Safety protection system based on dust collector pressure monitoring

Abstract

The present application relates to the technical field of environmental protection equipment safety control, and particularly relates to a safety protection system based on dust remover pressure monitoring, comprising a safety protection center, an error identification module, an error judgment module, a reference acquisition module, an electric spark correction module, a risk evaluation module, a flue gas backstepping module and an adjustment execution module; the system is analyzed from the perspective of influencing factors to eliminate false alarms or missed alarms caused by monitoring errors, and is analyzed from two points of pressure reference range prediction and historical electric spark interference to obtain accurate pressure reference range, so as to reduce the misjudgment rate; meanwhile, according to the correlation coefficient table and the corrected true pressure value, the current flue gas composition state in the dust remover is backstepped, whether the flue gas composition meets the safe operation condition is judged, the unsafe factors are eliminated, and the equipment accidents of dust remover damage and the safety accidents of harming the on-site staff are avoided.

Description

Technical Field

[0001] This invention relates to the field of environmental protection equipment safety control technology, and in particular to a safety protection system based on dust collector pressure monitoring. Background Technology

[0002] With increasingly stringent environmental protection requirements in the steel industry, dust collectors are widely used in major steel companies. However, internal flue gas explosions in dust collectors are often relieved by explosion relief valves. An explosion relief valve is a mechanical spring-type pressure relief device, which is usually installed at the inlet and outlet of the electrostatic precipitator. When an explosion occurs inside the dust collector due to the composition of the flue gas reaching the explosive range, or due to factors such as electric sparks or frictional arcing, the explosion relief valve is passively opened to relieve pressure.

[0003] Currently, the actual operation in the industry is that the mechanical spring opening pressure of the explosion relief valve is a fixed value. If the pressure generated by the explosion does not reach the fixed opening pressure, the explosion relief valve cannot be opened. However, such small explosions that do not trigger explosion relief may still damage the internal structural components of the dust collector, interfere with subsequent early warning, and fail to trigger an alarm in time after a small explosion occurs, causing staff to miss the best opportunity to analyze the cause of the explosion, which may easily lead to a series of explosions.

[0004] Meanwhile, when a sudden explosion with a large equivalent force occurs, the pressure relief demand exceeds the pressure relief range of the explosion relief valve, which may cause serious equipment accidents and safety accidents, posing a significant threat to the lives of on-site personnel and enterprise production. Therefore, a solution is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a safety protection system based on dust collector pressure monitoring to address the aforementioned technical deficiencies.

[0006] The objective of this invention can be achieved through the following technical solution: a safety protection system based on dust collector pressure monitoring, comprising a safety protection center, an error identification module, an error judgment module, a reference acquisition module, an electrical spark correction module, a risk assessment module, a flue gas reverse propagation module, and an adjustment execution module;

[0007] The error identification module is used to evaluate and analyze the data errors of the collected parameters from multiple sources of the dust collector to obtain invalid or valid signals;

[0008] The error judgment module is used to obtain and analyze the error contribution value of the valid data group in the retrieved historical database, and obtain the error contribution value corresponding to each influencing factor.

[0009] The benchmark acquisition module is used to divide the smelting cycle of a single furnace and perform interval benchmark acquisition analysis on parameters from multiple sources to obtain the pressure benchmark range for each processing period.

[0010] The electric spark correction module is used to perform electric spark impact correction analysis on the collected electric spark discharge current and vibration frequency, obtain the pre-set pressure threshold adjustment scheme corresponding to the current pressure impact level, and adjust the pressure reference range based on the pre-set pressure threshold adjustment scheme to obtain the adjusted pressure reference range.

[0011] The risk assessment module is used to perform risk warning assessment and analysis on the collection time corresponding to the number of valid collections. The obtained warning signals of level 1, level 1, level 2 and level 3 are collectively referred to as warning signals.

[0012] The flue gas reverse module is used to retrieve a pre-set correlation coefficient table of flue gas component concentration during the pressure-processing period, calculate the pressure deviation value, and accompany the flue gas component evaluation process.

[0013] The adjustment and execution module is used to respond to various signals and provide warnings through text feedback.

[0014] Preferably, the analysis process of the error identification module is as follows:

[0015] Set a monitoring period, collect multi-source parameters of the dust collector during the monitoring period, input the collected multi-source parameters into the error identification model to obtain the output analysis result list, which includes the accuracy results, the name of each source parameter and its corresponding value;

[0016] The accuracy of the analysis results is then assessed to determine whether the signal is invalid or valid.

[0017] Preferably, the analysis process of the error judgment module is as follows:

[0018] Retrieve six months of pressure monitoring data, sensor maintenance records, environmental monitoring data, and equipment operating parameters from the historical database, and filter out valid data sets that meet the constraints.

[0019] Using the controlled variable method, other influencing factor parameters were kept constant while only the target factor parameters were changed. The correlation between the changes in the target factor and the pressure monitoring error was analyzed, and a table of the correspondence between the target factor and the error rate was obtained.

[0020] Preferably, a regression model is constructed with pressure monitoring error = α × sensor aging coefficient + β × transmission distance attenuation coefficient + γ × electromagnetic interference intensity coefficient + constant term. The regression coefficients are solved by the least squares method to realize the quantitative calculation of the comprehensive influence of multiple factors, and finally output the error contribution value corresponding to each influencing factor.

[0021] Preferably, the analysis process of the benchmark acquisition module is as follows:

[0022] Using the single-furnace steel smelting cycle as the time unit, clearly defined processing time intervals are divided, and the start time, duration, and corresponding process operations of each processing time interval are recorded.

[0023] The multi-source parameters collected for each processing time period are preprocessed, and the preprocessed multi-source parameters for each processing time period are used for feature extraction to extract the pressure features within each time period.

[0024] The fan speed change rate and the deviation of molten iron carbon content are used as correlation features and combined with pressure features to form a multi-dimensional feature vector of time period-pressure-process.

[0025] Based on the linear correlation between dust collector pressure and smelting cycle, a pre-set multiple linear regression algorithm model is used to calculate the predicted value of dust collector outlet pressure at time t. Based on the predicted value of dust collector outlet pressure at time t, the pressure reference range for each processing period is obtained.

[0026] Preferably, the analysis process of the electrical spark correction module is as follows:

[0027] The electric spark discharge current and vibration frequency of the dust collector are collected during the monitoring period. The collected electric spark discharge current and vibration frequency are processed to obtain the effective collection count. Based on the effective collection count, the proportion of the impact duration and the cumulative impact count in the historical operation period of the dust collector are obtained.

[0028] The impact duration percentage and cumulative impact frequency were analyzed to determine low-risk, medium-risk, and high-risk impacts. These three impact levels were collectively referred to as the stress impact level.

[0029] The pressure impact level of the current dust collector is obtained. Based on the current pressure impact level, a pre-set pressure threshold adjustment scheme is obtained. The pressure reference range is adjusted based on the pre-set pressure threshold adjustment scheme to obtain the adjusted pressure reference range.

[0030] Preferably, the analysis process of the risk assessment module is as follows:

[0031] The sampling time corresponding to the number of valid samplings is set as the spark time. The sampling time of the real-time pressure data is compared with the spark time to obtain the valid pressure data.

[0032] The quantified error contribution value is back-compensated into the effective pressure data to obtain the corrected true pressure value. The true pressure value is then processed to obtain the Level 1 warning signal, Level 1 warning signal, Level 2 warning signal, and Level 3 warning signal. These Level 1, Level 1, Level 2, and Level 3 warning signals are collectively referred to as warning signals.

[0033] Preferably, the analysis process of the flue gas reverse propagation module is as follows:

[0034] The effective pressure data of the dust collector corresponding to the warning signal and the pressure deviation value of the pressure reference value of the corresponding processing time interval are obtained.

[0035] Retrieve the pre-set correlation coefficient table of flue gas component concentration during the pressure-processing period, obtain the correlation coefficient of O2 and CO in the flue gas component concentration based on the current pressure deviation value, and then obtain the real-time O2 concentration and real-time CO concentration by using the formula: real-time concentration of target component O2 or CO = preset safety benchmark concentration of the component + (pressure deviation value × correlation coefficient).

[0036] The real-time O2 concentration and real-time CO concentration are processed to obtain alarm signals or safety signals.

[0037] The beneficial effects of this invention are as follows:

[0038] This invention effectively eliminates monitoring errors caused by factors such as sensor drift and electromagnetic interference by real-time acquisition of multi-dimensional parameters and setting of error identification models, ensuring the accuracy of pressure data and providing a reliable data foundation for subsequent anomaly judgment and strategy execution. At the same time, it analyzes from the perspective of influencing factors, providing error compensation basis for pressure data correction, accurate prediction of relationship models, and subsequent safety strategy execution, eliminating false alarms or missed alarms caused by monitoring errors.

[0039] This invention analyzes the pressure reference range from two aspects: prediction and historical electrical spark interference, to obtain an accurate pressure reference range. This provides an accurate evaluation standard for subsequent pressure analysis, reduces the misjudgment rate, and improves the timeliness and effectiveness of early warning. Furthermore, based on the correlation coefficient table and the corrected actual pressure value, the current flue gas composition state inside the dust collector is inferred to determine whether the flue gas composition meets the safe operating conditions. This addresses the deficiency in existing technologies where pressure data cannot be correlated with the safety status of flue gas, enabling on-site personnel to react quickly, eliminate unsafe factors, and avoid equipment accidents that damage the dust collector and safety accidents that injure on-site personnel. Attached Figure Description

[0040] The invention will now be further described with reference to the accompanying drawings;

[0041] Figure 1 This is a partial reference diagram of the present invention;

[0042] Figure 2 This is a flowchart of the system of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0044] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments;

[0045] Example 1: Please refer to Figures 1 to 2 As shown, the present invention is a safety protection system based on dust collector pressure monitoring, including a safety protection center, an error identification module, an error judgment module, a reference acquisition module, an electric spark correction module, a risk assessment module, a flue gas back-pushing module, and an adjustment execution module. The error identification module and the error judgment module have a one-way communication connection, the error judgment module has a one-way communication connection with the safety protection center, the safety protection center has a two-way communication connection with the reference acquisition module, the reference acquisition module has a two-way communication connection with the electric spark correction module, the safety protection center has a one-way communication connection with the risk assessment module, the risk assessment module has a one-way communication connection with both the flue gas back-pushing module and the adjustment execution module, and the flue gas back-pushing module has a one-way communication connection with the adjustment execution module.

[0046] The error identification module is used to evaluate and analyze the data errors of the collected parameters from multiple sources of the dust collector. The specific data error evaluation and analysis process is as follows:

[0047] Set a monitoring period and collect multi-source parameters of the dust collector during the monitoring period. The multi-source parameters include, but are not limited to, dust collector inlet pressure parameters, dust collector outlet pressure parameters, and fan speed parameters at each stage of the smelting cycle.

[0048] Using the parameter data corresponding to the pressure fluctuation curve of the dust collector under normal operating conditions as positive samples and the historical abnormal pressure fluctuation data as negative samples, multiple sets of positive and negative samples are obtained for preprocessing, including cleaning and enhancement. An error identification model is constructed based on the preprocessed positive and negative samples.

[0049] The collected parameters from multiple sources are input into the error identification model to obtain a list of output analysis results. The list of analysis results includes accuracy results (data interference (including sensor drift error, data transmission delay error and external electromagnetic interference error) and data normality), names of each source parameter and corresponding values, etc.

[0050] The system also performs a judgment process on the accuracy of the analysis results. If the accuracy result is data interference, an invalid signal is generated; if the accuracy result is data normal, a valid signal is generated.

[0051] The adjustment and execution module is used to respond to invalid or valid signals and immediately display the pre-warning text corresponding to the invalid or valid signal, so as to intuitively understand whether there are errors in the collected multi-source parameters and ensure the accuracy of subsequent analysis;

[0052] When an invalid signal is generated, the error judgment module is used to obtain and analyze the error contribution value of the valid data group retrieved from the historical database. The specific error contribution value acquisition and analysis process is as follows:

[0053] Retrieve six months of pressure monitoring data, sensor maintenance records (such as calibration time and replacement cycle), environmental monitoring data (temperature and dust concentration), and equipment operating parameters (fan speed and high-voltage electric field strength) from the historical database, and filter out valid data groups that meet the constraints (no process abnormalities (such as no small explosions or seal leaks) but monitoring errors exist) (each data group includes "actual monitoring pressure value - standard pressure value - potential influencing factor parameters").

[0054] Using the controlled variable method, other influencing factor parameters (such as temperature, fan speed, etc.) are kept constant, while only the target factor parameter (such as dust concentration) is changed. The correlation between the change of the target factor and the pressure monitoring error is analyzed, and a target factor-error rate correspondence table is obtained.

[0055] For example, with a fixed transmission distance and constant ambient temperature, the pressure error value under different usage times is recorded by adjusting the sensor's usage time, and a correspondence between "sensor aging coefficient and error rate" is established (e.g., aging coefficient of 0.1 and error rate of 2% after 6 months of use; aging coefficient of 0.2 and error rate of 5% after 12 months of use).

[0056] Based on the single-factor quantification results, a regression model was constructed using multiple linear regression analysis, which is: pressure monitoring error = α × sensor aging coefficient + β × transmission distance attenuation coefficient + γ × electromagnetic interference intensity coefficient + constant term. The regression coefficients (α, β, γ) were solved by the least squares method to achieve quantitative calculation of the comprehensive impact of multiple factors. Finally, the error contribution value corresponding to each influencing factor (i.e., the specific value of the pressure monitoring error caused by the factor) was output, and the obtained error contribution value corresponding to each influencing factor was sent to the safety protection center for storage.

[0057] Example 2: The benchmark acquisition module is used to divide the smelting cycle of a single furnace and perform interval benchmark acquisition analysis on parameters from multiple sources. The specific interval benchmark acquisition analysis process is as follows:

[0058] Using the single-furnace steel smelting cycle as the time unit, clearly defined processing time intervals are divided (such as charging period 0-10min, smelting period 10-40min, and tapping period 40-50min). The start time, duration, and corresponding process operations (such as auxiliary material addition and fan speed adjustment) of each processing time interval are recorded.

[0059] The multi-source parameters collected for each processing time period are preprocessed, and the preprocessed multi-source parameters for each processing time period are then used to extract features, including the pressure characteristics within each time period, including the average pressure, pressure fluctuation amplitude (the difference between the maximum and minimum pressure within the time period), and pressure change rate (the amount of pressure change per unit time, such as Pa / s).

[0060] Data such as the rate of change of fan speed and the deviation of carbon content in molten iron are used as correlation features and combined with pressure features to form a multi-dimensional feature vector of time period-pressure-process (e.g., "15min smelting period - average pressure - 450Pa - pressure fluctuation range 20Pa - fan speed change rate 5%), to ensure that the model can distinguish between "normal process fluctuations" and "abnormal explosion fluctuations".

[0061] Based on the linear correlation between dust collector pressure and smelting cycle, a pre-set multiple linear regression algorithm model is selected for calculation. The model expression is as follows: The predicted value of the dust collector outlet pressure at time t is obtained.

[0062] Where P(t) is the predicted value of the dust collector outlet pressure at time t (t>0), T1, T2, and T3 are the time proportions of the charging period, smelting period, and tapping period, respectively, S is the fan speed deviation value, C is the iron carbon content deviation value, and a1, a2, a3, b1, and b2 are regression coefficients (the regression coefficients are solved by substituting the above expression into the pre-set training set and using the least squares method). This is the error term (error contribution value, which is directly substituted into the model for compensation).

[0063] Based on the predicted value of the dust collector outlet pressure at time t, the pressure reference range for each processing period is obtained, and the pressure reference range is sent to the safety protection center for storage.

[0064] The spark correction module is used to perform spark impact correction analysis on the collected spark discharge current and vibration frequency. The specific spark impact correction analysis process is as follows:

[0065] The electric spark discharge current of the dust collector is collected during the monitoring period by installing a current sensor in the high-voltage electric field circuit of the dust collector, and the vibration frequency of the dust collector is collected by deploying a vibration sensor.

[0066] The collected electric spark discharge current and vibration frequency are processed for discrimination. If the electric spark discharge current is greater than the preset electric spark discharge current threshold, the number of times the corresponding electric spark discharge current is collected is determined to be a valid number of times. If the vibration frequency is greater than the preset vibration frequency threshold, the number of times the vibration frequency is collected is determined to be a valid number of times.

[0067] The total number of valid data collections and the corresponding collection duration within the historical operating period of the dust collector are obtained. Then, the total collection duration corresponding to the valid data collections is obtained, and the ratio between the total collection duration and the monitoring period is obtained. The ratio between the total collection duration and the monitoring period is set as the percentage of the impact duration, and the total number of valid data collections is set as the cumulative impact count.

[0068] The system also performs discrimination processing on the percentage of impact duration and the cumulative number of impacts. Simultaneously, it retrieves the preset impact duration percentage range (SCmax and SCmin) and the preset cumulative number of impacts range (LCmax and LCmin) for the percentage of impact duration and the cumulative number of impacts. The impact duration percentage and the cumulative number of impacts are compared and analyzed with the preset impact duration percentage range and the preset cumulative number of impacts range. If the impact duration percentage is < SCmin and the cumulative number of impacts is < LCmin, it is judged as a low-risk impact. If SCmin ≤ impact duration percentage ≤ SCmax, or LCmax ≤ cumulative number of impacts ≤ LCmin, it is judged as a medium-risk impact. If the impact duration percentage is > SCmax and the cumulative number of impacts is > LCmax, it is judged as a high-risk impact.

[0069] Low-risk impact, medium-risk impact, and high-risk impact are collectively referred to as stress impact levels;

[0070] Obtain the current pressure impact level of the dust collector, and obtain a pre-set pressure threshold adjustment scheme based on the current pressure impact level;

[0071] The pressure reference range is adjusted based on the pre-set pressure threshold adjustment scheme to obtain the adjusted pressure reference range, and the adjusted pressure reference range is sent to the safety protection center for storage.

[0072] For example: If the current pressure impact level of the dust collector is medium risk, the corresponding pre-set pressure threshold adjustment scheme is to reduce the pressure fluctuation threshold (e.g., adjust it from 0.5 kPa to 0.4 kPa) to improve the sensitivity of identifying small-yield explosions.

[0073] Example 3: The risk assessment module is used to perform risk warning assessment and analysis on the collection time corresponding to the number of valid collections. The specific risk warning assessment and analysis process is as follows:

[0074] The collection time corresponding to the number of valid collections is set as the spark time. The collection time of real-time pressure data is compared with the spark time to determine whether there is a spark time within the collection time of real-time pressure data. If there is, the corresponding real-time pressure data is set as suspected interference data. The average value of the real-time pressure data collected in the first 3 seconds is used to replace the suspected interference data as the valid pressure data.

[0075] The quantified error contribution value is back-compensated into the effective pressure data to obtain the corrected true pressure value (e.g., monitoring pressure value -380Pa, error contribution value +4Pa, corrected true pressure value -384Pa).

[0076] This ensures the accuracy of pressure anomaly detection and avoids alarm delays or false alarms caused by electromagnetic interference from electric sparks (such as electric sparks causing the pressure value to momentarily appear normal, masking the actual explosion pressure fluctuations).

[0077] Obtain the pre-set pressure threshold adjustment scheme corresponding to the current pressure impact level, and the adjusted pressure reference ranges: E1, E2, and E3, where E1 < E2 < E3. Compare and analyze the effective pressure data with E1, E2, and E3.

[0078] If the effective pressure data is less than E1, a Level 1 warning signal is generated; if E1 is less than or equal to the effective pressure data and less than E2, a Level 1 warning signal is generated; if E2 is less than or equal to the effective pressure data and less than E3, a Level 2 warning signal is generated; if the effective pressure data is greater than or equal to E3, a Level 3 warning signal is generated. The warning intensity of the Level 1, Level 1, Level 2, and Level 3 warning signals decreases sequentially.

[0079] The Level 1, Level 1, Level 2, and Level 3 early warning signals are collectively referred to as early warning signals. The adjustment and execution module is used to respond to the early warning signals and immediately display the corresponding pre-warning text to remind operation and management personnel to make reasonable management based on the feedback information, so as to improve the pressure safety monitoring effect.

[0080] When an early warning signal is generated, the flue gas reverse calculation module retrieves a pre-set correlation coefficient table of flue gas component concentrations during the pressure-processing period, calculates the pressure deviation value, and performs a flue gas component evaluation process. The specific flue gas component evaluation process is as follows:

[0081] The deviation between the effective pressure data of the dust collector corresponding to the warning signal and the pressure reference value of the corresponding processing time period is obtained. The formula is: Pressure deviation value = Corrected actual pressure value - The median value of the adjusted pressure reference range of the corresponding processing time period (half of the sum of the maximum and minimum values).

[0082] Retrieve a pre-set correlation coefficient table for the concentration of flue gas components during the pressure-processing period. For example: O2 concentration correlation coefficient: for every 10 Pa increase in pressure deviation (i.e., pressure 10 Pa higher than the baseline value), the corresponding O2 concentration increases by 0.5%; for every 10 Pa decrease in pressure deviation (i.e., pressure 10 Pa lower than the baseline value), the corresponding O2 concentration decreases by 0.3%. CO concentration correlation coefficient: for every 15 Pa increase in pressure deviation, the corresponding CO concentration increases by 0.2%; for every 15 Pa decrease in pressure deviation, the corresponding CO concentration decreases by 0.1%.

[0083] Based on the current pressure deviation value, the correlation coefficient between O2 and CO in the flue gas component concentration is obtained. Then, the real-time O2 concentration and real-time CO concentration are obtained by calculating the real-time concentration of the target component (O2 and CO) = the preset safety benchmark concentration of the component + (pressure deviation value × correlation coefficient).

[0084] The system performs discrimination processing on real-time O2 concentration and real-time CO concentration. If the O2 concentration is less than the preset O2 concentration threshold or the CO concentration is less than the preset CO concentration threshold, an alarm signal is generated. If the O2 concentration is greater than or equal to the preset O2 concentration threshold and the CO concentration is greater than or equal to the preset CO concentration threshold, a safety signal is generated.

[0085] The adjustment and execution module is used to respond to alarm signals or safety signals and immediately display the pre-warning text corresponding to the alarm signal or safety signal. For example, based on the pre-warning text "normal" corresponding to the safety signal, it maintains the current operating status and continuously monitors the data. Based on the pre-warning text "high risk" corresponding to the alarm signal, it records the time of the abnormality, pressure change curve, and flue gas composition back-inferring data, and forms an abnormality report which is stored in the historical database for subsequent cause analysis.

[0086] In summary, by collecting multi-dimensional parameters in real time and setting up an error identification model, monitoring errors caused by factors such as sensor drift and electromagnetic interference can be effectively eliminated, ensuring the accuracy of pressure data and providing a reliable data foundation for subsequent anomaly judgment and strategy execution. At the same time, analysis from the perspective of influencing factors provides a basis for error compensation for pressure data correction, accurate prediction of relationship models, and subsequent safety strategy execution, eliminating false alarms or missed alarms caused by monitoring errors.

[0087] Furthermore, by analyzing both the predicted pressure reference range and historical electrical spark interference, a precise pressure reference range is obtained. This provides accurate evaluation criteria for subsequent pressure analysis, reduces the misjudgment rate, and improves the timeliness and effectiveness of early warnings. Based on the correlation coefficient table and the corrected actual pressure value, the current flue gas composition state inside the dust collector can be deduced to determine whether the flue gas composition meets the safe operating conditions. This addresses the deficiency in existing technologies where pressure data cannot be correlated with the safety status of flue gas, enabling on-site personnel to react quickly, eliminate unsafe factors, and avoid equipment accidents that damage the dust collector and safety accidents that injure on-site personnel.

[0088] The threshold is set for comparative analysis of results to determine whether they are good or bad. The value of the threshold is determined by a combination of large-scale model analysis of sample data and human experience. It can also be adjusted appropriately based on seasonal or common-sense influencing factors.

[0089] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A safety protection system based on dust collector pressure monitoring, characterized in that, It includes a safety protection center, an error identification module, an error judgment module, a benchmark acquisition module, an electrical spark correction module, a risk assessment module, a flue gas reverse propagation module, and a regulation execution module; The error identification module is used to evaluate and analyze the data errors of the collected parameters from multiple sources of the dust collector to obtain invalid or valid signals; The error judgment module is used to obtain and analyze the error contribution value of the valid data group in the retrieved historical database, and obtain the error contribution value corresponding to each influencing factor. The benchmark acquisition module is used to divide the single-furnace steel smelting cycle and perform interval benchmark acquisition analysis on parameters from multiple sources to obtain the pressure benchmark range for each processing period. The electric spark correction module is used to perform electric spark impact correction analysis on the collected electric spark discharge current and vibration frequency, obtain the pre-set pressure threshold adjustment scheme corresponding to the current pressure impact level, and adjust the pressure reference range based on the pre-set pressure threshold adjustment scheme to obtain the adjusted pressure reference range. The risk assessment module is used to perform risk warning assessment and analysis on the collection time corresponding to the number of valid collections. The obtained warning signals of level 1, level 1, level 2 and level 3 are collectively referred to as warning signals. The flue gas reverse module is used to retrieve a pre-set correlation coefficient table of flue gas component concentration during the pressure-processing period, calculate the pressure deviation value, and accompany the flue gas component evaluation process. The adjustment and execution module is used to respond to various signals and provide warnings through text feedback.

2. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the error identification module is as follows: Set a monitoring period, collect multi-source parameters of the dust collector during the monitoring period, input the collected multi-source parameters into the error identification model to obtain the output analysis result list, which includes the accuracy results, the name of each source parameter and its corresponding value; The accuracy of the analysis results is then assessed to determine whether the signal is invalid or valid.

3. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the error judgment module is as follows: Retrieve six months of pressure monitoring data, sensor maintenance records, environmental monitoring data, and equipment operating parameters from the historical database, and filter out valid data sets that meet the constraints. Using the controlled variable method, other influencing factor parameters were kept constant while only the target factor parameters were changed. The correlation between the changes in the target factor and the pressure monitoring error was analyzed, and a table of the correspondence between the target factor and the error rate was obtained.

4. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, A regression model is constructed with pressure monitoring error = α × sensor aging coefficient + β × transmission distance attenuation coefficient + γ × electromagnetic interference intensity coefficient + constant term. The regression coefficients are solved by the least squares method to achieve quantitative calculation of the comprehensive influence of multiple factors, and finally the error contribution value corresponding to each influencing factor is output.

5. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the benchmark acquisition module is as follows: Using the single-furnace steel smelting cycle as the time unit, clearly defined processing time intervals are divided, and the start time, duration, and corresponding process operations of each processing time interval are recorded. The multi-source parameters collected for each processing time period are preprocessed, and the preprocessed multi-source parameters for each processing time period are used for feature extraction to extract the pressure features within each time period. The fan speed change rate and the deviation of molten iron carbon content are used as correlation features and combined with pressure features to form a multi-dimensional feature vector of time period-pressure-process. Based on the linear correlation between dust collector pressure and smelting cycle, a pre-set multiple linear regression algorithm model is used to calculate the predicted value of dust collector outlet pressure at time t. Based on the predicted value of dust collector outlet pressure at time t, the pressure reference range for each processing period is obtained.

6. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the electrical discharge correction module is as follows: The electric spark discharge current and vibration frequency of the dust collector are collected during the monitoring period. The collected electric spark discharge current and vibration frequency are processed to obtain the effective collection count. Based on the effective collection count, the proportion of the impact duration and the cumulative impact count in the historical operation period of the dust collector are obtained. The impact duration percentage and cumulative impact frequency were analyzed to determine low-risk, medium-risk, and high-risk impacts. These three impact levels were collectively referred to as the stress impact level. The pressure impact level of the current dust collector is obtained. Based on the current pressure impact level, a pre-set pressure threshold adjustment scheme is obtained. The pressure reference range is adjusted based on the pre-set pressure threshold adjustment scheme to obtain the adjusted pressure reference range.

7. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the risk assessment module is as follows: The sampling time corresponding to the number of valid samplings is set as the spark time. The sampling time of the real-time pressure data is compared with the spark time to obtain the valid pressure data. The quantified error contribution value is back-compensated into the effective pressure data to obtain the corrected true pressure value. The true pressure value is then processed to obtain the Level 1 warning signal, Level 1 warning signal, Level 2 warning signal, and Level 3 warning signal. These Level 1, Level 1, Level 2, and Level 3 warning signals are collectively referred to as warning signals.

8. The safety protection system based on dust collector pressure monitoring according to claim 1, characterized in that, The analysis process of the flue gas reverse propagation module is as follows: The effective pressure data of the dust collector corresponding to the warning signal and the pressure deviation value of the pressure reference value of the corresponding processing time interval are obtained. Retrieve the pre-set correlation coefficient table of flue gas component concentration during the pressure-processing period, obtain the correlation coefficient of O2 and CO in the flue gas component concentration based on the current pressure deviation value, and then obtain the real-time O2 concentration and real-time CO concentration by using the formula: real-time concentration of target component O2 or CO = preset safety benchmark concentration of the component + (pressure deviation value × correlation coefficient). The real-time O2 concentration and real-time CO concentration are processed to obtain alarm signals or safety signals.

Images