Whole-course tracking system and method for cold-chain logistics

Abstract

The invention relates to the technical field of data processing, in particular to a cold-chain logistics whole-course tracking system and method. Obtaining a fluctuation point according to the discrete characteristic of the temperature time sequence, and obtaining a noise point according to the amplitude characteristic of the fluctuation point and the data difference characteristic of the fluctuation point and the adjacent data point; taking a vacancy moment and a moment of a noise point in the temperature time sequence as a target moment; taking the time period from the target moment to the historical nearest fluctuation point as a first time period, and taking the time period from the target moment to the future nearest fluctuation point as a second time period; when the temperature change trends of the first time period and the second time period are opposite, the correction degree is obtained according to the length features and the data difference features of the first time period and the second time period; and obtaining a correction temperature according to the data change characteristics and the correction degree of the moments before and after the target moment. A correction temperature time sequence is obtained according to all correction temperatures; monitoring is carried out according to the temperature correction time sequence, and the accuracy of temperature correction and temperature monitoring is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a whole-process tracking system and method for cold chain logistics. Background Art

[0002] Cold chain logistics refers to a logistics system that always maintains a specific low-temperature environment throughout the whole process of product production, storage, transportation and distribution. In order to ensure the quality of products, it is necessary to monitor data such as temperature and humidity inside the carriage during transportation and remotely transmit it to the cold chain data center, so that relevant personnel can understand the temperature status and make adjustments. Due to the expansion of the coverage of the cold chain transportation industry, more and more remote areas and complex environments are involved, and the stability of data transmission cannot be determined; especially during long-distance transportation, problems such as short-term signal interruption, signal loss and signal noise may be encountered, resulting in the loss and inaccuracy of data at some moments, and it is necessary to supplement the missing and abnormal data for temperature monitoring.

[0003] Currently, linear interpolation method is usually used to interpolate abnormal or missing data, and the average value of adjacent data points is taken as the replacement value. When there are abnormal temperature fluctuations inside the carriage, it is necessary to focus on whether there is a situation such as temperature runaway, etc. However, linear interpolation is easy to smooth the part of abnormal temperature fluctuations and difficult to reflect the real temperature changes, resulting in a low accuracy of the interpolation result and affecting the accuracy of temperature monitoring. Summary of the Invention

[0004] In order to solve the technical problem that supplementing missing and abnormal temperature data by the linear interpolation method is likely to lead to a low accuracy of the interpolation result and affect the accuracy of temperature monitoring, the purpose of the present invention is to provide a whole-process tracking system and method for cold chain logistics, and the specific technical solutions adopted are as follows: Obtain the temperature time series transmitted from the logistics vehicle to the cold chain center; Obtain the fluctuation points according to the data discrete characteristics of the temperature time series; obtain the noise level according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and adjacent data points; obtain the noise points according to the noise level of the fluctuation points; perform data replacement on the noise points and repeat to obtain new noise points; take the vacant moments and the moments of all noise points in the temperature time series as the target moments; Take the temperature data from the target moment to the nearest historical fluctuation point as the first period, and take the temperature data from the target moment to the nearest future fluctuation point as the second period; when the temperature change trends of the first period and the second period are opposite, obtain the correction degree of the target moment according to the length characteristics of the first period and the second period and the data difference characteristics of adjacent moments; obtain the corrected temperature according to the data change characteristics of the moments before and after the target moment and the correction degree; when the temperature change trends of the first period and the second period are the same, obtain the corrected temperature according to the data characteristics of the moments before and after the target moment; Obtain the corrected temperature time series according to the corrected temperatures of all target moments; perform monitoring according to the corrected temperature time series.

[0005] Further, the step of obtaining the fluctuation points according to the data discrete characteristics of the temperature time series includes: Obtain the peak points and valley points of the temperature time series through the AMPD peak detection algorithm; take all the peak points and all the valley points as the fluctuation points.

[0006] Further, the step of obtaining the noise degree according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and adjacent data points includes: Calculate the absolute value of the difference between the fluctuation point and the previous fluctuation point to obtain the first fluctuation value; calculate the absolute value of the difference between the fluctuation point and the next fluctuation point to obtain the second fluctuation value; calculate the average value of the first fluctuation value and the second fluctuation value and normalize it to obtain the discrete degree value; calculate the absolute value of the difference between the fluctuation point and the previous adjacent data point to obtain the first adjacent difference; calculate the ratio of the first adjacent difference to the first fluctuation value to obtain the first mutation degree; calculate the absolute value of the difference between the fluctuation point and the next adjacent data point to obtain the second adjacent difference; calculate the ratio of the second adjacent difference to the second fluctuation value to obtain the second mutation degree; calculate the sum value of the first mutation degree and the second mutation degree and normalize it to obtain the local change value; calculate the average value of the discrete degree value and the local change value to obtain the noise degree of the fluctuation point.

[0007] Further, the step of obtaining the noise points according to the noise degree of the fluctuation points includes: When the noise degree exceeds the preset threshold, the fluctuation point is a noise point.

[0008] Further, the step of replacing the data of the noise points and repeating to obtain new noise points includes: Calculate the average value of the non-fluctuating points in the temperature sequence to obtain the average temperature; replace the values of the noise points with the average temperature and obtain a new temperature time series; re-obtain the new noise points in the new temperature time series until the acquisition of the noise points stops after repeating a preset number of times, and obtain all the noise points in the temperature time series.

[0009] Further, the step of obtaining the correction degree of the target moment according to the length characteristics of the first period and the second period and the data difference characteristics of adjacent moments includes: , where R represents the correction degree of the target moment, S represents the total data length of the first period and the second period, represents normalization, represents the trend length characteristic value, N represents the number of data points in the first period, represents the value of the nth data point in the first period starting from the target moment, represents the value of the data point, represents the distance weight, represents the adjacent change value, represents the weighted change value, represents the comprehensive change value of the first period, M represents the number of data points in the second period, represents the value of the mth data point in the second period starting from the target moment, represents the value of the data point, represents the distance weight, represents the adjacent change value, represents the weighted change value, represents the comprehensive change value of the second period, represents the overall change value.

[0010] Further, the step of obtaining the corrected temperature according to the data change characteristics of the moments before and after the target moment and the correction degree includes: Calculate the average value of the adjacent data points before and after the target moment to obtain the local value; calculate the difference between the adjacent data point before the target moment and the previous data point to obtain the first local change value; calculate the difference between the adjacent data point after the target moment and the next data point to obtain the second local change value; calculate the average value of the first local change value and the second local change value to obtain the local change characteristic value; calculate the sum value of the constant 1 and the correction degree to obtain the correction coefficient; calculate the product of the correction coefficient and the local change characteristic value to obtain the correction amount; calculate the sum value of the correction amount and the local value to obtain the corrected temperature of the target moment.

[0011] Further, when the temperature change trends in the first time period and the second time period are the same, the step of obtaining the corrected temperature according to the data characteristics of the moments before and after the target moment includes: Calculating the average temperature of the adjacent data points before and after the target moment to obtain the corrected temperature of the target moment.

[0012] Further, the step of obtaining the corrected temperature time series according to the corrected temperatures of all target moments includes: Replacing the temperature data of the target moment in the temperature time series with the corresponding corrected temperature to obtain the corrected temperature time series.

[0013] The present invention also provides a whole-process tracking system for cold-chain logistics, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps of any one of the whole-process tracking methods for cold-chain logistics.

[0014] The present invention has the following beneficial effects: In the present invention, since the noise in the acquisition process is relatively discrete, obtaining the fluctuation points can initially determine the range of the noise points. Since the values of the noise points are significantly different from the normal temperature and the temperature change characteristics with adjacent data points are relatively obvious, obtaining the noise degree can characterize the possibility that the fluctuation points are noise points; the target moment represents the moment in the temperature time series where temperature correction is required. Obtaining the first time period and the second time period is used to analyze the temperature change characteristics before and after the target moment; since the correction methods for the target moment under different temperature change trends are different, analyzing whether the temperature change trends in the first time period and the second time period are the same can initially determine the temperature correction method. Obtaining the correction degree can determine the magnitude of the temperature correction according to the temperature change degree near the target moment, improve the correction accuracy, and avoid the situation of weakening the temperature fluctuation characteristics of the target moment. Obtaining the corrected temperature according to the data change characteristics and the correction degree of the moments before and after the target moment makes the corrected temperature closer to the true temperature of the logistics carriage at the target moment, thereby improving the accuracy of temperature monitoring. Obtaining the corrected temperature time series can replace and correct the temperature data of the missing and noise points in the temperature time series, and finally monitor according to the corrected temperature time series, improving the accuracy of the cold-chain center's temperature monitoring of the logistics carriage. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 The flowchart of a full - process tracking method for cold - chain logistics provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific implementation manners, structures, features and effects of a full - process tracking system and method for cold - chain logistics proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solutions of a full - process tracking system and method for cold - chain logistics provided by the present invention with reference to the accompanying drawings.

[0020] Please refer to Figure 1 , which shows the flowchart of a full - process tracking method for cold - chain logistics provided by an embodiment of the present invention. The method includes the following steps: Step S1: Obtain the temperature time series transmitted from the logistics vehicle to the cold - chain center.

[0021] In the embodiment of the present invention, the implementation scenario is to replace the missing and abnormal data transmitted from the logistics vehicle to the cold - chain center to improve the accuracy of cold - chain logistics temperature monitoring. First, obtain the temperature time series transmitted from the logistics vehicle to the cold - chain center. In the embodiment of the present invention, the temperature data inside the carriage is collected every 5 seconds and transmitted to the cold - chain center. In order to accurately monitor the temperature of the transported products in the recent period, the cold - chain center processes the temperature time series within the current 20 - minute period every 20 minutes. The embodiment of the present invention performs data analysis based on temperature characteristics, and the implementer can determine the collection object, collection frequency and data duration according to the implementation scenario.

[0022] Step S2: Obtain the fluctuation points according to the data discrete characteristics of the temperature time series; obtain the noise degree according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and adjacent data points; obtain the noise points according to the noise degree of the fluctuation points; perform data replacement on the noise points and repeat to obtain new noise points; use the vacant moments and the moments of all noise points in the temperature time series as the target moments.

[0023] Since noise is likely to occur during the acquisition of the temperature inside the carriage, and data loss may occur at individual moments due to poor signal during transmission; such noise data and lost data will both result in low reliability and integrity of the temperature time series. Therefore, it is necessary to replace and supplement such data. If the temperature data at a certain moment is lost, the value at that moment will not be displayed in the temperature time series, and such moments are regarded as vacant moments. For the noise data in the temperature time series, since the temperature change inside the carriage is relatively gentle and there will be no obvious fluctuations, while the randomness of the noise data is relatively strong, so the noise data are mostly data points with obvious fluctuations. Therefore, the fluctuation points are obtained according to the data dispersion characteristics of the temperature time series; preferably, in the embodiment of the present invention, the steps of obtaining the fluctuation points include: obtaining the peak points and valley points of the temperature time series through the AMPD peak detection algorithm; taking all the peak points and all the valley points as the fluctuation points; it should be noted that the AMPD peak detection algorithm belongs to the prior art, and the specific steps will not be elaborated. The fluctuation points are the data points with relatively obvious fluctuations in the temperature time series. However, under normal circumstances, the temperature will also change, such as changes in the refrigeration power and the opening and closing of the carriage door, and some temperature inflection points will also be identified as fluctuation points. However, such normal temperature changes are slow, while the data of the noise points usually show mutation characteristics; therefore, the noise level can be obtained according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and the adjacent data points.

[0024] Preferably, in the embodiments of the present invention, the steps of obtaining the noise level include: calculating the absolute value of the difference between a fluctuation point and the previous fluctuation point to obtain a first fluctuation value; calculating the absolute value of the difference between the fluctuation point and the next fluctuation point to obtain a second fluctuation value; since adjacent fluctuation points are valley points and peak points, if the fluctuation point is a peak point, the adjacent fluctuation points before and after are valley points; when the fluctuation point is a noise point, the values are relatively discrete, and the first fluctuation value and the second fluctuation value are relatively large. Calculate the average value of the first fluctuation value and the second fluctuation value and normalize it to obtain a dispersion value; the larger the dispersion value, the more likely it means that the fluctuation point is a random noise point. Calculate the absolute value of the difference between the fluctuation point and the previous adjacent data point to obtain a first adjacent difference; the larger the first adjacent difference, the greater the temperature difference between the fluctuation point and the previous data point, the more it does not conform to the characteristic of smooth temperature change, and the more likely it is a noise point. Calculate the ratio of the first adjacent difference to the first fluctuation value to obtain a first mutation degree; the larger the first mutation degree, the more likely the fluctuation point is a noise point; when the fluctuation point is a normal fluctuation point, the temperature difference between the fluctuation point and the adjacent data points is small, conforming to the characteristic of smooth change, and the first mutation degree is small. Calculate the absolute value of the difference between the fluctuation point and the next adjacent data point to obtain a second adjacent difference; calculate the ratio of the second adjacent difference to the second fluctuation value to obtain a second mutation degree; similarly, when the fluctuation point is a noise data point, the greater the temperature difference from the next adjacent data point, the greater the second mutation degree. Calculate the sum value of the first mutation degree and the second mutation degree and normalize it to obtain a local change value; the larger the local change value, the more likely it means that the fluctuation point is a noise point. Calculate the average value of the dispersion value and the local change value to obtain the noise level of the fluctuation point; the greater the noise level, the more likely it means that the fluctuation point is a noise point. The formula for obtaining the noise level includes:

[0025] In the formula, W represents the noise level of the fluctuation point, represents the normalization function, represents the first fluctuation value, represents the second fluctuation value, represents the dispersion value, represents the first adjacent difference, represents the first mutation degree, represents the second adjacent difference, represents the second mutation degree, represents the local change value.

[0026] Further, noise points can be obtained according to the noise level of the fluctuation points; preferably, in the embodiments of the present invention, the steps of obtaining noise points include: when the noise level exceeds a preset threshold, the fluctuation points are noise points. After calculating the noise levels of all the fluctuation points, since linear normalization will place the noise levels of normal fluctuation points and noise fluctuation points near 0 and 1 respectively, the noise level of normal fluctuation points is close to 0, while the noise level of noise points is close to 1. Therefore, in the embodiments of the present invention, the preset threshold is taken as 0.5, and the implementer can determine it according to the implementation scenario. Since there may be a situation of continuous noise points in the temperature time series, for example, the adjacent data points of a certain peak point are noise points. Although there may also be a large difference in the values between the noise points, these adjacent data points may not be detected as fluctuation points. Therefore, in order to improve the accuracy of subsequent data processing, it is necessary to replace the data of the noise points and repeat to obtain new noise points.

[0027] Preferably, in the embodiments of the present invention, the steps of obtaining new noise points include: calculating the average value of the non-fluctuation points in the temperature sequence to obtain the average temperature; the average temperature characterizes the temperature level under normal conditions. Replace the value of the noise point with the average temperature and obtain a new temperature time series. The noise points obtained in one acquisition will be removed from this new temperature time series. Then, new noise points can be re-obtained in the new temperature time series through the same steps until the acquisition of noise points stops after repeating a preset number of times, and all the noise points in the temperature time series are obtained; since the situation of continuous noise points is less, and the number of continuous noise points is small. Therefore, in the embodiments of the present invention, the preset number of times is taken as 3 times, that is, after repeating the steps of obtaining noise points 3 rounds, all the noise points are obtained. Further, the vacant moments in the temperature time series and the moments of all the noise points can be used as target moments, and the data at the target moments needs to be replaced, so as to improve the accuracy of temperature monitoring in the logistics carriage.

[0028] Step S3, take the temperature data from the target moment to the nearest historical fluctuation point as the first period, and take the temperature data from the target moment to the nearest future fluctuation point as the second period; when the temperature change trends of the first period and the second period are opposite, obtain the correction degree of the target moment according to the length characteristics of the first period and the second period and the data difference characteristics of adjacent moments; obtain the corrected temperature according to the data change characteristics of the moments before and after the target moment and the correction degree; when the temperature change trends of the first period and the second period are the same, obtain the corrected temperature according to the data characteristics of the moments before and after the target moment.

[0029] When the temperature in the logistics carriage is relatively stable or shows a fixed changing trend, data filling can be performed through the existing interpolation method. The average value of the adjacent data points before and after the target moment is used as the replacement value, and the replacement value at this time has high accuracy. However, when the carriage door is opened, external hot air enters the carriage, and the temperature time series will show short-term fluctuating changes; or when the refrigeration equipment in the carriage runs unstably, it will also cause the temperature to show fluctuating changes. If the target moment occurs during the temperature fluctuating changes, the accuracy of the replacement value obtained through the existing interpolation method is relatively low, which is likely to blur the specific temperature change details and even miss the temperature anomaly data, resulting in the inability to adjust the temperature in a timely manner and affecting the monitoring accuracy. Therefore, in order to improve the temperature monitoring accuracy, it is necessary to analyze the temperature change situation at the target moment; when the temperature change trends before and after the target moment are opposite, it means that the temperature has fluctuated, and it is necessary to further analyze the temperature fluctuation characteristics near the target moment to improve the accuracy of temperature correction; it should be noted that due to the possible occurrence of continuous moment missing data or continuous noise points, continuous target moments may occur; however, in this scenario, the temperature fluctuation characteristics are not drastic, and the acquisition moments are close, so the continuous target moments can be regarded as one target moment for temperature correction. In the embodiments of the present invention, when the number of continuous target moments does not exceed 3, the continuous target moments are regarded as one target moment for analysis. When it exceeds, it means that there is a failure in the temperature acquisition or transmission during this period, and no temperature correction will be performed, and the temperature data during this period will be left blank. First, the temperature data from the target moment to the nearest historical fluctuation point is used as the first period, and the temperature data from the target moment to the nearest future fluctuation point is used as the second period; the temperature change trends within the first period are close, and the temperature change trends within the second period are close; when the temperature change trends of the first period and the second period are opposite, for example, the temperature in the first period shows an upward trend and the temperature in the second period shows a downward trend, it means that the temperature at this target moment has fluctuated, and it is necessary to further analyze the correction degree. In the embodiments of the present invention, the data of the first period and the second period are linearly fitted. When the positive and negative of the slopes of the two fitted straight lines are different, it means that the temperature change trends of the first period and the second period are opposite. Furthermore, the correction degree of the target moment can be obtained according to the length characteristics of the first period and the second period and the data difference characteristics of adjacent moments; preferably, in the embodiments of the present invention, the steps of obtaining the correction degree include:

[0030] , where R represents the correction degree at the target moment, and S represents the total data length of the first period and the second period. When the carriage door is opened, the outside hot air contacts the cold air inside the carriage, forming heat conduction into the carriage. As the opening time increases, the contact proportion of the outside hot air inside the carriage becomes larger, and at this time, the temperature change degree will increase with the passage of time. At the same time, when the refrigeration equipment operates unstably or fails, with the passage of time and the influence of the outside temperature, the temperature change degree will also gradually increase. represents normalization, represents the trend length eigenvalue. Therefore, when the trend length eigenvalue is larger, the temperature change duration is longer, the change degree is larger, and thus the temperature correction degree is larger. N represents the number of data points in the first period, represents the distance weight, represents the value of the nth data point in the first period starting from the target moment; represents the value of the data point, that is, the value of the data point before the nth data point; represents the adjacent change value, represents the weighted change value. When the distance from the target moment is closer and n is smaller, the distance weight is larger, and the adjacent change value is more important. represents the comprehensive change value of the first period. When the comprehensive change value of the first period is larger, it means that the cumulative temperature change degree within the first period is larger, and the temperature decrease or increase trend from the first period to this target moment is larger, then the temperature correction degree is larger. M represents the number of data points in the second period, represents the value of the mth data point in the second period starting from the target moment; represents the value of the data point, that is, the value of the data point after the mth data point; represents the distance weight, represents the adjacent change value. When the distance from the target moment is farther and m is larger, the distance weight is smaller. represents the weighted change value. The adjacent change value closer to the target moment is more important. represents the comprehensive change value of the second period. Similarly, when the comprehensive change value of the second period is larger, it means that the cumulative temperature change degree within the second period is larger, and the temperature decrease or increase trend from the target moment to the second period is larger. represents the overall change value; when the overall change value is larger, it means that the temperature change degree from this target moment to the previous and subsequent periods is larger, the temperature increase or decrease degree is larger, and the temperature correction degree is larger.

[0031] Further, after obtaining the correction degree at the target moment, the corrected temperature can be obtained according to the data change characteristics and the correction degree of the moments before and after the target moment; preferably, in the embodiments of the present invention, the steps of obtaining the corrected temperature include: calculating the average value of the adjacent data points before and after the target moment to obtain a local value; the closer the adjacent data points before and after are to the target moment, and thus the local value is closest to the data at the target moment, and the local value can be used as the base value when correcting the temperature at the target moment. Calculating the difference between the adjacent data point before the target moment and the previous data point to obtain a first local change value; calculating the difference between the adjacent data point after the target moment and the next data point to obtain a second local change value; the first local change value and the second local change value reflect the temperature change degree at the adjacent data points before and after. Calculating the average value of the first local change value and the second local change value to obtain a local change characteristic value; when the local change characteristic value is larger, it means that the temperature change degree at the adjacent data points before and after the target moment is larger. Since the acquisition moments are close and the temperature change is regular, the temperature change degree between the adjacent data points before and after and the target moment can be analyzed based on this local change characteristic value. Calculating the sum value of the constant 1 and the correction degree to obtain a correction coefficient; when the correction degree is larger, it means that the temperature change degree from the adjacent data points before and after to the target moment is larger, and thus it is necessary to increase the temperature change degree on the basis of the local change characteristic value to reflect the more obvious temperature change characteristic at the target moment. Calculating the product of the correction coefficient and the local change characteristic value to obtain a correction amount; calculating the sum value of the correction amount and the local value to obtain the corrected temperature at the target moment; the corrected temperature can accurately correct the temperature value at the target moment based on the temperature change degree characteristics near the target moment, avoiding the situation where the existing interpolation method smooths the temperature characteristics at the target moment and weakens the fluctuation characteristics, and improving the accuracy of temperature monitoring. The formula for obtaining the corrected temperature includes:

[0032] In the formula, T represents the corrected temperature, K represents the local value, F represents the local change characteristic value, R represents the correction degree, represents the correction coefficient, represents the correction amount.

[0033] Further, when the temperature change trends in the first period and the second period are the same, the corrected temperature is obtained according to the data characteristics of the moments before and after the target moment; specifically, it includes: calculating the temperature average value of the adjacent data points before and after the target moment to obtain the corrected temperature at the target moment. Since the temperature change trends before and after the target moment are the same at this time and the temperature change is regular in this scenario, it is highly accurate to represent the temperature at the target moment according to the temperature average value of the adjacent data points before and after.

[0034] Step S4: Obtain a corrected temperature time series based on the corrected temperatures at all target times; perform monitoring according to the corrected temperature time series.

[0035] After obtaining the corrected temperatures at all target times, a corrected temperature time series can be obtained based on the corrected temperatures at all target times. Replace the temperature data at the target times in the temperature time series with the corresponding corrected temperatures to obtain the corrected temperature time series. Furthermore, monitoring can be performed according to the corrected temperature time series; the corrected temperature time series can be accurately corrected according to the temperature change characteristics at the target times, so that the corrected temperature time series can more accurately represent the true temperature change characteristics inside the logistics carriage, improving the accuracy of temperature monitoring of the logistics carriage in the cold chain center.

[0036] In summary, the embodiment of the present invention provides a method for tracking the whole process of cold chain logistics; obtaining fluctuation points according to the data discrete characteristics of the temperature time series, and obtaining noise points according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and adjacent data points; taking the vacant times in the temperature time series and the times of all noise points as target times; taking the period from the target time to the nearest historical fluctuation point as the first period, and taking the period from the target time to the nearest future fluctuation point as the second period; when the temperature change trends of the first period and the second period are opposite, obtaining the correction degree according to the length characteristics and data difference characteristics of the first period and the second period; obtaining the corrected temperature according to the data change characteristics and the correction degree of the times before and after the target time. The present invention obtains a corrected temperature time series based on all corrected temperatures; performs monitoring according to the corrected temperature time series, improving the accuracy of temperature correction and temperature monitoring.

[0037] The present invention also proposes a system for tracking the whole process of cold chain logistics, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps of any one of the methods for tracking the whole process of cold chain logistics.

[0038] It should be noted that: the above sequence of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0039] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A cold chain logistics tracking method, characterized in that: The method comprises the following steps: Obtain the temperature time series transmitted from logistics vehicles to the cold chain center; Obtain fluctuation points according to the discrete characteristics of the temperature time series data; obtain the noise degree according to the amplitude characteristics of the fluctuation points and the data difference characteristics between the fluctuation points and the adjacent data points; obtain noise points according to the noise degree of the fluctuation points; replace the data of the noise points and repeatedly obtain new noise points; take the vacant moments in the temperature time series and the moments of all noise points as target moments; The temperature data from the target moment to the most recent historical fluctuation point is taken as the first period, and the temperature data from the target moment to the most recent future fluctuation point is taken as the second period; when the temperature change trends of the first period and the second period are opposite, the correction degree of the target moment is obtained according to the length characteristics of the first period and the second period and the data difference characteristics of adjacent moments; the corrected temperature is obtained according to the data change characteristics of the moments before and after the target moment and the correction degree; when the temperature change trends of the first period and the second period are the same, the corrected temperature is obtained according to the data characteristics of the moments before and after the target moment; Obtain a corrected temperature time sequence according to the corrected temperatures at all target moments; and perform monitoring according to the corrected temperature time sequence.

2. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining the fluctuation point according to the discrete characteristics of the temperature time series data includes: The peak points and trough points of the temperature time series are obtained by using an AMPD peak detection algorithm; all the peak points and all the trough points are used as the fluctuation points.

3. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining the noise degree according to the amplitude characteristics of the fluctuation point and the data difference characteristics between the fluctuation point and the adjacent data points comprises: Calculate the absolute value of the difference between the fluctuation point and the previous fluctuation point to obtain the first fluctuation value; calculate the absolute value of the difference between the fluctuation point and the next fluctuation point to obtain the second fluctuation value; calculate the average value of the first fluctuation value and the second fluctuation value and normalize them to obtain a discreteness value; calculate the absolute value of the difference between the fluctuation point and the previous adjacent data point to obtain the first adjacent difference; calculate the ratio of the first adjacent difference to the first fluctuation value to obtain the first mutation degree; calculate the absolute value of the difference between the fluctuation point and the next adjacent data point to obtain the second adjacent difference; calculate the ratio of the second adjacent difference to the second fluctuation value to obtain the second mutation degree; calculate the sum of the first mutation degree and the second mutation degree and normalize them to obtain a local change value; calculate the average value of the discreteness value and the local change value to obtain the noise degree of the fluctuation point.

4. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining the noise point according to the noise degree of the fluctuation point comprises: When the noise level exceeds a preset threshold, the fluctuation point is a noise point.

5. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of replacing the noise points with data and repeatedly acquiring new noise points comprises: Calculate the average value of the non-fluctuating points in the temperature sequence to obtain the average temperature; replace the value of the noise point with the average temperature and obtain a new temperature time series; re-obtain new noise points in the new temperature time series until the acquisition of noise points is stopped after a preset number of repetitions, and obtain all noise points in the temperature time series.

6. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining the correction degree of the target time according to the length characteristics of the first time period and the second time period and the data difference characteristics of adjacent time periods comprises: , where R represents the correction degree of the target time, S represents the sum of the data lengths of the first and second time periods, represents normalization, represents the trend length characteristic value, N represents the number of data points in the first period, It represents the value of the nth data point in the first period starting from the target time. Indicates The value of the data point, represents the distance weight, represents the adjacent change value, represents the weighted change value, represents the comprehensive change value of the first period, M represents the number of data points in the second period, It represents the value of the mth data point in the second period starting from the target time. Indicates The value of the data point, represents the distance weight, represents the adjacent change value, represents the weighted change value, Represents the comprehensive change value of the second period, Indicates the overall change value.

7. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining the corrected temperature according to the data change characteristics before and after the target moment and the correction degree comprises: Calculate the average value of adjacent data points before and after the target moment to obtain a local numerical value; calculate the difference between the front adjacent data point and the previous data point at the target moment to obtain a first local change value; calculate the difference between the rear adjacent data point and the rear data point at the target moment to obtain a second local change value; calculate the average value of the first local change value and the second local change value to obtain a local change characteristic value; calculate the sum of constant 1 and the correction degree to obtain a correction coefficient; calculate the product of the correction coefficient and the local change characteristic value to obtain a correction amount; calculate the sum of the correction amount and the local numerical value to obtain the corrected temperature at the target moment.

8. A cold chain logistics full-process tracking method according to claim 1, characterized in that: When the temperature change trends of the first time period and the second time period are the same, the step of obtaining the corrected temperature according to the data features before and after the target moment comprises: The average temperature of adjacent data points before and after the target moment is calculated to obtain the corrected temperature of the target moment.

9. A cold chain logistics full-process tracking method according to claim 1, characterized in that: The step of obtaining a corrected temperature time sequence according to the corrected temperatures at all target moments comprises: The temperature data at the target time in the temperature time series is replaced with the corresponding corrected temperature to obtain the corrected temperature time series.

10. A cold chain logistics tracking system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.

Images