Method for predicting residual life of filter bag of bag type dust collector based on multiple sound sensors

Abstract

The invention relates to the technical field of bag type dust collectors, in particular to a method for predicting the residual life of a filter bag of a bag type dust collector based on multiple sound sensors. A plurality of sound sensors are uniformly distributed in the bag type dust collector, the sound sensors collect sound signals generated in the running process of the filter bag in real time, time-frequency analysis is carried out on the sound signals of the sound sensors by adopting wavelet transform, and filter bag aging characteristic signals are extracted from the sound signals; according to the filter bag structure parameters, the filter bag aging characteristic signals of the plurality of sound sensors are fused, and the difference between monitoring data and the real operation condition of the filter bag caused by the position and working condition difference of a single sound sensor is avoided, so that the accuracy of the filter bag aging characteristic signals can be remarkably improved; and the residual life of the filter bag at the current moment can be accurately predicted by combining the average aging rate of the filter bag.

Description

Technical Field

[0001] This invention relates to the field of bag filter technology, and in particular to a method for predicting the remaining lifespan of bag filter bags based on multiple sound sensors. Background Technology

[0002] Baghouse dust collectors, as important environmental protection equipment in industrial production, are widely used in industries such as steel, chemical, and building materials. They purify gas by capturing particulate matter from flue gas through filter bags. However, filter bags gradually age over time, leading to decreased filtration efficiency and even damage, causing dust leaks and resulting in serious environmental pollution and economic losses. Therefore, predicting the remaining lifespan of filter bags before they break has become a crucial issue in the maintenance and management of baghouse dust collectors.

[0003] Currently, filter bag life prediction mainly relies on periodic replacement or manual inspection based on experience. However, these methods often fail to accurately grasp the actual operating status of the filter bags, which may lead to premature or delayed replacement of the filter bags, increasing maintenance costs and potentially causing unplanned equipment downtime.

[0004] With the development of sensor technology, using sound sensors to monitor the operating status of filter bags in real time and combining this with data analysis techniques for lifespan prediction has gradually become an effective method. However, due to the different placement locations of individual sound sensors and the varying operating conditions at different parts of the filter bag, the monitoring data from a single sound sensor cannot accurately reflect the operating status of the filter bag, resulting in inaccurate predictions of the remaining lifespan of the filter bag. Furthermore, the signal fusion processing of multiple sound sensors and the optimization of lifespan prediction algorithms remain technical challenges. Summary of the Invention

[0005] The main objective of this invention is to provide a method for predicting the remaining lifespan of filter bags in baghouse dust collectors based on multiple sound sensors, in order to solve the technical problem that a single sound sensor cannot accurately predict the remaining lifespan of filter bags in the prior art.

[0006] To achieve the above objectives, the present invention provides a method for predicting the remaining lifespan of filter bags in a baghouse dust collector based on multiple acoustic sensors, comprising the following steps: S1. Obtain the sound signal of the i-th sound sensor at the current time, and preprocess the sound signal to obtain the effective filter bag sound signal of the i-th sound sensor; where i≥1, and i is a positive integer; S2. Obtain the scale parameter and time shift parameter of the effective filter bag sound signal of the i-th sound sensor, and perform time-frequency analysis on the effective filter bag sound signal through wavelet transform based on the scale parameter and time shift parameter to obtain the filter bag aging characteristic signal in the effective filter bag sound signal of the i-th sound sensor. S3. Determine if i is greater than or equal to n; if not, assign i+1 to i and return to step S1; if yes, output the filter bag aging characteristic signals of all sound sensors; where n is the number of sound sensors. S4. Obtain the filter bag structure parameters, and fuse the filter bag aging characteristic signals of all sound sensors according to the filter bag structure parameters to obtain the filter bag remaining life index at the current moment. S5. Obtain the average aging rate of the filter bag, and determine the remaining lifespan of the filter bag at the current moment based on the average aging rate and the remaining lifespan index of the filter bag at the current moment.

[0007] Furthermore, step S2 specifically includes the following steps: Obtain the scale parameter and time shift parameter of the effective filter bag sound signal of the i-th sound sensor, and determine the mother wavelet function based on the scale parameter and time shift parameter; A first model is obtained. Wavelet coefficients are determined based on the mother wavelet function and the effective filter bag sound signal. Wavelet amplitude values ​​are determined based on the wavelet coefficients. The filter bag aging characteristic signal in the effective filter bag sound signal of the i-th sound sensor is obtained using the wavelet amplitude value. The first model is: ; Where f(t) is the sound signal; Ψ(t) is the mother wavelet function; a is the scaling parameter, used to adjust the analysis frequency range; b is the time shift parameter, used to locate the instantaneous features of the signal; t is the time; W a,b These are wavelet coefficients.

[0008] More preferably, obtaining the filter bag aging characteristic signal from the effective filter bag sound signal of the i-th sound sensor through the small amplitude value specifically includes the following steps: The cumulative energy within a preset time window and scale range is obtained, a second model is established based on the cumulative energy, and the aging characteristic signal of the filter bag is determined based on the second model. The second model is: ; Among them, S i (t) represents the aging characteristic signal of the filter bag from the i-th sound sensor; α1, α2, and α3 are adjustment coefficients, which can be obtained by looking up a table; |W a,b | represents the amplitude of the wavelet coefficient at (a,b), indicating the instantaneous intensity of local acoustic energy during the operation of the filter bag; E(t) represents the cumulative energy within the preset time window and scale range.

[0009] More preferably, the step S2 is followed by the following steps: A third model is obtained. Based on the wavelet amplitude, the wavelet energy density is determined using the third model. The optimal values ​​of the scale parameter and time shift parameter are determined by the maximum value of the wavelet energy density. The third model is: ; Wherein, E(a,b) is the wavelet energy density at (a,b).

[0010] More preferably, step S4 specifically includes the following steps: The length and diameter of the filter bag are obtained. A fusion model is established based on the length, diameter, and the second model. The aging characteristic signals of the filter bag from the n sound sensors are fused using the fusion model to obtain the remaining lifespan index of the filter bag at the current moment. The fusion model is as follows: ; Among them, R t This refers to the remaining lifespan of the filter bag; w i λ is the weighting coefficient for the i-th sensor; i L is a correction factor used to adjust for differences in sensor mounting location and sensitivity. f and D f These are the length and diameter of the filter bag, respectively, and are key parameters characterizing the filter bag's structural parameters; t m For the maximum time.

[0011] More preferably, step S5 specifically includes the following steps: The average aging rate of the filter bag is obtained. Based on the average aging rate and the remaining lifespan index of the filter bag, a filter bag aging prediction model is established. The remaining lifespan of the filter bag at the current moment is determined based on the filter bag aging prediction model. The filter bag aging prediction model is as follows: ; Where t0 is the current time, and R(t0) = R t μ represents the average aging rate of the filter bag, which is obtained by curve fitting of historical operating data or accelerated aging experiments.

[0012] Furthermore, the following steps are included before step S1: Three layers of data monitoring points (upper, middle, and lower) are set along the height direction of the filter bag. The n sound sensors are evenly distributed within the data monitoring points. The upper data monitoring point corresponds to the pulse-jet cleaning area, the middle data monitoring point corresponds to the filter body area, and the lower data monitoring point corresponds to the area above the ash hopper. The bag filter includes a pulse-jet cleaning area, a filter body area, and an area above the ash hopper.

[0013] More preferably, the sound sensors are evenly distributed on the left and right sides of the housing of the bag filter or symmetrically distributed along the diagonal direction of the housing.

[0014] More preferably, it also includes a protective cover, wherein the sound sensors are evenly distributed on the outer surface of the housing, and the protective cover covers the outer periphery of the sound sensors.

[0015] Furthermore, the sampling frequencies of the n sound sensors are kept consistent, and the sampling times are synchronized.

[0016] Compared with the prior art, the present invention has the following beneficial effects: In this invention, multiple sound sensors are evenly distributed in the bag filter. The sound sensors collect the sound signals generated during the operation of the filter bags in real time. Wavelet transform is used to perform time-frequency analysis on the sound signals of each sound sensor to extract the filter bag aging characteristic signals. Based on the filter bag structural parameters, the filter bag aging characteristic signals of multiple sound sensors are fused, avoiding the discrepancy between the monitoring data and the actual operating condition of the filter bags caused by the differences in the position and operating conditions of individual sound sensors. This significantly improves the accuracy of the filter bag aging characteristic signals. Combined with the average aging rate of the filter bags, the remaining lifespan of the filter bags at the current moment can be accurately predicted. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a method for predicting the remaining lifespan of a bag filter bag based on multiple sound sensors, according to an embodiment of the present invention.

[0019] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0023] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0024] Please see Figure 1 This embodiment provides a method for predicting the remaining lifespan of filter bags in a baghouse dust collector based on multiple sound sensors, including the following steps: S1. Obtain the sound signal of the i-th sound sensor at the current time, and preprocess the sound signal to obtain the effective filter bag sound signal of the i-th sound sensor; where i≥1, and i is a positive integer;

[0025] Specifically, in this embodiment, multiple sound sensors are evenly distributed throughout the bag filter. The sound sensors collect the sound signals generated during the operation of the filter bags in real time. Environmental noise and mechanical background noise are removed through preprocessing methods such as bandpass filtering, mean normalization, and noise suppression, thereby extracting the effective filter bag sound signals actually generated during the operation of the filter bags.

[0026] S2. Obtain the scale parameter and time shift parameter of the effective filter bag sound signal of the i-th sound sensor, and perform time-frequency analysis on the effective filter bag sound signal through wavelet transform based on the scale parameter and time shift parameter to obtain the filter bag aging characteristic signal in the effective filter bag sound signal of the i-th sound sensor. S3. Determine if i is greater than or equal to n; if not, assign i+1 to i and return to step S1; if yes, output the filter bag aging characteristic signals of all sound sensors; where n is the number of sound sensors. S4. Obtain the filter bag structure parameters, and fuse the filter bag aging characteristic signals of all sound sensors according to the filter bag structure parameters to obtain the filter bag remaining life index at the current moment. S5. Obtain the average aging rate of the filter bag, and determine the remaining lifespan of the filter bag at the current moment based on the average aging rate and the remaining lifespan index of the filter bag at the current moment.

[0027] In this embodiment, multiple sound sensors are evenly distributed in the bag filter. The sound sensors collect the sound signals generated during the operation of the filter bags in real time. Wavelet transform is used to perform time-frequency analysis on the sound signals of each sound sensor to extract the filter bag aging characteristic signals. Based on the filter bag structural parameters, the filter bag aging characteristic signals of multiple sound sensors are fused, avoiding the difference between the monitoring data and the actual operating condition of the filter bags caused by the difference in the position and operating conditions of individual sound sensors. This can significantly improve the accuracy of the filter bag aging characteristic signals. Combined with the average aging rate of the filter bags, the remaining life of the filter bags at the current moment can be accurately predicted.

[0028] In one embodiment, step S2 further includes the following steps: Obtain the scale parameter and time shift parameter of the effective filter bag sound signal of the i-th sound sensor, and determine the mother wavelet function based on the scale parameter and time shift parameter; A first model is obtained. Wavelet coefficients are determined based on the mother wavelet function and the effective filter bag sound signal. Wavelet amplitude values ​​are determined based on the wavelet coefficients. The filter bag aging characteristic signal in the effective filter bag sound signal of the i-th sound sensor is obtained using the wavelet amplitude value. The first model is: ; Where f(t) is the sound signal; Ψ(t) is the mother wavelet function; a is the scaling parameter, used to adjust the analysis frequency range; b is the time shift parameter, used to locate the instantaneous features of the signal; t is the time; W a,b These are wavelet coefficients.

[0029] In this embodiment, as a further preferred embodiment, obtaining the filter bag aging characteristic signal in the effective filter bag sound signal of the i-th sound sensor through the small amplitude value specifically includes the following steps: The cumulative energy within a preset time window and scale range is obtained, a second model is established based on the cumulative energy, and the aging characteristic signal of the filter bag is determined based on the second model. The second model is: ; Among them, S i (t) represents the aging characteristic signal of the filter bag from the i-th sound sensor; α1, α2, and α3 are adjustment coefficients, which are summarized in a table through cumulative experiments and can be obtained by looking up the table; |W a,b | represents the amplitude of the wavelet coefficient at (a,b), indicating the instantaneous intensity of local acoustic energy during the operation of the filter bag; E(t) represents the cumulative energy within the preset time window and scale range.

[0030] This embodiment also includes the following steps: A third model is obtained. Based on the wavelet amplitude, the wavelet energy density is determined using the third model. The optimal values ​​of the scale parameter and time shift parameter are determined by the maximum value of the wavelet energy density. The third model is: ; Wherein, E(a,b) is the wavelet energy density at (a,b).

[0031] The optimal values ​​of the scaling parameter and time translation parameter are determined by the maximum value of the wavelet energy density, and its expression is as follows: ;where a For the optimal value of the scaling parameter, b This represents the optimal value for the time shift parameter.

[0032] The principle of this embodiment is as follows: When the filter bag surface becomes clogged or dust accumulates, the local resistance to airflow through the filter bag increases, causing the filter bag's vibration frequency to decrease, the sound energy distribution to tend towards lower frequencies, and the wavelet amplitude to significantly increase in the low-scale region, thereby affecting S... i (t) increases; When the filter bag material hardens or experiences fiber fatigue, the filter bag's flexibility decreases, altering the sound wave propagation speed and reflection characteristics. This leads to increased high-frequency components and a higher rate of sound energy change, resulting in increased S... i (t) exhibits abrupt changes at high-frequency scales; When the airflow distribution inside the filter bag becomes turbulent or locally damaged, periodic impacts and pulsating noise appear in the flow field, resulting in a wideband energy increase and a significant rise in wavelet energy density E(t).

[0033] In one embodiment, considering the differences in operating conditions at different locations of the bag filter, to improve the accuracy of prediction, the characteristic signals S from multiple sensors are... i (t) Perform weighted fusion. Step S4 specifically includes the following steps: The length and diameter of the filter bag are obtained. A fusion model is established based on the length, diameter, and the second model. The aging characteristic signals of the filter bag from the n sound sensors are fused using the fusion model to obtain the remaining lifespan index of the filter bag at the current moment. The fusion model is as follows: ; Among them, R t This refers to the remaining lifespan of the filter bag; w i λ is the weighting coefficient for the i-th sensor; i L is a correction factor used to adjust for differences in sensor mounting location and sensitivity. f and D f These are the length and diameter of the filter bag, respectively, and are key parameters characterizing the filter bag's structural parameters; t m For the maximum time.

[0034] It is worth noting that L in the filter bag structural parameters f and D f The introduction of this feature allows the fusion model to be applied to baghouse dust collectors of different specifications. It should be noted that the weighting coefficient w in the steps... i The signal fusion weights are only related to the wavelet coefficients W in wavelet analysis. a,b There is no direct mathematical connection. The output value R in this step of this embodiment... t This is the remaining lifespan index of the filter bag at the current moment, representing the health status of the filter bag at time t.

[0035] More preferably, step S5 specifically includes the following steps: The average aging rate of the filter bag is obtained. Based on the average aging rate and the remaining lifespan index of the filter bag, a filter bag aging prediction model is established. The remaining lifespan of the filter bag at the current moment is determined based on the filter bag aging prediction model. The filter bag aging prediction model is as follows: ; Where t0 is the current time, and R(t0) = R t μ represents the average aging rate of the filter bag, obtained through curve fitting of historical operating data or accelerated aging experiments. The filter bag aging prediction model reflects the dynamic change in filter bag lifespan over time.

[0036] In the above embodiments, to ensure the representativeness of the signals collected by multiple sound sensors and the effectiveness of the fusion calculation, the arrangement of the sound sensors meets the following requirements: Three layers of data monitoring points (upper, middle, and lower) are set along the height of the filter bag. The n sound sensors are evenly distributed within the data monitoring points. The upper data monitoring point corresponds to the pulse-jet cleaning area, the middle data monitoring point corresponds to the filter body area, and the lower data monitoring point corresponds to the area above the ash hopper, thereby effectively reflecting the dust removal effect, filtration resistance, and ash accumulation status of the filter bag. The bag filter includes a pulse-jet cleaning area, a filter body area, and an area above the ash hopper.

[0037] The sound sensors are evenly distributed on the left and right sides of the housing of the bag filter or symmetrically distributed along the diagonal of the housing. The sound sensors are used to capture acoustic difference information in different airflow channels and improve the spatial resolution of the prediction model.

[0038] In one embodiment, a protective cover and a vibration-damping mounting structure are also provided. The sound sensor is mounted on the outer surface of the housing through the vibration-damping mounting structure. The protective cover covers the outer periphery of the sound sensor, which serves to resist high temperatures and prevent dust, thereby ensuring the long-term stable operation of the sound sensor.

[0039] It should be noted that the sampling frequencies of the n sound sensors are kept consistent and the sampling times are synchronized to ensure the accuracy of wavelet time-frequency analysis and feature fusion calculation.

[0040] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for predicting the residual life of a filter bag of a bag-type dust collector based on a plurality of sound sensors, characterized by, The method comprises the following steps: S1, obtaining a sound signal of an i th sound sensor at a current time, and preprocessing the sound signal to obtain an effective filter bag sound signal of the i th sound sensor; wherein i≥1, i is a positive integer; S2, obtaining a scale parameter and a time translation parameter of the effective filter bag sound signal of the i th sound sensor, performing time-frequency analysis on the effective filter bag sound signal through wavelet transform according to the scale parameter and the time translation parameter, and obtaining a filter bag aging characteristic signal in the effective filter bag sound signal of the i th sound sensor; S3, determining whether i is greater than or equal to n; if not, assigning i+1 to i, and returning to step S1; if yes, outputting the filter bag aging characteristic signals of all sound sensors; wherein n is the number of sound sensors; S4, obtaining a filter bag structure parameter, and fusing the filter bag aging characteristic signals of all sound sensors according to the filter bag structure parameter to obtain a filter bag remaining life index at the current time; S5, obtaining an average aging rate of the filter bag, and determining the filter bag remaining life at the current time according to the average aging rate and the filter bag remaining life index at the current time.

2. The method of claim 1, wherein the method further comprises: The step S2 specifically comprises the following steps: obtaining a scale parameter and a time translation parameter of the effective filter bag sound signal of the i th sound sensor, and determining a mother wavelet function according to the scale parameter and the time translation parameter; obtaining a first model, determining a wavelet coefficient through the first model according to the mother wavelet function and the effective filter bag sound signal, determining a wavelet amplitude according to the wavelet coefficient, and obtaining the filter bag aging characteristic signal in the effective filter bag sound signal of the i th sound sensor through the wavelet amplitude; wherein the first model is: ； where f(t) is the sound signal; Ψ(t) is the mother wavelet function; a is a scale parameter for adjusting the analysis frequency range; b is a time translation parameter for positioning the instantaneous features of the signal; t is the time; W a,b are the wavelet coefficients.

3. The method of claim 2, wherein the method further comprises: The step of obtaining the filter bag aging characteristic signal in the effective filter bag sound signal of the i th sound sensor through the wavelet amplitude specifically comprises the following steps: obtaining accumulated energy in a preset time window and a scale range, establishing a second model according to the accumulated energy, and determining the filter bag aging characteristic signal according to the second model; wherein the second model is: ； Wherein, S i (t) is the filter bag aging characteristic signal of the i th sound sensor; α1, α2 and α3 are all adjustment coefficients, which can be obtained by looking up table; |W a,b | is the wavelet coefficient amplitude at (a,b), which represents the instantaneous intensity of local sound energy during the operation of the filter bag; E(t) is the cumulative energy in the preset time window and scale range.

4. The method of claim 3, wherein the method further comprises: The step S2 further comprises the following steps: obtaining a third model, determining a wavelet energy density through the third model according to the wavelet amplitude, and determining the optimal values of the scale parameter and the time translation parameter through the maximum value of the wavelet energy density; wherein the third model is: ； wherein E(a, b) is the wavelet energy density at (a, b).

5. The method of claim 3, wherein the method further comprises: The step S4 specifically comprises the following steps: obtaining the length and diameter of the filter bag, establishing a fusion model according to the length, diameter and second model, fusing the filter bag aging characteristic signals of the n sound sensors through the fusion model, and obtaining the filter bag remaining life index at the current time; wherein the fusion model is: ； wherein R t is the filter bag remaining life index; w i is the weighting coefficient of the ith sensor; λ i is the correction coefficient, used to correct the sensor installation position and sensitivity difference; L f and D f are the length and diameter of the filter bag, respectively, and are key parameters representing the filter bag structure parameters; t m is the maximum time.

6. The method of claim 5, wherein the method further comprises: The step S5 specifically comprises the following steps: obtaining an average aging rate of the filter bag, establishing a filter bag aging prediction model according to the average aging rate and the filter bag remaining life index, and determining the filter bag remaining life at the current time according to the filter bag aging prediction model; wherein the filter bag aging prediction model is: ； where t0 is the current time, and R(t0) = R t ; μ is the average aging rate of the filter bag, which is obtained by curve fitting of historical operation data or accelerated aging experiment.

7. The method of claim 1, wherein the method further comprises: The step S1 further comprises the following steps: The upper, middle and lower three layers of data monitoring points are arranged along the height direction of the filter bag, the n sound sensors are uniformly arranged in the data monitoring points, the upper layer of data monitoring points corresponds to the blowing area, the middle layer of data monitoring points corresponds to the filter main body area, and the lower layer of data monitoring points corresponds to the area above the ash bucket; wherein the bag type dust collector comprises a blowing area, a filter main body area and an area above an ash bucket.

8. The method of claim 7, wherein the method further comprises: The sound sensors are uniformly arranged on the left and right sides of the shell of the bag type dust collector or are symmetrically arranged along the diagonal direction of the shell.

9. The method of claim 8, wherein the method further comprises: It also includes a protective cover, the sound sensors are uniformly arranged on the outer surface of the shell, and the protective cover is covered on the outer periphery of the sound sensors.

10. The method of claim 1, wherein the method further comprises: The sampling frequencies of the n sound sensors are consistent, and the sampling times are synchronized.

Images