Search

CN-121980360-A - Container transportation state monitoring method based on multi-parameter sensing

CN121980360ACN 121980360 ACN121980360 ACN 121980360ACN-121980360-A

Abstract

The embodiment of the application provides a container transportation state monitoring method based on multi-parameter sensing, which is applied to the technical field of logistics monitoring and comprises the steps of taking a Hall switch trigger signal as a time reference and a logic starting point, and collecting acceleration, attitude angle, position and temperature and humidity signals in a time window before and after triggering; the method comprises the steps of carrying out time sequence association analysis on acceleration, attitude angle and position signals to generate initial event judgment, extracting impact attenuation time constant and attitude angle deviation maintaining quantity when the initial event is judged to be the start of physical door opening or hoisting, constructing a two-dimensional characteristic space to carry out mechanical impact classification, asynchronously retrieving temperature and humidity signals according to judgment results to carry out root-cause verification, generating joint event codes and executing differential resource control. The application can accurately distinguish trigger events with different properties, realizes risk classification of mechanical impact and root tracing of environmental abnormality, obviously prolongs the endurance time of the terminal and ensures the monitoring reliability.

Inventors

  • Zheng Chongqing
  • LIN XINGYIN

Assignees

  • 上海文施光电科技有限公司

Dates

Publication Date
20260505
Application Date
20260407

Claims (10)

  1. 1. A method for monitoring a shipping status of a container based on multi-parameter sensing, the method comprising: Taking a Hall switch trigger signal as a time reference and a logic starting point, and collecting acceleration signals, attitude angle signals, position signals and temperature and humidity signals in a time window before and after the trigger time; performing time sequence association analysis on the acceleration signal, the attitude angle signal and the position signal in the time window to generate an initial event judgment result; when the initial event judging result is that the physical door opening or hoisting is started, respectively extracting characteristics of acceleration signals and attitude angle signals in the time window to obtain an impact attenuation time constant and an inclination maintaining degree, and determining a mechanical impact state classifying result according to the impact attenuation time constant and the inclination maintaining degree; According to the time sequence correlation between the time sequence change characteristics and the mechanical impact state classification result, carrying out root verification on the abnormality of the box body to generate a root verification result; And generating a joint event code according to the initial event judging result, the mechanical impact state classifying result and the root cause checking result, and executing a classifying response mechanism according to the joint event code.
  2. 2. The method of claim 1, wherein the performing a time-series correlation analysis on the acceleration signal, the attitude angle signal, and the position signal within the time window generates an initial event determination result, comprising: Intercepting an acceleration signal, an attitude angle signal and a position signal in a preset time period before and after triggering by taking the triggering moment of the Hall switch as a reference; respectively extracting time domain response characteristics, gesture evolution characteristics and space migration characteristics related to trigger time from the acceleration signals, the gesture angle signals and the position signals, performing time sequence correspondence on the extracted time domain response characteristics, gesture evolution characteristics and space migration characteristics, and constructing triggered local response characteristics and whole migration characteristics; And determining an initial event judgment result according to the coupling relation between the local response characteristic and the whole migration characteristic.
  3. 3. The method of claim 2, wherein determining an initial event decision result comprises: after the Hall switch is triggered, extracting acceleration transient response in a local direction, gesture deflection response in a single-side direction and returning stability response for restoring an initial gesture after triggering in a corresponding period; the acceleration transient response, the attitude deflection response and the return response are subjected to time sequence association, and a short-range mechanical release process caused by local constraint release in the door opening process is identified; The short-range mechanical release process forms the local response characteristic, and the continuous position migration process in the corresponding period is not detected to form the integral migration characteristic deletion; And when the short-range mechanical release process is identified and the continuous position migration process is not detected, determining that the initial event judgment result is a physical door opening according to the coupling relation of the existence of the local response characteristic and the lack of the integral migration characteristic.
  4. 4. The method of claim 2, wherein determining an initial event decision result comprises: After the Hall switch is triggered, extracting a continuous position migration process, a gesture reciprocating swing process and an acceleration ground-off transition process in a corresponding period; the continuous position migration process, the gesture reciprocating swing process and the acceleration ground-off transition process are coupled in time sequence, and an integral lifting transmission process formed when the box body is converted from a supporting state to a hanging state is identified; And when the whole lifting transmission process is identified, judging that the Hall switch trigger is caused by lifting operation, and determining that the initial event judgment result is lifting start.
  5. 5. The method of claim 2, wherein determining an initial event decision result comprises: After the Hall switch is triggered, the acceleration signal, the attitude angle signal and the position signal in the preset time period before and after the triggering are subjected to joint verification to generate a verification result; When the verification result is that the time domain response feature, the gesture evolution feature and the space migration feature are all absent, determining that the Hall switch trigger does not cause the mechanical state change and the space position change of the box body, and determining that the initial event determination result is magnetic field interference.
  6. 6. The method of claim 1, wherein said determining a mechanical impact state classification result based on said impact decay time constant and said degree of tilt retention comprises: When the initial event judging result is that the physical door opening or hoisting is started, determining a mechanical response section from acceleration signals and attitude angle signals in the time window according to the impact attenuation time constant and the inclination maintaining degree; Extracting an acceleration response feature sequence from acceleration signals in the mechanical response section, extracting a gesture response feature sequence from gesture angle signals in the mechanical response section, performing correlation analysis on the acceleration response feature sequence, the gesture response feature sequence, the impact attenuation time constant and the inclination maintenance degree, and constructing a mechanical response combination feature representing the motion stability after impact; And determining a mechanical impact state classification result according to the mechanical response combination characteristics.
  7. 7. The method of claim 6, wherein the mechanical impact state classification result comprises: if the gesture generates limited deflection after impact and enters a returning process, and the vibration duration time corresponding to the returning process is shorter than the gesture holding time, determining the mechanical impact state classification result as a carrying collision state; If the post-impact posture deviates from the initial state and remains deviated in a subsequent period of time and continuous vibration still exists after impact attenuation, determining that the mechanical impact state classification result is a falling impact state; If the gesture is deflected in the same direction after impact or the gesture recovery process is not carried out after impact attenuation, determining that the mechanical impact state classification result is a flip instability state.
  8. 8. The method of claim 1, wherein the performing root-cause verification on the case anomaly based on the time-series correlation of the time-series variation characteristic and the mechanical impact state classification result to generate a root-cause verification result comprises: determining a Hall switch triggering corresponding period according to an initial event judging result, and determining an impact response corresponding period according to a mechanical impact state classifying result; the Hall switch triggers a temperature signal and a humidity signal in a corresponding time period and a corresponding time period of the impact response; extracting a temperature change characteristic sequence of the temperature signal and extracting a humidity change characteristic sequence of the humidity signal; and determining a box body abnormality verification result according to the time sequence correlation between the temperature change characteristic sequence and the humidity change characteristic sequence.
  9. 9. The method of claim 8, wherein determining the box anomaly verification result comprises: if the initial event judging result is that the door is opened physically, and the Hall switch triggers the temperature and humidity signals in the corresponding time period to be suddenly changed and form a stable state, the checking result is that the door is opened abnormally; If the mechanical impact state classification result is a falling impact state or a turnover unsteady state, and the humidity signal after the impact response corresponding period rises and lags behind the mechanical response, the verification result is that the tank body is permeated; if the temperature and humidity signals in the corresponding time periods of the initial event and the mechanical impact are abnormally changed, but the position signals do not represent the continuous position migration process, the verification result is that the terminal seal is invalid.
  10. 10. The method of claim 1, wherein said performing a classification response mechanism in accordance with said joint event code comprises: When the mechanical impact state classification result is a falling impact state, the acquisition time length is prolonged to a first preset time length; when the mechanical impact state classification result is a carrying collision state, shortening the acquisition time length to a second preset time length; When the root cause verification result is that the tank body water seepage or the terminal sealing fails, adopting local persistent storage and triggering remote alarm; and when the result of the root verification is that the box door is abnormally opened, adopting cyclic storage and triggering local alarm.

Description

Container transportation state monitoring method based on multi-parameter sensing Technical Field The application relates to the technical field of logistics monitoring, in particular to a container transportation state monitoring method based on multi-parameter sensing. Background With the development of global trade, the transportation range of freight containers covers the world, and the number of containers operated by a single logistics main body can reach millions, and the freight containers are distributed on all continents and major trade airlines of the world. In the container transportation process, the position, state and environmental parameters of the container need to be monitored in real time so as to ensure the safety and transportation efficiency of the goods. At present, container monitoring equipment generally operates in a globalization mode along with a box body, flows along with the box after installation, and returns to a maintenance point for a long time on average, part of airlines do not return to home even for a long time, and the equipment has little possibility of on-site maintenance. Therefore, the monitoring device must have high reliability and autonomous decision capability, and be able to accurately identify various types of abnormal events during transportation in the absence of human intervention. The mobile communication networks in different areas have different standards, the international roaming protocol is complex, and the equipment needs to keep stable connection under various network environments and frequency point switching conditions. Meanwhile, the transportation line covers various extreme environments such as high-temperature equator, severe cold arctic, high-humidity islands, inland temperature difference change and the like, and higher requirements are put on the adaptability and monitoring accuracy of the equipment. In the prior art, the container monitoring method is mostly triggered by a single sensor or manually inspected, so that different events such as physical door opening, lifting, collision, falling and water seepage are difficult to accurately distinguish under the conditions of globalization, multiple environments and low maintenance, false alarm or missing alarm is easy to generate, and the intelligent management requirement of large-scale globalization container transportation cannot be met. Disclosure of Invention The embodiment of the application provides a container transportation state monitoring method based on multi-parameter sensing, which realizes the accurate classification, risk classification and root cause tracing of container transportation events, dynamically allocates resources based on event levels, reduces false alarm rate and improves endurance. In order to achieve the above purpose, the application adopts the following technical scheme: a method of monitoring a shipping status of a container based on multi-parameter awareness, the method comprising: Taking a Hall switch trigger signal as a time reference and a logic starting point, and collecting acceleration signals, attitude angle signals, position signals and temperature and humidity signals in a time window before and after the trigger time; performing time sequence association analysis on the acceleration signal, the attitude angle signal and the position signal in the time window to generate an initial event judgment result; when the initial event judging result is that the physical door opening or hoisting is started, respectively extracting characteristics of acceleration signals and attitude angle signals in the time window to obtain an impact attenuation time constant and an inclination maintaining degree, and determining a mechanical impact state classifying result according to the impact attenuation time constant and the inclination maintaining degree; According to the time sequence correlation between the time sequence change characteristics and the mechanical impact state classification result, carrying out root verification on the abnormality of the box body to generate a root verification result; And generating a joint event code according to the initial event judging result, the mechanical impact state classifying result and the root cause checking result, and executing a classifying response mechanism according to the joint event code. In some possible implementations, the performing timing correlation analysis on the acceleration signal, the attitude angle signal, and the position signal in the time window generates an initial event determination result, including: Intercepting an acceleration signal, an attitude angle signal and a position signal in a preset time period before and after triggering by taking the triggering moment of the Hall switch as a reference; respectively extracting time domain response characteristics, gesture evolution characteristics and space migration characteristics related to trigger time from the acceleration signals, the ges