CN-122022647-A - Cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion

CN122022647ACN 122022647 ACN122022647 ACN 122022647ACN-122022647-A

Abstract

The invention provides a cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion, which relates to the technical field of cold chain transportation monitoring, and comprises the steps of constructing a temperature field topology network by collecting sensor data in a cold chain transportation carriage, calculating a temperature ripple propagation path based on the temperature field topology network, analyzing the complexity characteristic of the ripple propagation path through a map entropy algorithm, correcting the complexity characteristic by combining humidity sensor data, acquiring a steady state characteristic value of a temperature field, further extracting a propagation mode of a temperature ripple effect, and identifying the temperature control abnormality by adopting a network synchronism analysis method; and (3) combining the carriage door opening and closing state sensor data, predicting the diffusion trend of the temperature abnormality by adopting a temperature field reconstruction algorithm, and outputting temperature control abnormality early warning information. The invention greatly improves the timeliness and reliability of the temperature control abnormality early warning, effectively improves the accuracy of temperature control abnormality identification, and better ensures the quality and safety of cold chain transported goods.

Inventors

ZHENG QUANJUN
ZHANG SONGTAO
GU CHANGQING
ZHENG LULU
CHEN LEI
YANG HAINING
LI WENJIAO
ZHANG LEI

Assignees

山东荣庆物流科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (10)

1. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion is characterized by comprising the following steps of: Collecting sensor data in a cold chain transportation carriage, constructing a temperature field topology network based on the spatial distribution of temperature sensors by using a Deltay triangulation method, calculating the temperature association strength of adjacent nodes in the network, and constructing a weighted topology model; calculating a temperature ripple propagation path based on the temperature field topology network, analyzing the complexity characteristic of the ripple propagation path through a map entropy algorithm, and correcting the complexity characteristic by combining humidity sensor data to obtain a steady state characteristic value of the temperature field; Based on the steady state characteristic value, extracting a propagation mode of a temperature ripple effect, and identifying temperature control abnormality by adopting a network synchronism analysis method; Based on the temperature control abnormality, the temperature field reconstruction algorithm is adopted to predict the diffusion trend of the temperature abnormality by combining with the carriage door opening and closing state sensor data, and the temperature control abnormality early warning information is output.
2. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein the construction of the temperature field topology network is based on a delusian triangulation method to generate a node connection relation of a three-dimensional space, and the edge weights among the nodes are obtained through pearson correlation coefficient calculation.
3. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein the propagation resistance coefficient of the temperature ripple propagation path is calculated by the following formula: , Wherein, the Is a node And The edge weight of the edge between the two, As a directional weight for the temperature propagation, Is a node Is used for the temperature disturbance intensity of the (c), Is a node Is used for the temperature disturbance intensity of the (c), Is a time decay factor.
4. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein the formula for calculating the complexity characteristics of the ripple propagation path by using the map entropy algorithm is as follows: , Wherein, the In order to normalize the characteristic values, For the total number of nodes in the temperature field topology network, As a value of a characteristic of the local structure, Represent the first And characteristic values.
5. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein a humidity correction coefficient is introduced based on humidity sensor data, and a calculation formula of the humidity correction coefficient is as follows: , Wherein, the And Is a node And The relative humidity value at which the temperature and humidity values, For the reference humidity difference value, For adjusting the coefficients.
6. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 5, wherein a steady state characteristic value is calculated based on the humidity correction coefficient in combination with a complexity characteristic, and a formula of the complexity characteristic is expressed as: , Wherein, the In order to normalize the characteristic values, Is the first The value of the characteristic is a value of, As a value of a characteristic of the local structure, Is the total number of monitoring points.
7. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 6, wherein the calculation formula of the steady state characteristic value is expressed as: , Wherein, the As a final temperature field steady state characteristic value, To normalize the trace of the adjacency matrix, To normalize the first in the adjacency matrix Row of lines Column elements.
8. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein the temperature field reconstruction algorithm predicts the diffusion trend of temperature abnormality through a probability field model, and the probability field model is: , Wherein, the In the form of a spatial coordinate system, As the location of the source of the anomaly, In order for the diffusion coefficient to be the same, As a result of the normalization factor, The weight is influenced for the status of the vehicle door, Is a reference characteristic value when the system is operating normally, Is an influence function of the state of the vehicle door.
9. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 8 is characterized in that early warning information of temperature control abnormality is divided into three stages of short-term early warning, medium-term early warning and long-term early warning, wherein the short-term early warning is based on a temperature gradient change rate and a temperature abnormality diffusion rate, the medium-term early warning is based on a temperature standard deviation and a cold energy loss rate, and the long-term early warning is based on characteristic values of temperature difference stability and insufficient refrigerating capacity.
10. The cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion according to claim 1, wherein the prediction of the temperature abnormality diffusion trend is combined with the sensor data of the door opening and closing state of the carriage, and the door opening and closing time and the opening area of the door state are used for dynamically correcting the temperature diffusion model.

Description

Cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion Technical Field The invention relates to the technical field of cold chain transportation monitoring, in particular to a cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion. Background With the rapid development of global cold chain logistics industry, cold chain transportation technology has become an important guarantee means for transporting high-value goods such as foods, medicines and biological products. The core goal of cold chain transportation is to ensure the stability of the temperature and humidity environment required by the goods in the transportation process, thereby avoiding the phenomenon of the quality reduction or even scrapping of the goods caused by the failure of temperature control. At present, temperature monitoring in a cold chain transportation carriage mainly depends on real-time acquisition data of distributed temperature sensors, and the data realize monitoring of the temperature of the cold chain carriage through a centralized control system. However, the conventional monitoring method usually only focuses on single-point temperature change or simple statistical mean analysis, and fails to fully consider the spatial distribution characteristics and dynamic evolution rules of the temperature field in the compartment. In a complex transportation environment with multiple temperature areas and multiple cargo types, temperature control abnormality gradually affects the internal environment of the whole carriage in a diffusion or local ripple propagation mode, so that in a traditional method, the identification of the temperature control abnormality usually has hysteresis, and the temperature abnormality propagation trend in the complex environment cannot be reflected timely and accurately. In addition, the influence of other environmental factors such as humidity, carriage door state and the like on temperature abnormality is not fully considered, so that the prior art has great limitation on the accuracy and early warning efficiency of abnormality identification. The existing cold chain temperature control monitoring technology still has a plurality of defects in the aspects of real-time monitoring and dynamic early warning, namely, firstly, the existing technology mostly adopts a simple threshold comparison or statistical analysis method of single temperature sensor data, lacks global modeling capability of a temperature field in a carriage, is difficult to reflect a propagation rule of temperature anomaly in space, secondly, the existing system mostly ignores complexity of temperature anomaly and multi-factor coupling effect, particularly, the existing technology cannot effectively fuse multi-sensor data under the scene that external environment parameters such as humidity, carriage door switch and the like have significant influence on a temperature control state, so that the applicability of an early warning model is poor, and furthermore, the existing technology lacks a deep analysis means of a temperature anomaly propagation mode, and cannot effectively characterize a propagation path of temperature disturbance and dynamic characteristics of the propagation path under a complex environment. These deficiencies not only reduce the efficiency of identifying temperature control anomalies, but may also result in irreversible loss of cargo due to temperature control failure. Disclosure of Invention The present invention has been made to solve the above-mentioned technical problems. The invention provides a cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion, which can solve the problems of temperature control abnormality recognition lag caused by incapability of accurately acquiring the spatial distribution and dynamic evolution rule of a temperature field in a cold chain transportation carriage, and inaccurate temperature control abnormality early warning caused by insufficient consideration of the influence of factors such as humidity, carriage door opening and closing states and the like on temperature disturbance to a certain extent. According to one aspect of the invention, a cold chain transportation temperature control abnormality early warning method based on multi-sensor data fusion is provided, which comprises the following steps: Collecting sensor data in a cold chain transportation carriage, constructing a temperature field topology network based on the spatial distribution of temperature sensors by using a Deltay triangulation method, calculating the temperature association strength of adjacent nodes in the network, and constructing a weighted topology model; calculating a temperature ripple propagation path based on the temperature field topology network, analyzing the complexity characteristic of the ripple propagation path through a m