Temperature and humidity whole-process monitoring system and detection method for cold-chain logistics

Abstract

The invention provides a temperature and humidity whole-process monitoring system and detection method for cold-chain logistics, and the method comprises the steps: S1, collecting the temperature and humidity data in a carriage and a vehicle door opening and closing state signal in real time; s3, when the fluctuation event coincides with the opening state of the vehicle door in time, a causal association strength coefficient of opening of the vehicle door and temperature and humidity fluctuation is calculated; S4, when the causal association strength coefficient does not exceed a threshold value, the temperature and humidity data in the compartment are continuously monitored; and S5, when the causal association strength coefficient exceeds a threshold value, starting a staged dynamic alarm threshold value mechanism, and when the temperature and humidity deviations simultaneously meet the conditions that the dynamic alarm threshold value of the current stage is continuously exceeded and the duration exceeds the recovery duration predicted based on historical recovery data, triggering temperature and humidity abnormal alarm. According to the invention, through multi-dimensional data fusion and intelligent analysis, the problems of high false alarm rate and untimely response in traditional cold chain temperature and humidity detection are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of cold chain logistics, and in particular to a temperature and humidity full-process monitoring system and detection method for cold chain logistics. Background Art

[0002] When loading and unloading at logistics nodes, refrigerated trucks frequently open their doors, allowing hot air to flow in, causing brief, dramatic fluctuations in temperature and humidity inside the vehicle. Traditional detection solutions that directly use fixed thresholds for alarms are prone to generating numerous false alarms due to transient fluctuations, while excessively loosening the thresholds can mask true anomalies. Conventional temperature and humidity sensors have a response time that cannot capture instantaneous temperature and humidity changes, potentially simplifying the complete "door opening and closing" fluctuation process into a single peak, leading to false positives. Delayed alarms can delay response to actual faults, while relying solely on absolute thresholds makes it impossible to distinguish between normal loading and unloading fluctuations and equipment anomalies. Summary of the Invention

[0003] Based on the problems existing in the above background technology, the present invention proposes a temperature and humidity full-process monitoring system and detection method for cold chain logistics, and the technical solutions adopted are as follows:

[0004] A temperature and humidity detection method for cold chain logistics, the temperature and humidity detection method comprising:

[0005] S1: Real-time collection of temperature and humidity data in the vehicle cabin and door opening and closing status signals;

[0006] S2: Slide the window temperature and humidity data and identify the instantaneous fluctuation events of the temperature and humidity data;

[0007] S3: When the fluctuation event coincides with the door opening state in time, calculate the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuations;

[0008] S4: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated;

[0009] S5: When the temperature and humidity deviations simultaneously meet the following conditions: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, a temperature and humidity anomaly alarm is triggered.

[0010] Preferably, the instantaneous fluctuation time of S2 includes: the rate of increase of temperature and humidity, the total duration of the fluctuation, and the time it takes for the temperature and humidity to return to the baseline.

[0011] Preferably, the sliding window detection of S2 includes:

[0012] Set up a sliding window including short-term window, medium-term window and long-term window. When the temperature and humidity rising rate in any window exceeds the preset threshold of each window, it is marked as the starting point of the fluctuation event.

[0013] The inflection point where the temperature and humidity gradient changes from positive to negative is taken as the peak point of the potential fluctuation event;

[0014] The end point of the fluctuation event was when the temperature and humidity stabilized within the baseline.

[0015] Preferably, the method for calculating the causal association strength coefficient of S3 includes:

[0016] Calculate the ratio of the door opening time to the total duration of the fluctuation event;

[0017] Calculate the temperature and humidity changes during the door opening period;

[0018] Calculate the time it takes for temperature and humidity to recover from peak values ​​to baseline;

[0019] All parameters are combined to output coefficient values ​​on a 0-1 scale.

[0020] Preferably, when the causal correlation strength coefficient does not exceed the threshold value, the step S4 of continuously monitoring the temperature and humidity data in the vehicle compartment specifically includes:

[0021] A: Immediately sound and light alarm;

[0022] B: Start quick diagnosis, estimate the refrigerant pressure through virtual pressure, evaluate the cabin air tightness through temperature and humidity recovery rate, and notify the driver of the diagnosis results.

[0023] Preferably, the S4 staged dynamic alarm threshold mechanism includes:

[0024] The first time period is after the peak of the fluctuation Seconds, the first level dynamic alarm threshold is the standard threshold floating ℃;

[0025] The second time period is after the peak of the fluctuation Seconds, the second level dynamic alarm threshold is the upper limit of the standard threshold ℃;

[0026] The third time period is after the peak of fluctuation seconds to restore the standard threshold.

[0027] Preferably, the recovery time predicted by the historical recovery data of S5 is obtained as follows:

[0028] Calculate the specific heat capacity of cargo based on its mass and type;

[0029] Match the preset heat transfer coefficient through the carriage ID;

[0030] Obtain the effective heat exchange area after loading through laser point cloud scanning;

[0031] The recovery time predicted by historical recovery data is calculated based on all parameter fitting.

[0032] Preferably, the S5 further includes:

[0033] Discretize the temperature and humidity rising curve of the fluctuation event into multiple characteristic sequences;

[0034] Generate a digital fingerprint of the feature sequence by a shift hash algorithm;

[0035] When the similarity between the digital fingerprint of the new fluctuation event and the digital fingerprint of the historical fluctuation event exceeds the similarity threshold, the decision parameters of the associated historical fluctuation event are directly called.

[0036] Preferably, the similarity is obtained by the following method:

[0037] Align the temperature and humidity rising curves of new fluctuation events with those of historical fluctuation events;

[0038] Calculate the sum of squares of temperature difference and humidity difference at the corresponding timestamps;

[0039] The similarity score of 0-100% is obtained through normalization.

[0040] A temperature and humidity monitoring system for cold chain logistics, comprising:

[0041] Real-time acquisition system: real-time acquisition of temperature and humidity data in the vehicle compartment and door opening and closing status signals;

[0042] Fluctuation event recognition system: uses sliding window temperature and humidity data to identify instantaneous fluctuation events of temperature and humidity data;

[0043] Causal correlation analysis system: When the fluctuation event coincides with the door opening state in time, the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuation is calculated;

[0044] Dynamic alarm threshold system: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated;

[0045] Abnormal alarm system: When the temperature and humidity deviations meet the following conditions simultaneously: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, the temperature and humidity abnormal alarm is triggered.

[0046] Beneficial effects of the present invention: The present invention solves the problems of high false alarm rate and untimely response in traditional cold chain temperature and humidity detection through multi-dimensional data fusion and intelligent analysis mechanism. At the data acquisition end, the temperature and humidity are acquired synchronously with the door opening and closing status, and the sliding window detection is combined to accurately identify instantaneous fluctuation events to make up for the sensor response lag defect; the calculation of the causal correlation strength coefficient quantifies the relationship between door opening and temperature and humidity fluctuations to achieve the distinction between normal loading and unloading fluctuations and equipment anomalies; the staged dynamic alarm threshold mechanism avoids false alarms or missed alarms caused by fixed thresholds, and dynamically adjusts the threshold according to the fluctuation process to improve the accuracy of early warning; the recovery time is predicted based on the cargo attributes and carriage parameters, so that the alarm triggering conditions are more in line with the actual working conditions; the feature sequence digital fingerprint and similarity matching technology can quickly call historical decision parameters and greatly shorten the abnormal response time. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a temperature and humidity detection method for cold chain logistics described in the present invention. DETAILED DESCRIPTION

[0048] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0049] One embodiment of the present invention provides a temperature and humidity detection method for cold chain logistics, the temperature and humidity detection method comprising:

[0050] S1: Real-time collection of temperature and humidity data in the vehicle cabin and door opening and closing status signals;

[0051] S2: Slide the window temperature and humidity data and identify the instantaneous fluctuation events of the temperature and humidity data;

[0052] S3: When the fluctuation event coincides with the door opening state in time, calculate the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuations;

[0053] S4: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated;

[0054] S5: When the temperature and humidity deviations simultaneously meet the following conditions: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, a temperature and humidity anomaly alarm is triggered.

[0055] The working principle and effect of the above technical solution are as follows: First, sensors collect real-time temperature and humidity data within the vehicle compartment, as well as door opening and closing status signals, and process the temperature and humidity data using a sliding window technique. When a fluctuation event coincides with a door opening state, the system calculates the causal correlation strength coefficient by counting parameters such as the proportion of door opening time, the temperature and humidity change during this period, and the temperature and humidity recovery time. This quantitatively determines whether the temperature and humidity fluctuation is caused by the door opening. If the coefficient does not exceed the threshold, an abnormality likely exists that is not caused by loading or unloading. An audible and visual alarm is immediately issued, and a rapid diagnosis is initiated to assess refrigerant pressure and vehicle compartment airtightness. If the coefficient exceeds the threshold, the fluctuation is likely caused by loading or unloading, and a phased dynamic alarm threshold mechanism is activated, dynamically adjusting the alarm threshold by time period. Finally, the current dynamic threshold and the predicted recovery time based on the cargo specific heat capacity, the vehicle compartment heat transfer coefficient, and the effective heat transfer area are combined to determine whether the temperature and humidity deviation meets the abnormal alarm conditions. If the temperature and humidity continuously exceed the dynamic threshold for a period exceeding the predicted recovery time, an abnormal alarm is triggered.

[0056] The calculation of the causal correlation strength coefficient realizes the distinction between normal loading and unloading fluctuations and equipment abnormalities, reducing the false alarm rate caused by traditional fixed thresholds; the phased dynamic alarm threshold mechanism flexibly adjusts the threshold according to the fluctuation process, which not only avoids false triggering of transient fluctuations, but also ensures timely response to real abnormalities such as equipment failures; the recovery time is predicted based on cargo characteristics and carriage parameters, making the alarm judgment more in line with actual working conditions and reducing the risk of missed reports; digital fingerprint and similarity matching technology greatly shortens the abnormal response time and improves processing efficiency by reusing historical experience.

[0057] In one embodiment of the present invention, the method for calculating the causal association strength coefficient of S3 includes:

[0058] Calculate the ratio of the door opening time to the total duration of the fluctuation event;

[0059] Calculate the temperature and humidity changes during the door opening period;

[0060] Calculate the time it takes for temperature and humidity to recover from peak values ​​to baseline;

[0061] All parameters are integrated to output a coefficient value of 0-1 scale, and the causal correlation strength coefficient is obtained by the following formula:

[0062]

[0063] Among them, D represents the temperature-related causal strength coefficient; R represents the time correlation, that is, the ratio of the door opening time to the total duration of the fluctuation event; S represents the fluctuation synchronization, that is, the ratio of the temperature change value during the door opening time period to the total temperature change value; H represents the recovery hysteresis, that is, the ratio of the time for the temperature to return to the baseline to the temperature rise time; similarly, the humidity-related causal strength coefficient is obtained. , D and Take the average value to get the causal correlation strength coefficient related to temperature and humidity

[0064] The working principle and effectiveness of the above technical solution are as follows: In the calculation of the causal-effect intensity coefficient, the essence of introducing temporal correlation is to quantify the temporal contribution of door opening and closing to the fluctuation event. Traditional methods only determine whether the door is open, while R captures the "percentage of door opening time," avoiding misjudgments caused by short-term door openings and closings while also identifying sustained temperature changes caused by prolonged door openings. Fluctuation synchronization is introduced to quantify the actual proportion of temperature change occurring during the door opening period. Traditional methods ignore the temporal coupling of energy generation, while S locks temperature changes to the door opening period, accurately locating the true cause of the energy change. Recovery hysteresis is introduced to quantify the health of the system's self-recovery capabilities. Traditional methods only monitor whether recovery has occurred, while H assesses the relative hysteresis of the recovery speed, dynamically distinguishing between the rapid response of normal loading and unloading and the slow recovery caused by equipment aging. The three parameters are mutually adapted to reduce false alarm rates.

[0065] In one embodiment of the present invention, when the causal correlation strength coefficient does not exceed the threshold value, the step S4 of continuously monitoring the temperature and humidity data in the vehicle compartment specifically includes:

[0066] A: Immediately sound and light alarm;

[0067] B: Start quick diagnosis, estimate the refrigerant pressure through virtual pressure, evaluate the cabin air tightness through temperature and humidity recovery rate, and notify the driver of the diagnosis results.

[0068] The working principle and effect of the above technical solution are as follows: When the causal correlation strength coefficient does not exceed the threshold, the system enters a continuous monitoring state. First, the causal correlation strength coefficient threshold is set to 0.7. At this time, the system immediately triggers an audible and visual alarm to indicate an abnormality and simultaneously initiates a rapid diagnosis process: by establishing a thermodynamic model of the vehicle cabin, the refrigerant pressure is virtually estimated based on the current temperature and humidity data and environmental parameters to determine whether there is a leak or pressure abnormality in the refrigeration system. Simultaneously, the temperature and humidity recovery rate within a short period of time (such as 5-10 minutes) is analyzed, and the airtightness status is evaluated in combination with the vehicle cabin structural parameters. Finally, diagnostic information such as the virtual estimated refrigerant pressure value and airtightness assessment results are pushed to the driver's terminal in real time.

[0069] When the causal correlation strength coefficient is lower than the threshold, an audible and visual alarm is immediately triggered to ensure that the driver and surrounding personnel are aware of the abnormality at the first time, avoiding delays due to negligence. At the same time, virtual pressure is used to estimate the refrigerant pressure, and the air tightness of the compartment is assessed based on the temperature and humidity recovery rate to quickly locate potential fault points, upgrading traditional passive monitoring to active diagnosis. Compared with a single alarm mode, this solution not only provides timely warnings, but also provides a preliminary basis for judging faults such as refrigerant leakage and decreased air tightness, significantly shortening troubleshooting time and reducing the risk of cargo loss due to equipment abnormalities. The diagnostic results are pushed to the driver in real time, facilitating the implementation of targeted measures, effectively improving the efficiency and safety of handling temperature control anomalies during cold chain transportation, reducing operation and maintenance costs, and ensuring stable cargo quality.

[0070] In one embodiment of the present invention, the S4 staged dynamic alarm threshold mechanism includes:

[0071] The first time period is after the peak of the fluctuation Seconds, the first level dynamic alarm threshold is the standard threshold floating ℃;

[0072] The second time period is after the peak of the fluctuation Seconds, the second level dynamic alarm threshold is the upper limit of the standard threshold ℃;

[0073] The third time period is after the peak of fluctuation seconds to restore the standard threshold.

[0074] The working principle and effect of the above technical solution are: set the first time period The value range is 10-15 seconds after the peak value of the fluctuation. At this time, the first level dynamic alarm threshold is the standard threshold. =6 - 10℃, during this stage, a brief overshoot of temperature and humidity is allowed to avoid false alarms caused by transient errors of sensors; the second time period 30-40 seconds after the peak of the fluctuation, the second level dynamic alarm threshold is set to the standard threshold upper =2-5℃. If the temperature and humidity have not returned to the dynamic threshold during this stage, it indicates that there is a persistent abnormality. The third time period If the temperature and humidity have not returned to the standard threshold within 60-90 seconds after the peak fluctuation, the system will trigger the regular alarm process.

[0075] The phased dynamic alarm threshold mechanism is scientifically set 、 、 Time nodes and 、 The threshold gradient effectively balances the accuracy and timeliness of cold chain temperature and humidity monitoring. During the time period, the threshold value rises significantly Provide the system with a short-term overshoot buffer space to avoid false alarms caused by normal transient fluctuations such as sensor response delays and hot swaps during loading and unloading; stage, the appropriately narrowed threshold rises It can keenly capture persistent abnormal trends and issue early warnings for potential hidden dangers such as refrigeration equipment failure and decreased compartment sealing; The standard thresholds are restored in stages to ensure strict enforcement of the alarm mechanism under long-term abnormal conditions. This time-based, differentiated threshold adjustment strategy not only reduces invalid alarms caused by normal loading and unloading, but also shortens response time when real abnormalities occur.

[0076] In one embodiment of the present invention, the recovery time predicted by the historical recovery data of S5 is obtained as follows:

[0077] Calculate the specific heat capacity of cargo based on its mass and type;

[0078] Match the preset heat transfer coefficient through the carriage ID;

[0079] Obtain the effective heat exchange area after loading through laser point cloud scanning;

[0080] The predicted recovery time of historical recovery data is calculated based on all parameter fitting, and the predicted recovery time of historical recovery data is obtained by the following formula:

[0081]

[0082] Among them, m represents the mass of the goods, Indicates the specific heat capacity of the cargo (kJ / (kg·K)), Indicates the change in total temperature, h indicates the heat transfer coefficient (W / (m²·K), and A indicates the heat transfer area;

[0083] The working principle and effect of the above technical solution are as follows: the mass of the cargo is read in real time through the on-board weighing sensor; the specific heat capacity of the cargo is obtained through intelligent matching of the cargo type library; the heat transfer coefficient is obtained by matching the material database through the carriage ID; and the heat transfer area of ​​the cargo is calculated through laser point cloud scanning after loading.

[0084] This formula is based on the principles of thermodynamics and heat transfer. By quantifying the heat capacity of the cargo and the heat transfer efficiency of the compartment interface, it accurately calculates the characteristic time required for the temperature rise of the cargo to reach a specific amplitude during cold chain transportation. According to the first law of thermodynamics, the total heat absorbed by the cargo during the heating process is expressed as follows: Jointly determine the system thermal inertia and temperature rise It directly affects the total amount of heat to be absorbed; the denominator Following Newton's law of cooling, it represents the heat exchange power between the vehicle compartment and the external environment. The heat transfer coefficient h depends on the vehicle compartment material and sealing performance. The effective heat exchange area A is dynamically acquired through laser point cloud scanning technology, reflecting the actual heat transfer area under loading conditions.

[0085] Compared to traditional methods, temperature change calculations used fixed parameters set in advance, without considering the cargo load and stacking conditions during each shipment, resulting in significant errors. In this solution, cargo mass is directly measured in real time using onboard weighing sensors, specific heat capacity is accurately matched to cargo type from a database, and effective heat exchange area is scanned using laser point clouds. This real-time data allows calculations to more accurately reflect actual transportation conditions, significantly reducing errors. It also predicts temperature trends in advance, providing earlier warnings and preventing cargo damage due to temperature anomalies.

[0086] In one embodiment of the present invention, the S5 further includes:

[0087] Discretize the temperature and humidity rising curve of the fluctuation event into multiple characteristic sequences;

[0088] Generate a digital fingerprint of the feature sequence by a shift hash algorithm;

[0089] When the similarity between the digital fingerprint of the new fluctuation event and the digital fingerprint of the historical fluctuation event exceeds the similarity threshold, the decision parameters of the associated historical fluctuation event are directly called.

[0090] The working principle and effect of the above technical solution are as follows: the system first discretizes the temperature and humidity rise curve of a fluctuation event, breaking the continuous curve into multiple discrete feature sequences based on time intervals or data features. Each sequence contains key data points of temperature and humidity changes within a specific time interval. These feature sequences are then encoded using a shift hashing algorithm, mapping the complex temperature and humidity data into a fixed-length digital fingerprint. This fingerprint serves as the "identifier" of the fluctuation event, preserving the core characteristics of the data while facilitating rapid retrieval and comparison. When a new fluctuation event occurs, the system also generates its digital fingerprint and compares it with the digital fingerprint database of historical fluctuation events. During the comparison, the temperature and humidity rise curves of the new and historical events are aligned based on timestamps to ensure data dimensionality consistency. The sum of the squares of the temperature and humidity differences at the corresponding timestamps is then calculated. This operation amplifies the weight of the discrepancies and highlights key differences between the curves. Finally, the calculated results are normalized and mapped to a range of 0-100% to form a similarity score. If the score exceeds the preset similarity threshold (90%), it means that the temperature and humidity change pattern of the new fluctuation event is highly similar to that of a historical event, and the system directly calls the decision parameters of the associated historical event;

[0091] This solution significantly improves the efficiency and accuracy of cold chain temperature and humidity anomaly responses through a digital fingerprint and similarity matching mechanism. It converts rising temperature and humidity curves into digital fingerprints, effectively compressing data dimensions and enabling rapid retrieval and comparison while preserving fluctuation characteristics, significantly reducing data processing volume. A similarity scoring method based on timestamp alignment and sum-of-squares calculations captures subtle differences in temperature and humidity change patterns, avoiding the omissions and false alarms that are common with traditional threshold judgments. When new events are highly similar to historical events, historical decision parameters are directly called upon, enabling the system to reuse experience and shortening anomaly response time.

[0092] One embodiment of the present invention provides a temperature and humidity monitoring system for cold chain logistics, the temperature and humidity monitoring system comprising:

[0093] Real-time acquisition system: real-time acquisition of temperature and humidity data in the vehicle compartment and door opening and closing status signals;

[0094] Fluctuation event recognition system: uses sliding window temperature and humidity data to identify instantaneous fluctuation events of temperature and humidity data;

[0095] Causal correlation analysis system: When the fluctuation event coincides with the door opening state in time, the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuation is calculated;

[0096] Dynamic alarm threshold system: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated;

[0097] Abnormal alarm system: When the temperature and humidity deviations meet the following conditions simultaneously: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, the temperature and humidity abnormal alarm is triggered.

[0098] The working principle and effect of the above technical solution are as follows: First, sensors collect real-time temperature and humidity data within the vehicle compartment, as well as door opening and closing status signals, and process the temperature and humidity data using a sliding window technique. When a fluctuation event coincides with a door opening state, the system calculates the causal correlation strength coefficient by counting parameters such as the proportion of door opening time, the temperature and humidity change during this period, and the temperature and humidity recovery time. This quantitatively determines whether the temperature and humidity fluctuation is caused by the door opening. If the coefficient does not exceed the threshold, an abnormality likely exists that is not caused by loading or unloading. An audible and visual alarm is immediately issued, and a rapid diagnosis is initiated to assess refrigerant pressure and vehicle compartment airtightness. If the coefficient exceeds the threshold, the fluctuation is likely caused by loading or unloading, and a phased dynamic alarm threshold mechanism is activated, dynamically adjusting the alarm threshold by time period. Finally, the current dynamic threshold and the predicted recovery time based on the cargo specific heat capacity, the vehicle compartment heat transfer coefficient, and the effective heat transfer area are combined to determine whether the temperature and humidity deviation meets the abnormal alarm conditions. If the temperature and humidity continuously exceed the dynamic threshold for a period exceeding the predicted recovery time, an abnormal alarm is triggered.

[0099] The calculation of the causal correlation strength coefficient realizes the distinction between normal loading and unloading fluctuations and equipment abnormalities, reducing the false alarm rate caused by traditional fixed thresholds; the phased dynamic alarm threshold mechanism flexibly adjusts the threshold according to the fluctuation process, which not only avoids false triggering of transient fluctuations, but also ensures timely response to real abnormalities such as equipment failures; the recovery time is predicted based on cargo characteristics and carriage parameters, making the alarm judgment more in line with actual working conditions and reducing the risk of missed reports; digital fingerprint and similarity matching technology greatly shortens the abnormal response time and improves processing efficiency by reusing historical experience.

[0100] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A temperature and humidity detection method for cold chain logistics, characterized in that: The temperature and humidity detection method comprises: S1: Real-time collection of temperature and humidity data in the vehicle cabin and door opening and closing status signals; S2: Slide the window temperature and humidity data and identify the instantaneous fluctuation events of the temperature and humidity data; S3: When the fluctuation event coincides with the door opening state in time, calculate the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuations; S4: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated; S5: When the temperature and humidity deviations simultaneously meet the following conditions: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, a temperature and humidity anomaly alarm is triggered.

2. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The instantaneous fluctuation time of S2 includes: the rate of increase of temperature and humidity, the total duration of the fluctuation, and the time it takes for the temperature and humidity to return to the baseline.

3. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The sliding window detection of S2 includes: Set up a sliding window including short-term window, medium-term window and long-term window. When the temperature and humidity rising rate in any window exceeds the preset threshold of each window, it is marked as the starting point of the fluctuation event. The inflection point where the temperature and humidity gradient changes from positive to negative is taken as the peak point of the potential fluctuation event; The end point of the fluctuation event was when the temperature and humidity stabilized within the baseline.

4. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The method for calculating the causal correlation strength coefficient of S3 includes: Calculate the ratio of the door opening time to the total duration of the fluctuation event; Calculate the temperature and humidity changes during the door opening period; Calculate the time it takes for temperature and humidity to recover from peak values ​​to baseline; All parameters are integrated to output a causal correlation strength coefficient with a scale of 0-1.

5. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: When the causal correlation strength coefficient does not exceed the threshold, the step S4 continuously monitors the temperature and humidity data in the vehicle compartment, specifically including: A: Immediately sound and light alarm; B: Start quick diagnosis, estimate the refrigerant pressure through virtual pressure, evaluate the cabin air tightness through temperature and humidity recovery rate, and notify the driver of the diagnosis results.

6. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The S4 staged dynamic alarm threshold mechanism includes: The first time period is after the peak of the fluctuation Seconds, the first level dynamic alarm threshold is the standard threshold floating ℃; The second time period is after the peak of the fluctuation Seconds, the second level dynamic alarm threshold is the upper limit of the standard threshold ℃; The third time period is after the peak of fluctuation seconds to restore the standard threshold.

7. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The recovery time predicted by the historical recovery data of S5 is obtained as follows: Calculate the specific heat capacity of cargo based on its mass and type; Match the preset heat transfer coefficient through the carriage ID; Obtain the effective heat exchange area after loading through laser point cloud scanning; The recovery time predicted by historical recovery data is calculated based on all parameter fitting.

8. A temperature and humidity detection method for cold chain logistics according to claim 1, characterized in that: The S5 further includes: Discretize the temperature and humidity rising curve of the fluctuation event into multiple characteristic sequences; Generate a digital fingerprint of the feature sequence by a shift hash algorithm; When the similarity between the digital fingerprint of the new fluctuation event and the digital fingerprint of the historical fluctuation event exceeds the similarity threshold, the decision parameters of the associated historical fluctuation event are directly called.

9. A temperature and humidity detection method for cold chain logistics according to claim 8, characterized in that: The similarity is obtained by the following method: Align the temperature and humidity rising curves of new fluctuation events with those of historical fluctuation events; Calculate the sum of squares of temperature difference and humidity difference at the corresponding timestamps; The similarity score of 0-100% is obtained through normalization.

10. A temperature and humidity monitoring system for cold chain logistics, characterized in that: The temperature and humidity full-process monitoring system includes: Real-time acquisition system: real-time acquisition of temperature and humidity data in the vehicle compartment and door opening and closing status signals; Fluctuation event recognition system: uses sliding window temperature and humidity data to identify instantaneous fluctuation events of temperature and humidity data; Causal correlation analysis system: When the fluctuation event coincides with the door opening state in time, the causal correlation strength coefficient between the door opening and the temperature and humidity fluctuation is calculated; Dynamic alarm threshold system: When the causal correlation strength coefficient does not exceed the threshold, the temperature and humidity data in the vehicle compartment are continuously monitored; when the causal correlation strength coefficient exceeds the threshold, the staged dynamic alarm threshold mechanism is activated; Abnormal alarm system: When the temperature and humidity deviations meet the following conditions simultaneously: continuously exceeding the dynamic alarm threshold of the current stage, and lasting longer than the recovery time predicted based on historical recovery data, the temperature and humidity abnormal alarm is triggered.

Images