extracts similar paths based on time continuity and regional consistency, merges and classifies them according to the consistency of time series between paths and the number of segment intersections, and generates risk perception identification results.

Images