CN-122009625-A - Dynamic monitoring method and related device for explosion bags and coal sample scattering of packer

CN122009625ACN 122009625 ACN122009625 ACN 122009625ACN-122009625-A

Abstract

The invention discloses a dynamic monitoring method and a related device for the scattering of a baling press explosion bag and a coal sample, belonging to the technical field of power plant fuel management; the method comprises the steps of collecting coal sample transportation images in real time through visual perception equipment on a transportation path after coal sample packaging, analyzing the coal sample transportation images in real time based on an image recognition algorithm, detecting whether a coal sample scattering area exists, estimating scattering quantity according to image characteristics of the scattering area when the coal sample scattering area is detected, and generating a bag explosion warning signal if the scattering quantity exceeds a preset threshold value. The invention solves the problems of lag detection, high false alarm rate and insufficient data reliability in the prior art through the three-level architecture of the perception layer, the algorithm layer and the service layer, and realizes the dynamic, accurate and reliable monitoring of the whole flow of coal sample encapsulation and transportation.

Inventors

WANG PENGHUI
CHEN JUNQIANG
FANG YUCHEN
CHEN YE
CHEN YIMING

Assignees

西安热工研究院有限公司
华能上海石洞口发电有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (10)

1. A dynamic monitoring method for the scattering of a baling press blast bag and a coal sample is characterized by comprising the following steps: On a transportation path of the packaged coal sample, acquiring a coal sample transportation image in real time through visual perception equipment; analyzing the coal sample transportation image in real time based on an image recognition algorithm, and detecting whether a coal sample scattering area exists or not; when the coal sample scattering area is detected, estimating the scattering quantity according to the image characteristics of the scattering area; And if the scattering quantity exceeds a preset threshold value, generating a bag explosion warning signal.
2. The dynamic monitoring method for the scattering of the explosion bags and the coal samples of the packer according to claim 1, wherein the visual sensing equipment comprises a plurality of cameras arranged at the outlet of the packer, the middle section of belt transportation and the inlet of the sample collecting barrel.
3. The method for dynamically monitoring the scattering of the coal sample and the explosive bag of the packer according to claim 1, wherein the step of analyzing the coal sample transportation image in real time based on the image recognition algorithm to detect whether the scattering area of the coal sample exists comprises the following steps: based on YOLOv s model, adding a channel attention module, and constructing a coal sample scattering recognition model; collecting images of normal transportation, bag explosion scattering and mechanical vibration dust raising scenes, marking a scattering area pixel level mask, and training a coal sample scattering identification model; Based on the collected coal sample transportation image, recognizing a coal sample scattering area according to a trained coal sample scattering recognition model; And judging the coal sample scattering area when the detected scattering area in the single frame image is more than or equal to a first area threshold or the accumulated detected scattering area in the continuous multi-frame image is more than or equal to a second area threshold.
4. The dynamic monitoring method for the scattering of the coal sample and the explosive bag of the packer according to claim 1, wherein the step of analyzing the coal sample transportation image in real time based on the image recognition algorithm to detect whether the scattering area of the coal sample exists, further comprises: and a physical sensor is deployed on the transportation path to acquire a physical signal, and the physical signal is combined with the coal sample transportation image to perform bag explosion detection through a multi-mode fusion algorithm, wherein the physical sensor is a laser thickness sensor and a vibration sensor.
5. The method for dynamically monitoring the scattering of a baling press blast bag and a coal sample according to claim 1, wherein when the scattering area of the coal sample is detected, the step of estimating the scattering amount according to the image characteristics of the scattering area comprises the following steps: Extracting a pixel level mask of a coal sample scattering area from a coal sample transportation image through an image segmentation algorithm; Converting the pixel area of the scattering area into an actual scattering area based on a mapping relation between a pre-calibrated pixel area and an actual area; and calculating the scattering quantity according to the actual scattering area and the preset weight of the coal sample in unit area.
6. The method for dynamically monitoring the scattering of the explosive bags and the coal samples of the packer according to claim 1, wherein the step of generating the explosive bag warning signal if the scattering amount exceeds a preset threshold value specifically comprises: Carrying out grading alarm according to the scattering quantity to generate an alarm signal; And synchronously pushing the alarm signal to a fuel management system, wherein the alarm information comprises an associated time stamp and video data fragments, and uploading the encrypted alarm information to a blockchain for certification.
7. The method for dynamically monitoring the scattering of the baler blast bags and the coal samples according to claim 6, wherein the content of the grading alarm comprises: primary alarming, namely detecting scattered coal samples but not exceeding a threshold value, and triggering local audible and visual alarming; And (5) high-level warning, namely, stopping the belt and lifting the baffle when the scattering quantity exceeds a threshold value or the scattering time continuously exceeds the threshold value.
8. A dynamic monitoring system for a baling press blast bag and coal sample scatter, comprising: The data acquisition module is used for acquiring coal sample transportation images in real time through the visual perception equipment on the transportation path after the coal samples are packed; The coal sample scattering identification module is used for analyzing the coal sample transportation image in real time based on an image identification algorithm and detecting whether a coal sample scattering area exists or not; The scattering amount calculating module is used for estimating the scattering amount according to the image characteristics of the scattering area when the scattering area of the coal sample is detected; And the explosion bag warning module is used for generating an explosion bag warning signal if the scattering quantity exceeds a preset threshold value.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of a method for dynamically monitoring the dispersion of baler bags and coal samples according to any one of claims 1-7.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of a method for dynamically monitoring the dispersion of baler bags and coal samples according to any one of claims 1-7.

Description

Dynamic monitoring method and related device for explosion bags and coal sample scattering of packer Technical Field The invention belongs to the technical field of power plant fuel management, and relates to a dynamic monitoring method and a related device for a baling press explosion bag and coal sample scattering. Background Under the background of increasingly serious competition in the energy industry, thermal power plants face multiple challenges of improving production efficiency, strictly controlling cost and guaranteeing safe operation. The entering, sampling, sample preparation, testing and metering (abbreviated as "collecting and preparing meter") of fuel (such as coal) are a critical ring in the fuel management system, and the accuracy, fairness and transparency of the process are directly related to the power generation cost accounting, supplier settlement and internal corrosion control encourage honesty. In the collecting and preparing link, the reliable encapsulation of the collected coal sample is a key step for ensuring that the sample is not polluted, lost and exchanged in the transmission process. At present, a mechanical packer is commonly used in the industry to automatically seal coal bags. Existing technologies for detecting the explosion bags of the packer and monitoring the scattering of the coal sample mainly depend on the following modes, but all the technologies have obvious defects and shortcomings. First, detection methods based on pressure thresholds. The method is to install a pressure sensor in the sealing link of the packer, and judge the tightness by monitoring whether the sealing pressure reaches a preset threshold (such as 6-8 MPa). And if the pressure does not reach the standard, triggering an alarm. However, the method can only detect the abnormal static pressure at the moment of sealing, can not identify the problems of bag body breakage or sealing failure in dynamic scenes such as transportation, falling, collision and the like after sealing is finished, and has serious limitation on the detection range. Second, detection methods based on static weighing. According to the method, weighing equipment is arranged on a conveyor belt behind a packer, the weights of coal samples before and after the packer are compared, and if the error exceeds a certain threshold (such as + -2 kg), the leakage is judged. The method has the defects that firstly, natural volatilization or slight dust emission of the coal sample is likely to occur, misjudgment is easy to occur, secondly, weighing is off-line and batch, real-time monitoring cannot be realized, serious hysteresis of problems is found, and finally, the method cannot locate the position and the reason of a leakage point, so that subsequent improvement is not facilitated. Third, a manual spot check based method. The method relies on the operator to periodically visually inspect and manually press the packaged sample bags. The method has the advantages of extremely low sampling inspection coverage rate (generally less than 10 percent), incapability of comprehensive monitoring, large influence of human judgment on subjective factors, high omission rate, increased labor cost and difficulty in realizing traceability of the whole process. Fourth, a monitoring method based on vibration sensors. In this method, a vibration sensor is mounted on a conveyor belt, and a bag breakage is estimated by analyzing an abnormality in a vibration spectrum. However, in the actual industrial environment, the packer and the coal conveying belt have great interference such as mechanical vibration, environmental noise and the like, and false alarm is easy to be caused, so that the accuracy of the method in the actual application is very low and the reliability is not enough. In summary, the existing monitoring method has low detection precision and reliability, relies on offline weighing or manual sampling inspection, has long response delay, distributes detection data in a local controller, is easy to tamper or lose information, and lacks a reliable audit tracing chain. What is needed is a solution for dynamically monitoring the explosion of a packer and the scattering of a coal sample, which can realize real-time, accurate and reliable monitoring of the whole process, can be deeply linked with a service system and provide a trusted data storage certificate, so as to make up for the defects in the prior art. Disclosure of Invention The invention aims to provide a dynamic monitoring method and a related device for the scattering of a baling press explosion bag and a coal sample, which are used for solving the technical problems of low detection precision and reliability, long response time and influence on production and management efficiency in the prior art. In order to achieve the purpose, the invention is realized by adopting the following technical scheme: in a first aspect, the invention provides a dynamic monitoring method for the scattering of a