Intelligent dust cleaning control method and system for bag filter

Abstract

The application relates to the technical field of data acquisition control, and discloses an intelligent ash removal control method and system for a cloth bag dust remover. The method comprises the following steps: constructing a partition disturbance coupling degree matrix; compensating the temperature and viscosity of each partition pressure difference to obtain an equivalent resistance coefficient; identifying an ash removal candidate area by using time sequence prediction and multi-dimensional joint determination; generating an ash removal instruction based on coupling degree constraint scheduling and online stop spraying decision; and adaptively adjusting a threshold value and evaluating a filter bag health degree. Compared with the prior art which depends on a single-point total pressure difference to trigger ash removal, especially under the conditions of high temperature, high dust and multi-filter bag partition parallel filtration, the prior art cannot distinguish the resistances of the partitions, resulting in the technical problems of ash removal delay or excessive blowing. Since the application adopts a distributed differential pressure sensor array and partition independent control, independent sensing of the resistances of the partitions and on-demand ash removal are realized, and the ash removal precision and the service life of the filter bag are improved.

Description

Technical Field

[0001] This invention relates to the field of data acquisition and control technology, and in particular to an intelligent dust removal control method and system for bag filters. Background Technology

[0002] Currently, high-temperature metal bag dust collectors integrating dust and nitrogen removal are widely used in high-temperature flue gas treatment in metallurgy, chemical industry, and other fields. These collectors typically have hundreds of metal filter bags arranged in parallel. Dust-laden gas is introduced into each filtration zone through an airflow distribution device including guide plates and rectifier grids, and the dust layer accumulated on the filter bag surface is removed by pulse jet cleaning. Conventional dust removal control methods involve installing a single-point differential pressure transmitter between the dust collector's main inlet and outlet, using the total inlet-outlet pressure difference as the sole trigger for dust removal. When the total pressure difference exceeds a set upper threshold, the control system sends a synchronous action command to all pulse valves, performing a comprehensive pulse jet cleaning of all filter bags. Existing control optimization paths mainly focus on adjusting the jet pressure, setting fixed or variable-duration jet cycles, and time-series-based cyclic cleaning modes. However, at the sensing level, these methods are still limited to monitoring the single-point total pressure difference and lack independent observation methods for zoned resistance within the filter bag array, making it difficult to reflect the differences in actual dust accumulation and clogging levels in different areas of the filter bags.

[0003] For example, in actual operation, it is difficult for airflow distribution devices to achieve absolute uniformity in each flow channel. In addition, factors such as the initial adhesion state of dust on the filter bag surface and fluctuations in upstream operating conditions lead to significant asynchrony in the dust accumulation rate and resistance growth process of different filtration zones. Typically, this manifests as follows: a thick dust cake has accumulated on the filter bags in one or several zones, causing a sharp increase in local resistance and even airflow blockage. However, because other zones maintain lower resistance, the total pressure difference is averaged down and does not reach the dust removal trigger value. This causes the blocked zone to operate under high pressure difference conditions for a long time, exacerbating the mechanical fatigue of the metal filter bags and potentially causing dust penetration due to excessively high local filtration velocities, affecting emission compliance. When the total pressure difference finally accumulates to the trigger threshold, the dust removal system launches indiscriminate and powerful jet cleaning on all zones. Those zones that are still in a light dust state are forced to endure unnecessary repeated impacts from pulsed airflow, resulting in excessive cleaning of the filter bags. This not only wastes a large amount of compressed air and increases operating energy consumption but also accelerates fatigue damage to the filter bag material.

[0004] Therefore, there is an urgent need for an intelligent control method based on the equivalent resistance characteristics of zones to construct zone-triggered cleaning, which can promptly and directionally clean the dust when any zone shows signs of blockage, avoid ineffective blowing in zones with normal resistance, and effectively solve the problems of cleaning delay and excessive blowing caused by single-point measurement, so as to improve the targeting of cleaning control, system energy efficiency and service life of high-temperature metal filter bags. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to propose an intelligent dust removal control method for baghouse dust collectors. This method addresses the technical problem in existing technologies that rely on single-point total pressure difference to trigger dust removal, especially under conditions of high temperature, high dust, and parallel filtration of multiple filter bags in different zones. This problem results in the inability to distinguish the resistance of filter bags in each zone, leading to delayed dust removal or excessive blowing.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent dust removal control method for bag filters.

[0007] The intelligent dust removal control method for a bag filter includes: Step S10: Obtain the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipe. Based on the spatial layout information and the control range information of the pulse jet branch pipe, perform the distributed differential pressure signal acquisition and arrangement task using a partitioning and coupling calibration method, and output the partitioned sensing and coupling configuration data. Step S20: Based on the partitioned sensing and coupling configuration data, perform partitioned pressure difference signal processing using temperature compensation, viscosity compensation, and equivalent resistance coefficient conversion methods to output partitioned equivalent resistance analysis data; Step S30: Based on the partitioned equivalent resistance analysis data, the dust removal candidate area discrimination task is performed using a time-series prediction and multi-dimensional threshold joint judgment method, and the dust removal candidate area labeling data is output. Step S40: Based on the dust removal candidate area marking data, the partition dust removal instruction generation task is executed using coupling degree constraint priority sorting and online stop spraying decision method, and the partition dust removal execution record data is output; Step S50: Based on the partition cleaning execution record data, correct the cleaning parameters and determine the health status of the filter bags according to the historical efficiency feedback, and output maintenance check prompts.

[0008] Preferably, step S10, which involves acquiring the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipes, and performing a distributed differential pressure signal acquisition and deployment task based on the spatial layout information and the control range information of the pulse jet branch pipes using a partitioning and coupling calibration method, and outputting partitioned sensing and coupling configuration data, specifically includes: Step S101: The metal filter bags inside the dust collector are divided into multiple independent filtration zones according to the flue gas channel formed by the guide plate and the rectifier grid in the airflow distribution device, and each zone boundary corresponds to the independent control range of the pulse jet branch pipe. The total number of independent filtration zones is recorded as M. Step S102: Set an intake-side high-temperature differential pressure measurement point at the filter bag inlet of each independent filtration zone, and set an exhaust-side high-temperature differential pressure measurement point at the outlet of the purified gas manifold above the independent filtration zone. The intake-side high-temperature differential pressure measurement point and the exhaust-side high-temperature differential pressure measurement point form the differential pressure acquisition channel of the independent filtration zone and constitute a distributed differential pressure sensing array. Step S103: Through a cold airflow organization simulation experiment, the i-th independent filtration zone is purged individually in sequence, and the peak value of the instantaneous pressure fluctuation caused by the individual purging of the i-th independent filtration zone in the j-th independent filtration zone is measured. and in accordance with Calculate the matrix elements of the interval airflow disturbance coupling degree matrix The partition sensing and coupling configuration data is obtained. Where i and j are both independent filter partition numbers, and i and j are both integers from 1 to M; This represents the airflow disturbance intensity factor generated in the j-th independent filter zone when the i-th independent filter zone performs pulse cleaning; This represents the peak value of the instantaneous pressure fluctuation measured in its own differential pressure acquisition channel when the i-th independent filter zone is purged individually. To prevent pressure corrections where the denominator is zero; This means that the calculated value is limited to between 0 and 1.

[0009] Preferably, step S20, which involves performing partitioned pressure difference signal processing based on the partitioned sensing and coupling configuration data using temperature compensation, viscosity compensation, and equivalent drag coefficient conversion, and outputting partitioned equivalent drag analysis data, specifically includes: Step S201: According to the sampling period The differential pressure acquisition channels of each independent filtering zone are sampled synchronously to obtain the original differential pressure sequence of the i-th independent filtering zone. After sorting L consecutive sampled values, removing the maximum and minimum values, and taking the arithmetic mean, the filtered pressure difference sequence is obtained. , where t is the sampling time and L is the number of sampled values ​​participating in the sorting filter; Step S202: Obtain the absolute temperature of the insulated thermocouple in the exhaust manifold of the i-th independent filter zone in real time. Look up the table to get the corresponding absolute temperature The dynamic viscosity of dust-laden gas According to the dual-parameter compensation formula of temperature and viscosity Calculate the pressure difference after compensation ; in, For reference absolute temperature, To reference absolute temperature The dynamic viscosity of the dust-laden gas below; Step S203: Obtain the measured flow rate of the heated gas mass flow meter on the inlet branch of the i-th independent filtration zone. ,according to Calculate the instantaneous filtration velocity of the i-th independent filtration zone. and in accordance with Calculate the equivalent resistance coefficient of the i-th independent filter partition. The equivalent resistance coefficient of each independent filtration zone Compensated pressure difference Instantaneous filtration speed The partitioned equivalent drag analysis data is composed of the interval airflow disturbance coupling degree matrix C. in, Let be the inlet volumetric flow rate of the i-th independent filtering partition; is the total filtration area of ​​the i-th independent filtration zone; n is the flow characteristic index, with a value of 1.15; To prevent abnormal wind speed correction when the filtration wind speed is zero.

[0010] Preferably, the reference absolute temperature Take 623 Kelvin, the reference absolute temperature The dynamic viscosity of dust-laden gas Pick Pascal second, the sampling period The interval is set to 0.5 seconds, and the number of sampled values ​​L participating in the sorting and filtering is set to 10.

[0011] Preferably, step S30, which involves performing the dust removal candidate area discrimination task based on the partitioned equivalent resistance analysis data using a time-series prediction and multi-dimensional threshold joint determination method, and outputting the dust removal candidate area labeling data, specifically includes: Step S301: For the i-th independent filtering partition, maintain an equivalent resistance time series sliding window of length W in memory, and store the equivalent resistance coefficients of the most recent W sampling times. Its first-order difference, using a smoothing factor The quadratic exponential smoothing algorithm for the equivalent drag coefficient Perform a forward H-step prediction to obtain the predicted equivalent drag coefficient. ; Where W is the sliding window length of the time series. H is a smoothing factor with a value ranging from 0.1 to 0.3, and H is the number of forward prediction steps. Step S302: According to Calculate the cumulative blocking index of the i-th independent filtering partition. and in accordance with Calculate the coefficient of variation for all independent filtering partitions. ; Where D is the cumulative time window length; For integration time; This serves as the early warning resistance threshold. The standard deviation of the equivalent resistance coefficient for all independent filtering zones at sampling time t; The average equivalent resistance coefficient of all independent filtering zones at sampling time t; To prevent abnormal denominator when the mean is zero; Step S303: When the condition is met , , Any of the conditions in, or satisfying and When the condition is the maximum equivalent drag coefficient among all independent filter zones, the i-th independent filter zone is marked as a dust removal candidate zone, and dust removal candidate zone marking data is generated, which includes candidate zone number, triggering reason, current equivalent drag coefficient, predicted equivalent drag coefficient, cumulative blockage index and inter-zone airflow disturbance coupling degree matrix. in, The initial dust removal start-up resistance threshold, , This is the equivalent resistance coefficient under clean filter bag conditions. The value ranges from 1.8 to 2.2; This represents the lower limit of the cumulative congestion index. The threshold for emergency dust removal resistance, and Not less than 3 times ; This represents the upper limit of the zone resistance dispersion.

[0012] Preferably, step S40, which involves generating a zone cleaning instruction based on the candidate cleaning area marking data using a coupling constraint priority sorting and online stop spraying decision method, and outputting the zone cleaning execution record data, specifically includes: Step S401: Calculate the dynamic cleaning priority for each cleaning candidate area based on the cleaning candidate area marking data. And according to the dynamic dust removal priority A dynamic priority queue is formed from high to low, where the dynamic dust removal priority is... according to calculate; Where q is the sequence number of the independent filter partition that is currently in the dust removal state or the cooling waiting state; The airflow disturbance intensity factor between the i-th independent filter zone and the q-th independent filter zone; This is the cumulative blocking weight coefficient; The coupling penalty weight coefficient; Step S402: Perform constraint verification on the first dust removal candidate area in the dynamic priority queue; the interval between its current position and the end time of the last dust removal of any independent filter partition is not less than the minimum recovery time. The airflow disturbance intensity factor between the filter and an independent filter section that is currently being cleaned or is in a cooling waiting state shall not exceed the upper limit of coupling. And when no other independent filter partition is in the cleaning state at the same time, a pulse cleaning command is generated for the first cleaning candidate area; when the airflow disturbance intensity factor is greater than the coupling limit. At that time, the corresponding dust removal candidate area will be delayed until the cooling waiting time. Re-verify after completion; Step S403: For the equivalent drag coefficient Not less than the emergency dust removal resistance threshold Each independent filter zone skips the dynamic priority queue sorting and generates forced dust removal commands and triggers alarm signals; the preset number of pulses for a single dust removal cycle for each independent filter zone is [number missing]. Delay stabilization time after the kth pulse jet Wait for the airflow to stabilize, then collect the compensated pressure difference for this independent filtration zone. and the stop spray threshold In comparison, when Less than The subsequent pulses are terminated at the designated time, and partition cleaning execution record data is generated, which includes partition number, cleaning start time, actual number of pulses, equivalent resistance coefficient before cleaning, equivalent resistance coefficient after cleaning, and stop spray judgment result. in, Minimum recovery time; This represents the upper limit of coupling. Cooling down waiting time; The preset number of pulses; k is the actual pulse number; To delay the stabilization time; This indicates the differential pressure threshold at which the corresponding independent filtration zone has reached the end of the dust removal process.

[0013] Preferably, step S50, which involves correcting the cleaning parameters and determining the health status of the filter bags based on the historical efficiency feedback according to the partitioned cleaning execution record data, and outputting maintenance check prompts, specifically includes: Step S501: Create an independent historical cleaning file for each individual filter zone, recording the equivalent resistance coefficient before cleaning. Equivalent resistance coefficient after dust removal Actual number of pulses And the runtime after dust removal to the next dust removal trigger. and in accordance with Calculate the efficiency of a single dust removal cycle ; in, The equivalent resistance coefficient before dust removal. This is the equivalent resistance coefficient after dust removal. The actual number of pulses executed. To restore uptime, This refers to the efficiency of a single dust removal cycle; Step S502: Take the efficiency of the last K cleaning cycles moving average efficiency ,when Continuously below the efficiency health lower limit At that time, the dust removal start resistance threshold of the corresponding independent filter zone will be set. Lower and preset number of pulses Increase by 1 to 2 times, among which The current dust removal start-up resistance threshold 5% to 8%; Step S503: After two consecutive adaptive adjustments It has not yet recovered to the lower limit of efficiency and health. When the above is the case, the output points to the maintenance and inspection prompts of the corresponding independent filtration zone. The maintenance and inspection prompts include filter bag damage inspection prompts, pulse valve failure inspection prompts, and blowpipe alignment misalignment inspection prompts.

[0014] The present invention also provides an intelligent dust removal control system for a bag filter, comprising: The partition sensing and coupling configuration module is used to acquire the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipe. Based on the spatial layout information and the control range information of the pulse jet branch pipe, it performs the distributed differential pressure signal acquisition and layout task by partitioning and coupling degree calibration, and outputs partition sensing and coupling configuration data. The differential pressure signal processing module is used to perform the partition differential pressure signal processing task based on the partition sensing and coupling configuration data, using temperature compensation, viscosity compensation and equivalent drag coefficient conversion methods, and output partition equivalent drag analysis data. The dust removal candidate area discrimination module is used to perform the dust removal candidate area discrimination task based on the partition equivalent resistance analysis data and adopt a time-series prediction and multi-dimensional threshold joint judgment method, and output the dust removal candidate area labeling data. The dust removal instruction generation module is used to perform the partition dust removal instruction generation task based on the dust removal candidate area marking data, using coupling degree constraint priority sorting and online stop spraying decision method, and output partition dust removal execution record data; The parameter correction and maintenance prompt module is used to correct the cleaning parameters and determine the health status of the filter bags based on the historical efficiency feedback of the partition cleaning execution record data, and output maintenance check prompts.

[0015] The present invention also provides an intelligent dust removal control device for a bag filter, the intelligent dust removal control device for a bag filter includes: a memory, a processor, and an intelligent dust removal control program for a bag filter stored in the memory and executable on the processor, wherein the intelligent dust removal control program for a bag filter implements the above method when executed by the processor.

[0016] The present invention also provides a computer program product, the computer program product including an intelligent dust removal control program for a bag filter, the intelligent dust removal control program for a bag filter implementing the above method when executed by a processor.

[0017] The beneficial effects of this invention are as follows: 1. By constructing a distributed differential pressure sensing array and combining temperature and viscosity dual-parameter compensation with equivalent resistance coefficient conversion, the interference of gas viscosity changes at high temperatures on resistance monitoring is eliminated. This can accurately reflect the real-time clogging degree of each filter bag zone, providing a true and reliable basis for independent cleaning of each zone. It solves the inherent drawback of traditional single-point total differential pressure measurement being unable to locate local blockages, and avoids high differential pressure operation of filter bags and decreased filtration efficiency caused by delayed cleaning.

[0018] 2. By constraining the cleaning schedule through the interval airflow disturbance coupling degree matrix, the superposition of airflow disturbances caused by simultaneous blowing in adjacent zones is avoided, reducing unnecessary swinging and collision of filter bags. After each pulse, the invalid pulse is terminated in a timely manner by online stop-spray decision, thereby significantly reducing compressed air consumption while ensuring cleaning effect. At the same time, the start threshold and pulse parameters are adaptively adjusted by historical cleaning efficiency feedback, and the health status of filter bags is evaluated and maintenance is prompted in a timely manner, realizing refined protection of the filter bags throughout their entire life cycle, extending the filter bag replacement cycle and reducing the overall operation and maintenance cost of the system. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the first embodiment of an intelligent dust removal control method for a bag filter according to the present invention.

[0020] Figure 2 This is a schematic diagram of the distributed differential pressure sensor array layout in the first embodiment of the intelligent dust removal control method for a bag filter according to the present invention.

[0021] Figure 3 This is a schematic diagram showing the comparison before and after partitioned original differential pressure sliding filtering in the first embodiment of the intelligent dust removal control method for a bag filter according to the present invention.

[0022] Figure 4 This is a schematic diagram comparing the temperature and viscosity dual-parameter compensation effects of the first embodiment of the intelligent dust removal control method for a bag filter according to the present invention.

[0023] Figure 5 This is a schematic diagram illustrating the normalization effect of the equivalent resistance coefficient and filtration velocity in the first embodiment of an intelligent dust removal control method for a bag filter according to the present invention. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0025] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1: As Figure 1 The diagram shown is a flowchart illustrating the first embodiment of an intelligent dust removal control method for a bag filter according to the present invention.

[0027] In the first embodiment, the intelligent dust removal control method for a bag filter includes: Step S10: Obtain the spatial layout information of all metal filter bags inside the dust collector and the independent control range information of the pulse jet branch pipes. Based on the information, use the partitioning and coupling degree calibration mechanism to execute the distributed differential pressure sensor array deployment and the construction of the inter-regional airflow disturbance coupling degree matrix, and output the partitioned sensing and coupling configuration information. The distributed differential pressure sensing array consists of M zone differential pressure acquisition channels. Its layout is based on the spatial arrangement information of all metal filter bags inside the dust collector and the range information of independent control of each pulse jet branch pipe. Within each independent filtration zone, a high-temperature micro differential pressure transmitter on the inlet side is installed behind the airflow distribution plate and before the filter bag inlet. Simultaneously, a high-temperature micro differential pressure transmitter on the exhaust side is installed at the outlet of the purified gas manifold above this zone. The differential pressure between the two measuring points represents the total resistance of the filter bag and its surface powder layer in that zone. The zone-specific airflow disturbance coupling degree matrix C is an M×M dimensional matrix. Its elements represent the airflow disturbance intensity factor generated in zone j when zone i performs pulse cleaning, with values ​​limited to 0 to 1. The initial assignment of this matrix was completed through a cold-state airflow organization simulation experiment: each zone was individually pulse-cleaned sequentially, and the instantaneous pressure fluctuation peak value of the remaining zones at the time of pulse cleaning was collected one by one. Normalization calculations were performed based on the maximum instantaneous back pressure of the cleaning zone itself. The output information includes partitioned sensing configuration and coupling configuration information. The former specifies the location of the measurement points and the corresponding acquisition channels for each partition, while the latter provides the initial values ​​of each element in the matrix and the rules for possible subsequent updates. This information provides a spatial mapping and mutual interference quantification basis for the targeted acquisition of partitioned pressure differentials in the S20 stage and the coupling constraints of the dust removal scheduling in the S40 stage.

[0028] By using the flue gas duct defined by the guide vanes and rectifier grilles in the airflow distribution device as physical boundaries to divide the filtration zones, and ensuring that the boundaries of each zone perfectly match the control range of the pulse jet branch pipes, this step structurally ensures a one-to-one correspondence between the monitored objects and the objects to be cleaned, eliminating signal crosstalk that may result from boundary ambiguity. The deployed distributed differential pressure sensor array enables the system to independently acquire the differential pressure signal of each filtration zone, no longer relying on the overall total differential pressure of the entire equipment. The coupling degree matrix, through cold-state simulation experiments, transforms the flow field disturbance intensity caused by the cleaning action in adjacent areas into calculable numerical constraints. The subsequent scheduler can directly use these values ​​as judgment parameters without performing complex online flow field calculations. Thus, the previously invisible resistance differences in the filter bag area and the mutual interference relationship during cleaning are transformed into structured, perceptible, and constrainable information, laying the foundation for accurate extraction of resistance signals and spatiotemporal isolation control of time-sharing and zoned cleaning in subsequent steps.

[0029] Traditional dust collector cleaning control typically involves installing a single-point differential pressure transmitter between the main inlet and outlet of the equipment. This system can only sense the overall differential pressure, failing to detect differences in resistance distribution across different filter bag areas, and cannot assess the degree of airflow disturbance to adjacent areas during cleaning in one zone. In high-temperature, multi-zone parallel filtration scenarios, if the total differential pressure has not yet reached the trigger threshold, locally clogged zones will continue to operate under high differential pressure. Once the total differential pressure triggers cleaning, the control system sends synchronous commands to all pulse valves, performing indiscriminate overall blowing across all zones. This crude control method cannot identify the local sources of resistance growth, nor can it predict or avoid inter-zone interference caused by cleaning. This step, through proactive physical zoning and coupling calibration, constructs a resistance sensing network covering the entire space and a zone-specific disturbance quantification matrix. This enables the control system to independently observe the state of each execution unit and quantify the interactive effects of cleaning actions in advance, fundamentally overcoming the shortcomings of traditional single-point measurement methods from an information perspective.

[0030] For example, a high-temperature metal bag dust collector integrating dust and nitrification is divided into six independent filtration zones, each containing approximately 50 metal filter bags. Each zone corresponds to an independent pulse jet cleaning branch pipe. In implementation, a high-temperature differential pressure transmitter is installed behind the airflow distribution plate and at the outlet of the upper manifold in zone 3, forming an independent differential pressure acquisition channel for zone 3. During the cold airflow organization simulation experiment, when zone 1 is pulse-jet cleaned alone, the peak instantaneous pressure fluctuation of zone 3 is measured to be 120 Pa, while the maximum instantaneous back pressure generated by zone 1 itself is 400 Pa. The normalized coupling degree C13 is then 0.3. This method is repeated pairwise to obtain a complete 6×6 coupling degree matrix. In subsequent actual operation, if zone 1 is listed as a candidate zone for dust removal and is prepared for dust removal, the system can assess the degree of disturbance to zone 3 based on the C13 value and use it as one of the scheduling constraints. This deployment and calibration method ensures a clear correspondence between the sensing range and control range of each zone, and the disturbance relationship is clearly quantified.

[0031] Specifically, such as Figure 2As shown in the diagram, this sub-figure illustrates the independent filtration zones inside the dust collector using six rectangular areas. Each zone has an inlet-side high-temperature differential pressure transmitter (red dot) located behind the airflow distribution plate and before the filter bag inlet, and an exhaust-side high-temperature differential pressure transmitter (blue dot) located at the outlet of the purified gas manifold above the zone, forming a zone differential pressure acquisition channel. This arrangement allows the total resistance of the filter bag and its surface powder cake layer in each zone to be measured independently, no longer relying on the overall average pressure difference. Traditional single-point total pressure difference measurement mixes the resistance states of the six zones together, and local blockages in individual zones can easily be leveled out by other cleaner zones, leading to delayed dust removal. This application, through a distributed sensor array, can promptly capture subtle increases in the resistance of each zone, providing an accurate and spatially resolved raw signal basis for subsequent calculation of the equivalent resistance coefficient and identification of dust removal candidate zones, enhancing the specificity and timeliness of the dust removal response.

[0032] Step S20: Based on the partition sensing and coupling configuration information, the partition pressure difference signal acquisition and normalization processing task is performed using a temperature and viscosity dual-parameter compensation and equivalent drag coefficient conversion mechanism, and the equivalent drag coefficient of each partition and the airflow disturbance coupling degree matrix between partitions are output. The equivalent drag coefficient for each zone is the core output of this step. Its calculation is based on the zone differential pressure after temperature and viscosity dual-parameter compensation, and the instantaneous filtration velocity converted from the measured flow rate of a thermal gas mass flow meter. The compensation process first obtains the real-time absolute temperature through a sheathed thermocouple attached to the exhaust manifold of each zone. Then, the dynamic viscosity of the dust-laden gas at that temperature is obtained from a table. The filtered differential pressure value is corrected according to the dual-parameter compensation formula to eliminate pressure drift caused by temperature fluctuations. The filtering process uses an averaging method after removing the maximum and minimum values ​​within a sliding window, which effectively suppresses spike interference caused by pulse jets and valve actions. The equivalent drag coefficient is obtained by dividing the compensated differential pressure by the flow characteristic exponent of the filtration velocity. This exponent is 1.15, reflecting the characteristic that the high-temperature dust-laden airflow does not completely obey linear seepage laws in a non-laminar flow state. The partitioned sensing configuration output in step S10 provides a channel mapping for differential pressure acquisition. The equivalent drag coefficient of each partition and the inter-partition airflow disturbance coupling matrix output in this step are directly used as the real-time status observation quantity for the dust removal candidate area discrimination in S30.

[0033] This step, through a series of signal conditioning steps, converts the original differential pressure signal, which is affected by multiple factors such as temperature, viscosity, and airflow, into an equivalent resistance coefficient that reflects only the intrinsic resistance of the filter bag's powder cake layer. The temperature and viscosity dual-parameter compensation stage uniformly corrects the differential pressure values ​​measured under different temperature conditions to the same design reference condition, ensuring temporal consistency and comparability of differential pressure data at different times and eliminating resistance distortion caused by furnace temperature changes. Dividing by the flow characteristic exponent of the filtration velocity eliminates differential pressure changes caused by fluctuations in the processed airflow due to process output adjustments. Even during airflow increases or decreases, the equivalent resistance coefficient can still stably characterize the actual degree of filter bag blockage. Sliding window mean filtering removes the contamination of the signal by sudden transients such as jetting and valve actions. After this chain of processing, the data sequence sent to the S30 timing prediction module becomes smooth and has a clear physical meaning. The prediction algorithm will not be falsely triggered by high-frequency glitches or ambient temperature drift, thus improving the reliability of dust removal judgment.

[0034] Traditional dust removal control systems typically compare the total differential pressure value directly to a fixed threshold, with almost no independent compensation for temperature or viscosity. Within the operating temperature range of 300-400 degrees Celsius for dust-laden gas, the dynamic viscosity of the gas increases significantly with increasing temperature. Without viscosity correction for the measured differential pressure, even if the filter bags are clean at higher temperatures, the measured differential pressure may be too high, easily triggering unnecessary dust removal. Conversely, at lower temperatures, the differential pressure may be too low, delaying dust removal in clogged zones. Furthermore, directly using the raw differential pressure without considering changes in filtration velocity can misinterpret airflow fluctuations as changes in filter bag resistance, leading to inconsistent dust removal timing. Frequent load adjustments render the dust removal initiation threshold meaningless. This step introduces dual-parameter compensation for temperature and viscosity, and normalizes the airflow velocity, actively eliminating these two sources of interference. This ensures that the dust removal judgment threshold remains consistent with the actual degree of clogging within the normal operating condition fluctuation range, reducing misjudgments and missed judgments caused by changes in environmental parameters.

[0035] Specifically, such as Figure 3 As shown, this sub-figure plots the original differential pressure signal (gray curve, containing random spikes) and the differential pressure signal after sliding window sorting and extreme value removal averaging filtering (blue curve) on the same time axis. Transient spikes in the original differential pressure caused by electromagnetic pulse valve action or local airflow impact are effectively removed, resulting in a smooth curve with a clear trend. Traditional control systems, if directly including these spikes in the judgment, may lead to false triggering of dust removal or obscuring the true trend. The extreme value removal averaging filtering of this application, while retaining continuous resistance change information, significantly suppresses non-operating condition noise, providing a higher quality input signal for subsequent temperature and viscosity compensation and equivalent resistance coefficient calculation, thereby reducing the risk of misjudgment caused by data spikes and improving the system's robustness. Figure 4As shown, this subplot displays the evolution of the differential pressure before compensation (gray curve) and after compensation (red curve) over time on the left vertical axis, while the right vertical axis simultaneously plots the absolute temperature change at the corresponding moment (orange dashed line). It can be seen that when the temperature fluctuates by an order of ±40K, the differential pressure before compensation fluctuates accordingly. However, after compensation using both temperature and viscosity parameters, the temperature component of the differential pressure curve is significantly eliminated, retaining only the signal characteristics caused by changes in filter bag resistance. Traditional control ignores the influence of temperature, directly comparing the original differential pressure with a fixed threshold. When the temperature rises, the differential pressure reading increases, easily triggering unnecessary cleaning prematurely, resulting in wasted compressed air and excessive filter bag wear; conversely, when the temperature decreases, it may trigger delayed, leading to increased clogging. This application normalizes the differential pressure values ​​collected under different temperature conditions by introducing a reference temperature of 623K and a reference viscosity, ensuring a unified standard for the cleaning trigger threshold across the entire temperature range and improving the environmental adaptability of the cleaning decision. Figure 5 As shown in the figure, this subplot compares the changes in equivalent resistance coefficients (blue, green, and red dots) calculated after temperature and viscosity compensation and the uncompensated original equivalent resistance coefficients (gray crosses) with filtration velocity at different temperatures (623K, 653K, and 593K). The compensated equivalent resistance coefficients are basically unaffected by wind velocity, changing only with the filter bag resistance state, with concentrated and narrowly distributed scatter points; while the uncompensated coefficients fluctuate significantly with wind velocity, showing obvious deviations, especially in low and high wind velocity regions. This is because the uncompensated scheme includes differential pressure changes caused by airflow fluctuations. When process adjustments lead to changes in the treated airflow, traditional control systems easily misinterpret airflow increases or decreases as resistance increases or decreases, thus triggering incorrect dust removal commands. This application successfully normalizes the wind velocity by dividing the compensated differential pressure by the 1.15th power of the filtration velocity, ensuring that the equivalent resistance coefficient remains stable within the normal operating condition fluctuation range.

[0036] Step S30: Based on the equivalent drag coefficient of each partition and the airflow disturbance coupling degree matrix of the interval, the time-series prediction and multi-dimensional joint judgment mechanism is used to perform the dust removal candidate area discrimination task, and output the dust removal candidate area mark and the airflow disturbance coupling degree matrix of the interval; The inputs to this step are the equivalent drag coefficients of each partition and the inter-partition airflow disturbance coupling matrix. The outputs are the candidate area markers for dust removal and the updated coupling matrix. Time-series prediction employs a quadratic exponential smoothing algorithm with a forgetting factor. For each partition, a time-series sliding window of length W is maintained in memory, storing the equivalent drag coefficients and their first-order differences for the most recent W sampling times. The smoothing factor is then used... The trend and level are smoothed simultaneously between 0.1 and 0.3 to predict the equivalent resistance coefficient for the next H steps. The cumulative over-limit index is obtained by integrating the deviation of the equivalent resistance coefficient from the warning threshold within a moving time window D, and is used to quantify the severity of persistent over-limit in a partition. The coefficient of variation of the equivalent resistance coefficient of all partitions reflects the unevenness of the resistance distribution across all partitions, and is used to identify the overall distribution distortion caused by abnormally high resistance in individual partitions. The joint determination of the dust removal candidate area includes three triggering conditions: the predicted value reaches the regular triggering threshold and the cumulative over-limit index is not lower than the lower limit; the current value reaches the emergency threshold; or the coefficient of variation of all partitions exceeds the limit and the resistance of that partition is at its maximum value. The output dust removal candidate area markers are directly submitted to S40 for scheduling decisions.

[0037] This step, by introducing a predictive mechanism and multi-dimensional joint judgment, shifts the decision-making node for initiating dust removal from post-event detection to in-event prediction. This allows the control system to lock onto potential dust removal targets before the resistance actually reaches the trigger threshold, reserving time margin for the S40 scheduler to handle coupling constraints and cooling waiting. The quadratic exponential smoothing algorithm balances data trend capture with random noise suppression, effectively tracking the nonlinear resistance growth process. The cumulative over-limit index filters out short-term, minor over-limits, responding only to continuous accumulation of over-limits, avoiding misjudgments caused by transient sensor disturbances or short-term airflow fluctuations. The coefficient of variation condition, from a global system perspective, actively extracts the most contributing abnormal zones when the overall resistance distribution is severely uneven, preventing local extreme cases from being masked by the average value. Through these processes, the calibration of dust removal candidate zones becomes more focused, reducing missed and delayed judgments, enabling the S40 to allocate dust removal resources to the truly needed zones at the correct time.

[0038] Traditional dust removal control relies almost entirely on whether the current total pressure difference exceeds a fixed threshold, neglecting both the rate and trend of resistance changes and the difficulty in distinguishing whether resistance increases originate from individual zones or a uniform overall rise. In actual operation, filter bag resistance often experiences a period of accelerated increase. By the time the total pressure difference finally reaches the trigger value, some zones may have already been operating under high pressure differentials for an extended period, leading to increased mechanical fatigue of the filter bags. Furthermore, for sudden changes in local resistance caused by unexpected filter bag damage or pulse valve leakage, the fixed threshold either responds slowly or fails to recognize them at all. Under multi-zone parallel filtration conditions, the total pressure difference at a single point cannot reflect internal non-uniformity. An abnormally high resistance in one zone can easily be averaged down by other low-resistance zones, creating a blind spot in judgment. This step, through a combination of prediction and multi-dimensional criteria, incorporates future trends, continuous accumulation, and relative anomaly levels into the dust removal triggering logic. This overcomes the slow judgment and structural blind spots of traditional methods in the face of dynamic changes and local anomalies, making the determination of dust removal initiation timing more forward-looking and discriminative.

[0039] For example, during the operation of a dust collector, the equivalent resistance coefficient of zone 5 showed a gradual increasing trend over 15 consecutive minutes. Calculations using a quadratic exponential smoothing algorithm with a smoothing coefficient of 0.2 predicted that the coefficient would reach 2150 in 3 minutes, exceeding the normal start-up threshold of 2000. Simultaneously, the cumulative over-limit index of this zone over the past 5 minutes had already exceeded the system's set lower limit. Therefore, while the actual resistance coefficient was still at 1950 and had not yet reached the threshold, zone 5 was pre-marked as a candidate zone for dust removal and sent to the scheduling queue. At the same time, zone 2 experienced filter bag damage, causing its equivalent resistance coefficient to jump instantly from 2100 to 3600, exceeding the emergency threshold of 3100. The system immediately marked it as a candidate zone for dust removal and triggered an alarm; the scheduler could prioritize its handling without waiting for the normal queue. Furthermore, when the coefficient of variation for all six zones rises to 0.35, exceeding the set upper limit of 0.3, and zone 7 has the highest resistance coefficient, even though it has not yet reached the normal trigger line, the system identifies it as the main contributor to unevenness and lists it as a candidate zone for dust removal. This multi-dimensional joint decision-making method makes the dust removal triggering timing closer to the actual needs of the filter zones.

[0040] Step S40: Based on the candidate cleaning area markers and the inter-area airflow disturbance coupling degree matrix, the coupling degree constraint priority scheduling and online stop spray decision mechanism are used to execute the inter-area cleaning instruction generation task, and output the cleaning execution instructions of each area and the equivalent resistance coefficient change data before and after cleaning. The dynamic priority queue maintained by the dust removal scheduler is arranged from largest to smallest according to the current equivalent resistance coefficient of each partition. The partition at the head of the queue receives the highest dust removal priority, and the control system issues an execution command of a preset pulse sequence to the electromagnetic pulse valve corresponding to that partition. The constraint verification process includes three conditions: First, the elapsed time since the end of the last dust removal of any partition must be greater than the minimum recovery time; second, the corresponding element values ​​in the coupling degree matrix C between the candidate partition and all partitions that are currently in the dust removal state or have just ended dust removal and are still within the cooling time protection period must not exceed the upper limit of the coupling degree set by the system; third, at most one filter partition is allowed to be in the dust removal execution state at the same time. After each pulse jet is completed, the online stop-jet decision mechanism waits for the airflow to recover after a stable delay time, and then collects the post-compensation differential pressure of the partition. If it has fallen below the stop threshold, the subsequent pulses that have not yet been executed are immediately stopped. The details of the dust removal execution command output in this step and the change data of the equivalent resistance coefficient before and after dust removal are the direct basis for evaluating the dust removal efficiency and making adaptive parameter adjustments in the S50 stage.

[0041] Traditional pulse cleaning control typically uses a fixed number of pulses and a fixed rotation cycle to sequentially clean all filter bags, or simultaneously opens all pulse valves for overall cleaning once the total pressure difference is triggered. Neither of these methods can allocate cleaning energy according to the actual degree of clogging in each area. Areas with severe clogging may not be thoroughly cleaned due to insufficient pulse energy, leaving a thicker powder cake layer, while cleaner areas are forced to endure excessive mechanical impact and compressed air consumption. The coupling effect is almost completely ignored; adjacent filter bags are continuously blown in the same time period, and the mutual impact and flow field oscillation exacerbate the fatigue accumulation of the filter bag material. This step uses priority scheduling to concentrate the limited cleaning time window and number of pulses on high-resistance zones. Coupling constraints implement mutual exclusion of cleaning actions in space and time. Online stop-pulse operation cuts off pulses as needed within each zone, thereby ensuring cleaning effectiveness while reducing the number of pulses per filter bag and the ineffective consumption of compressed air, thus protecting the filter bags from mechanical deformation and material fatigue.

[0042] For example, in a certain operation, the dust removal candidate queue includes partitions 2, 4, and 6, with current equivalent resistance coefficients of 2300, 2150, and 2080, respectively. The scheduler selects partition 2 as the first partition to receive dust removal rights. It checks the coupling degree C21 between partition 2 and partition 1, which has just finished dust removal but has not yet reached its cooling time. This C21 is 0.25, which does not exceed the system's set upper limit of 0.5. Furthermore, the time since the last dust removal end time for any partition in the entire system has exceeded the minimum recovery time. Therefore, the electromagnetic pulse valve corresponding to partition 2 is activated, with a preset pulse count of 3. After the first pulse jet is completed, a stabilization time is allowed until the airflow recovers. The system then collects the compensated differential pressure and finds that it has dropped below the stop threshold. The subsequent second and third pulse jets are immediately stopped, and the actual execution count is recorded as 1. Partition 2 then enters the cooling period. The second-best partition 4 waits to meet the minimum recovery time and, through a coupling degree check with partition 2, finds C42 to be 0.3, which is within the limit. It then receives the dust removal rights and is activated. When partition 6 in the queue suddenly experienced a sharp increase in its equivalent resistance coefficient, exceeding the emergency threshold, the system immediately initiated a pulse-jet cleaning and triggered an alarm, disregarding the current queue order and coupling constraints. This control method ensures that the number of pulse-jet cleaning cycles for each partition matches the actual dust removal requirements.

[0043] Step S50: Based on the cleaning execution instructions of each zone and the change data of the equivalent resistance coefficient before and after cleaning, the historical efficiency feedback adaptive mechanism is used to perform the task of optimizing the cleaning parameters of the zone and assessing the health of the filter bag, and output the adjusted cleaning start threshold and maintenance inspection prompts.

[0044] Historical dust removal records are established independently for each filter zone. The records primarily include the equivalent resistance coefficient before and after each dust removal cycle, the actual number of pulses executed, and the recovery time from the end of the current dust removal cycle to the next dust removal cycle for that zone. Dust removal efficiency is defined as the ratio of the actual resistance reduction during the current cycle to the theoretically maximum resistance reduction that could occur when restoring the filter zone from its pre-duration state to its initial clean state. The system uses the moving average of the most recent dust removal efficiencies as a continuous indicator of the current dust removal capability of that zone, smoothing out fluctuations caused by the randomness of individual operations. When the moving average efficiency continuously falls below the preset efficiency health lower limit, the system determines that the dust removal parameters for that zone are no longer suitable for the actual characteristics of the powder cake or the equipment status. It automatically adjusts the dust removal initiation threshold for that zone downwards by 5% to 8% of the current value and increases the preset number of pulses by 1 to 2 times to advance the dust removal intervention time and enhance the blowing force. If the moving average efficiency fails to recover to above the healthy lower limit after two consecutive adaptive parameter adjustments, the system will issue specific maintenance and inspection prompts, pointing to possible mechanical causes such as filter bag damage, pulse valve failure, or misalignment of the blowpipe.

[0045] This step utilizes data on changes in the equivalent resistance coefficient before and after dust removal to construct a progressive feedback loop, from performance monitoring to parameter self-adjustment and then to health diagnosis. When a continuous decline in the dust removal efficiency of a certain zone is detected, the system presumes that the underlying cause may be an excessively high start-up threshold setting leading to delayed dust removal intervention, or insufficient preset pulse counts to effectively remove the increasingly dense powder cake. Therefore, it proactively lowers the start-up threshold and appropriately increases the pulse supply to restore dust removal efficiency. This adjustment process is completed automatically, and timely compensation can be provided when the residual resistance on the filter bag surface gradually increases or minor leaks occur. When continuous adaptive adjustment still cannot reverse the downward trend in efficiency, the system rules out simple parameter mismatch issues and instead pinpoints faults at the component or structural level, providing maintenance personnel with clear troubleshooting directions, such as checking for mechanical damage to the corresponding filter bag, whether the electromagnetic pulse valve is opening and closing normally, and whether the blow-off branch pipe is misaligned. This tiered response mechanism enables dust removal control to adapt to performance changes throughout the filter bag's entire lifecycle and transforms maintenance from passive emergency response to state-based proactive prevention.

[0046] Traditional dust removal systems often use fixed dust removal thresholds and a fixed number of pulses, failing to respond to changes in residual resistance of the filter media after long-term operation. As operating time accumulates, a difficult-to-remove residue layer gradually forms on the filter bag surface. At this point, the same pulse energy is no longer as effective as with new bags in removing the powder cake, resulting in a progressively smaller resistance reduction after dust removal. This forces a continuous increase in the dust removal frequency, creating a vicious cycle and leading to premature filter bag failure. Traditional controls lack the ability to detect and adjust for this, simply continuing to execute the original parameter settings. This step, by continuously tracking the actual efficiency of each dust removal cycle, automatically compensates the control parameters appropriately in the early stages of continuous efficiency degradation, which can slow down the rate of deterioration in the dust removal effect to some extent. When the compensation measures reach their limit, the system can proactively alert potential fault points, avoiding continued blind operation under ineffective parameters. Compared to fixed parameter strategies, this adaptive and health assessment mechanism helps extend the effective service life of filter bags and shifts maintenance from reactive response to status-based alerts.

[0047] For example, during the stable operation phase, the sliding average efficiency of a certain section after the last five cleaning cycles remained at 0.88, higher than the system's set lower limit of 0.6. As a difficult-to-remove chemical deposit gradually formed on the filter bag surface, the sliding average efficiency of the last five cleaning cycles dropped to 0.52. The system determined that the cleaning capacity was insufficient and automatically lowered the cleaning start threshold for that section by approximately 7% from 2150 to 2000, and increased the preset number of pulses from 2 to 3. After the next cleaning cycle following the parameter adjustment, the sliding average efficiency rebounded to 0.65, indicating that the adjustment was effective. Another section, after undergoing the same adaptive adjustment, still had a sliding average efficiency below 0.6 for two consecutive cycles (e.g., 0.55 and 0.50 after the two adjustments). The system then generated a maintenance check prompt text, prompting maintenance personnel to check for mechanical damage to the metal filter bags in that section, whether the corresponding electromagnetic pulse valves were functioning correctly, and whether the blow-through branch pipes were misaligned. On-site inspection confirmed a micro-leak in the pulse valve diaphragm, causing insufficient actual blowing pressure. After replacing the component, the dust removal efficiency returned to normal. This mechanism enables the system to proactively identify performance degradation and promptly indicate the direction of intervention.

[0048] Example 2: Furthermore, the present invention provides an intelligent dust removal control system for baghouse dust collectors, employing an intelligent dust removal control method for baghouse dust collectors as described in the above embodiments, which can solve the technical problems of intelligent dust removal control for baghouse dust collectors. The beneficial effects of the intelligent dust removal control system for baghouse dust collectors provided by the present invention are the same as those of the intelligent dust removal control method for baghouse dust collectors provided in the above embodiments, and other technical features in the intelligent dust removal control system for baghouse dust collectors are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0049] Example 3: This invention provides an intelligent dust removal control device for baghouse dust collectors. The device includes at least one processor and a memory communicatively connected to the processor. The memory stores instructions executable by the processor, which are then executed to enable the processor to perform the intelligent dust removal control method for baghouse dust collectors described in Example 1. The intelligent dust removal control device for baghouse dust collectors in this invention can include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and vehicle terminals (e.g., vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This intelligent dust removal control device for baghouse dust collectors is merely an example and should not limit the functionality or scope of the invention. A smart dust removal control device for a bag filter may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory or a program loaded from a storage device into a random access memory. The random access memory also stores various programs and data required for the operation of the smart dust removal control device for a bag filter. The processing unit, the read-only memory, and the random access memory are interconnected via a bus. An I / O interface is also connected to the bus. Typically, the following systems can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. The communication device allows the smart dust removal control device for a bag filter to communicate wirelessly or wiredly with other devices to exchange data. Although a smart dust removal control device for a bag filter with various systems has been described, it should be understood that it is not required to implement or possess all the systems described. It can be implemented alternatively or with more or fewer systems.

[0050] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the intelligent dust removal control method for a bag filter as described above. The computer program product provided by this invention can solve the technical problem of intelligent dust removal control for a bag filter. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as the beneficial effects of the intelligent dust removal control method for a bag filter provided in the above embodiments, and will not be repeated here.

[0051] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a read-only memory. When the computer program is executed by a processing device, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0052] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0053] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the present invention and its equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A smart dust removal control method for a bag filter, characterized in that, The method includes: Step S10: Obtain the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipe. Based on the spatial layout information and the control range information of the pulse jet branch pipe, perform the distributed differential pressure signal acquisition and arrangement task using a partitioning and coupling calibration method, and output the partitioned sensing and coupling configuration data. Step S20: Based on the partitioned sensing and coupling configuration data, perform partitioned pressure difference signal processing using temperature compensation, viscosity compensation, and equivalent resistance coefficient conversion methods to output partitioned equivalent resistance analysis data; Step S30: Based on the partitioned equivalent resistance analysis data, the dust removal candidate area discrimination task is performed using a time-series prediction and multi-dimensional threshold joint judgment method, and the dust removal candidate area labeling data is output. Step S40: Based on the dust removal candidate area marking data, the partition dust removal instruction generation task is executed using coupling degree constraint priority sorting and online stop spraying decision method, and the partition dust removal execution record data is output; Step S50: Based on the partition cleaning execution record data, correct the cleaning parameters and determine the health status of the filter bags according to the historical efficiency feedback, and output maintenance check prompts.

2. The intelligent dust removal control method for a bag filter as described in claim 1, characterized in that, Step S10 involves acquiring the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipes. Based on the spatial layout information and the control range information of the pulse jet branch pipes, a distributed differential pressure signal acquisition and deployment task is performed using a partitioning and coupling calibration method. The step of outputting partitioned sensing and coupling configuration data specifically includes: Step S101: The metal filter bags inside the dust collector are divided into multiple independent filtration zones according to the flue gas channel formed by the guide plate and the rectifier grid in the airflow distribution device, and each zone boundary corresponds to the independent control range of the pulse jet branch pipe. The total number of independent filtration zones is recorded as M. Step S102: Set an intake-side high-temperature differential pressure measurement point at the filter bag inlet of each independent filtration zone, and set an exhaust-side high-temperature differential pressure measurement point at the outlet of the purified gas manifold above the independent filtration zone. The intake-side high-temperature differential pressure measurement point and the exhaust-side high-temperature differential pressure measurement point form the differential pressure acquisition channel of the independent filtration zone and constitute a distributed differential pressure sensing array. Step S103: Through a cold airflow organization simulation experiment, the i-th independent filtration zone is purged individually in sequence, and the peak value of the instantaneous pressure fluctuation caused by the individual purging of the i-th independent filtration zone in the j-th independent filtration zone is measured. And calculate the matrix elements of the interval airflow disturbance coupling degree matrix. The partition sensing and coupling configuration data is obtained.

3. The intelligent dust removal control method for a bag filter as described in claim 1, characterized in that, Step S20, which involves performing partitioned pressure difference signal processing based on the partitioned sensing and coupling configuration data using temperature compensation, viscosity compensation, and equivalent drag coefficient conversion, and outputting partitioned equivalent drag analysis data, specifically includes: Step S201: According to the sampling period The differential pressure acquisition channels of each independent filtering zone are sampled synchronously to obtain the original differential pressure sequence of the i-th independent filtering zone. After sorting L consecutive sampled values, removing the maximum and minimum values, and taking the arithmetic mean, the filtered pressure difference sequence is obtained. , where t is the sampling time and L is the number of sampled values ​​participating in the sorting filter; Step S202: Obtain the absolute temperature of the insulated thermocouple in the exhaust manifold of the i-th independent filter zone in real time. Look up the table to get the corresponding absolute temperature The dynamic viscosity of dust-laden gas According to the dual-parameter compensation formula of temperature and viscosity Calculate the pressure difference after compensation ; in, For reference absolute temperature, To reference absolute temperature The dynamic viscosity of the dust-laden gas below; Step S203: Obtain the measured flow rate of the heated gas mass flow meter on the inlet branch of the i-th independent filtration zone. ,according to Calculate the instantaneous filtration velocity of the i-th independent filtration zone. and in accordance with Calculate the equivalent resistance coefficient of the i-th independent filter partition. The equivalent resistance coefficient of each independent filtration zone Compensated pressure difference Instantaneous filtration speed The partitioned equivalent drag analysis data is composed of the interval airflow disturbance coupling degree matrix C. in, Let be the inlet volumetric flow rate of the i-th independent filtering partition; is the total filtration area of ​​the i-th independent filtration zone; n is the flow characteristic index, with a value of 1.15; To prevent abnormal wind speed correction when the filtration wind speed is zero.

4. The intelligent dust removal control method for a bag filter as described in claim 3, characterized in that, In step S202, the reference absolute temperature Take 623 Kelvin, the reference absolute temperature The dynamic viscosity of dust-laden gas Pick Pascal second, the sampling period The interval is set to 0.5 seconds, and the number of sampled values ​​L participating in the sorting and filtering is set to 10.

5. The intelligent dust removal control method for a bag filter as described in claim 1, characterized in that, Step S30, which involves performing a dust removal candidate area discrimination task based on the partitioned equivalent resistance analysis data using a time-series prediction and multi-dimensional threshold joint determination method, and outputting dust removal candidate area labeling data, specifically includes: Step S301: For the i-th independent filtering partition, maintain an equivalent resistance time series sliding window of length W in memory, and store the equivalent resistance coefficients of the most recent W sampling times. Its first-order difference, using a smoothing factor The quadratic exponential smoothing algorithm for the equivalent drag coefficient Perform a forward H-step prediction to obtain the predicted equivalent drag coefficient. ; Where W is the sliding window length of the time series. H is a smoothing factor with a value ranging from 0.1 to 0.3, and H is the number of forward prediction steps. Step S302: Calculate the cumulative blocking index of the i-th independent filtering partition. And calculate the coefficient of variation for all independent filtering partitions. ; Step S303: When the condition is met , , Any of the conditions in, or satisfying and When the condition is the maximum equivalent drag coefficient among all independent filter zones, the i-th independent filter zone is marked as a dust removal candidate zone, and dust removal candidate zone marking data is generated, which includes candidate zone number, triggering reason, current equivalent drag coefficient, predicted equivalent drag coefficient, cumulative blockage index and inter-zone airflow disturbance coupling degree matrix. in, The initial dust removal start-up resistance threshold. , This is the equivalent resistance coefficient under clean filter bag conditions. The value ranges from 1.8 to 2.2; This represents the lower limit of the cumulative congestion index. The threshold for emergency dust removal resistance, and Not less than 3 times ; This represents the upper limit of the zone resistance dispersion.

6. The intelligent dust removal control method for a bag filter as described in claim 1, characterized in that, Step S40, which involves generating a zone cleaning instruction based on the candidate cleaning area marking data using a coupling constraint priority sorting and online stop spraying decision method, and outputting the zone cleaning execution record data, specifically includes: Step S401: Calculate the dynamic cleaning priority for each cleaning candidate area based on the cleaning candidate area marking data. And according to the dynamic dust removal priority A dynamic priority queue is formed from high to low. Step S402: Perform constraint verification on the first dust removal candidate area in the dynamic priority queue; the interval between its current position and the end time of the last dust removal of any independent filter partition is not less than the minimum recovery time. The airflow disturbance intensity factor between the filter and an independent filter section that is currently being cleaned or is in a cooling waiting state shall not exceed the upper limit of coupling. And when no other independent filter partition is in the cleaning state at the same time, a pulse cleaning command is generated for the first cleaning candidate area; when the airflow disturbance intensity factor is greater than the coupling limit. At that time, the corresponding dust removal candidate area will be delayed until the cooling waiting time. Re-verify after completion; Step S403: For the equivalent drag coefficient Not less than the emergency dust removal resistance threshold Each independent filter zone skips the dynamic priority queue sorting and generates forced dust removal commands and triggers alarm signals; the preset number of pulses for a single dust removal cycle for each independent filter zone is [number missing]. Delay stabilization time after the kth pulse jet Wait for the airflow to stabilize, then collect the compensated pressure difference for this independent filtration zone. And the stop spray threshold that meets the preset dust removal end conditions. In comparison, when Less than The subsequent pulses are terminated at the designated time, generating partition cleaning execution record data that includes the partition number, cleaning start time, actual number of pulses, equivalent resistance coefficient before cleaning, equivalent resistance coefficient after cleaning, and stop spray judgment result.

7. The intelligent dust removal control method for a bag filter as described in claim 1, characterized in that, Step S50, which involves correcting cleaning parameters and determining filter bag health status based on historical efficiency feedback according to the partitioned cleaning execution record data, and outputting maintenance check prompts, specifically includes: Step S501: Create an independent historical cleaning file for each individual filter zone, recording the equivalent resistance coefficient before cleaning. Equivalent resistance coefficient after dust removal Actual number of pulses And the runtime after dust removal to the next dust removal trigger. And calculate the single cleaning efficiency. ; Step S502: Take the efficiency of the last K cleaning cycles moving average efficiency ,when Continuously below the efficiency health lower limit At that time, the dust removal start resistance threshold of the corresponding independent filter zone will be set. Lower and preset number of pulses Increase by 1 to 2 times, among which The current dust removal start-up resistance threshold 5% to 8%; Step S503: After two consecutive adaptive adjustments It has not yet recovered to the lower limit of efficiency and health. When the above is the case, the output points to the maintenance and inspection prompts of the corresponding independent filtration zone. The maintenance and inspection prompts include filter bag damage inspection prompts, pulse valve failure inspection prompts, and blowpipe alignment misalignment inspection prompts.

8. An intelligent dust removal control system for a bag filter, applied to the intelligent dust removal control method for a bag filter according to any one of claims 1 to 7, characterized in that, The system includes: The partition sensing and coupling configuration module is used to acquire the spatial layout information of the metal filter bags inside the dust collector and the control range information of the pulse jet branch pipe. Based on the spatial layout information and the control range information of the pulse jet branch pipe, it performs the distributed differential pressure signal acquisition and layout task by partitioning and coupling degree calibration, and outputs partition sensing and coupling configuration data. The differential pressure signal processing module is used to perform the partition differential pressure signal processing task based on the partition sensing and coupling configuration data, using temperature compensation, viscosity compensation and equivalent drag coefficient conversion methods, and output partition equivalent drag analysis data. The dust removal candidate area discrimination module is used to perform the dust removal candidate area discrimination task based on the partition equivalent resistance analysis data and adopt a time-series prediction and multi-dimensional threshold joint judgment method, and output the dust removal candidate area labeling data. The dust removal instruction generation module is used to perform the partition dust removal instruction generation task based on the dust removal candidate area marking data, using coupling degree constraint priority sorting and online stop spraying decision method, and output partition dust removal execution record data; The parameter correction and maintenance prompt module is used to correct the cleaning parameters and determine the health status of the filter bags based on the historical efficiency feedback of the partition cleaning execution record data, and output maintenance check prompts.

9. An intelligent dust removal control device for a bag filter, characterized in that, The intelligent dust removal control device for baghouse dust collectors includes: a memory, a processor, and an intelligent dust removal control program for baghouse dust collectors stored in the memory and executable on the processor. When the intelligent dust removal control program for baghouse dust collectors is executed by the processor, it implements an intelligent dust removal control method for baghouse dust collectors according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes an intelligent dust removal control program for bag filters, which, when executed by a processor, implements an intelligent dust removal control method for bag filters according to any one of claims 1 to 7.

Images