Remote temperature calibration method and system for vaccine cold chain system

Abstract

The invention discloses a vaccine cold chain system remote temperature calibration method and system, and relates to the technical field of remote calibration, and the method comprises the steps: obtaining environment data; environment change features are extracted, dimensionality reduction is carried out through PCA, and temperature feature vectors are obtained; constructing a vaccine storage environment temperature prediction model; acquiring a real-time temperature value, performing cross verification on the real-time temperature value and the predicted temperature value, and judging whether the temperature sensor has deviation or not; based on the judgment result, a calibration program is automatically triggered to calibrate the temperature sensor; according to the method, the vaccine storage environment temperature prediction model is constructed and the temperature sensor is dynamically calibrated based on PCA dimension reduction purification key information, so that the limitation caused by pure dependence on single-point temperature reading in the traditional scheme is effectively overcome, the safety and reliability of the whole cold-chain logistics chain are enhanced, and the method is suitable for large-scale popularization and application. And a firmer guarantee is provided for maintaining the activity of the vaccine.

Description

Technical Field

[0001] The present invention relates to the field of remote calibration technology, and in particular to a remote temperature calibration method and system for a vaccine cold chain system. Background Art

[0002] With the development of biopharmaceutical technology, vaccine cold chain transportation and storage have become key links to ensure the effectiveness of vaccines. Temperature control is particularly important during the storage of vaccines, because inappropriate temperature fluctuations may cause the vaccine to fail, thus affecting public health and safety. Existing vaccine cold chain monitoring systems mostly use a single temperature sensor for real-time monitoring, and use a preset threshold alarm mechanism to prevent the temperature from deviating from the specified range. However, such systems usually have several significant defects: First, the impact of environmental changes (such as humidity, light, etc.) is not sufficiently considered, and the actual temperature conditions cannot be accurately reflected; second, there is a lack of effective detection and automatic calibration mechanism for the temperature sensor's own drift or failure. Once the sensor deviates, it may lead to erroneous data reports without being discovered in time. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a remote temperature calibration method and system for a vaccine cold chain system to solve the problem of measurement errors caused by environmental factors and the degradation of the temperature sensor's own performance.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a method for remote temperature calibration of a vaccine cold chain system, which comprises deploying a sensor array, acquiring environmental data, and performing data cleaning;

[0007] Extract environmental change features and use PCA to reduce dimension to obtain temperature feature vectors;

[0008] Construct a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment;

[0009] Get the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value to determine whether the temperature sensor has deviation;

[0010] Based on the judgment result, the calibration program is automatically triggered to calibrate the temperature sensor. The current ambient temperature is collected through the calibrated temperature sensor to achieve temperature calibration.

[0011] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, wherein: the sensor array includes a temperature sensor, a pressure sensor, a light sensor and a humidity sensor;

[0012] The environmental data includes temperature, humidity, air pressure and light intensity.

[0013] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, wherein: the data cleaning is performed as follows:

[0014] All sensor data are timestamped, the Z-score method is applied to identify outliers, outliers are marked, and marked outliers are replaced with sliding averages. Interpolation is used to handle missing data points of sensors. Data from different sensors are converted into a unified data format and timestamps are added.

[0015] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, the environmental change characteristics are extracted and PCA is used for dimension reduction to obtain the environmental feature vector. The specific steps are as follows:

[0016] Environmental change characteristics are calculated for the temperature, humidity, pressure, and light intensity at each time point, and the difference between the temperature, humidity, pressure, and light intensity at the previous time point is then divided by the time interval to obtain the rate of change of temperature, humidity, pressure, and light intensity over time;

[0017] The extracted environmental change features are standardized to have the same scale. Based on the standardized feature data, first, the covariance matrix of feature X is calculated, and then the eigenvalue Λ and eigenvector matrix V are obtained by eigendecomposition, which is expressed as,

[0018]

[0019] C(X)=VΛV T

[0020] Where C(X) is the covariance matrix, X is the data matrix with samples as rows and features as columns, n is the number of samples, and X T is the transpose of X;

[0021] Select the column vectors corresponding to the largest k eigenvalues ​​in Λ to form the matrix V k , and select the first k largest eigenvalues ​​from Λ to form the matrix Λ k , expressed as,

[0022] V k =V[:,:k]

[0023] Λ k =Λ[:k]

[0024] Among them, k represents the number of features after dimensionality reduction, [:,:k] means selecting the first k columns from all rows of V, V kis the matrix of the first k eigenvalues, [:k] means selecting the first k eigenvalues ​​from all rows of Λ, Λ k is the diagonal matrix of the first k eigenvalues;

[0025] The temperature feature vector F is expressed as,

[0026]

[0027] Here, X represents the data matrix.

[0028] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, wherein: the vaccine storage environment temperature prediction model is constructed to predict the temperature fluctuation trend of the vaccine storage environment, and the specific steps are as follows:

[0029] Based on the temperature feature vector F, a vaccine storage environment temperature prediction model is constructed, and the expression is:

[0030]

[0031] Among them, T pred (t) is the predicted temperature value at time t, n is the total number of features in the temperature feature vector F, and w i is the weight coefficient, g(E(τ)) is the environmental impact function, E(τ) is the environmental data vector at the historical time point τ, τ represents the historical time point, exponential decay term, ρ is the time decay constant;

[0032] The environmental impact function is expressed as,

[0033]

[0034] Among them, α represents the weight coefficient of the ambient temperature logarithm term, β represents the weight coefficient of the ambient humidity power law term, γ represents the weight coefficient of the periodic effect of light intensity, δ represents the weight coefficient of the temperature and humidity interaction term, η represents the weight coefficient of the ratio of ambient humidity to corrected light intensity, κ represents the smoothing factor of ambient temperature, ε represents the smoothing factor of the ratio of ambient humidity to corrected light intensity, φ is the frequency of controlling the periodic effect of light intensity, E1(T) is the ambient temperature at the historical time point τ, E2(τ) is the ambient humidity at the historical time point T, and E3(T) is the light intensity at the historical time point τ;

[0035] For each time point t i , apply the vaccine storage environment temperature prediction model, calculate the predicted temperature, and record the predicted temperature T at each time point pred (t i ), construct the time series T pred =[T pred (t1), T pred(t2), ..., T pred (t n )], using the time series T pred Draw a graph to visually show the temperature change trend over time.

[0036] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, wherein: the real-time temperature value is obtained, the real-time temperature value is cross-checked with the predicted temperature value, and whether the temperature sensor has a deviation is determined, and the specific steps are as follows:

[0037] Define the collected real-time temperature value T real (t) and transmit the real-time data collected by the temperature sensor to a remote server;

[0038] The real-time temperature value is cross-checked with the predicted temperature value, expressed as,

[0039] D(t)=T real (t)-T pred (t)

[0040] Among them, T pred (t) represents the predicted temperature value at time t, T real (t) represents the real-time temperature value at time t, and D(t) represents the deviation at time t;

[0041] Set the warning threshold of the deviation D(t) to ±A℃;

[0042] When D(t)<±A℃, it means the temperature sensor is normal;

[0043] When D(t)≥±A℃, it indicates that the temperature sensor is abnormal.

[0044] As a preferred solution of the remote temperature calibration method of the vaccine cold chain system of the present invention, wherein: based on the judgment result, the calibration program is automatically triggered to calibrate the temperature sensor, and the current ambient temperature is collected by the calibrated temperature sensor to achieve temperature calibration. The specific steps are as follows:

[0045] When the sensor is normal, the vaccine cold chain system does not need to calibrate the temperature sensor and directly uses the temperature sensor to collect the current ambient temperature;

[0046] When the sensor is abnormal, the vaccine cold chain system immediately adjusts the calibration coefficient of the temperature sensor, receives instructions from the server, and corrects its measurement error. After the calibration is completed, the vaccine cold chain system continuously monitors the output value of the temperature sensor and compares it with the prediction model.

[0047] In a second aspect, the present invention provides a remote temperature calibration system for a vaccine cold chain system, including a data acquisition module for deploying a sensor array, acquiring environmental data, and performing data cleaning;

[0048] The data processing module is used to extract the characteristics of environmental changes and use PCA to reduce the dimension to obtain the environmental feature vector;

[0049] The environmental temperature prediction model building module is used to build a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment;

[0050] A cross-check module is used to obtain the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value, and determine whether the temperature sensor has a deviation;

[0051] The intelligent calibration control module automatically triggers the calibration procedure based on the judgment result, calibrates the temperature sensor, and collects the current ambient temperature through the calibrated temperature sensor to achieve temperature calibration.

[0052] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the remote temperature calibration method for a vaccine cold chain system as described in the first aspect of the present invention is implemented.

[0053] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the remote temperature calibration method for a vaccine cold chain system as described in the first aspect of the present invention.

[0054] The beneficial effects of the present invention are as follows: by processing the data collected by the sensor array deployed in the cold chain environment, including data cleaning to eliminate noise interference, feature extraction and using principal component analysis (PCA) to achieve dimensionality reduction and simplify the model input, a mathematical model capable of predicting the temperature change trend in the future is constructed. This model not only takes into account the changing law of the temperature itself, but also combines the possible comprehensive effects of factors such as humidity, air pressure and light intensity on the temperature. When there is a significant difference between the actual monitored temperature and the predicted value, it indicates that there may be a sensor misalignment. At this time, the system will automatically start the calibration process to ensure the accuracy of the recorded data. This method effectively overcomes the limitations of the traditional solution that relies solely on single-point temperature readings, enhances the safety and reliability of the entire cold chain logistics chain, and provides a more solid guarantee for maintaining the activity of vaccines. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0056] Figure 1 This is a flow chart of the remote temperature calibration method for the vaccine cold chain system in Example 1. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a method for remote temperature calibration of a vaccine cold chain system, comprising the following steps:

[0061] S1, deploy sensor array, obtain environmental data, and perform data cleaning, including the following steps:

[0062] The sensor array includes temperature sensors, pressure sensors, light sensors, and humidity sensors;

[0063] Environmental data includes temperature, humidity, air pressure, and light intensity.

[0064] All sensor data are timestamped, the Z-score method is applied to identify outliers, outliers are marked, and marked outliers are replaced with sliding averages. Interpolation is used to handle missing data points of sensors. Data from different sensors are converted into a unified data format and timestamps are added.

[0065] It should be noted that by deploying a highly sensitive sensor array and implementing a sophisticated data cleaning process, this step ensures the accuracy and consistency of the collected environmental data. The application of timestamp alignment and outlier identification technology effectively eliminates noise interference in the data, while sliding average substitution and interpolation methods make up for data missing and enhance data continuity. The addition of a unified data format and timestamps improves data processing efficiency and the convenience of subsequent analysis.

[0066] S2. Extract environmental change features and use PCA to reduce the dimension to obtain environmental feature vectors. The specific steps are as follows:

[0067] Environmental change characteristics are calculated for the temperature, humidity, pressure, and light intensity at each time point, and the difference between the temperature, humidity, pressure, and light intensity at the previous time point is then divided by the time interval to obtain the rate of change of temperature, humidity, pressure, and light intensity over time;

[0068] The extracted environmental change features are standardized to have the same scale. Based on the standardized feature data, the covariance matrix of feature X is calculated to reflect the linear correlation between different features. Then, feature decomposition is performed to obtain the eigenvalue Λ and eigenvector matrix V, which are expressed as,

[0069]

[0070] C(X)=VΛV T

[0071] Where C(X) is the covariance matrix, which is used to represent the linear correlation between different features, X is the data matrix, with samples as rows and features as columns, n is the number of samples, and X T X is transposed, which is used to exchange features and samples;

[0072] It should be noted that each element in the covariance matrix represents the covariance between two features. If the element is positive, it means that the two features are positively correlated; if it is negative, it means negative correlation; if it is close to zero, it means that the two features have almost no correlation. The eigenvalue matrix is ​​a diagonal matrix whose diagonal elements are the eigenvalues ​​of the covariance matrix, representing the variance or degree of variation of the data in a specific direction. In principal component analysis (PCA), the size of the eigenvalue determines which principal components are selected. The directions corresponding to larger eigenvalues ​​are more important and can be retained, while the principal components corresponding to smaller eigenvalues ​​can be ignored, thereby achieving dimensionality reduction; each column of the eigenvector matrix is ​​an eigenvector of the covariance matrix. The eigenvectors can be used to project the data onto these new basis vectors for principal component analysis or feature extraction;

[0073] Select the column vectors corresponding to the largest k eigenvalues ​​in Λ to form the matrix V k, and select the first k largest eigenvalues ​​from Λ to form the matrix Λ k , expressed as,

[0074] V k =V[:,:k]

[0075] Λ k =Λ[:k]

[0076] Among them, k represents the number of features after dimensionality reduction, [:,:k] means selecting the first k columns from all rows of V, V k is the matrix of the first k eigenvalues, each column is an eigenvector of the covariance matrix C(X), [:k] means selecting the first k eigenvalues ​​from all rows of Λ, Λ k It is the diagonal matrix of the first k eigenvalues, and the eigenvalue represents the variance of the data in the direction of the corresponding eigenvector;

[0077] It should be noted that the matrix V k Each eigenvector in represents the change of the data in an independent direction. When selecting the first k eigenvectors, we are actually selecting the first k principal component directions of the data. These principal component directions are those that maximize the variance of the data (i.e., the directions in which the data changes most significantly), so these directions can effectively describe the main structure of the data. The matrix Λ k The eigenvalues ​​in reflect the degree of variation of the data in each principal component direction. Selecting the first k largest eigenvalues ​​means selecting the directions that explain the most data variance. By retaining the first k eigenvalues, the main information of the data can be retained as much as possible during the dimensionality reduction process, and those directions that contribute less to the data variance can be ignored.

[0078] The temperature feature vector F is expressed as,

[0079]

[0080] Among them, X represents the data matrix. By multiplying X by the eigenvector matrix and the eigenvalue matrix, the original data can be projected into a new feature space, thereby extracting the main features of the data.

[0081] It should be noted that the use of PCA dimensionality reduction technology combined with feature extraction of environmental changes has achieved effective compression of data dimensions through standardization of features and refinement of feature vectors.

[0082] S4. Construct a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment. The specific steps are as follows:

[0083] Based on the temperature feature vector F, a vaccine storage environment temperature prediction model is constructed, and the expression is:

[0084]

[0085] Among them, T pred (t) is the predicted temperature value at time t, which indicates the temperature prediction of the vaccine storage environment at time t. n is the total number of features in the environmental feature vector F, which indicates how many environmental features (such as temperature, humidity, light, etc.) affect the temperature prediction. w i is the weight coefficient, which is used to measure the impact of each feature on temperature prediction. The weight of each feature is w i Determines the contribution of this feature to the final temperature prediction. A higher weight means that this feature has a more important impact on temperature changes. g(E(τ)) is the environmental impact function, which is used to describe the impact of environmental data at historical time τ on temperature. E(τ) is the environmental data vector at historical time point τ, which contains historical environmental characteristics such as temperature, humidity, and light intensity. It is used to calculate the impact of the environment on the current temperature, thereby reflecting the dynamic impact of historical data. τ represents the historical time point. The exponential decay term, ρ, is the time decay constant, which is used to control the decay rate of the impact of historical data on the current forecast;

[0086] It should be noted that It represents the cumulative impact of historical environmental data on the predicted temperature at the current time t. By integrating the historical environmental data over time, it reflects the comprehensive impact of the environment on the current temperature at each time τ in the past. The exponential decay term ensures that the data at a distant time point has less impact. Represents the sum of feature weights, which is a normalization term used to normalize the feature weights to ensure that the sum of the weights is 1, so that the predicted value represents a reasonable temperature range.

[0087] The environmental impact function is expressed as,

[0088]

[0089] Among them, α represents the weight coefficient of the logarithmic term of ambient temperature, which controls the influence of ambient temperature on temperature prediction after logarithmic transformation; β represents the weight coefficient of the power law term of ambient humidity, which controls the influence of power law transformation of ambient humidity on temperature; γ represents the weight coefficient of the periodic influence of light intensity, which controls the periodic influence of light intensity on temperature prediction; δ represents the weight coefficient of the interaction term of temperature and humidity, which describes the mutual influence of temperature and humidity; η represents the weight coefficient of the ratio term of ambient humidity and corrected light intensity, which describes the influence of the ratio between humidity and light intensity on temperature; κ represents The smoothing factor of the ambient temperature is added before taking the logarithm of the ambient temperature to avoid logarithmic operation errors when the temperature is non-positive and ensure the numerical stability of the model. ε represents the smoothing factor of the ratio of ambient humidity to corrected light intensity to avoid the problem of the denominator being zero. φ is the frequency of controlling the periodic effect of light intensity, which controls the frequency of the periodic effect of light intensity on temperature and reflects the diurnal periodic change of light. E1(τ) is the ambient temperature at the historical time point τ, E2(τ) is the ambient humidity at the historical time point τ, and E3(τ) is the light intensity at the historical time point T.

[0090] It should be noted that the environmental impact function designs a complex model by integrating the nonlinear and interactive effects of factors such as temperature, humidity, and light, combined with periodicity and detail adjustments, aiming to accurately reflect the comprehensive impact of environmental factors on biological or emotional states;

[0091] For each time point t i , apply the vaccine storage environment temperature prediction model, calculate the predicted temperature, and record the predicted temperature T at each time point pred (t i ), construct the time series T pred =[T pred (t1), T pred (t2), ..., T pred (t n )], using the time series T pred Draw a graph to visually show the temperature change trend over time.

[0092] It should be noted that the constructed vaccine storage environment temperature prediction model, by designing the environmental impact function, not only takes into account the current environmental status, but also fully integrates the dynamic impact of historical environmental data, forming a forward-looking temperature fluctuation prediction framework.

[0093] S5. Obtain the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value, and determine whether the temperature sensor has deviation. The specific steps are as follows:

[0094] The real-time temperature value collected by the temperature sensor at time t is defined as T real (t);

[0095] Predict the temperature value at time t through the constructed temperature prediction model (based on historical environmental data);

[0096] The real-time temperature value is cross-checked with the predicted temperature value, expressed as,

[0097] D(t)=T real (t)-T pred (t)

[0098] Among them, T pred (t) represents the predicted temperature value at time t, T real (t) represents the real-time temperature value at time t, and D(t) represents the deviation at time t;

[0099] Set the warning threshold value of the deviation D(t) ±A℃ to ±0.5℃;

[0100] When D(t)<±0.5℃, it means the temperature sensor is normal;

[0101] When D(t)≥±0.5℃, it indicates that the temperature sensor is abnormal and an early warning is issued.

[0102] It should be noted that the cross-check mechanism between real-time temperature values ​​and predicted temperature values ​​provides immediate supervision and feedback for the safety of the vaccine storage environment. By setting a reasonable warning threshold, this step can quickly identify temperature deviations and initiate response measures in a timely manner, greatly reducing the possibility of vaccine failure due to temperature out of control.

[0103] S6. Based on the judgment result, the calibration procedure is automatically triggered to calibrate the temperature sensor. The current ambient temperature is collected by the calibrated temperature sensor to achieve temperature calibration, including the following steps:

[0104] When the sensor is normal, the vaccine cold chain system does not need to calibrate the temperature sensor. The temperature sensor works normally and the real-time temperature measured by the sensor is directly used as the current ambient temperature for temperature control and monitoring in the cold chain system. The normal temperature data of the sensor will be fed back to the system to update the temperature prediction model and improve the accuracy of future temperature predictions.

[0105] When the sensor is abnormal, the vaccine cold chain system automatically triggers the calibration procedure, receives the calibration instructions from the server through the remote control module, enters the calibration process, immediately adjusts the calibration coefficient of the temperature sensor, receives instructions from the server, and corrects its measurement error. After the calibration is completed, the vaccine cold chain system continuously monitors the output value of the temperature sensor, compares it with the prediction model, calculates the calibrated error, and ensures that the calibrated sensor error is always within an acceptable range. If the calibrated error still exceeds the warning threshold, the calibration is triggered again or a warning is issued to notify the administrator that the sensor may have a hardware failure.

[0106] It should be noted that the calculation of the calibration coefficient is based on the trend, historical data and real-time data of the sensor error. The server will calculate the appropriate calibration coefficient based on these data and send it to the local controller.

[0107] It should be noted that the automatically triggered calibration procedure achieves rapid response and automatic adjustment of the vaccine storage environment temperature based on the deviation judgment result, ensuring immediate action when temperature anomalies are detected, effectively preventing the normal operation of the temperature sensor.

[0108] This embodiment also provides a vaccine cold chain system remote temperature calibration system, including a data acquisition module for deploying a sensor array, acquiring environmental data, and performing data cleaning;

[0109] The data processing module is used to extract the characteristics of environmental changes and use PCA to reduce the dimension to obtain the temperature feature vector;

[0110] The environmental temperature prediction model building module is used to build a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment;

[0111] A cross-check module is used to obtain the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value, and determine whether the temperature sensor has a deviation;

[0112] The intelligent calibration control module automatically triggers the calibration procedure based on the judgment result, calibrates the temperature sensor, and collects the current ambient temperature through the calibrated temperature sensor to achieve temperature calibration.

[0113] This embodiment also provides a computer device, which is suitable for the case of a remote temperature calibration method for a vaccine cold chain system, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the remote temperature calibration method for a vaccine cold chain system proposed in the above embodiment.

[0114] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0115] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by the processor, it implements the remote temperature calibration method of the vaccine cold chain system proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination of them, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0116] In summary, the beneficial effect of the present invention is to process the data collected by the sensor array deployed in the cold chain environment, including data cleaning to eliminate noise interference, feature extraction and use of principal component analysis (PCA) to achieve dimensionality reduction and simplify model input, and then construct a mathematical model that can predict the temperature change trend in the future. This model not only takes into account the changing law of temperature itself, but also combines the possible comprehensive effects of factors such as humidity, air pressure and light intensity on temperature. When there is a significant difference between the actual monitored temperature and the predicted value, it indicates that there may be a sensor misalignment. At this time, the system will automatically start the calibration process to ensure the accuracy of the recorded data. This method effectively overcomes the limitations of the traditional solution that relies solely on single-point temperature readings, enhances the safety and reliability of the entire cold chain logistics chain, and provides a more solid guarantee for maintaining vaccine activity.

[0117] It should be noted that 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A remote temperature calibration method for a vaccine cold chain system, characterized in that: include, Deploy sensor arrays, acquire environmental data, and perform data cleaning; Extract environmental change features and use PCA to reduce dimension to obtain temperature feature vectors; Construct a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment; Get the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value to determine whether the temperature sensor has deviation; Based on the judgment result, the calibration program is automatically triggered to calibrate the temperature sensor. The current ambient temperature is collected through the calibrated temperature sensor to achieve temperature calibration.

2. The remote temperature calibration method for a vaccine cold chain system according to claim 1, characterized in that: The sensor array includes a temperature sensor, a pressure sensor, a light sensor and a humidity sensor; The environmental data includes temperature, humidity, air pressure and light intensity.

3. The remote temperature calibration method for a vaccine cold chain system according to claim 2, characterized in that: The data cleaning is performed as follows: The environmental data are timestamped, the Z-score method is applied to identify outliers, outliers are marked, and the marked outliers are replaced with sliding averages. The missing data points of the sensors are processed by interpolation. The data from different sensors are converted into a unified data format and timestamps are added.

4. The remote temperature calibration method for a vaccine cold chain system according to claim 3, characterized in that: The environmental change features are extracted and PCA is used for dimensionality reduction to obtain the environmental feature vector. The specific steps are as follows: Environmental change characteristics are calculated for the temperature, humidity, pressure, and light intensity at each time point, and the difference between the temperature, humidity, pressure, and light intensity at the previous time point is then divided by the time interval to obtain the rate of change of temperature, humidity, pressure, and light intensity over time; The extracted temperature trend features and environmental change features are standardized. Based on the standardized feature data, first, the covariance matrix of feature X is calculated, and then the feature decomposition is performed to obtain the eigenvalue Λ and the eigenvector matrix V, which are expressed as, C(X)=VAV T Where C(X) is the covariance matrix, X is the data matrix with samples as rows and features as columns, n is the number of samples, and X T is the transpose of X; Select the column vectors corresponding to the largest k eigenvalues ​​in Λ to form the matrix V k , and select the first k largest eigenvalues ​​from Λ to form the matrix Λ k , expressed as, V k =V[:,:k] L k =Λ[:k] Among them, k represents the number of features after dimensionality reduction, [:,:k] means selecting the first k columns from all rows of V, V k is the matrix of the first k eigenvalues, [:k] means selecting the first k eigenvalues ​​from all rows of Λ, Λ k is the diagonal matrix of the first k eigenvalues; The temperature feature vector F is expressed as, Here, X represents the data matrix.

5. The remote temperature calibration method for a vaccine cold chain system according to claim 4, characterized in that: The vaccine storage environment temperature prediction model is constructed to predict the temperature fluctuation trend of the vaccine storage environment. The specific steps are as follows: Based on the temperature feature vector F, a vaccine storage environment temperature prediction model is constructed, and the expression is: Among them, T pred (t) is the predicted temperature value at time t, n is the total number of features in the temperature feature vector F, and w i is the weight coefficient, g(E(τ)) is the environmental impact function, E(τ) is the environmental data vector at the historical time point τ, τ represents the historical time point, exponential decay term, ρ is the time decay constant; The environmental impact function is expressed as, Among them, α represents the weight coefficient of the logarithmic term of ambient temperature, β represents the weight coefficient of the power law term of ambient humidity, γ represents the weight coefficient of the periodic effect of light intensity, δ represents the weight coefficient of the interaction term of temperature and humidity, η represents the weight coefficient of the ratio of ambient humidity to corrected light intensity, κ represents the smoothing factor of ambient temperature, ε represents the smoothing factor of the ratio of ambient humidity to corrected light intensity, φ is the frequency of controlling the periodic effect of light intensity, E1(τ) is the ambient temperature at the historical time point T, E2(τ) is the ambient humidity at the historical time point T, and E3(τ) is the light intensity at the historical time point T; For each time point t i , apply the vaccine storage environment temperature prediction model, calculate the predicted temperature, and record the predicted temperature T at each time point pred (t i ), construct the time series T pred =[T pred (t1), T pred (t2), ..., T pred (t n )], using the time series T pred Draw a graph to visually show the temperature change trend over time.

6. The remote temperature calibration method for a vaccine cold chain system according to claim 5, characterized in that: The real-time temperature value is obtained, and the real-time temperature value is cross-checked with the predicted temperature value to determine whether the temperature sensor has a deviation. The specific steps are as follows: Define the collected real-time temperature value T real (t) and transmit the real-time data collected by the temperature sensor to a remote server; The real-time temperature value is cross-checked with the predicted temperature value, expressed as, D(t)=T real (t)-T pred (t) Among them, T pred (t) represents the predicted temperature value at time t, T real (t) represents the real-time temperature value at time t, and D(t) represents the deviation at time t; Set the warning threshold of the deviation D(t) to ±A℃; When D(t)<±A℃, it means the temperature sensor is normal; When D(t)≥±A℃, it indicates that the temperature sensor is abnormal.

7. The remote temperature calibration method for a vaccine cold chain system according to claim 6, characterized in that: Based on the judgment result, the calibration procedure is automatically triggered to calibrate the temperature sensor, and the current ambient temperature is collected by the calibrated temperature sensor to achieve temperature calibration. The specific steps are as follows: When the sensor is normal, the vaccine cold chain system does not need to calibrate the temperature sensor and directly uses the temperature sensor to collect the current ambient temperature; When the sensor is abnormal, the vaccine cold chain system immediately adjusts the calibration coefficient of the temperature sensor, receives instructions from the server, and corrects its measurement error. After the calibration is completed, the vaccine cold chain system continuously monitors the output value of the temperature sensor and compares it with the prediction model.

8. A vaccine cold chain system remote temperature calibration system, based on the vaccine cold chain system remote temperature calibration method according to any one of claims 1 to 7, characterized in that: include, Data acquisition module, used to deploy sensor arrays, obtain environmental data, and perform data cleaning; The data processing module is used to extract the characteristics of environmental changes and use PCA to reduce the dimension to obtain the environmental feature vector; The environmental temperature prediction model building module is used to build a vaccine storage environment temperature prediction model to predict the temperature fluctuation trend of the vaccine storage environment; A cross-check module is used to obtain the real-time temperature value, cross-check the real-time temperature value with the predicted temperature value, and determine whether the temperature sensor has a deviation; The intelligent calibration control module is used to automatically trigger the calibration program based on the judgment result, calibrate the temperature sensor, collect the current ambient temperature through the calibrated temperature sensor, and realize temperature calibration.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the vaccine cold chain system remote temperature calibration method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the remote temperature calibration method for a vaccine cold chain system according to any one of claims 1 to 7 are implemented.

Images