Spray tower filter screen self-cleaning control system based on Internet of Things

Abstract

The invention discloses a spray tower filter screen self-cleaning control system based on Internet of Things, and relates to the technical field of spray towers, and the spray tower filter screen self-cleaning control system comprises an acquisition module for acquiring a main airflow parameter of a real-time operation state of a spray tower, a filter screen blockage influence parameter, and a filter wire density and a filter wire diameter of a to-be-cleaned filter screen; marking a filter wire cross node area of the to-be-cleaned filter screen according to the main airflow parameter, the filter wire density and the filter wire diameter to obtain a blockage risk detection point; extracting blockage risk distribution characteristics of the filter screen influenced by different blockage conditions in a historical period from historical filter screen blockage and self-cleaning monitoring data; according to the blockage risk distribution characteristics, obtaining risk correlation factors among the blockage risk coefficient, the to-be-detected risk spacing and the filter screen blockage influence parameters; extracting a target association factor from the risk association factors according to the filter screen blockage influence parameters and the blockage risk distribution characteristics of the spray tower; the cleaning device has the advantages that the cleaning effect is guaranteed, and damage to the filter screen is reduced to the maximum extent.

Description

Technical Field

[0001] This invention relates to the field of spray tower technology, and more specifically, to a self-cleaning control system for spray tower filters based on the Internet of Things. Background Technology

[0002] In actual operation of spray towers, filter clogging is a key issue affecting their filtration efficiency and operational stability. Traditional filter cleaning methods rely heavily on manual periodic inspections or fixed-cycle cleaning, making it difficult to accurately monitor the clogging status in real time. This leads to a significant decrease in spray tower operating efficiency and can even cause equipment failure. Existing technologies generally assess filter clogging risk based on a single parameter, such as the concentration of dust in the inlet air or the pressure difference on the filter surface. This fails to comprehensively consider the synergistic effects of the main airflow moisture content, pressure, filter fiber density, and diameter, resulting in inaccurate risk assessments and affecting the effectiveness of cleaning solutions. Furthermore, the inability to establish a correlation between clogging risk distribution characteristics and current operating parameters using historical data means that the development of cleaning solutions lacks data support, making it difficult to achieve self-cleaning control. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a self-cleaning control system for spray tower filters based on the Internet of Things.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A self-cleaning control system for spray tower filters based on the Internet of Things includes: Acquisition module: Acquires the main airflow parameters, filter screen clogging impact parameters, and filter fiber density and diameter of the filter screen to be cleaned in real time; Marking module: Based on the main airflow parameters, filter wire density, and filter wire diameter, the cross-node area of ​​the filter wires to be cleaned is marked to obtain the clogging risk detection points; Extraction module: Extracts the distribution characteristics of clogging risk of filters under different clogging conditions in historical periods from historical filter clogging and self-cleaning monitoring data; The first processing module: Based on the characteristics of the clogging risk distribution, it obtains the risk correlation factor between the clogging risk coefficient, the distance of the risk to be measured, and the filter clogging impact parameter; Based on the filter clogging impact parameter and the clogging risk distribution characteristics of the spray tower, it extracts the target correlation factor from the risk correlation factor; The second processing module: obtains the comprehensive blockage risk factor of the blockage risk detection point based on the target correlation factor; and obtains the current efficiency degradation value of the filter affected by the blockage risk based on the comprehensive blockage risk factor. Output module: Receives a self-cleaning start command based on the current performance degradation value, and outputs the optimal self-cleaning scheme based on the current blockage risk status.

[0005] Preferably, the main airflow parameters include airflow moisture content and airflow pressure; The parameters affecting filter clogging include inlet dust concentration, spray liquid flow rate, and filter surface temperature.

[0006] Preferably, the clogging risk detection points are obtained by marking the cross-node areas of the filter fibers in the filter screen to be cleaned according to the main airflow parameters, filter fiber density, and filter fiber diameter. This specifically includes the following steps: Based on the main airflow parameters, the first risk detection point is obtained by marking the cross nodes of the filter wires in the air inlet side edge of the filter screen to be cleaned. Based on the main airflow parameters, the cross-node area of ​​the filter filament cross-node of the filter screen to be cleaned is marked with the cross-node of the air outlet side edge of the filter screen to obtain the second risk detection point; Based on the filter wire density, filter wire diameter, first risk detection point, and second risk detection point, risk detection points are marked in the filter wire intersection area to obtain the blockage risk detection points.

[0007] Preferably, it further includes: The distance between adjacent congestion risk detection points is calculated to obtain the risk interval to be measured.

[0008] Preferably, the distribution characteristics of clogging risk affecting the filter under different clogging conditions in historical periods are extracted from historical filter clogging and self-cleaning monitoring data, specifically including the following steps: Extract the airflow moisture content and airflow pressure of the spray tower in historical periods from historical filter clogging and self-cleaning monitoring data; Extract the location and risk coefficient of filter clogging risk in the spray tower at different inlet dust concentrations from historical filter clogging and self-cleaning monitoring data; The distribution characteristics of clogging risk are obtained by integrating the airflow moisture content, airflow pressure, filter clogging risk location and risk coefficient.

[0009] Preferably, the risk correlation factor between the clogging risk coefficient, the measured risk interval, and the filter clogging impact parameters is obtained based on the clogging risk distribution characteristics, specifically including the following steps: Based on the characteristics of clogging risk distribution, historical clogging risk coefficients of clogging risk detection points are extracted from historical filter clogging and self-cleaning monitoring data. Based on the distance between the risks to be measured and the basic parameters of the impact of blockage, the risk coefficient difference is obtained by subtracting the risk coefficients of adjacent blockage risk detection points in the historical blockage risk coefficients. The risk correlation value is obtained by fitting the risk coefficient difference, the distance between the risks to be measured, and the basic parameters of the blockage impact; The risk correlation factor is obtained by weighting all risk correlation values.

[0010] Preferably, the target correlation factor is extracted from the risk correlation factors based on the filter clogging impact parameters and clogging risk distribution characteristics of the spray tower, specifically including the following steps: The matching results are obtained by matching the filter clogging impact parameters with the clogging risk distribution characteristics corresponding to the basic clogging impact parameters. Based on the matching results, target correlation factors that are compatible with the basic parameters of the current congestion impact are extracted from the risk correlation factors.

[0011] Preferably, the comprehensive congestion risk factor for the congestion risk detection point is obtained based on the target correlation factor, specifically including the following steps: The first pre-treatment risk interval is obtained by calculating the straight-line distance between the blockage risk detection point and the first risk detection point; The first risk coefficient value of the blockage risk detection point is obtained based on the first pre-processing risk interval, the risk coefficient value to be measured, and the risk correlation factor. The second pre-processing risk interval is obtained by calculating the straight-line distance between the target blockage risk detection point and the second risk detection point. The second risk coefficient value of the blockage risk detection point is obtained based on the second pre-processing risk interval, the risk coefficient value to be measured, and the risk correlation factor. The risk growth rate value is obtained by monitoring the blockage risk growth rate at the first risk detection point. The first risk factor is obtained by multiplying the first risk coefficient value and the risk growth rate value. The second risk factor is obtained by multiplying the second risk coefficient value and the risk growth rate value. The first risk factor and the second risk factor are summed to obtain the comprehensive congestion risk factor.

[0012] Preferably, the current performance degradation value of the filter due to clogging risk is obtained based on the comprehensive clogging risk factor, specifically including the following steps: Obtain a reference performance database of filter performance degradation values ​​under different clogging risk coefficients, different risk growth rates, and different basic parameters of clogging impact; The current performance degradation value of the filter affected by the clogging risk is obtained by matching the comprehensive clogging risk factor with the reference performance database.

[0013] Based on the current performance degradation value, a self-cleaning start command is obtained. Based on the self-cleaning start command, the optimal self-cleaning scheme for the current blockage risk status is output, which specifically includes the following steps: If the current performance degradation value of the integrated data is greater than or equal to the filter self-cleaning warning threshold, then a self-cleaning start command will be output. After retrieving the pre-stored cleaning solution library based on the self-cleaning start command, an optimized self-cleaning solution adapted to the current clogging risk status is output. The cleaning time and filter damage rate corresponding to the optimized self-cleaning solution were tested to obtain the solution performance data; Determine the allowable cleaning time range and filter damage rate value for the spray tower to obtain operational constraint data; The optimal self-cleaning scheme is extracted from the scheme performance data that matches the operational constraint data.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention achieves comprehensive risk control through multi-module collaboration. It comprehensively collects main airflow parameters, filter clogging impact parameters, and inherent filter properties, providing ample basic data for risk analysis. The marking module combines this data to accurately mark clogging risk detection points in the filter filament intersection areas. For example, when handling airflows with high moisture content and high dust concentration, it can accurately locate nodes on the inlet and outlet edges that are clogged due to particulate matter accumulation. The extraction and processing modules construct a risk correlation mechanism through historical data mining and real-time parameter modeling. For instance, when the inlet dust concentration suddenly increases, it calculates the corresponding comprehensive clogging risk factor, assesses the current degree of filter clogging, and achieves intelligent adaptation of cleaning timing and solutions through quantitative analysis of efficiency decay values. When the efficiency decay value reaches the warning threshold, a self-cleaning command is triggered, preventing premature cleaning that wastes resources and delayed cleaning that leads to excessive clogging or even damage to the filter. Meanwhile, the output module can select the most suitable solution from the pre-stored cleaning solution library to match the current clogging risk state. If the current clogging is a mixture of wet and dust, it will prioritize the cleaning solution that combines high-pressure washing and hot air drying to ensure the cleaning effect and minimize damage to the filter. Attached Figure Description

[0015] Figure 1 A schematic diagram of a self-cleaning control system for a spray tower filter based on the Internet of Things is provided for this invention; Figure 2 This invention presents a schematic diagram illustrating the steps involved in obtaining risk correlation factors in an IoT-based self-cleaning control system for spray tower filters. Detailed Implementation

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0018] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0019] Reference Figures 1-2 As shown.

[0020] The embodiments further illustrate the Internet of Things-based self-cleaning control system for spray tower filters proposed in this invention.

[0021] A self-cleaning control system for spray tower filters based on the Internet of Things includes: Acquisition module: Acquires the main airflow parameters, filter screen clogging impact parameters, and filter fiber density and diameter of the filter screen to be cleaned in real time; Marking module: Based on the main airflow parameters, filter wire density, and filter wire diameter, the cross-node area of ​​the filter wires to be cleaned is marked to obtain the clogging risk detection points; Extraction module: Extracts the distribution characteristics of clogging risk of filters under different clogging conditions in historical periods from historical filter clogging and self-cleaning monitoring data; The first processing module: Based on the characteristics of the clogging risk distribution, it obtains the risk correlation factor between the clogging risk coefficient, the distance of the risk to be measured, and the filter clogging impact parameter; Based on the filter clogging impact parameter and the clogging risk distribution characteristics of the spray tower, it extracts the target correlation factor from the risk correlation factor; The second processing module: obtains the comprehensive blockage risk factor of the blockage risk detection point based on the target correlation factor; and obtains the current efficiency degradation value of the filter affected by the blockage risk based on the comprehensive blockage risk factor. Output module: Receives a self-cleaning start command based on the current performance degradation value, and outputs the optimal self-cleaning scheme based on the current blockage risk status.

[0022] The acquisition module deploys multiple sensors to collect multi-dimensional data. Temperature and humidity sensors and pressure sensors are installed at the air inlet of the spray tower to obtain the moisture content and pressure of the main airflow. The temperature and humidity sensor is a model SHT40, and the pressure sensor is a model BMP280. A laser dust sensor (model PM2008) is installed in the air inlet dust channel. An electromagnetic flow meter (model LDG-S) is installed in the spray liquid pipeline, and a thermocouple temperature sensor (model K) is attached to the filter screen surface. The electromagnetic flow meter and thermocouple temperature sensor respectively collect the inlet dust concentration, spray liquid flow rate, and filter screen surface temperature. An optical microscopic detection unit (model VHX-6000) is also provided to measure the filter filament density and diameter of the filter screen to be cleaned. All sensors are connected to the system core controller via a Modbus-RTU bus to achieve real-time data transmission.

[0023] The marking module uses an industrial-grade embedded controller, model STM32H743, which has a built-in point marking algorithm developed based on C language. After receiving the main airflow parameters, filter wire density and diameter data transmitted by the acquisition module, it intelligently marks the filter wire cross node area and outputs the coordinate information of the blockage risk detection point. This module interacts with the first processing module through a USB interface.

[0024] The extraction module relies on a cloud server to extract risk distribution characteristics under different blockage conditions from historical data, such as the correlation pattern between airflow parameters and filter blockage location. The data is synchronized to the first processing module via a 5G communication module.

[0025] The first processing module is configured with an edge computing server, model Advantech UNO-4673. Combining the clogging risk distribution characteristics of the extraction module, it calculates the clogging risk coefficient, the risk spacing to be measured, and the risk correlation factor of the filter clogging impact parameter. Through parameter matching logic, it extracts the target correlation factor adapted to the current working condition from the correlation factor set. This module is connected to the second processing module through an Ethernet interface.

[0026] The second processing module uses a digital signal processor, model TITMS320C6678, which receives the target correlation factor from the first processing module and calculates the comprehensive blockage risk factor of the blockage risk detection point. At the same time, it calls the pre-trained efficiency decay model, based on the TensorFlowLite framework, to map the comprehensive risk factor to the current efficiency decay value of the filter. The data is transmitted to the output module via the CAN bus.

[0027] The output module is based on a PLC controller and has built-in self-cleaning threshold judgment logic. When the efficiency decay value reaches the warning threshold, it outputs a self-cleaning start command. It retrieves the cleaning scheme library pre-stored in the local storage module through the Ethernet interface, and combines the cleaning time and filter damage rate fed back by the timing module and damage detection sensor to select the optimal self-cleaning scheme that meets the operating constraints of the spray tower. Finally, it triggers the cleaning actuator through the relay output interface.

[0028] The main airflow parameters include airflow moisture content and airflow pressure; The parameters affecting filter clogging include inlet dust concentration, spray fluid flow rate, and filter surface temperature.

[0029] The main airflow parameters and filter clogging impact parameters are collected during the operation of the spray tower. The filter filament density and diameter of the filter to be cleaned are also obtained. The main airflow parameters include airflow moisture content and airflow pressure; for example, high airflow moisture content and unstable pressure increase the risk of filter clogging. Filter clogging impact parameters include inlet dust concentration, spray liquid flow rate, and filter surface temperature data. For example, high inlet dust concentration, slow spray liquid flow rate, or abnormal filter surface temperature will affect the degree of filter clogging.

[0030] Based on the main airflow parameters, filter fiber density, and filter fiber diameter, the cross-node areas of the filter fibers to be cleaned are marked to obtain the clogging risk detection points. The specific steps include: Based on the main airflow parameters, the first risk detection point is obtained by marking the cross nodes of the filter wires in the air inlet side edge of the filter screen to be cleaned. Based on the main airflow parameters, the cross-node area of ​​the filter filament cross-node of the filter screen to be cleaned is marked with the cross-node of the air outlet side edge of the filter screen to obtain the second risk detection point; Based on the filter wire density, filter wire diameter, first risk detection point, and second risk detection point, risk detection points are marked in the filter wire intersection area to obtain the blockage risk detection points.

[0031] The first risk detection point is obtained by marking the filter filament cross-nodes on the inlet side edge of the filter screen based on the main airflow parameters. The main airflow parameters include airflow moisture content and airflow pressure. These parameters affect the airflow state on the filter inlet side, thus affecting the clogging risk of the filter filament cross-nodes. When the airflow moisture content is high and the pressure is high, the filter filament cross-nodes on the inlet side edge are more prone to clogging due to the accumulation of moisture and particulate matter. Therefore, the first risk detection point is marked by judging the main airflow parameters.

[0032] The second risk detection point is obtained by marking the filter filament cross nodes on the outlet side edge of the filter screen based on the main airflow parameters. The flow state of the airflow on the outlet side after filtration is affected by the main airflow parameters, and the filter filament cross nodes on the outlet side edge are also at risk of blockage. The second risk detection point is determined by the main airflow parameters.

[0033] Combining filter fiber density, filter fiber diameter, and the previously obtained first and second risk detection points, risk detection points are marked in the filter fiber cross-node area to obtain the clogging risk detection points. The higher the filter fiber density, the smaller the diameter, the more numerous the cross-nodes, and the smaller the gaps, the higher the clogging risk. Taking filter fiber density ρ and filter fiber diameter d as an example, where the unit of filter fiber density is fibers / square centimeter and the unit of filter fiber diameter is millimeters, this can be determined using the formula... The impact of filter wire structure on the clogging risk detection points was calculated. ,in , These are weighting coefficients, obtained by fitting historical data, based on the degree of influence. The locations of the first and second risk detection points determine the location of the blockage risk detection points.

[0034] The first risk detection point is located at the air inlet edge of the filter, which is the area where airflow carrying particles first comes into contact, resulting in a significant initial accumulation of clogging risk. The second risk detection point is located at the air outlet edge of the filter, where residual particles after filtration are prone to secondary deposition. A preset impact threshold is determined based on historical data. When the impact level R is greater than or equal to the preset threshold, clogging risk detection points are densely marked inside the filter, using the first and second risk detection points as a benchmark (e.g., one point every 4 millimeters). If the impact level R is less than the preset threshold, the spacing is appropriately increased, such as one point every 10 millimeters. This method covers the easily clogged areas of the filter, providing accurate detection points for subsequent assessment of filter clogging risk and development of self-cleaning solutions.

[0035] Also includes: The distance between adjacent congestion risk detection points is calculated to obtain the risk interval to be measured.

[0036] After marking the clogging risk detection points, the straight-line distance between adjacent clogging risk detection points is measured and statistically analyzed. For example, three clogging risk detection points, A, B, and C, are marked in the filter filament intersection area of ​​a spray tower filter. The straight-line distance between points A and B is measured to be 5 mm, and the straight-line distance between points B and C is 6 mm. These distances are the risk intervals to be measured. The straight-line distance data between all adjacent clogging risk detection points is obtained in this way.

[0037] Extracting the distribution characteristics of filter clogging risk under different clogging conditions in historical periods from historical filter clogging and self-cleaning monitoring data, specifically including the following steps: Extract the airflow moisture content and airflow pressure of the spray tower in historical periods from historical filter clogging and self-cleaning monitoring data; Extract the location and risk coefficient of filter clogging risk in the spray tower at different inlet dust concentrations from historical filter clogging and self-cleaning monitoring data; The distribution characteristics of clogging risk are obtained by integrating the airflow moisture content, airflow pressure, filter clogging risk location and risk coefficient.

[0038] Historical data on filter clogging and self-cleaning monitoring were used to extract the airflow moisture content and pressure of the spray tower over historical periods. For example, the moisture content of the airflow treated by the spray tower in different months was extracted from the monitoring data of the past year; for instance, the average moisture content in January was 8 g / m³, and the airflow pressure was 0.3 MPa; while the average moisture content in July was 12 g / m³, and the airflow pressure was 0.4 MPa. These data reflect the different airflow conditions throughout history.

[0039] The location and risk coefficient of filter clogging risk for the spray tower at different inlet dust concentrations were extracted from historical filter clogging and self-cleaning monitoring data. For example, when the inlet dust concentration is 100 mg / m³, the central area of ​​the filter is prone to clogging, with a corresponding risk coefficient of 0.6; when the inlet dust concentration is 150 mg / m³, the edge area of ​​the filter has a higher risk of clogging, with a risk coefficient of 0.8.

[0040] The clogging risk distribution characteristics are obtained by integrating airflow moisture content, airflow pressure, filter clogging risk location, and risk coefficient. In the data matrix, rows represent different historical operating conditions, composed of airflow moisture content and pressure, columns represent different filter locations, and the matrix values ​​are the corresponding risk coefficients. For example, when the airflow moisture content is 10 g / m³, the pressure is 0.35 MPa, and the inlet dust concentration is 120 mg / m³, the risk coefficient for filter location (x=5cm, y=5cm) is 0.7, and the risk coefficient for location (x=10cm, y=10cm) is 0.5. This clearly shows the clogging risk distribution at different filter locations under different airflow parameters and dust concentrations, thus forming a clogging risk distribution characteristic. This characteristic can be used to match current data to determine the current filter clogging risk status.

[0041] Based on the distribution characteristics of clogging risk, the risk correlation factor between the clogging risk coefficient, the measured risk interval, and the filter clogging impact parameters is obtained, specifically including the following steps: Based on the characteristics of clogging risk distribution, historical clogging risk coefficients of clogging risk detection points are extracted from historical filter clogging and self-cleaning monitoring data. Based on the distance between the risks to be measured and the basic parameters of the impact of blockage, the risk coefficient difference is obtained by subtracting the risk coefficients of adjacent blockage risk detection points in the historical blockage risk coefficients. The risk correlation value is obtained by fitting the risk coefficient difference, the distance between the risks to be measured, and the basic parameters of the blockage impact; The risk correlation factor is obtained by weighting all risk correlation values.

[0042] Based on the distribution characteristics of clogging risk, historical clogging risk coefficients for detection points are extracted from historical filter clogging and self-cleaning monitoring data. For example, in historical data, the risk coefficient for a certain clogging risk detection point is 0.6 under specific airflow moisture content, pressure, and dust concentration, while the risk coefficient for another adjacent point is 0.8.

[0043] The risk coefficient difference is obtained by subtracting the risk coefficients of adjacent blockage risk detection points in the historical blockage risk coefficient data based on the distance between the detected risks and the basic parameters of blockage impact. The basic parameters of blockage impact are the inlet dust concentration and the spray liquid flow rate. Assuming the distance between the detected risks is 5 mm, the inlet dust concentration of the basic parameters of blockage impact is 100 mg / m³, and the historical risk coefficients of adjacent points are 0.6 and 0.8 respectively, then the risk coefficient difference is 0.8 - 0.6 = 0.2.

[0044] The risk correlation value is obtained by fitting the risk coefficient difference, the distance between the risks to be measured, and the basic parameters of the blockage impact. A multiple linear fitting formula is used, such as... Where V is the risk correlation value, ΔR is the risk coefficient difference, L is the risk interval to be measured, C is the inlet dust concentration, and a, b, and c are the coefficients obtained from the fitting. If a=0.5, b=0.02, c=0.001, ΔR=0.2, L=5, and C=100, then V=0.5×0.2+0.02×5+0.001×100=0.1+0.1+0.1=0.3.

[0045] In this multiple linear fitting, although the risk coefficient difference ΔR, the distance between the measured risks L, and the inlet dust concentration C have different dimensions, they quantify the risk from three dimensions—the difference characteristics, spatial distribution characteristics, and the intensity characteristics of the inducing factors—in a physical sense. The fitting coefficients a, b, and c serve as weight calibration, converting the characteristics of these three different dimensions into a quantitative index for comprehensive evaluation under the same dimension (risk correlation value V). This enables a coordinated judgment of multiple factors affecting filter clogging risk and provides a unified quantitative basis for subsequent risk assessment.

[0046] Finally, a weighted average of all risk correlation values ​​is used to obtain the risk correlation factor. Assuming three risk correlation values ​​are 0.3, 0.4, and 0.5, with corresponding weights of 0.3, 0.4, and 0.3 respectively, and the weights determined by the sample size or importance of each correlation value, the risk correlation factor is 0.3×0.3 + 0.4×0.4 + 0.5×0.3 = 0.09 + 0.16 + 0.15 = 0.4. By establishing a correlation model between the risk coefficient, the measured risk interval, and the clogging impact parameters, a crucial quantitative basis is provided for subsequent assessment of the current filter's clogging risk.

[0047] Based on the filter clogging impact parameters and clogging risk distribution characteristics of the spray tower, the target correlation factor is extracted from the risk correlation factors, specifically including the following steps: The matching results are obtained by matching the filter clogging impact parameters with the clogging risk distribution characteristics corresponding to the basic clogging impact parameters. Based on the matching results, target correlation factors that are compatible with the basic parameters of the current congestion impact are extracted from the risk correlation factors.

[0048] The current filter clogging impact parameters are matched with the basic clogging impact parameters corresponding to the clogging risk distribution characteristics. Filter clogging impact parameters include inlet dust concentration, spray fluid velocity, and filter surface temperature data, while the basic clogging impact parameters are the baseline values ​​corresponding to these parameters from historical data. For example, if the inlet dust concentration is 120 mg / m³, the spray fluid velocity is 2 m / s, and the filter surface temperature is 30℃, the basic clogging impact parameter set corresponding to inlet dust concentrations of 100-140 mg / m³, spray fluid velocities of 1.8-2.2 m / s, and filter surface temperatures of 28-32℃ is found in the historical data of the clogging risk distribution characteristics, and the matching result is obtained.

[0049] Based on the matching results, target correlation factors that are compatible with the current basic parameters of blockage impact are extracted from the risk correlation factors. It is assumed that the risk correlation factors are a series of pre-generated values ​​based on different sets of basic parameters of blockage impact; for example, for the matched set of basic parameters mentioned above, the corresponding risk correlation factor is 0.75. This matching and extraction process ensures that the correlation factors used in subsequent calculations accurately adapt to the current operating conditions, laying the foundation for calculating the comprehensive blockage risk factor of the blockage risk detection points.

[0050] Through calculation formula Calculate the matching degree ,in, This represents the parameter affecting the current i-th filter blockage. The corresponding basic parameters affecting congestion, where n is the number of parameters, when When the value is less than a set threshold (which can be set to 0.1 based on historical data), a successful match is considered, and the corresponding target correlation factor is extracted. For example, if the differences between the current three parameters and the base parameters are 5%, 3%, and 4%, then M = (0.05 + 0.03 + 0.04) / 3 = 0.04. Since this is less than the threshold of 0.1, a successful match is considered, and the corresponding target correlation factor is extracted.

[0051] The comprehensive congestion risk factor for congestion risk detection points is obtained based on the target correlation factors, specifically including the following steps: The first pre-treatment risk interval is obtained by calculating the straight-line distance between the blockage risk detection point and the first risk detection point; The first risk coefficient value of the blockage risk detection point is obtained based on the first pre-processing risk interval, the risk coefficient value to be measured, and the risk correlation factor. The second pre-processing risk interval is obtained by calculating the straight-line distance between the target blockage risk detection point and the second risk detection point. The second risk coefficient value of the blockage risk detection point is obtained based on the second pre-processing risk interval, the risk coefficient value to be measured, and the risk correlation factor. The risk growth rate value is obtained by monitoring the blockage risk growth rate at the first risk detection point. The first risk factor is obtained by multiplying the first risk coefficient value and the risk growth rate value. The second risk factor is obtained by multiplying the second risk coefficient value and the risk growth rate value. The first risk factor and the second risk factor are summed to obtain the comprehensive congestion risk factor.

[0052] First, the straight-line distance between the blockage risk detection point and the first risk detection point is calculated to obtain the first pretreatment risk distance. For example, if the straight-line distance between a blockage risk detection point and the first risk detection point is 8 mm, this distance reflects the spatial relationship between this point and the high-risk area on the intake side. Then, based on the first pretreatment risk distance, the measured risk coefficient value, and the risk correlation factor, the first risk coefficient value of the blockage risk detection point is calculated using the formula... The first risk coefficient value was calculated. ,in, As the first pre-treatment risk interval, The value of the risk coefficient to be measured. This is a risk-related factor. Assume the risk coefficient value to be measured is... The risk association factor is 0.6. If the value is 0.5, then the first risk coefficient value is... It is 9.6.

[0053] Multi-dimensional data is collected by various sensors on the spray tower, including airflow moisture content collected by the SHT40 temperature and humidity sensor, main airflow parameters such as airflow pressure collected by the BMP280 pressure sensor, inlet dust concentration collected by the PM2008 laser dust sensor, spray liquid flow rate collected by the LDG-S electromagnetic flowmeter, filter clogging impact parameters such as filter surface temperature collected by the K-type thermocouple temperature sensor, and filter self-property parameters such as filter filament density and diameter obtained by the VHX-6000 optical microscopic detection unit. This real-time data is matched with clogging risk distribution characteristics extracted from historical filter clogging and self-cleaning monitoring data in the cloud server, such as the correlation between clogging location and risk coefficient under different parameter combinations. A risk analysis model suitable for the current operating conditions is selected. The risk analysis model is obtained through machine learning training on a large amount of historical data. Based on the risk analysis model, the collected data is calculated to obtain the test risk coefficient value C, which reflects the risk level of the current clogging risk detection point.

[0054] Next, the straight-line distance between the target blockage risk detection point and the second risk detection point is calculated to obtain the second pretreatment risk interval. For example, if the straight-line distance between this point and the second risk detection point is 6 millimeters, this reflects the spatial connection with the risk area on the outlet side. Then, based on the second pretreatment risk interval, the measured risk coefficient value, and the risk correlation factor, the second risk coefficient value is calculated according to the formula. The second risk coefficient value was calculated. ,in, This is the second pretreatment risk interval. If the second pretreatment risk interval... If the value is 6, then the second risk coefficient value is... It is 7.2.

[0055] The risk growth rate is obtained by monitoring the congestion risk growth rate at the first risk detection point. Assuming the monitored risk growth rate is 0.3, the first risk factor is obtained by multiplying the first risk coefficient value by this rate value, according to the formula... The first risk factor was calculated. The value is 2.88. Multiplying the second risk coefficient value and the risk growth rate value yields the second risk factor, according to the formula... The first risk factor was calculated. It is 2.16.

[0056] Time-related growth rate v and first risk coefficient value Or the second risk coefficient value They can be directly multiplied because they have undergone dimensional normalization and physical meaning coupling. (Growth rate v, first risk coefficient value) Second risk coefficient value During the model training phase, the dimensional differences have been eliminated through standardized calculations, placing them in the same quantization dimension.

[0057] The first and second risk factors are summed to obtain the comprehensive congestion risk factor, which is then calculated using the formula... The comprehensive congestion risk factor was calculated. The value is 5.04. The comprehensive clogging risk factor reflects the spatial relationship between the location and the risk detection points on both sides, the basic risk coefficient, and the risk growth rate. It can reflect the current comprehensive clogging risk level of the clogging risk detection point and provide a key basis for subsequent assessment of filter performance degradation and development of self-cleaning solutions.

[0058] The current performance degradation value of the filter due to clogging risk is obtained based on the comprehensive clogging risk factor, specifically including the following steps: Obtain a reference performance database of filter performance degradation values ​​under different clogging risk coefficients, different risk growth rates, and different basic parameters of clogging impact; The current performance degradation value of the filter affected by the clogging risk is obtained by matching the comprehensive clogging risk factor with the reference performance database.

[0059] First, a reference performance database needs to be obtained showing the corresponding filter performance degradation values ​​under different clogging risk coefficients, risk growth rates, and basic parameters related to clogging impact. For example, when the clogging risk coefficient is 0.6, the risk growth rate is 0.2, and the inlet dust concentration in the basic parameters of clogging impact is 100 mg / m³, the corresponding filter performance degradation value is 15% obtained through extensive experiments or historical data statistics; when the clogging risk coefficient is 0.8, the risk growth rate is 0.3, and the inlet dust concentration is 120 mg / m³, the performance degradation value is 25%. The relationship between these different parameter combinations and their corresponding performance degradation values ​​is compiled into a database to provide a basis for subsequent matching.

[0060] The comprehensive clogging risk factor is matched with a reference performance database to obtain the current performance degradation value of the filter affected by clogging risk. For example, if the comprehensive clogging risk factor is 5.04, this factor takes into account the effects of factors such as the comprehensive clogging risk coefficient, risk growth rate, and basic parameters affecting clogging. The reference performance database is then used to find the performance degradation value corresponding to the parameter combination that is closest to this comprehensive clogging risk factor. If a linear interpolation formula is used for matching, assuming there are two adjacent data points in the reference database, the comprehensive clogging risk factor... A value of 5.0 corresponds to an effectiveness decay value. The overall congestion risk factor is 20%. If it is 5.1, then the corresponding performance degradation value is... The current comprehensive congestion risk factor is 22%. It is 5.04, according to the formula. The corresponding performance degradation value was calculated. The figure is 20.8%. This matching process can accurately determine the degree of performance degradation of the filter due to the risk of clogging, providing key data for subsequent decisions on whether to activate self-cleaning and selecting a cleaning solution.

[0061] Based on the current performance degradation value, a self-cleaning start command is obtained. Based on the self-cleaning start command, the optimal self-cleaning scheme for the current blockage risk status is output, which specifically includes the following steps: If the current performance degradation value of the integrated data is greater than or equal to the filter self-cleaning warning threshold, then a self-cleaning start command will be output. After retrieving the pre-stored cleaning solution library based on the self-cleaning start command, an optimized self-cleaning solution adapted to the current clogging risk status is output. The cleaning time and filter damage rate corresponding to the optimized self-cleaning solution were tested to obtain the solution performance data; Determine the allowable cleaning time range and filter damage rate value for the spray tower to obtain operational constraint data; The optimal self-cleaning scheme is extracted from the scheme performance data that matches the operational constraint data.

[0062] First, determine if the current filter performance degradation value has reached the self-cleaning warning threshold. If the calculated current performance degradation value is greater than or equal to this threshold, then a self-cleaning start command is output. For example, if the filter self-cleaning warning threshold is set to 25%, then when the current performance degradation value is 30%, the self-cleaning start command is triggered.

[0063] The self-cleaning startup command retrieves a pre-stored cleaning solution library and outputs an optimized self-cleaning solution adapted to the current clogging risk level. The cleaning solution library stores various cleaning strategies for different levels of clogging risk, such as a high-pressure water flushing solution for high-dust clogging and a hot air drying-assisted cleaning solution for wet clogging.

[0064] The cleaning time and filter damage rate corresponding to the optimized self-cleaning scheme are tested to obtain the scheme performance data. Assume that the cleaning time of a certain optimized self-cleaning scheme is 2 hours and the filter damage rate is 3%. The allowable cleaning time range and filter damage rate value for the spray tower are determined to obtain operational constraint data. For example, the allowable cleaning time range for the spray tower is 1-3 hours, and the filter damage rate value does not exceed 5%.

[0065] The optimal self-cleaning scheme is selected from those whose performance data meets the operational constraints. If a scheme's cleaning time is within the range of 1-3 hours (2 hours), and the filter damage rate is 3% and does not exceed 5%, then this scheme is the optimal self-cleaning scheme.

[0066] Let the cleaning time constraint satisfy the following condition: ,in To the minimum allowable cleaning time, For the maximum allowable cleaning time, The filter damage rate constraint must be met to ensure the cleaning time of the solution is satisfactory. ,in For the maximum allowable filter damage rate, This represents the filter damage rate of the proposed solution. A solution is selected as the preferred self-cleaning solution if it simultaneously meets all these criteria.

[0067] The experimental group used the high-pressure rinsing and hot air drying cleaning scheme described in this application. The high-pressure rinsing pressure was set at 0.8 MPa, and the rinsing time was 5 minutes. The hot air drying temperature was controlled at 60℃, and the drying time was 8 minutes. The control group used the traditional low-pressure rinsing scheme, with a rinsing pressure of 0.3 MPa and a rinsing time of 10 minutes.

[0068] Table 1 shows the comparison of indicators between the experimental group and the control group.

[0069]

[0070] Table 1 shows the quantitative indicators and experimental data. In terms of efficiency recovery rate, the experimental group reached 95%, which is much higher than the control group's 78%, indicating that the patented solution can almost completely restore the filtration efficiency of the filter. In terms of filter damage rate, the experimental group was only 2.1%, which is significantly lower than the control group's 5.3%, demonstrating that the patented solution has a better protective effect on the filter during the cleaning process. The total cleaning time was 13 minutes for the experimental group and 10 minutes for the control group. Although the patented solution takes a little longer, its overall performance is significantly better than the traditional solution due to the advantages of combined efficiency recovery and filter damage control.

[0071] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0072] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An Internet of Things based control system for self-cleaning of filter screen of a spray tower, characterized in that, The method comprises the following steps: An acquisition module: acquiring main airflow parameters of a spray tower in real-time running state, filter screen blockage influence parameters, filter wire density and filter wire diameter of a filter screen to be cleaned; A marking module: marking filter wire cross node areas of the filter screen to be cleaned according to the main airflow parameters, the filter wire density and the filter wire diameter to obtain blockage risk detection points; An extraction module: extracting blockage risk distribution characteristics of the filter screen influenced by different blockage conditions in a historical period from historical filter screen blockage and self-cleaning monitoring data; A first processing module: obtaining a blockage risk coefficient, a risk correlation factor between the to-be-measured risk interval and the filter screen blockage influence parameters according to the blockage risk distribution characteristics; and obtaining a target correlation factor from the risk correlation factor according to the filter screen blockage influence parameters and the blockage risk distribution characteristics of the spray tower; A second processing module: obtaining a comprehensive blockage risk factor of the blockage risk detection points according to the target correlation factor; Obtaining a current performance attenuation value of the filter screen influenced by the blockage risk according to the comprehensive blockage risk factor; An output module: obtaining a self-cleaning starting instruction according to the current performance attenuation value, and outputting an optimal self-cleaning scheme of a current blockage risk state according to the self-cleaning starting instruction.

2. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 1, characterized in that, The main airflow parameters include airflow humidity and airflow pressure. The filter screen blockage influence parameters include inlet dust concentration, spray liquid flow rate and filter screen surface temperature data.

3. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 1, characterized in that, The marking module comprises the following steps: Marking filter screen inlet side edge cross node areas of the filter screen to be cleaned according to the main airflow parameters to obtain first risk detection points; Marking filter screen outlet side edge cross node areas of the filter screen to be cleaned according to the main airflow parameters to obtain second risk detection points; Marking risk detection points of the filter wire cross node areas according to the filter wire density, the filter wire diameter, the first risk detection points and the second risk detection points to obtain the blockage risk detection points.

4. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 1, characterized in that, The method further comprises the following steps: Obtaining the to-be-measured risk interval by calculating the straight-line distance between adjacent blockage risk detection points.

5. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 4, characterized in that, The extraction module comprises the following steps: Extracting airflow humidity and airflow pressure of the spray tower in the historical period from the historical filter screen blockage and self-cleaning monitoring data; Extracting filter screen blockage risk positions and risk coefficients of the spray tower in the inlet dust concentration from the historical filter screen blockage and self-cleaning monitoring data; Integrating the airflow humidity, the airflow pressure, the filter screen blockage risk positions and the risk coefficients to obtain the blockage risk distribution characteristics.

6. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 5, characterized in that, The first processing module comprises the following steps: Extracting historical blockage risk coefficients of the blockage risk detection points from the historical filter screen blockage and self-cleaning monitoring data according to the blockage risk distribution characteristics; Obtaining risk coefficient differences between risk coefficients of adjacent blockage risk detection points in the historical blockage risk coefficients by difference processing according to the to-be-measured risk interval and the blockage influence basic parameters; and The risk correlation value is obtained by fitting and calculating the risk coefficient difference value, the to-be-measured risk interval, and the clogging influence basic parameter; The risk correlation factor is obtained by weighted average processing of all risk correlation values.

7. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 6, characterized in that, The target correlation factor is extracted from the risk correlation factor according to the filter screen clogging influence parameter and the clogging risk distribution characteristic, and specifically includes the following steps: The clogging influence basic parameter corresponding to the filter screen clogging influence parameter and the clogging risk distribution characteristic is matched to obtain a matching result; According to the matching result, the target correlation factor that is adapted to the current clogging influence basic parameter is extracted from the risk correlation factor.

8. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 7, characterized in that, The comprehensive clogging risk factor of the clogging risk detection point is obtained according to the target correlation factor, and specifically includes the following steps: The straight-line distance between the clogging risk detection point and the first risk detection point is counted to obtain a first pre-processing risk interval; The first risk coefficient value of the clogging risk detection point is obtained according to the first pre-processing risk interval, the to-be-measured risk coefficient value, and the risk correlation factor; The straight-line distance between the target clogging risk detection point and the second risk detection point is counted to obtain a second pre-processing risk interval; The second risk coefficient value of the clogging risk detection point is obtained according to the second pre-processing risk interval, the to-be-measured risk coefficient value, and the risk correlation factor; The risk growth rate value is obtained by monitoring the clogging risk growth rate of the first risk detection point, the first risk coefficient value is multiplied by the risk growth rate value to obtain a first risk factor, and the second risk coefficient value is multiplied by the risk growth rate value to obtain a second risk factor; The first risk factor and the second risk factor are summed to obtain the comprehensive clogging risk factor.

9. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 8, characterized in that, The current performance decay value of the filter screen affected by the clogging risk is obtained according to the comprehensive clogging risk factor, and specifically includes the following steps: A reference performance database corresponding to the filter screen filtering performance decay value under different clogging risk coefficients, different risk growth rates, and different clogging influence basic parameters is obtained; The current performance decay value of the filter screen affected by the clogging risk is obtained by data matching processing of the comprehensive clogging risk factor and the reference performance database.

10. The self-cleaning control system of a spray tower filter screen based on the Internet of Things according to claim 9, characterized in that, The self-cleaning start instruction is obtained according to the current performance decay value, and the optimized self-cleaning scheme of the current clogging risk state is output according to the self-cleaning start instruction, and specifically includes the following steps: If the integrated data current performance decay value is greater than or equal to the filter screen self-cleaning warning threshold, the self-cleaning start instruction is output; The pre-stored cleaning scheme library is called according to the self-cleaning start instruction, and the optimized self-cleaning scheme adapted to the current clogging risk state is output; The scheme performance data is obtained by detecting the cleaning time and the filter screen damage rate corresponding to the optimized self-cleaning scheme; The cleaning time range allowed by the operation of the spray tower and the filter screen damage rate value are determined to obtain operation constraint data; The preferred self-cleaning scheme corresponding to the scheme performance data meeting the operation constraint data is extracted from the optimized self-cleaning scheme.

Images