Cold chain monitoring and early warning system and method based on Internet of Things

Abstract

The invention relates to the technical field of cold chain monitoring, in particular to a cold chain monitoring and early warning system and method based on the Internet of Things, and the method comprises the steps: obtaining historical temperature time sequence data in a refrigerator car before the current moment, and mapping the historical temperature time sequence data into a two-dimensional coordinate system; according to the slope between two adjacent data points in the two-dimensional coordinate system, obtaining at least one temperature rise trend section, when the number of the temperature rise trend sections is greater than a preset number threshold value, obtaining at least one suspected abnormal temperature rise trend section, according to all the suspected abnormal temperature rise trend sections, obtaining the abnormal risk degree of the initial temperature rise trend, and obtaining the abnormal risk degree of the initial temperature rise trend; correcting the initial temperature rise trend abnormal risk degree to obtain a temperature rise trend abnormal risk degree; and the initial temperature threshold value is adjusted according to the abnormal risk degree of the temperature rise trend, the self-adaptive temperature threshold value at the current moment is obtained, monitoring and early warning are conducted on the temperature of the refrigerator car at the current moment, and temperature abnormal monitoring caused by sensor noise is avoided through the self-adaptive temperature threshold value.

Description

Technical Field

[0001] The present invention relates to the technical field of cold chain monitoring, and in particular, to a cold chain monitoring and warning system and method based on the Internet of Things. Background Art

[0002] In cold chain logistics, refrigerated trucks play a dominant role in the cold chain transportation link. Cold chain logistics is crucial for ensuring the quality of temperature-sensitive goods, making it crucial to monitor and warn the temperature inside the refrigerated truck in real time. With the development of technology, the cold chain logistics industry has gradually become digital and intelligent. Specifically, a monitoring platform is constructed through the Internet of Things, and the temperature data collected by the temperature sensors inside the refrigerated truck is uploaded to the cold chain logistics information platform in real time to achieve real-time monitoring and warning of cold chain temperature parameters.

[0003] In the prior art, the temperature inside the refrigerated truck is monitored and warned in real time through the threshold method. When the real-time temperature exceeds the set fixed temperature threshold, a temperature anomaly warning is issued. However, it is difficult for the fixed temperature threshold to achieve accurate monitoring and early warning at the same time. For example, when a relatively high temperature threshold is set, the quality of temperature-sensitive goods is easily damaged; when a relatively low temperature threshold is set, it is easily affected by the noise of the temperature sensor, resulting in false monitoring and false warning, which affects the transportation efficiency of cold chain logistics.

[0004] Therefore, how to ensure the accuracy of the temperature threshold to improve the timeliness of temperature anomaly warning based on the threshold method has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a cold chain monitoring and warning system and method based on the Internet of Things to solve the problem of how to ensure the accuracy of the temperature threshold to improve the timeliness of temperature anomaly warning based on the threshold method.

[0006] In a first aspect, an embodiment of the present invention provides a cold chain monitoring and warning method based on the Internet of Things. The method includes the following steps: Obtain the historical temperature time series data inside the refrigerated truck within a preset historical period before the current moment, and map the historical temperature time series data to a two-dimensional coordinate system, where the horizontal axis of the two-dimensional coordinate system is time and the vertical axis is temperature; According to the slope between two adjacent data points in the two-dimensional coordinate system, obtain the rising data points, connect the adjacent and continuous rising data points to obtain at least one rising trend segment. When the number of rising trend segments is greater than the preset number threshold, obtain the non-continuous distribution degree of each rising trend segment according to the interval distribution characteristics of the rising trend segments; At least one suspected abnormal temperature rising trend segment is obtained according to the discontinuous distribution degree of each temperature rising trend segment. An initial abnormal risk degree of the temperature rising trend is obtained according to the trend segment lengths and trend segment slopes of all suspected abnormal temperature rising trend segments. The initial abnormal risk degree of the temperature rising trend is corrected according to the similarity between all suspected abnormal temperature rising trend segments to obtain the abnormal risk degree of the temperature rising trend. The initial temperature threshold is adjusted according to the abnormal risk degree of the temperature rising trend to obtain the adaptive temperature threshold at the current moment. The temperature of the refrigerated truck at the current moment is monitored and warned using the adaptive temperature threshold.

[0007] Preferably, the obtaining of the temperature rising data points according to the slope between two adjacent data points in the two-dimensional coordinate system includes: For any data point in the two-dimensional coordinate system except the first data point, the previous data point of the any data point is obtained. According to the coordinates of the any data point and the coordinates of the previous data point, the slope between the any data point and the previous data point is obtained, which is denoted as the temperature change trend value of the any data point. If the temperature change trend value of the any data point is greater than 0, then the any data point is determined to be a temperature rising data point.

[0008] Preferably, the obtaining of the discontinuous distribution degree of each temperature rising trend segment according to the interval distribution characteristics of the temperature rising trend segment includes: The absolute value of the difference in abscissa between the last data point of the previous temperature rising trend segment and the first data point of the next temperature rising trend segment is denoted as the interval distance between two adjacent temperature rising trend segments. For any temperature rising trend segment, the interval distance between the any temperature rising trend segment and its previous temperature rising trend segment is obtained, which is denoted as the left interval distance. The interval distance between the any temperature rising trend segment and its next temperature rising trend segment is obtained, which is denoted as the right interval distance. The left interval distance and the right interval distance are accumulated to obtain the interval characteristic value of the any temperature rising trend segment. The interval characteristic values of each temperature rising trend segment are obtained. The absolute value of the difference between the interval characteristic value of the any temperature rising trend segment and the interval characteristic value of each other temperature rising trend segment is calculated to obtain the average value of the absolute values of the differences, which is denoted as the discontinuous distribution degree of the any temperature rising trend segment.

[0009] Preferably, the obtaining of at least one suspected abnormal temperature rising trend segment according to the discontinuous distribution degree of each temperature rising trend segment includes: The discontinuous distribution degrees of all temperature rising trend segments are linearly normalized to obtain the normalized value of each temperature rising trend segment. If the normalized value of any temperature rising trend segment is less than a preset normalized threshold, then the any temperature rising trend segment is determined to be a suspected abnormal temperature rising trend segment.

[0010] Preferably, obtaining the initial abnormal risk degree of the temperature rising trend according to the trend segment lengths and trend segment slopes of all suspected abnormal temperature rising trend segments includes: For any suspected abnormal temperature rising trend segment, perform linear fitting on the any suspected abnormal temperature rising trend segment to obtain a trend segment slope, and obtain a trend segment length according to the horizontal span and vertical span of the any suspected abnormal temperature rising trend segment. Take the product between the trend segment slope and the trend segment length as the temperature rising trend estimation value of the any suspected abnormal temperature rising trend segment; Obtain the temperature rising trend estimation values of each suspected abnormal temperature rising trend segment to obtain the average value of the temperature rising trend estimation values. According to the interval distances between every two adjacent suspected abnormal temperature rising trend segments among all suspected abnormal temperature rising trend segments, obtain the average interval distance. Use the exponential function with the natural constant as the base to perform negative mapping on the average interval distance to obtain the temperature rising frequency. Obtain the initial abnormal risk degree of the temperature rising trend according to the product between the temperature rising frequency and the average value of the temperature rising trend estimation values.

[0011] Preferably, correcting the initial abnormal risk degree of the temperature rising trend according to the similarity between all suspected abnormal temperature rising trend segments to obtain the abnormal risk degree of the temperature rising trend includes: According to the trend segment slopes of each suspected abnormal temperature rising trend segment, calculate the absolute value of the difference between the trend segment slopes of every two adjacent suspected abnormal temperature rising trend segments respectively to obtain the average absolute value of the difference. Use a preset exponential function to perform negative mapping on the average absolute value of the difference to obtain the similarity degree; Obtain the abnormal risk degree of the temperature rising trend according to the product between the similarity degree and the initial abnormal risk degree of the temperature rising trend.

[0012] Preferably, adjusting the initial temperature threshold according to the abnormal risk degree of the temperature rising trend to obtain the adaptive temperature threshold at the current moment includes: Normalize the abnormal risk degree of the temperature rising trend to obtain the corresponding normalization result. Obtain the difference between the constant 1 and the normalization result of the preset ratio as the adjustment coefficient. Take the product between the initial temperature threshold and the adjustment coefficient as the adaptive temperature threshold at the current moment.

[0013] In a second aspect, an embodiment of the present invention provides an Internet of Things-based cold chain monitoring and early warning system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements an Internet of Things-based cold chain monitoring and early warning method as described in the first aspect.

[0014] The beneficial effects of the embodiments of the present invention compared with the prior art are: The present invention obtains historical temperature time-series data in a preset historical period before the current moment, and maps the historical temperature time-series data into a two-dimensional coordinate system, where the horizontal axis of the two-dimensional coordinate system is time and the vertical axis is temperature; obtains heating data points according to the slope between two adjacent data points in the two-dimensional coordinate system, connects adjacent and continuous heating data points to obtain at least one heating trend segment, and when the number of heating trend segments is greater than a preset number threshold, obtains the degree of discontinuous distribution of each heating trend segment according to the interval distribution characteristics of the heating trend segments; obtains at least one suspected abnormal heating trend segment according to the degree of discontinuous distribution of each heating trend segment, obtains the initial abnormal risk degree of the heating trend according to the trend segment length and trend segment slope of all suspected abnormal heating trend segments, and corrects the initial abnormal risk degree of the heating trend according to the similarity between all suspected abnormal heating trend segments to obtain the abnormal risk degree of the heating trend; adjusts the initial temperature threshold according to the abnormal risk degree of the heating trend to obtain the adaptive temperature threshold at the current moment, and uses the adaptive temperature threshold to monitor and give an early warning of the temperature of the refrigerated truck at the current moment. Among them, according to the change trend of historical temperature data, it is judged whether there is a risk degree of abnormal temperature rise (abnormal risk degree of heating trend) at the current moment, and the initial temperature threshold is adjusted according to the risk degree to obtain the adaptive temperature threshold at the current moment, which is used to monitor whether the temperature at the current moment is abnormal. By combining the temperature change characteristics and the sensor mechanical stress drift characteristics of the refrigerated truck during transportation, the setting of the adaptive temperature threshold is carried out, avoiding the abnormal temperature monitoring caused by sensor noise, and improving the timeliness and accuracy of temperature sensor monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0016] Figure 1 It is a flowchart of a cold chain monitoring and early warning method based on the Internet of Things provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following will describe in detail the embodiments of the present disclosure, and the examples of the embodiments are shown in the drawings. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present disclosure and should not be construed as a limitation of the present disclosure.

[0018] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described here can be implemented in an order other than those illustrated or described here. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are only examples of devices and methods consistent with some aspects of the present disclosure.

[0019] In order to illustrate the technical solution of the present invention, the following will be described through specific embodiments.

[0020] The specific scenario targeted by the present invention is as follows: During the transportation of a cold chain refrigerated truck, a temperature sensor is used to monitor the temperature inside the refrigerated truck in real time. When the temperature reaches the set temperature threshold, an abnormal temperature alarm for the refrigerated truck is triggered. However, the warning effect of a fixed temperature threshold is poor and it is easily interfered by sensor noise during the transportation of the refrigerated truck. Therefore, it is necessary to improve the warning ability based on threshold monitoring to reduce the impact of temperature changes on cold chain transported items and ensure the quality of cold chain transported items.

[0021] See Figure 1 , which is a method flowchart of a cold chain monitoring and warning method based on the Internet of Things provided in Embodiment 1 of the present invention. As Figure 1 shown, the method may include: Step S101, obtain the historical temperature time series data inside the refrigerated truck within a preset historical period before the current moment, and map the historical temperature time series data into a two-dimensional coordinate system, where the horizontal axis of the two-dimensional coordinate system is time and the vertical axis is temperature.

[0022] Install a temperature sensor inside the refrigerated truck compartment to monitor the temperature data inside the compartment in real time. Among them, the sampling frequency of the temperature sensor is once per second, and the acquisition duration is from the start to the end of the refrigerated truck's engine shutdown. That is, the temperature data inside the compartment is collected starting from the start of the refrigerated truck, and each collected temperature data is uploaded to the monitoring terminal in real time through the Internet of Things. The monitoring terminal analyzes and warns the received temperature data, and the temperature data is stored in the format of a time series sequence.

[0023] Taking the current moment as an example, the historical temperature time series data composed of the temperature data collected within 5 minutes before the current moment is used as the analysis object at the current moment to obtain the adaptive temperature threshold at the current moment and monitor and warn the temperature at the current moment. It should be noted that considering that if the length of the historical temperature time series data is too large, it will lead to excessive calculation, and if the length of the historical temperature time series data is too small, there will be insufficient effective temperature information in the carriage. Therefore, in the present invention, the temperature data within 5 minutes is used as the historical temperature time series data. There is no requirement here, and the implementer can set it by himself.

[0024] After obtaining the historical temperature time series data before the current moment, the historical temperature time series data is mapped into a two-dimensional coordinate system. A temperature data in the historical temperature time series data corresponds to a data point in the two-dimensional coordinate system for subsequent temperature change analysis. Among them, the horizontal axis of the two-dimensional coordinate system is time, the time interval on the horizontal axis is 1 second, and the vertical axis is the temperature data.

[0025] Step S102: Obtain the rising temperature data points according to the slope between two adjacent data points in the two-dimensional coordinate system, connect the adjacent and continuous rising temperature data points, and obtain at least one rising trend segment. When the number of rising trend segments is greater than the preset number threshold, obtain the non-continuous distribution degree of each rising trend segment according to the interval distribution characteristics of the rising trend segments.

[0026] In the refrigerated truck, the temperature needs to be strictly maintained at a low temperature state. If there is a node where the temperature in the refrigerated truck rises, there is a certain possibility of temperature abnormality. Therefore, according to the change trend between adjacent data points in the two-dimensional coordinate system, the rising temperature data points are determined. The specific determination method is as follows: for any data point in the two-dimensional coordinate system except the first data point, obtain the previous data point of the any data point, and according to the coordinates of the any data point and the coordinates of the previous data point, obtain the slope between the any data point and the previous data point, which is recorded as the temperature change trend value of the any data point; if the temperature change trend value of the any data point is greater than 0, then determine the any data point as a rising temperature data point.

[0027] According to the method of obtaining the rising temperature data points, traverse each data point in the two-dimensional coordinate system except the first data point to obtain a plurality of rising temperature data points. If there are no rising temperature data points, it means that the temperature in the refrigerated truck is in a normal low temperature state, and the adaptive adjustment of the temperature threshold is not required. Using the initial temperature threshold to perform abnormal warning on the temperature at the current moment can meet the requirements. If there are rising temperature data points, connect the adjacent and continuous rising temperature data points to form a rising trend segment, and obtain at least one rising trend segment.

[0028] The main reason for the temperature sensor in the refrigerated truck to collect temperature data showing an upward trend is due to abnormal temperature inside the compartment. For example, poor thermal insulation or sealing performance of the compartment, aging or damage of the sealing strip, and possible infiltration of external hot air into the compartment can all cause the temperature inside the compartment to rise. Or, a malfunction in the refrigeration system or insufficient power leading to a decrease in refrigeration efficiency, and the interior of the compartment is heated by external temperature conduction. Thus, when the temperature in the compartment is abnormal, the temperature inside the compartment is mainly caused by the infiltration of external heat, and the air and goods in the refrigerated truck have a certain thermal inertia, and it takes some time for the heat to be evenly conducted throughout the compartment. Therefore, the temperature will rise gradually and be accompanied by certain fluctuations, forming multiple segments of upward temperature trends when mapped in a two-dimensional coordinate system.

[0029] On the other hand, the temperature sensor in the compartment collecting temperature data showing an upward trend is also affected by the noise of the sensor itself. When the refrigerated truck passes through bumpy ground during driving, due to the large volume of the refrigerated truck, strong vibrations will be generated. When the vibrations are transmitted to the temperature sensor installed in the compartment, it may cause mechanical stress drift of the temperature sensor, that is, the collected temperature data shows a drift trend, and it will also generate temperature data showing an increase. Moreover, this type of drift may generate a segment of upward temperature fluctuation.

[0030] Therefore, based on the above characteristics, analyze the distribution of the upward temperature trend segments to obtain real temperature abnormal data and exclude the temperature increase data caused by sensor stress drift. Specifically, count the number of upward temperature trend segments in the two-dimensional coordinate system. When the number of upward temperature trend segments is less than or equal to 2, it is considered that the number of upward temperature trend segments in the historical temperature time series data is small and there is no abnormal temperature increase, and the upward temperature trend segments are caused by the stress drift of the sensor. On the contrary, when the number of upward temperature trend segments is greater than 2, it indicates that the number of upward temperature trend segments is large and there is a trend of temperature increase inside the compartment. Further temperature change analysis is needed to lower the temperature threshold at the current moment and give an early warning of the abnormal temperature at the current moment, improving the timeliness and accuracy of the temperature warning at the current moment.

[0031] Since the stress drift of the sensor will gradually recover after it occurs and there is no subsequent upward trend, the positions of the upward temperature trend segments caused by the stress drift of the sensor are relatively randomly isolated. However, the real temperature abnormality caused by external temperature conduction has a continuous upward characteristic. Therefore, there will be multiple upward temperature trend segments with relatively close intervals. In the embodiment of the present invention, according to the interval distribution characteristics of the upward temperature trend segments, it is judged whether each upward temperature trend segment is caused by the stress drift of the sensor or by real temperature abnormality, so as to obtain the upward temperature trend segments caused by real temperature abnormality.

[0032] Specifically, first, the absolute value of the difference in the abscissa between the last data point of the previous temperature increase trend segment and the first data point of the next temperature increase trend segment is denoted as the interval distance between two adjacent temperature increase trend segments. Then, for any temperature increase trend segment, the interval distance between the any temperature increase trend segment and its previous temperature increase trend segment is obtained and denoted as the left interval distance, and the interval distance between the any temperature increase trend segment and its next temperature increase trend segment is obtained and denoted as the right interval distance. The left interval distance and the right interval distance are accumulated to obtain the interval characteristic value of the any temperature increase trend segment. The interval characteristic value of each temperature increase trend segment is obtained, and the absolute value of the difference between the interval characteristic value of the any temperature increase trend segment and the interval characteristic value of each other temperature increase trend segment is calculated to obtain the average value of the absolute values of the differences, which is denoted as the degree of non - continuous distribution of the any temperature increase trend segment.

[0033] In one embodiment, taking the i - th temperature increase trend segment as an example, the calculation expression for the degree of non - continuous distribution of the i - th temperature increase trend segment is:

[0034] Among them, represents the degree of non - continuous distribution of the i - th temperature increase trend segment, n represents the number of temperature increase trend segments, represents the interval characteristic value of the i - th temperature increase trend segment, represents the interval characteristic value of the j - th temperature increase trend segment other than the i - th temperature increase trend segment, and | | represents the absolute value symbol.

[0035] It should be noted that the larger the value of , the greater the difference in the interval characteristic values between the i - th temperature increase trend segment and the j - th temperature increase trend segment. Corresponding to the average value of the differences in the interval characteristic values between the i - th temperature increase trend segment and each of the remaining temperature increase trend segments, that is , the larger the value of

[0036] the greater the difference in the interval distribution between the i - th temperature increase trend segment and other temperature increase trend segments, the more independent and random the distribution characteristics are, and the more likely it is that the temperature increase trend segment is caused by the stress drift of the sensor. Correspondingly, the greater the degree of non - continuous distribution of the i - th temperature increase trend segment. On the contrary,

[0037] Step S103: Obtain at least one suspected abnormal temperature rise trend segment according to the degree of discontinuous distribution of each temperature rise trend segment. Obtain the initial abnormal risk degree of the temperature rise trend according to the trend segment lengths and trend segment slopes of all suspected abnormal temperature rise trend segments. Modify the initial abnormal risk degree of the temperature rise trend according to the similarity between all suspected abnormal temperature rise trend segments to obtain the abnormal risk degree of the temperature rise trend.

[0038] After obtaining the degree of discontinuous distribution of each temperature rise trend segment, perform linear normalization processing on the degrees of discontinuous distribution of all temperature rise trend segments to obtain the normalized value of each temperature rise trend segment. Among them, linear normalization processing belongs to the prior art and will not be elaborated here. If there is a discontinuous temperature rise trend segment caused by the stress drift of the sensor, it will be distributed at both ends of [0, 1] respectively from the degree of discontinuous distribution of the continuous temperature rise trend segment. Therefore, set 0.5 as the normalization threshold. There is no limitation here. If the normalized value of any temperature rise trend segment is less than the preset normalization threshold, determine that the any temperature rise trend segment is a suspected abnormal temperature rise trend segment. On the contrary, if the normalized value of any temperature rise trend segment is greater than or equal to 0.5, it is considered that the temperature rise trend segment is an abnormal temperature rise trend segment caused by the sensor stress drift due to the sudden bump during the driving of the refrigerated truck, and thus this temperature rise trend segment is excluded. Similarly, compare the normalized value of each temperature rise trend segment with 0.5 to obtain at least one suspected abnormal temperature rise trend segment. At this time, obtain the continuous temperature rise trend segment with temperature change characteristics through the interval distribution characteristics of the temperature rise trend segment, that is, the suspected abnormal temperature rise trend segment, and initially exclude the temperature rise data caused by the stress drift of the sensor.

[0039] After obtaining the suspected abnormal temperature rise trend segment, the abnormal degree of the temperature in the refrigerated truck can be evaluated according to the distribution of all suspected abnormal temperature rise trend segments. When the interval distance between adjacent trend segments in the suspected abnormal temperature rise trend segment is closer and the temperature rise trend segment is longer, it indicates that the temperature reduction system has a lower degree of inhibition of the temperature rise, resulting in a faster temperature rise fluctuation frequency and a greater abnormal risk of the temperature rise trend. When the slope of the suspected abnormal temperature rise trend segment is larger, it indicates that the temperature rises faster and the abnormal risk of the temperature rise trend is greater. Therefore, in the embodiment of the present invention, the initial abnormal risk degree of the temperature rise trend is obtained according to the trend segment lengths and trend segment slopes of all suspected abnormal temperature rise trend segments. The specific obtaining method is as follows: For any suspected abnormal temperature rise trend segment, perform linear fitting on the any suspected abnormal temperature rise trend segment to obtain the trend segment slope. Obtain the trend segment length according to the horizontal span and vertical span of the any suspected abnormal temperature rise trend segment. Take the product between the trend segment slope and the trend segment length as the temperature rise trend estimation value of the any suspected abnormal temperature rise trend segment; Obtain the estimated value of the temperature increase trend for each suspected abnormal temperature increase trend segment, and get the average value of the estimated values of the temperature increase trend. According to the interval distance between every two adjacent suspected abnormal temperature increase trend segments among all suspected abnormal temperature increase trend segments, obtain the average interval distance. Use the exponential function with the natural constant as the base to perform a negative mapping on the average interval distance to obtain the temperature increase frequency. According to the product between the temperature increase frequency and the average value of the estimated values of the temperature increase trend, obtain the initial risk level of the temperature increase trend anomaly.

[0040] In one embodiment, taking one suspected abnormal temperature increase trend segment as an example, use the least squares method to perform a linear fitting on the suspected abnormal temperature increase trend segment, and take the slope of the fitted line as the trend segment slope k of the suspected abnormal temperature increase trend segment. Among them, the least squares method belongs to the prior art and will not be elaborated here. Calculate the absolute value of the difference in abscissas between the two endpoints of the suspected abnormal temperature increase trend segment, and record it as the horizontal span of the suspected abnormal temperature increase trend segment , calculate the absolute value of the difference in ordinates between the lowest point and the highest point of the suspected abnormal temperature increase trend segment, and record it as the vertical span of the suspected abnormal temperature increase trend segment , using the calculation method of the Pythagorean theorem, obtain the trend segment length L of the suspected abnormal temperature increase trend segment as:

[0041] Similarly, obtain the trend segment slope and trend segment length of each suspected abnormal temperature increase trend segment, and then combine the trend segment slope and trend segment length of each suspected abnormal temperature increase trend segment, as well as the interval distance between the suspected abnormal temperature increase trend segments, to obtain the calculation expression for the initial risk level of the temperature increase trend anomaly as:

[0042] Among them, represents the initial risk level of the temperature increase trend anomaly, M represents the number of interval distances between the suspected abnormal temperature increase trend segments, represents the p-th interval distance, N represents the number of suspected abnormal temperature increase trend segments, represents the trend segment length of the i-th suspected abnormal temperature increase trend segment, represents the trend segment slope of the i-th suspected abnormal temperature increase trend segment, represents the exponential function with the natural constant as the base.

[0043] It should be noted that the larger the interval distance between the suspected abnormal temperature increase trend segments, the worse the temperature increase persistence and the lower the temperature increase frequency, corresponding to the smaller the initial risk level of the temperature increase trend anomaly; the larger the trend segment length and the trend segment slope of the suspected abnormal temperature increase trend segment, the faster the frequency of continuous temperature increase in the refrigerated vehicle and the greater the temperature increase degree, corresponding to the greater the initial risk level of the temperature increase trend anomaly and the more likely there is a potential temperature anomaly risk.

[0044] Further, considering that when the refrigerated truck passes through a section of the road that is potholed and bumpy, due to frequent oscillations, the temperature sensor may repeatedly show a temperature increase trend segment caused by mechanical stress offset. Therefore, there may also be a possibility of abnormal sensor stress offset in the trend segments with similar intervals between the suspected abnormal temperature increase trend segments. When the refrigerated truck passes through a potholed and bumpy road section, the randomness of each oscillation degree is relatively strong, and the bumpy degrees of strong and weak are different, which further leads to different rising rates of the temperature increase trend segments where the sensor repeatedly shows stress offset. In contrast, the heat conduction from the outside temperature to the inside of the refrigerated truck is more stable. Therefore, the similarity of the rising rates of the suspected abnormal temperature increase trend segments will be higher. Therefore, in the embodiments of the present invention, according to the similarity between all suspected abnormal temperature increase trend segments, the initial risk degree of abnormal temperature increase trend is corrected. The specific correction method is as follows: According to the trend segment slope of each suspected abnormal temperature increase trend segment, calculate the absolute value of the difference between the trend segment slopes of every two adjacent suspected abnormal temperature increase trend segments respectively to obtain the average absolute value of the difference. Use a preset exponential function to perform a negative mapping on the average absolute value of the difference to obtain the similarity; obtain the risk degree of abnormal temperature increase trend according to the product of the similarity and the initial risk degree of abnormal temperature increase trend.

[0045] In an embodiment, the calculation expression of the similarity is:

[0046] Wherein, represents the similarity, represents the exponential function with the natural constant as the base, N represents the number of suspected abnormal temperature increase trend segments, represents the trend segment slope of the i-th suspected abnormal temperature increase trend segment, represents the trend segment slope of the (i + 1)-th suspected abnormal temperature increase trend segment, and | | represents the absolute value symbol.

[0047] It should be noted that the greater the difference in the trend segment slopes between the suspected abnormal temperature increase trend segments, the more inconsistent the rising rates between the suspected abnormal temperature increase trend segments, and the smaller the similarity.

[0048] After obtaining the similarity, use the similarity as the correction value to correct the initial risk degree of abnormal temperature increase trend. The correction formula is:

[0049] Wherein, represents the risk degree of abnormal temperature increase trend, R represents the initial risk degree of abnormal temperature increase trend, represents the similarity.

[0050] Step S104: Adjust the initial temperature threshold according to the abnormal risk degree of the temperature rising trend to obtain the adaptive temperature threshold at the current moment, and use the adaptive temperature threshold to monitor and give early warnings about the temperature of the refrigerated truck at the current moment.

[0051] The abnormal risk degree of the temperature rising trend excludes the abnormal stress offset of the sensor under continuous bumpy sections of the refrigerated truck, and then adaptively adjusts the initial temperature threshold using the abnormal risk degree of the temperature rising trend. The greater the abnormal risk degree of the temperature rising trend, it indicates that there is a greater possibility of a temperature rise risk in the historical temperature data before the current moment, and the rising degree is severe. Correspondingly, it is more necessary to adjust the initial temperature threshold downward to give early warnings about abnormal temperature rises in advance.

[0052] Specifically, adjusting the initial temperature threshold according to the abnormal risk degree of the temperature rising trend to obtain the adaptive temperature threshold at the current moment includes: Normalize the abnormal risk degree of the temperature rising trend to obtain the corresponding normalization result, obtain the difference between the constant 1 and the normalization result of the preset ratio as the adjustment coefficient, and take the product of the initial temperature threshold and the adjustment coefficient as the adaptive temperature threshold at the current moment.

[0053] In an embodiment, the calculation expression of the adaptive temperature threshold at the current moment is:

[0054] Wherein, is the adaptive temperature threshold at the current moment, is the initial temperature threshold, 1 represents a constant, 0.2 represents the preset ratio, norm() represents the normalization function, represents the abnormal risk degree of the temperature rising trend.

[0055] It should be noted that the greater the abnormal risk degree of the temperature rising trend, the greater the degree of reduction of the temperature threshold. The preset ratio of 0.2 is used to control the appropriate amount of reduction. For the initial temperature threshold, the corresponding initial temperature threshold can be set according to the specific goods in the refrigerated truck. For example, when the transported goods in the refrigerated truck are perishable foods, the initial temperature threshold in the carriage can be set to 12 degrees Celsius, 0 degrees Celsius, -10 degrees Celsius, -20 degrees Celsius, etc. according to different food types (fruits and vegetables, refrigerated foods, frozen foods). The setting of the temperature threshold in the refrigerated truck belongs to the prior art and will not be elaborated here.

[0056] After obtaining the adaptive temperature threshold at the current moment, collect the real-time temperature data at the current moment, compare the real-time temperature data with the adaptive temperature threshold at the current moment. When the real-time temperature data exceeds the adaptive temperature threshold, remotely transmit a warning signal through the Internet of Things, and at the same time, send an alarm notification to the refrigerated truck driver and relevant personnel in the cold chain transportation for timely inspection and handling.

[0057] Based on the same inventive concept as the above method, an embodiment of the present invention further provides an Internet of Things-based cold chain monitoring and warning system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above Internet of Things-based cold chain monitoring and warning methods.

[0058] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A cold chain monitoring and early warning method based on the Internet of Things, characterized in that, The method includes: Obtaining historical temperature time series data in a refrigerated truck within a preset historical period before the current moment, and mapping the historical temperature time series data onto a two-dimensional coordinate system, where the abscissa of the two-dimensional coordinate system is time and the ordinate is temperature; Obtaining heating data points according to the slope between two adjacent data points in the two-dimensional coordinate system, connecting adjacent and consecutive heating data points to obtain at least one heating trend segment. When the number of heating trend segments is greater than a preset number threshold, obtaining the degree of non - continuous distribution of each heating trend segment according to the interval distribution characteristics of the heating trend segments; Obtaining at least one suspected abnormal heating trend segment according to the degree of non - continuous distribution of each heating trend segment, obtaining an initial heating trend abnormal risk degree according to the trend segment length and trend segment slope of all suspected abnormal heating trend segments, and correcting the initial heating trend abnormal risk degree according to the similarity between all suspected abnormal heating trend segments to obtain a heating trend abnormal risk degree; Adjusting an initial temperature threshold according to the heating trend abnormal risk degree to obtain an adaptive temperature threshold at the current moment, and using the adaptive temperature threshold to monitor and give early warnings about the temperature of the refrigerated truck at the current moment.

2. The cold chain monitoring and early warning method based on the Internet of Things according to claim 1, characterized in that, The obtaining heating data points according to the slope between two adjacent data points in the two - dimensional coordinate system includes: For any data point in the two - dimensional coordinate system except the first data point, obtaining the previous data point of the any data point, and obtaining the slope between the any data point and the previous data point according to the coordinates of the any data point and the coordinates of the previous data point, which is denoted as the temperature change trend value of the any data point; if the temperature change trend value of the any data point is greater than 0, then determining the any data point as a heating data point.

3. The cold chain monitoring and warning method based on the Internet of Things according to claim 1, characterized in that The obtaining the degree of non - continuous distribution of each heating trend segment according to the interval distribution characteristics of the heating trend segments includes: Denoting the absolute value of the difference in abscissa between the last data point of the previous heating trend segment and the first data point of the next heating trend segment as the interval distance between two adjacent heating trend segments; For any heating trend segment, obtaining the interval distance between the any heating trend segment and its previous heating trend segment, which is denoted as the left interval distance, obtaining the interval distance between the any heating trend segment and its next heating trend segment, which is denoted as the right interval distance, and accumulating the left interval distance and the right interval distance to obtain the interval characteristic value of the any heating trend segment; Obtaining the interval characteristic value of each heating trend segment, calculating the absolute value of the difference between the interval characteristic value of the any heating trend segment and the interval characteristic value of each other heating trend segment to obtain an average value of the absolute values of the differences, which is denoted as the degree of non - continuous distribution of the any heating trend segment.

4. The cold chain monitoring and early warning method based on the Internet of Things according to claim 1, characterized in that The obtaining at least one suspected abnormal heating trend segment according to the degree of non - continuous distribution of each heating trend segment includes: Perform linear normalization on the degree of discontinuous distribution for all warming trend segments to obtain the normalized value for each warming trend segment. If the normalized value of any warming trend segment is less than the preset normalized threshold, determine the any warming trend segment as a suspected abnormal warming trend segment.

5. The cold chain monitoring and warning method based on the Internet of Things according to claim 3, wherein The obtaining of the initial warming trend abnormal risk degree according to the trend segment lengths and trend segment slopes of all suspected abnormal warming trend segments includes: For any suspected abnormal warming trend segment, perform linear fitting on the any suspected abnormal warming trend segment to obtain the trend segment slope. According to the horizontal span and vertical span of the any suspected abnormal warming trend segment, obtain the trend segment length, and take the product between the trend segment slope and the trend segment length as the warming trend estimated value of the any suspected abnormal warming trend segment; Obtain the warming trend estimated values of each suspected abnormal warming trend segment to get the average value of the warming trend estimated values. According to the interval distances between every two adjacent suspected abnormal warming trend segments among all suspected abnormal warming trend segments, obtain the average interval distance value. Use the exponential function with the natural constant as the base to perform negative mapping on the average interval distance value to obtain the warming frequency. According to the product between the warming frequency and the average value of the warming trend estimated values, obtain the initial warming trend abnormal risk degree.

6. The cold chain monitoring and early warning method based on the Internet of Things according to claim 5, characterized in that, The correcting of the initial warming trend abnormal risk degree according to the similarity between all suspected abnormal warming trend segments to obtain the warming trend abnormal risk degree includes: According to the trend segment slopes of each suspected abnormal warming trend segment, calculate the absolute value of the difference between the trend segment slopes of every two adjacent suspected abnormal warming trend segments respectively to obtain the average absolute difference value. Use the preset exponential function to perform negative mapping on the average absolute difference value to obtain the similarity degree; according to the product between the similarity degree and the initial warming trend abnormal risk degree, obtain the warming trend abnormal risk degree.

7. The cold chain monitoring and early warning method based on the Internet of Things according to claim 1, characterized in that The adjusting of the initial temperature threshold according to the warming trend abnormal risk degree to obtain the adaptive temperature threshold at the current moment includes: Normalize the warming trend abnormal risk degree to obtain the corresponding normalized result. Obtain the difference between the constant 1 and the normalized result of the preset ratio as the adjustment coefficient, and take the product between the initial temperature threshold and the adjustment coefficient as the adaptive temperature threshold at the current moment.

8. An Internet of Things-based cold chain monitoring and early warning system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for cold chain monitoring and warning based on the Internet of Things according to any one of claims 1 - 7.

Images