CN-121977670-A - Magnetic-sensing monitoring method and system for ash bucket material level of metal smelting dust remover

CN121977670ACN 121977670 ACN121977670 ACN 121977670ACN-121977670-A

Abstract

The invention provides a magnetic-sensing monitoring method and a magnetic-sensing monitoring system for a metal smelting dust remover ash bucket material level, which comprise the steps of obtaining a material level assessment model through pre-training, monitoring the magnetic field intensity in the ash bucket in real time, obtaining a monitored material level value of a current time node according to the magnetic field intensity in the ash bucket of the current time node, inputting the monitored material level value into the material level assessment model to obtain an assessment result, obtaining an accurate material level value according to the assessment result, regulating and controlling the dust remover according to the accurate material level value of the current time node, obtaining the monitored material level value through monitoring the magnetic field intensity in the ash bucket, and evaluating the monitored material level value through the material level assessment model, so that the accuracy of material level monitoring in the ash bucket is improved.

Inventors

XU HOUYOU
LIU YALONG
ZHAN ZHIFA
ZHOU YAN
GUO KUN
WANG MENGTING

Assignees

中钢武汉安环院绿世纪安全管理顾问有限公司

Dates

Publication Date: 20260505
Application Date: 20251217

Claims (10)

1. The magnetic-sensing monitoring method for the ash bucket material level of the metal smelting dust remover is characterized by comprising the following steps of: Monitoring the magnetic field intensity in the ash bucket in real time, and obtaining a monitored material level value of the current time node according to the magnetic field intensity in the ash bucket of the current time node; inputting the monitored material level value into the material level evaluation model to obtain an evaluation result, and obtaining an accurate material level value according to the evaluation result; And regulating and controlling the dust remover according to the accurate material level value of the current time node.
2. The magnetic-sensing monitoring method for the ash bucket material level of the metal smelting dust remover according to claim 1, wherein the method is characterized in that the monitored material level value of the current time node is obtained according to the strength of the magnetic field in the ash bucket of the current time node, and specifically comprises the following steps: the expression of the level is: H=( -Z 0 ）/U m ×H max ; Wherein R is a Hall coefficient, I is a constant current of the magneto-dependent sensor, B is the magnetic field intensity in the ash bucket, d is the thickness of the Hall element, H is the material level, Z 0 is zero compensation, U m is full range voltage, and H max is the maximum ash level height.
3. The magnetic-sensing monitoring method for the ash bucket level of the metal smelting dust remover according to claim 1, wherein the step of inputting the monitored level value into the level evaluation model to obtain an evaluation result specifically comprises the following steps: inputting the monitored material level value of the corresponding time node into the material level evaluation model; acquiring all accurate material level values in a front sliding window of a corresponding time node, and acquiring the material level value reliability of the monitored material level value of the corresponding time node by the material level evaluation model according to the monitored material level value of the corresponding time node and all the accurate material level values in the front sliding window of the corresponding time node; and taking the reliability of the material level value as an evaluation result corresponding to the monitored material level value of the corresponding time node.
4. The magnetic-sensing monitoring method for the ash bucket level of the metal smelting dust remover according to claim 3, wherein the level evaluation model obtains the level value reliability of the monitored level value of the corresponding time node according to the monitored level value of the corresponding time node and all accurate level values in a front sliding window of the corresponding time node, and specifically comprises the following steps: Inputting the monitored material level value of the corresponding time node into a time sequence feature extraction model to obtain a hidden time sequence feature corresponding to the corresponding time node; according to all accurate material level values in the front sliding window of the corresponding time node, corresponding feature vectors of the corresponding time node; Inputting the hidden time sequence features and the feature vectors into a tree model, and respectively outputting corresponding output weights by a plurality of decision trees in the tree model; obtaining an original score corresponding to each decision tree according to the corresponding output weight; And obtaining the reliability of the material level value of the monitored material level value of the corresponding time node according to the original scores corresponding to all the decision trees.
5. The method for magnetically monitoring the material level of the ash bucket of the metal smelting dust remover according to claim 4, wherein the monitored material level value of the corresponding time node is input into a time sequence feature extraction model to obtain the hidden time sequence feature corresponding to the corresponding time node, and the method specifically comprises the following steps: the expression of the hidden time sequence feature is: ; Wherein, the For the hidden timing feature corresponding to the t-time node, For the information output degree weight, For the monitored fill level value of the t-time node, In order to adjust the activation threshold of the output gate, And updating the value for the state corresponding to the t time node.
6. The magnetic sensing method for monitoring the material level of the ash bucket of the metal smelting dust remover according to claim 4, wherein the obtaining the reliability of the material level value of the monitored material level value of the corresponding time node according to the original scores corresponding to all the decision trees specifically comprises the following steps: Probability conversion is carried out on the original scores corresponding to the corresponding decision trees to obtain probability values corresponding to the corresponding decision trees, and the probability values are expressed as follows: ; Wherein, the Is the probability value of the kth decision tree, An original score for the kth decision tree; and taking the maximum probability value in all the decision trees as the level value reliability of the monitored level value of the corresponding time node.
7. The magnetic-sensing monitoring method for the ash bucket level of the metal smelting dust remover according to claim 4, wherein the feature vectors corresponding to the corresponding time nodes according to all accurate level values in the front sliding window of the corresponding time nodes specifically comprise: according to all accurate material level values in a front sliding window of the corresponding time node, obtaining trend slope, fluctuation coefficient, mutation frequency and signal response time matching degree coefficient in the sliding window; And obtaining the feature vector corresponding to the corresponding time node according to the trend slope, the fluctuation coefficient, the mutation frequency and the signal response time matching degree coefficient in the sliding window.
8. The magnetic-sensing monitoring method for the ash bucket material level of the metal smelting dust remover according to claim 1, wherein the accurate material level value is obtained according to the evaluation result, and specifically comprises the following steps: when the evaluation result is greater than or equal to a first preset value, judging the monitored material level value as the accurate material level value; and when the evaluation result is smaller than a first preset value, judging that the monitored material level value is not compliant, measuring the material level value by other preset modes to obtain a preliminary monitoring value, and taking the preliminary monitoring value as the accurate material level value.
9. The magnetic-sensing monitoring method for the ash bucket material level of the metal smelting dust remover according to claim 1, wherein the dust remover is regulated and controlled according to the accurate material level value of the current time node, and specifically comprises the following steps: Acquiring an accurate material level value of a current time node; When the accurate material level value is larger than the first preset material level, the monitoring frequency for monitoring the magnetic field intensity in the ash bucket is increased; when the accurate material level value is larger than or equal to the second preset material level, starting the ash discharging module by a preset switching value signal, and when the accurate material level value is smaller than or equal to the third preset material level, stopping the ash discharging module.
10. A magnetic-sensing monitoring system for the material level of an ash bucket of a metal smelting dust remover, which is used for applying the magnetic-sensing monitoring method for the material level of the ash bucket of the metal smelting dust remover according to any one of claims 1-9, and is characterized by comprising a material level monitoring module, a data processing module, a monitoring reliability judging module, a dynamic working condition adapting module and an ash discharging control module, wherein: The material level monitoring module is used for monitoring and acquiring the magnetic field intensity in the ash bucket in real time and sending the magnetic field intensity to the data processing module; The data processing module is used for obtaining a monitored material level value of the current time node according to the magnetic field intensity in the ash bucket of the current time node, and sending the monitored material level value to the monitoring reliability judging module; The monitoring reliability judging module is used for inputting the monitoring material level value into the material level evaluation model to obtain an evaluation result, and obtaining an accurate material level value according to the evaluation result output by the material level evaluation model; The dynamic working condition adapting module is used for regulating and controlling the monitoring frequency of the dust remover on the magnetic field intensity in the ash bucket, and the ash discharging control module is used for regulating and controlling the dust remover to start ash discharging or stop ash discharging.

Description

Magnetic-sensing monitoring method and system for ash bucket material level of metal smelting dust remover Technical Field The invention belongs to the field of metal smelting, and in particular relates to a magnetic-sensing monitoring method and a magnetic-sensing monitoring system for a material level of an ash bucket of a metal smelting dust remover. Background In the metallurgical production process, a large amount of dust that produces in the production process needs to be collected and get rid of, is used for collecting the dust through the ash bucket in actual scene generally, and the material level in the ash bucket is used for representing the bearing accumulation's in the ash bucket height, can judge the bearing condition of ash bucket according to the material level, and the monitoring of ash bucket material level directly influences dust pelletizing system's stability and production continuity. Because the dust environment in the ash bucket has the characteristics of high temperature, high humidity, corrosiveness and the like, the inspection is difficult to be directly carried out by personnel, the existing charge level indicator for monitoring the material level is usually required to be in direct contact with dust in the ash bucket, the high temperature, high humidity and the dust environment containing corrosiveness in the ash bucket are easy to interfere and damage the charge level indicator, the monitoring data of the charge level indicator are inaccurate and easy to break down, the accuracy of the monitoring of the material level in the ash bucket is difficult to meet the requirement, and the problems of false alarm, false alarm and the like caused by the accuracy are likely to cause the ash bucket to exceed the upper limit of the bearing capacity, so that the collapse accident of the ash bucket is caused. In view of this, overcoming the drawbacks of the prior art is a problem to be solved in the art. Disclosure of Invention The problem to be solved by the invention is how to improve the accuracy of monitoring the material level in the ash bucket. In a first aspect, a magnetic-sensing monitoring method for a dust hopper level of a metal smelting dust remover is provided, which comprises the following steps: Monitoring the magnetic field intensity in the ash bucket in real time, and obtaining a monitored material level value of the current time node according to the magnetic field intensity in the ash bucket of the current time node; inputting the monitored material level value into the material level evaluation model to obtain an evaluation result, and obtaining an accurate material level value according to the evaluation result; And regulating and controlling the dust remover according to the accurate material level value of the current time node. Preferably, the obtaining the monitored material level value of the current time node according to the magnetic field intensity in the ash bucket of the current time node specifically includes: the expression of the level is: H=(-Z0）/Um×Hmax; Wherein R is a Hall coefficient, I is a constant current of the magneto-dependent sensor, B is the magnetic field intensity in the ash bucket, d is the thickness of the Hall element, H is the material level, Z 0 is zero compensation, U m is full range voltage, and H max is the maximum ash level height. Preferably, the inputting the monitored level value into the level evaluation model to obtain an evaluation result specifically includes: inputting the monitored material level value of the corresponding time node into the material level evaluation model; acquiring all accurate material level values in a front sliding window of a corresponding time node, and acquiring the material level value reliability of the monitored material level value of the corresponding time node by the material level evaluation model according to the monitored material level value of the corresponding time node and all the accurate material level values in the front sliding window of the corresponding time node; and taking the reliability of the material level value as an evaluation result corresponding to the monitored material level value of the corresponding time node. Preferably, the level evaluation model obtains the level value reliability of the monitored level value of the corresponding time node according to the monitored level value of the corresponding time node and all accurate level values in the front sliding window of the corresponding time node, and specifically includes: Inputting the monitored material level value of the corresponding time node into a time sequence feature extraction model to obtain a hidden time sequence feature corresponding to the corresponding time node; according to all accurate material level values in the front sliding window of the corresponding time node, corresponding feature vectors of the corresponding time node; Inputting the hidden time sequence features and the feature vectors into a