CN113256150A

CN113256150A - Refrigerated container monitoring system based on edge calculation and visual analysis

Info

Publication number: CN113256150A
Application number: CN202110654305.4A
Authority: CN
Inventors: 汪宝; 李华; 王亚臣; 敦文斌; 杨学文; 佟志国; 王大伟; 于鹏飞
Original assignee: Tianjin Port Electric Power Co ltd
Current assignee: Tianjin Port Electric Power Co ltd
Priority date: 2021-06-11
Filing date: 2021-06-11
Publication date: 2021-08-13

Abstract

The invention discloses a refrigerated container monitoring system based on edge calculation and visual analysis. The invention sets up a network camera on the back of the refrigerated container opposite to the display at an interval of 1.5 meters, and collects the information presented by the display; the intelligent miniature circuit breaker is adopted as the miniature circuit breaker for supplying power to the cold box compressor, and the intelligent miniature circuit breaker has the function of switching on and switching off a power supply and collects the information of the electric quantity of the cold box compressor; the collected video information and the collected electric quantity information are identified through the edge computing AI, the processed reasonable and effective data are processed into digital information through protocol conversion, and then the digital information is transmitted to a monitoring center through a local area network to monitor the operation condition of the refrigerated container. The invention improves the goods-handling efficiency of the cold box storage yard, can effectively reduce the possibility of goods loss, and benefits the logistics bodies of suppliers, goods owners, wharf companies and the like.

Description

Refrigerated container monitoring system based on edge calculation and visual analysis

Technical Field

The invention relates to the technical management field of refrigerated container storage yards, in particular to a refrigerated container monitoring system based on edge calculation and visual analysis.

Background

With the continuous increase of the degree of openness of China to the outside, more and more varieties of cargos are transported by containers at ports and exits in China, wherein the transportation and stacking quantity of medicines and refrigerated and fresh-keeping foods is increased in a geometric number. All the goods need to be shipped and stored between two places by a professional refrigerated container after the storage temperature is set. After the fridge arrives at the wharf storage yard, in order to avoid the situation that the goods in the fridge are deteriorated and damaged due to the storage temperature exceeding the limit, the internal temperature and the operation condition of the compressor need to be patrolled and examined, and the internal temperature is ensured to be kept in the set temperature range.

According to the regulations of the national standard GB/T13145-2018 (technical management requirements of refrigerated container storage yards), the tallying personnel of the cold box storage yard needs to perform operation patrol on the cold box in electric operation every 2 to 4 hours and record the operation temperature every 4 to 6 hours, and the manual patrol interval of the storage yard with the temperature centralized monitoring system can be 8 to 12 hours.

In recent two years, due to the influence of new crown epidemic situations, the cold box storage pressure and the inspection time of the lagged ports of container terminals of all ports in China are greatly improved, and the daily tallying number and the inspection workload of tallying personnel are increased. The traditional tally mode close to manual inspection and temperature reading has the defects of low efficiency, poor real-time performance, incapability of tracing data and the like, and cannot meet the requirement of large-batch cold box inspection monitoring management. Meanwhile, in the period of preventing and controlling the new crown epidemic situation, certain health risks also exist in the period that tally personnel frequently enter and exit the cold box storage yard.

Disclosure of Invention

The invention aims to solve the technical problem of overcoming the defects in the prior art and provides a refrigerated container monitoring system based on edge calculation and visual analysis.

The invention relates to a refrigerated container monitoring system based on edge calculation and visual analysis, which is realized by the following technical scheme that a network camera is erected at the position, opposite to a display, of the back of a refrigerated container (a cold box) at an interval of 1.5 meters, and information presented by the display is collected; the miniature circuit breaker for supplying power to the cold box compressor adopts an intelligent miniature circuit breaker, and can acquire electric quantity information such as working voltage, working current, leakage current and the like of the cold box compressor while having the function of switching on and off a power supply. In order to more effectively utilize data, collected video information and electric quantity information data are subjected to correlation processing, a correlation fusion algorithm is applied, multi-source data are integrated and used for next decision making, and real-time and complete information is obtained through information fusion and AI analysis. The process not only comprises the association and fusion of multi-source data, but also comprises the whole processes of data acquisition, preprocessing, association fusion and comprehensive decision. The processed reasonable and effective data is converted through a protocol and then transmitted to a monitoring center through a local area network so as to monitor the running condition of the cold box.

The system also has an intelligent reporting function, and the refrigerating box which is electrically operated in the storage yard is filled with the operating temperature once per hour, so that the monitoring data can be traced.

Compared with the prior art, the invention has the beneficial effects that:

the invention realizes the remote monitoring of the operation condition of the refrigerated container by combining the visual identification technology and the artificial intelligence technology, greatly reduces the workload of tallying personnel, reduces the tallying risk, improves the tallying efficiency of a cold box storage yard, can effectively reduce the possibility of goods loss, and benefits the logistics bodies of suppliers, goods owners, wharf companies and the like.

Drawings

FIG. 1 is a field device layout diagram;

FIG. 2 is a schematic diagram of a system network architecture;

FIG. 3 is a schematic diagram of a video recognition and artificial intelligence algorithm flow.

Detailed Description

The present invention is described in further detail below with reference to the attached drawings and specific examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

As shown in figure 1, set up a web camera 2 just to display interval 1.5 meters department at refrigerated container 1 back, gather the information that the display appears, including the inside temperature of setting for of cold box, cold box actual operating temperature and warning indicator, the miniature circuit breaker who supplies power for the cold box compressor adopts intelligent miniature circuit breaker 3 (intelligent miniature circuit breaker 3 installs in the low-voltage distribution cabinet), when possessing the break-make power supply function, still can gather the operating voltage of cold box compressor, operating current, leakage current isoelectrical quantity information, and the excessive pressure is reported an emergency and asked for help or increased vigilance, the short circuit is reported an emergency and asked for help or increased vigilance information such as warning.

After the analysis task is started, video analysis is started on the one hand, the acquired video screenshots are transmitted to the on-site edge computing device through the Ethernet and serve as basic data of a visual identification technology, image information is converted into digital information through an artificial intelligence algorithm, and the temperature displayed by the liquid crystal of the refrigerated container is identified.

And on the other hand, the electric quantity detection is started, the collected electric quantity information is transmitted to the field edge computing device through the Ethernet, and the electric information of the operation of the cold box is obtained.

The edge computing device comprehensively analyzes the video identification information and the electrical parameters of the compressor of the refrigerating box, artificial intelligence judgment is carried out through the logic process shown in the figure 2, the electrical quantity information and the video identification information are comprehensively judged, and the video analysis result is discarded if the electrical quantity information and the video identification result are contradictory. And obtaining the identification result with high reliability by mutually verifying the rationality and the reliability of the identification result.

The electric quantity and the running temperature of the cold box identified by the video are displayed on a monitoring computer of a workstation together, and the identification result can be verified in an auxiliary way through the video display.

When the deviation of the operating temperature identified by the video and the set temperature is 5 ℃ or above, or the alarm indicator lamp is observed to be lightened, or the intelligent miniature circuit breaker sends alarm information, the system automatically sends a short message and an application program notice to the tally staff, and the tally staff is required to go to the site for maintenance and treatment.

Claims

1. A refrigerated container monitoring system based on edge calculation and visual analysis is characterized in that a network camera is erected at the position, opposite to a display, on the back of a refrigerated container at an interval of 1.5 meters, and information presented by the display is collected; the intelligent miniature circuit breaker is adopted as the miniature circuit breaker for supplying power to the cold box compressor, and the intelligent miniature circuit breaker has the function of switching on and switching off a power supply and collects the information of the electric quantity of the cold box compressor; identifying the collected video information and the collected electrical quantity information through edge calculation AI, carrying out correlation processing on the collected video information and the collected electrical quantity information data, integrating the multi-source data for the next decision by applying a correlation fusion algorithm, and obtaining real-time and complete information through information fusion and AI analysis; the processed reasonable and effective data is converted into digital information through a protocol, and then is transmitted to a monitoring center through a local area network to monitor the operation condition of the refrigerated container.

2. The intelligent monitoring and management system for refrigerated container based on visual identification and artificial intelligence technology as claimed in claim 1, wherein the information of the electrical quantity of the refrigerated container compressor comprises the working voltage, the working current and the leakage current of the cold box compressor.

3. The intelligent monitoring and management system for refrigerated container based on visual identification and artificial intelligence technology as claimed in claim 1, wherein the detected electrical parameters of the refrigerated container are intelligently and slightly cut off and connected to the edge computing device through RS485 communication line.