CN-122006366-A - Ash removal method and device for cement bag-type dust remover

CN122006366ACN 122006366 ACN122006366 ACN 122006366ACN-122006366-A

Abstract

The application relates to the technical field of industrial dust removal, in particular to a dust removing method and device for a cement bag-type dust remover. The method comprises the steps of collecting vibration sound waves of filter bags through an acoustic array during pulse ash removal, analyzing through a sound source positioning and voiceprint recognition model to obtain health scores, guiding an industrial camera to conduct image collection and recognition verification on target filter bags if the scores are abnormal, performing hierarchical ash removal control based on fusion diagnosis results of voiceprints and images, performing depressurization observation on suspected broken filter bags, immediately isolating and switching to a low-pressure maintenance mode on the sub-chambers where the confirmed broken filter bags are located, and dynamically adjusting ash removal frequencies of adjacent sub-chambers according to the damage degree and the pressure difference of a system to compensate filtering capacity. The device comprises an acoustic sensing array, a directional vision inspection unit, an intelligent analysis decision unit and a cooperative execution mechanism. The application realizes early accurate identification and intelligent ash removal control of breakage, remarkably prolongs the service life of the filter bag and improves the running stability and economy of the system.

Inventors

DUAN HONGLIANG
QIN JIANGUANG
ZHENG XINJUN
XU YUXI
WANG ZONGSHAN
YANG SHENGXI

Assignees

卫辉市春江水泥有限公司

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (8)

1. The ash removing method of the cement bag-type dust remover is characterized by comprising the following steps of: S1, in a pulse ash removal stage, collecting vibration acoustic wave signals of each filter bag or filter bag partition by utilizing a plurality of acoustic wave sensors which are arranged in an array in a dust remover air-purifying chamber, and determining the primary source position of an abnormal acoustic wave signal through an acoustic source positioning algorithm; s2, preprocessing the sound wave signals, extracting features, inputting the sound wave signals into a pre-trained voiceprint recognition model for analysis, and outputting scores representing the health states of the filter bags; s3, if the score is lower than a first threshold, judging that the filter bag is suspected to be damaged, and generating a control instruction according to the preliminary source position, and driving an image acquisition device to acquire a directional image of the specific target filter bag; S4, analyzing the collected directional image by utilizing an image recognition model, and outputting damage positioning information and damage degree evaluation if damage is confirmed; S5, executing a hierarchical ash removal control strategy based on the grading of the step S2, the damage positioning information and the damage degree evaluation of the step S4: S5a, if the score is lower than a first threshold value only and the image is not confirmed to be damaged, the ash removal pressure of the sub-chamber where the filter bag is positioned is adjusted to a first preset value, and the filter bag is marked as an observation state; s5b, if the image is confirmed to be damaged, immediately isolating a sub-chamber where the damaged filter bag is located, and switching the sub-chamber ash removal mode to a low-pressure maintenance mode; S5c, after isolating the damaged sub-chambers, dynamically calculating and adjusting the ash removal frequency of one or more adjacent intact sub-chambers to a second preset value according to the damage degree evaluation and the real-time pressure difference of the dust remover system so as to perform filtering capacity compensation.
2. The method of claim 1, wherein the training samples of the voiceprint recognition model include sonic data for different conditions of cement production, the model being capable of distinguishing between different voiceprint characteristics due to normal ash removal, cement caking drop, physical filter bag breakage, and high temperature filter bag burn.
3. The method according to claim 1, wherein the step S5c of "dynamic calculation" is specifically to establish a fuzzy control rule base with the total pressure difference, the position of damaged sub-chambers and the damage evaluation degree of the system as inputs and the up-regulation of the ash removal frequency of adjacent sub-chambers as outputs, and to perform real-time query and decision.
4. The method according to claim 1, further comprising the step of mapping the score of the step S2, the breakage positioning and degree evaluation of the step S4, and the execution strategy of the step S5 to a three-dimensional digital twin model of the dust remover in real time, wherein the model synchronously simulates the internal flow field and differential pressure distribution after the execution strategy, and compares the simulation prediction result with the sensor measured data to calibrate the control parameters.
5. An ash removal device for carrying out the method of any one of claims 1-4, comprising: the acoustic sensing array module consists of a plurality of acoustic wave sensors which are uniformly distributed at the top of the clean air chamber and have high directivity and is used for collecting spatially distributed acoustic wave signals; The directional vision inspection module comprises an industrial camera capable of moving on a track and a servo mechanism for driving the industrial camera to be accurately positioned to a preliminary source position indicated by the acoustic perception array module; The intelligent analysis decision unit is integrated with a voiceprint analysis module, an image analysis module and a hierarchical control strategy engine, wherein the voiceprint analysis module is internally provided with a sound source positioning algorithm and the voiceprint recognition model; The cooperative execution mechanism comprises a pulse valve group controlled by the intelligent analysis decision unit, a sub-chamber isolation valve and a servo driver used for driving the directional vision inspection module.
6. The apparatus of claim 5, wherein the acoustic wave sensor is a high temperature resistant beam forming MEMS microphone array unit having a specific direction angle arranged such that detection areas of adjacent sensors have a partial overlap.
7. The apparatus of claim 5, wherein the industrial camera of the directional vision inspection module is equipped with an auto-focus lens and an annular light-compensating lamp, and an outer protective cover thereof is provided with an inert gas positive pressure protection cavity communicated with the interior of the dust remover clean air chamber for preventing cement dust from contaminating the lens.
8. The apparatus of claim 5, further comprising a model calibration module that receives analog prediction data of the digital twin model and actual precipitator operation sensor data and employs an adaptive filtering algorithm to fine tune parameters in the hierarchical control strategy engine on-line.

Description

Ash removal method and device for cement bag-type dust remover Technical Field The application relates to the technical field of industrial dust removal, in particular to a dust removing method and device for a cement bag-type dust remover. Background In the cement production process, high-temperature, high-concentration and high-humidity dust generated by kiln head and tail, coal mill, cement mill and other working procedures is mainly treated by a bag-type dust collector. Pulse blowing ash removal is a key technology for maintaining filtering performance. However, the prior art has obvious disadvantages: 1. The filter bag breakage detection is lag and unreliable, and the traditional method relies on manual periodic unpacking inspection or monitoring of emission concentration exceeding alarm, so that early warning can not be realized. By adopting a single sensor (such as acoustic emission or pressure difference) for online detection, the false alarm rate is high under the complex working conditions (high noise and material impact) of a cement plant, and the damage point cannot be accurately positioned. 2. The ash removal control strategy is rough and inefficient, and ash removal is generally performed by adopting timing or constant pressure difference, and belongs to a one-knife mode, so that the individual health state of the filter bag cannot be perceived. This results in an excessive ash removal (accelerated wear, wasteful energy consumption) from the intact filter bag, while the damaged filter bag is either insufficiently ash removed or continues to undergo a strong blowing, causing the damage to expand rapidly, causing a "bag-breaking effect". Therefore, the invention provides a dust removing method and device for a cement bag-type dust remover, which solve the problems in the background technology. Disclosure of Invention The invention aims to overcome the defects of the prior art and provides a dust removing method and device for a cement bag-type dust remover. The application provides a dust removing method of a cement cloth bag dust remover, which comprises the following steps: S1, in a pulse ash removal stage, collecting vibration acoustic wave signals of each filter bag or filter bag partition by utilizing a plurality of acoustic wave sensors which are arranged in an array in a dust remover air-purifying chamber, and determining the primary source position of an abnormal acoustic wave signal through an acoustic source positioning algorithm; S2, preprocessing the sound wave signals, extracting features, inputting the sound wave signals into a pre-trained voiceprint recognition model for analysis, and outputting scores representing the health states of the filter bags; S3, if the score is lower than a first threshold, judging that the filter bag is suspected to be damaged, and generating a control instruction according to the preliminary source position, and driving an image acquisition device to acquire a directional image of the specific target filter bag; s4, analyzing the collected directional image by using an image recognition model, and outputting damage positioning information and damage degree evaluation if damage is confirmed; s5, executing a hierarchical ash removal control strategy based on the grading of the step S2, the damage positioning information and the damage degree evaluation of the step S4: S5a, if the score is lower than a first threshold value only and the image is not confirmed to be damaged, the ash removal pressure of the sub-chamber where the filter bag is positioned is adjusted to a first preset value, and the filter bag is marked as an observation state; s5b, if the image is confirmed to be damaged, immediately isolating a sub-chamber where the damaged filter bag is located, and switching the sub-chamber ash removal mode to a low-pressure maintenance mode; S5c, after isolating the damaged sub-chambers, dynamically calculating and adjusting the ash removal frequency of one or more adjacent intact sub-chambers to a second preset value according to the damage degree evaluation and the real-time pressure difference of the dust remover system so as to perform filtering capacity compensation. Optionally, the training sample of the voiceprint recognition model comprises sonic data under different working conditions of cement production, and the model can distinguish different voiceprint characteristics generated by normal ash removal, cement caking drop, filter bag physical damage and filter bag high-temperature burn. Optionally, the "dynamic calculation" in step S5c specifically includes establishing a fuzzy control rule base with the total pressure difference, the position of damaged sub-chambers and the damage evaluation degree of the system as input and the up-regulation of the ash removal frequency of adjacent sub-chambers as output, and performing real-time query and decision. Optionally, the method further comprises a step S6 of mapping the scoring of the step S2, the