Intelligent regulation and control method for dust removal system based on data acquisition of Internet of Things

Abstract

The invention discloses a dust removal system intelligent regulation and control method based on Internet of Things data acquisition. According to the method, coal quality and flue gas conditioning influence factors are calculated by collecting fired coal quality, corrected sulfur trioxide concentration and electric field secondary voltage and current data, a working condition evaluation index is generated through fusion, and then an expected corona current interval is determined. The real-time characteristic current value is extracted, the electric field control margin is calculated, the characteristic current is compared with an expected interval, the voltage adjustment amount is determined under the safety constraint, and the electric field voltage is adjusted and controlled according to the voltage adjustment amount. Finally, the regulation and control effect is verified by re-collecting data, a closed loop is formed, and self-adaptive optimization and safe operation of the dust removal system on the working condition are achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control and Internet of Things (IoT) application technology for environmental protection equipment in thermal power plants, specifically to an intelligent control method for dust removal systems based on IoT data acquisition. Background Technology

[0002] Electrostatic precipitators, as core equipment for flue gas purification in coal-fired power plants, work by charging dust particles with a high-voltage electric field and then capturing them under the influence of the electric field force. The corona discharge intensity is crucial in determining dust removal efficiency, directly characterized by secondary voltage and secondary current. The resistivity of the dust particles is a key physical property parameter affecting corona discharge and collection effectiveness: excessively high resistivity easily triggers back corona, leading to secondary dust re-entrainment and reduced efficiency; excessively low resistivity makes it difficult for dust particles to charge and be captured. The ash and sulfur content in the coal fed into the furnace are intrinsic factors determining the initial resistivity of the dust particles; while the sulfur trioxide injected into the flue gas conditioning system, by adsorbing onto the dust surface to form a conductive layer, is a key external means of adjusting the dust resistivity to adapt to the electric field conditions.

[0003] Existing control systems mostly employ fixed parameters or hysteretic adjustments based on outlet dust concentration, making it difficult to coordinate responses to fluctuations in coal quality and changes in conditioning. This results in the system's inability to quantify the impact of coal quality in real time, correct sulfur trioxide sensor drift, and lack of safety margin constraints and closed-loop verification mechanisms for voltage adjustment, leading to fluctuations in dust removal efficiency and operational risks.

[0004] Therefore, there is a need for an intelligent control method for electrostatic precipitators that can integrate real-time coal quality data, corrected sulfur trioxide concentration, and electric field operating status, dynamically generate safe and optimal control commands through intelligent evaluation, and form a closed-loop verification and learning mechanism. Summary of the Invention

[0005] To overcome the aforementioned deficiencies in existing technologies, this invention provides an intelligent control method and system for dust removal systems based on Internet of Things (IoT) data acquisition, aiming to achieve adaptive and safe optimized operation of electrostatic precipitators in response to coal quality and conditioning conditions.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides an intelligent control method for a dust removal system based on Internet of Things (IoT) data acquisition, comprising the following steps: S1: Real-time acquisition of coal quality data entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide in flue gas conditioning system, as well as secondary voltage time series data and secondary current time series data of target electric field of electrostatic precipitator. S2: Based on historical standard data of sulfur trioxide, drift correction is performed on the real-time concentration data of sulfur trioxide to obtain the effective concentration of sulfur trioxide; S3: Based on the influence of ash and sulfur content in the coal quality data on dust resistivity, the coal quality dust removal influence factor is calculated; based on the effect of effective sulfur trioxide concentration on dust resistivity, the flue gas conditioning influence factor is calculated. S4: Integrate the influencing factors of coal dust removal and flue gas conditioning to generate a working condition assessment index that characterizes the resistivity of dust, and determine the corresponding expected corona current range based on the working condition assessment index. S5: Extract the characteristic current value based on the secondary current time series data, calculate the electric field control margin based on the secondary voltage time series data, compare the characteristic current value with the expected corona current range, and determine the target voltage adjustment amount in combination with the electric field control margin. S6: Adjust the operating voltage of the target electric field based on the target voltage adjustment amount, and re-collect data after adjustment to verify the adjustment effect.

[0007] According to the above technical solution, the real-time acquisition of coal quality data entering the boiler of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide from the flue gas conditioning system, and secondary voltage time-series data and secondary current time-series data of the target electric field of the electrostatic precipitator includes: The ash and sulfur content of the coal entering the furnace is obtained in real time from the online coal quality analyzer deployed in the coal conveying system; the real-time measurement data output by the sulfur trioxide concentration sensor installed on the flue is read; the historical standard data of sulfur trioxide that has been pre-calibrated and stored is retrieved from the system database; and the time-series waveform data of the secondary voltage and secondary current of the target electric field are collected in real time through the communication interface of the high-voltage power supply equipment of the electrostatic precipitator.

[0008] According to the above technical solution, the step of performing drift correction on the real-time concentration data of sulfur trioxide based on historical standard data of sulfur trioxide to obtain the effective concentration of sulfur trioxide includes: Within the stable time window of the current operating conditions, the average value of the real-time sulfur trioxide concentration data is calculated and compared with the historical standard data of sulfur trioxide to obtain the concentration measurement drift at the current moment. A preset compensation model is used to dynamically compensate and correct the real-time sulfur trioxide concentration measurement value using the concentration measurement drift. The compensated and corrected concentration value is output as the effective concentration of sulfur trioxide used in subsequent analysis.

[0009] According to the above technical solution, the coal quality dust removal influence factor is calculated based on the influence of ash and sulfur content in the coal quality data on dust resistivity; the flue gas conditioning influence factor is calculated based on the effect of effective sulfur trioxide concentration on dust resistivity, including: Based on the empirical relationship between the coupling effects of ash content and sulfur content on dust resistivity, a dimensionless coal dust removal influence factor is quantified into a characteristic of the inherent influence of coal quality through a pre-set normalized calculation model. Based on the saturation characteristic relationship between the effective concentration of sulfur trioxide and its conditioning effect, a dimensionless flue gas conditioning influence factor is calculated through a nonlinear mapping function to characterize the intensity of external conditioning effects.

[0010] According to the above technical solution, the fusion of coal dust removal influencing factors and flue gas conditioning influencing factors generates a working condition assessment index characterizing the resistivity of dust, and determines the corresponding expected corona current range based on the working condition assessment index, including: The coal dust removal influencing factors and the flue gas conditioning influencing factors are integrated and calculated according to their contribution weights to the dust resistivity to generate a comprehensive operating condition evaluation index. Based on a large amount of historical operating data, a knowledge model reflecting the relationship between the operating condition evaluation index and the optimal operating current is established. The corresponding benchmark expected current value is obtained by querying the current value based on the currently calculated operating condition evaluation index. Centered on the benchmark expected current value, an expected corona current operating range including an upper and lower limit is determined according to the allowable fluctuation range.

[0011] According to the above technical solution, the steps of extracting characteristic current values ​​based on secondary current time-series data, calculating electric field control margin based on secondary voltage time-series data, comparing the characteristic current values ​​with the desired corona current range, and determining the target voltage adjustment amount in conjunction with the electric field control margin include: Statistical distribution analysis is performed on the secondary current time-series data within a set analysis period. The median value of the statistical distribution is extracted as the characteristic current value reflecting the current stable discharge level. The average value of the secondary voltage within the same period is calculated and compared with the breakdown voltage safety threshold preset according to the flue gas conditions to calculate the electric field control margin characterizing the voltage adjustment safety space. It is determined whether the characteristic current value is within the desired corona current range. If it is, no adjustment is required. If it is not, the initial voltage adjustment requirement is calculated based on the deviation of the characteristic current value from the median of the desired corona current range. The electric field control margin is used to impose a safety limit constraint on the initial voltage adjustment requirement, and finally, the target voltage adjustment amount is generated.

[0012] According to the above technical solution, the operating voltage of the target electric field is adjusted based on the target voltage adjustment amount, and data is re-acquired after adjustment to verify the adjustment effect; if the verification result does not meet expectations, a new round of adjustment process is triggered, including: The target voltage adjustment is converted into a control command and sent to the high-voltage power supply equipment of the target electric field to perform voltage adjustment. After the electric field stabilizes, a complete cycle of secondary voltage and secondary current time-series data is collected again. Based on the newly collected data, the characteristic current value and electric field control margin are recalculated, and it is verified whether the new characteristic current value falls within the expected range and whether the control margin meets the safety requirements. If the verification is successful, the current parameters are maintained. If it fails, a new round of complete control process starting from data correction is triggered, with the new data as the starting point. This data is used as a sample to optimize the knowledge model.

[0013] Secondly, the present invention provides an intelligent control system for a dust removal system based on Internet of Things (IoT) data acquisition, used to implement the above method, the system comprising: The data acquisition module is used to collect real-time data on the coal quality entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide from the flue gas conditioning system, as well as the secondary voltage time series data and secondary current time series data of the target electric field of the electrostatic precipitator. The data correction module is used to perform drift correction on the real-time concentration data of sulfur trioxide based on historical standard data of sulfur trioxide, so as to obtain the effective concentration of sulfur trioxide. The influencing factor calculation module is used to calculate the coal quality dust removal influencing factor based on the influence of ash and sulfur content in the coal quality data on dust resistivity; and to calculate the flue gas conditioning influencing factor based on the effect of effective sulfur trioxide concentration on dust resistivity. The operating condition assessment module is used to integrate the influencing factors to generate an operating condition assessment index, and to determine the expected corona current range based on the index. The intelligent decision-making module is used to extract characteristic current values, calculate electric field control margin, and generate target voltage adjustment amount through comparison and constraint calculation. The execution feedback module is used to perform voltage regulation, re-acquire data to verify the effect, and trigger a new round of regulation if the expected results are not achieved.

[0014] Thirdly, the present invention provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the program to implement the above-mentioned intelligent control method for a dust removal system based on Internet of Things data acquisition.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the above-mentioned intelligent control method for a dust removal system based on Internet of Things data acquisition.

[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects: This invention ensures the reliability of key decision-making data by constructing an online drift correction mechanism for sulfur trioxide concentration; it achieves accurate assessment of flue gas conditions by innovatively integrating the comprehensive influence of inherent coal quality characteristics and external interventions in conditioning on dust resistivity; it dynamically generates an adaptive optimal current range based on this assessment result, overcoming the rigidity problem of fixed parameter modes; it introduces electric field control margin as a hard safety constraint in voltage adjustment decisions, effectively preventing flashover risks; and it enables the system to have continuous self-optimization capabilities through a closed-loop verification learning process. Ultimately, this invention can automatically adjust the electrostatic precipitator to the most efficient state matching the current coal quality and conditioning conditions while ensuring the safe and stable operation of the electric field, thereby improving dust removal efficiency while achieving energy saving and consumption reduction. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the overall process of the intelligent control method for a dust removal system based on Internet of Things data acquisition provided in the embodiments of this application; Figure 2 This is a detailed diagram of the data acquisition process provided in the embodiments of this application; Figure 3 This is a detailed diagram of the data correction process provided in the embodiments of this application; Figure 4 This is a detailed diagram of the impact factor calculation process provided in the embodiments of this application; Figure 5 This is a detailed flowchart of the working condition assessment and interval generation process provided in the embodiments of this application; Figure 6 This is a detailed diagram of the intelligent decision-making process provided in the embodiments of this application; Figure 7 This is a schematic diagram showing the relationship between the characteristic current value and the desired corona current range provided in the embodiments of this application; Figure 8 This is a detailed diagram of the execution and feedback verification process provided in the embodiments of this application; Figure 9 This is a schematic diagram of the structure of the intelligent control system for dust removal system based on Internet of Things data acquisition provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail and completely below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this invention.

[0019] Please see Figure 1 , Figure 1 This is a schematic diagram of the overall process of the intelligent control method for a dust removal system based on Internet of Things data acquisition provided in this application embodiment, which specifically includes the following steps: S1: Real-time acquisition of coal quality data entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide in flue gas conditioning system, as well as secondary voltage time series data and secondary current time series data of target electric field of electrostatic precipitator. In this embodiment, step S1 includes the following specific contents, and the process can be found in the attached document. Figure 2 , Figure 2 Here is a detailed diagram of the data acquisition process provided in the embodiments of this application: S110: Real-time acquisition of coal quality data entering the furnace of thermal power plants, including ash content and sulfur content; An online coal quality analyzer is installed on the coal conveyor belt of a thermal power plant for data acquisition. First, the analyzer emits a neutron beam into the transported raw coal. The neutrons interact with the atomic nuclei of various elements in the coal, releasing gamma rays with characteristic energies. The energy spectrum of these characteristic gamma rays is detected and analyzed to obtain the area count of each characteristic energy peak. This energy spectrum data is then input into the analyzer's built-in quantitative analysis model for calculation, thereby calculating and outputting the mass percentage content of ash and sulfur in the current coal stream in real time. The quantitative analysis model is pre-established and stored in the following manner. The process in the analyzer involves: preparing a series of standard coal samples with known and accurate ash and sulfur contents; measuring each standard coal sample using the online coal quality analyzer to obtain the corresponding characteristic gamma-ray energy spectrum; establishing a set of mathematical equations relating the peak area count of the characteristic energy peaks to the known ash and sulfur contents of the standard coal samples using multiple linear regression analysis; solving this set of equations to obtain the response coefficients of each characteristic peak to ash and sulfur, thus completing the calibration of the quantitative analysis model; and finally transmitting the calculated real-time ash and sulfur content data to the data server in the control center via industrial Ethernet in real time. S120: Real-time acquisition of sulfur trioxide concentration data output by the sulfur trioxide concentration sensor of the flue gas conditioning system; An online sulfur trioxide concentration monitor is installed on the flue gas duct before the electrostatic precipitator inlet for data acquisition. The monitor draws flue gas samples from the duct at a constant flow rate. After dust removal, the sampled flue gas is guided through a precisely temperature-controlled condenser. In the condenser, gaseous sulfur trioxide in the flue gas combines with trace amounts of moisture and condenses into sulfuric acid droplets. The gas flow carrying these droplets passes through an optical detection cell, which is then irradiated with a laser beam of a specific wavelength. By measuring the light intensity attenuation after the laser passes through the detection cell and calculating based on Beer-Lambert's law, the real-time mass concentration of sulfur trioxide in the flue gas is obtained. The monitor completes one measurement cycle per minute and continuously outputs the measured concentration value through its digital communication interface. S130: Obtain pre-calibrated historical standard data for sulfur trioxide; Data is retrieved from a pre-established dedicated calibration database. In this embodiment, the data acquisition process involves first determining and confirming that the boiler and flue gas conditioning system are in a stable operating state. A stable operating state means simultaneously meeting the following conditions: boiler load fluctuations remain within ±3% of rated output for more than one hour; fluctuations in ash and sulfur content in the coal feed data are less than 5% of their average values; and the sulfur trioxide injection flow rate setting of the flue gas conditioning system is constant, with actual flow rate fluctuations within ±5% of the setting value. Under this confirmed stable operating state, the online sulfur trioxide concentration monitor installed on the flue is manually calibrated on-site. During calibration, under the same conditions... At the flue gas measuring point, flue gas samples were manually collected using the national standard isopropanol absorption method and then subjected to laboratory chemical analysis to accurately determine the true value of sulfur trioxide concentration in the flue gas under stable operating conditions. The true concentration value obtained from this manual analysis, along with the key operating parameters during this calibration, including the average boiler load, flue gas temperature, oxygen content, and the average value of the coal quality data obtained this time, were stored in the database as a complete historical standard data record. When correction is required, the similarity between the current operating parameters and the operating parameters of all historical records in the database is calculated, and the most similar historical record is matched and retrieved. The stored true concentration value is extracted as the benchmark data for this drift correction. S140: Real-time acquisition of secondary voltage and secondary current time-series data of the target electric field of the electrostatic precipitator; Establish a communication connection with the controller of the high-voltage power supply equipment supporting the target electrostatic precipitator electric field for data collection; initiate a read request to the specific data register address of the high-voltage power supply equipment controller at a fixed frequency of ten times per second through the Modbus TCP industrial communication protocol; continuously receive and parse the data packets returned by the controller, and synchronously obtain the sequence of instantaneous secondary voltage values and the sequence of instantaneous secondary current values reflecting the instantaneous operating state of the electric field, thereby forming two columns of time-series data with a strict time correspondence relationship. S2: Based on the historical standard data of sulfur trioxide, perform drift correction on the real-time concentration data of sulfur trioxide to obtain the effective concentration of sulfur trioxide. In this embodiment, step S2 includes the following specific content. Please refer to the following for the process Figure 3 , Figure 3 is the detailed diagram of the data correction process provided by the embodiment of the present application: S210: Calculate the concentration measurement drift amount at the current moment according to the real-time concentration data of sulfur trioxide and the historical standard data of sulfur trioxide. First, monitor the boiler load and the data of the coal quality entering the furnace. When it is determined that the operating condition enters and maintains a stable operating state, select all the real-time concentration data of sulfur trioxide collected within the recent thirty-minute time window; calculate the arithmetic average of all the real-time concentration data within this time window; at the same time, retrieve the true concentration value from the historical standard data of sulfur trioxide retrieved from the database; perform a subtraction operation on the calculated real-time concentration average value and the historical standard true value, and the obtained difference is calculated and determined as the measurement drift amount of the online concentration monitor at the current moment. S220: Use the concentration measurement drift amount to compensate and correct the real-time concentration data of sulfur trioxide. Select all the real-time concentration data of sulfur trioxide within a recent time window, and calculate the statistical characteristic value of the real-time concentration data within this window; through non-linear drift correction of the original reading of the sensor at the current moment, calculate the effective concentration characterization index of sulfur trioxide. The calculation formula for the effective concentration characterization index of sulfur trioxide is: ; In the formula, [[ID=2)]5] represents the effective concentration characterization index of sulfur trioxide calculated at the sampling time sequence t; t represents the sequence number of the current sampling moment, and this sequence number is an integer obtained by continuously counting at a fixed sampling time interval starting from the start of the sensor. represents the original real-time concentration data of sulfur trioxide directly read from the online monitor at the moment t. This represents the reference concentration of sulfur trioxide. This value is a constant determined by statistically analyzing historical data on the sulfur trioxide concentration at the outlet of the flue gas conditioning system of this power plant unit during long-term normal operation. Long-term normal operation means simultaneously meeting the following three conditions: the boiler load is stable within ±3% of the rated output for more than 24 hours; the quality of the coal fed into the furnace meets the design coal type requirements and the fluctuation of ash and sulfur content is less than 5% of its average; and the set value of sulfur trioxide injection flow rate of the flue gas conditioning system is constant and the actual flow rate fluctuation is within ±5% of the set value. M represents the true value of sulfur trioxide concentration retrieved from the historical standard database and measured manually under similar stable operating conditions; M represents the statistical characteristic value of all real-time sulfur trioxide concentration data within a recent configurable time window, which is obtained in this embodiment by the following method: selecting all real-time sulfur trioxide concentration data within a recent configurable time window as a sample, and performing kernel density estimation on the sample data; kernel density estimation is a non-parametric statistical method that constructs an overall probability density distribution curve by setting a smoothing function, i.e., a kernel function, for each data point in the sample, and estimates the distribution by smoothing the contributions of all data points, which can effectively suppress accidental measurement noise or interference from outliers, thereby obtaining a continuous and smooth probability density curve that reflects the central trend of the real-time sulfur trioxide concentration data; finally, the peak value of this probability density curve is calculated and found, and the real-time sulfur trioxide concentration value corresponding to the peak value is determined as the statistical characteristic value M. The intensity coefficient, representing drift correction, is a dimensionless empirical constant. Its value is determined by analyzing the degree of agreement between a large number of historical sensor drift correction records and corresponding manual calibration results. In this embodiment, the following steps are used to determine it: Data from multiple historical manual calibration times are collected. For each data point, the natural logarithm of the ratio of the manual calibration true value to the reference concentration is calculated, and then the natural logarithm of the ratio of the original sensor reading to the reference concentration at the corresponding time is subtracted to obtain a value. Next, the difference between the ratios of the manual calibration true value and the statistical characteristic value relative to the reference concentration is calculated, and multiplied by the ratio of the time difference corresponding to that data point to the time normalization constant is used to obtain another value. Finally, the linear relationship between these two sets of values ​​is fitted using the least squares method, and the slope of the resulting straight line is the intensity coefficient. The value; This represents the time sequence number of the starting moment of the time window corresponding to the statistical characteristic value M selected in this calculation; T represents the time normalization constant, a large constant in units of sampling intervals, used to convert the time difference term into a dimensionless decimal; exp represents the natural exponential function, i.e., an exponential operation with the natural constant e as the base. In this formula, the role of the exponential function is to convert a correction quantity that integrates historical standard concentration, recent concentration statistical characteristics, and time cumulative effect into a smooth correction multiplier; this correction multiplier is multiplied by the ratio of the original reading to achieve a gradual correction of the sensor reading; when the correction quantity is positive, the correction multiplier is greater than 1, correcting the ratio of the original reading upward; when the correction quantity is negative, the correction multiplier is less than 1, correcting the ratio of the original reading downward; the nonlinear characteristics of the exponential function ensure that the correction process responds significantly when the deviation is large, and the adjustment tends to be gentle when approaching the target, thus effectively avoiding overshoot in the correction process; S230: Output the compensated and corrected concentration value as the effective concentration of sulfur trioxide; Multiply the calculated effective sulfur trioxide concentration characterization index by the baseline sulfur trioxide concentration to obtain the final effective sulfur trioxide concentration output after drift correction. S3: Based on the influence of ash and sulfur content in the coal quality data on dust resistivity, the coal quality dust removal influence factor is calculated; based on the effect of effective sulfur trioxide concentration on dust resistivity, the flue gas conditioning influence factor is calculated. In this embodiment, step S3 includes the following specific details, which can be found in the flowchart below. Figure 4 , Figure 4 Here is a detailed flowchart of the impact factor calculation process provided in the embodiments of this application: S310: Based on the relationship between ash content and sulfur content on dust resistivity, the coal dust removal influencing factor is calculated. An increase in ash content in coal usually significantly increases the resistivity of dust, while sulfur oxides generated after combustion can help reduce resistivity under certain conditions. The two have different mechanisms of influence and are coupled. In order to comprehensively quantify the inherent impact of coal quality on dust removal conditions, it is necessary to calculate the coal quality dust removal influence factor. The formula for calculating the influencing factors of coal dust removal is: ; In the formula, The value represents the coal quality dust removal influencing factor. It is a dimensionless scalar. Its value directly represents the strength of the trend of increased dust resistivity caused by the current coal quality. The larger the value, the more unfavorable the coal quality is to the operation of the electrostatic precipitator. This indicates the ash content of the coal received in real time from the online coal quality analyzer, expressed as a mass percentage. This ground state represents the component content calculated based on the actual received state of the coal. This indicates the sulfur content of the coal received in real time from the online coal quality analyzer, expressed as a mass percentage. This ground state represents the component content calculated based on the actual received state of the coal. This represents the ash content reference value, which is a dimensionless constant. The value is taken as the mass percentage of the received ash content of the design coal type selected during the boiler design of this thermal power unit. The design coal type refers to the typical coal quality determined during the boiler design stage based on the boiler structure, thermal calculations, and environmental protection requirements, which can ensure the safe, economical, and stable operation of the boiler. Its parameters are jointly determined by the boiler manufacturer and the power plant design unit and serve as the benchmark for equipment selection and operation adjustment. This represents the sulfur content reference value, which is a dimensionless constant. The value is taken as the mass percentage of sulfur content on the received basis of the design coal type selected during the boiler design of this thermal power unit. Its source and significance are the same as the ash content reference value. The ash content influence intensity coefficient is a dimensionless constant. In this embodiment, the determination method of the ash content influence intensity coefficient is as follows: Under laboratory conditions, a series of standard coal samples with the same sulfur content but different ash contents are prepared. Fly ash is generated in a standard combustion and dust collection device, and the resistivity of each fly ash sample is accurately measured using a high-resistivity meter. Then, the natural logarithm of the measured resistivity value is taken, and a linear regression is performed with the corresponding normalized ash value, i.e., the ratio of actual ash content to reference ash content. The slope of the obtained regression line is the ash content influence intensity coefficient. The value; and Both coefficients represent the sulfur conditioning effect coefficients and are dimensionless constants. The joint determination method for these two coefficients is as follows: Under laboratory conditions, a series of standard coal samples with the same ash content but different sulfur contents are prepared, fly ash is generated, and resistivity is measured; the normalized sulfur content (the ratio of actual sulfur content to reference sulfur content) is used as the independent variable, and the resistivity decrease rate is used as the dependent variable. The resistivity decrease rate is defined as the percentage decrease relative to the reference resistivity measured when the sulfur content is zero; a saturated growth curve model is fitted using the Levenberg-Marquardt algorithm. The parameters controlling the growth rate and curve shape in this saturated growth curve model are... and The value of exp represents the natural exponential function, which is an exponential operation with the natural constant e as the base. This natural exponential function plays a nonlinear mapping role in the formula, which can transform the linear combination of input variables into an output that grows rapidly or changes saturatedly, thereby accurately simulating the exponential enhancement effect of ash resistivity and the saturation characteristics of sulfur conditioning effect. The formula simulates the nonlinear characteristics of coal quality influence by combining two exponential functions. The first exponential term simulates the exponential enhancement effect of ash resistivity, and the second term, which contains the reciprocal of the exponential decay function, simulates the saturation mitigation effect of sulfur on reducing resistivity. S320: Based on the effect of effective sulfur trioxide concentration on dust resistivity, the flue gas conditioning influence factor is calculated. Sulfur trioxide injected into the flue gas forms a conductive layer on the surface of dust by adsorption, which effectively reduces the resistivity of the dust. However, its conditioning effect shows the characteristic of first rapidly increasing and then gradually saturating with increasing concentration. In order to quantify this external conditioning effect, it is necessary to calculate the flue gas conditioning influence factor. The formula for calculating the flue gas conditioning influencing factor is: ; In the formula, The flue gas conditioning influence factor is a dimensionless scalar between 0 and 1. The closer its value is to 1, the more sufficient the sulfur trioxide conditioning effect is, and the dust resistivity is reduced to the ideal range. This indicates the effective concentration of sulfur trioxide, which is obtained from the real-time readings of the online monitoring instrument after drift correction based on historical standard data. The sulfur trioxide characteristic concentration represents the rate at which the conditioning effect changes with concentration. Mathematically, it corresponds to the sulfur trioxide concentration required to achieve the maximum potential improvement of 63.2% in the conditioning effect. In this embodiment, 63.2% is chosen because it originates from the characteristic time constant theory of a first-order system step response and is widely used in engineering to characterize an intermediate characteristic point of a gradually saturating process. Concentration parameters defined by this intermediate characteristic point allow the mathematical model to have optimal stability and fit. The determination method is as follows: During the actual operation of the power plant, a period in which the boiler load, coal quality, and flue gas conditions remain stable is selected. During this period, the sulfur trioxide injection system is adjusted to a series of different and stable injection concentration gradients. For each injection concentration, after the sulfur trioxide injection system response stabilizes, the secondary voltage and secondary current time-series data of the target electric field of the electrostatic precipitator are recorded simultaneously. The effective driving velocity of the dust is calculated by analyzing the volt-ampere characteristic curve, and the percentage increase of this driving velocity relative to the baseline value when no sulfur trioxide is injected is used as a quantitative indicator of the conditioning effect at that sulfur trioxide concentration. Subsequently, with the measured sulfur trioxide concentration as the abscissa and the corresponding conditioning effect index as the ordinate, the nonlinear least squares method is used to fit the data points to the formula described above. and In the relationship curve, an optimization algorithm is used to find the solution that minimizes the sum of squared errors between the fitted curve and the measured data. and The value obtained from this The numerical value represents the characteristic concentration of sulfur trioxide in conditioning. The shape factor representing the tunable response is a dimensionless constant greater than zero. This tunable response shape factor is related to... Jointly determined during the same fitting process, it determines the steepness and shape of the conditioning effect curve transitioning from the initial stage to the saturation stage; exp represents the natural exponential function, which, combined with the outer negative sign, forms a saturation growth curve asymptotically from 0 to 1, used to simulate the physical process of the conditioning effect rapidly increasing with concentration and then tending to saturate; the concentration term in the formula participates in the calculation in the form of a ratio to the characteristic concentration, and is expressed through a power function. The shape of the curve is adjusted by exponentiation to precisely match the actual response characteristics of the conditioning effect under different coal qualities and operating conditions; S330: Output factors affecting coal dust removal and flue gas conditioning; The calculated coal dust removal influencing factor and flue gas conditioning influencing factor are output as two independent key indicators that respectively characterize the inherent characteristics and the intensity of external intervention; these two dimensionless factors provide quantitative input for the next step of comprehensive evaluation of the actual working condition of dust. S4: Integrate the influencing factors of coal dust removal and flue gas conditioning to generate a working condition assessment index that characterizes the resistivity of dust, and determine the corresponding expected corona current range based on the working condition assessment index. In this embodiment, step S4 includes the following specific details, which can be found in the flowchart below. Figure 5 , Figure 5 Here is a detailed flowchart of the working condition assessment and interval generation process provided in the embodiments of this application: S410: The working condition evaluation index is calculated based on the influence of coal dust removal influencing factors and flue gas conditioning influencing factors on dust resistivity. The effective resistivity of dust in an electric field is ultimately determined by both the inherent properties of coal and the effect of flue gas conditioning. The coal dust removal influencing factor reflects the potential trend of increasing resistivity, while the flue gas conditioning influencing factor reflects the actual degree of reduction in resistivity. To comprehensively characterize this dynamic equilibrium state, it is necessary to calculate the operating condition evaluation index. The formula for calculating the operating condition assessment index is: ; In the formula, G represents the working condition evaluation index, which is a dimensionless scalar with a value greater than zero. The higher the value, the higher the effective resistivity of the dust, and the stronger the corona discharge that the electric field needs to maintain to overcome the collection difficulties caused by the high resistivity. The lower the value, the more the dust resistivity is in the ideal range where it is easier to collect. Indicates the factors affecting coal dust removal; Indicates the factors affecting flue gas conditioning; The design benchmark value of the coal quality influencing factor is a dimensionless constant. This value is taken as the theoretical result obtained by the coal quality dust removal influencing factor calculation model when the boiler burns the design coal type, and serves as a reference benchmark for evaluating the current coal quality impact. The design benchmark value of the flue gas conditioning influence factor is a dimensionless constant. This value is taken as the theoretical result obtained by the flue gas conditioning influence factor calculation model under the rated sulfur trioxide conditioning concentration, and serves as a reference benchmark for evaluating the current conditioning effect. The fusion weight index, representing the impact of coal dust removal, is a dimensionless constant greater than zero. The fusion weighting index representing the influence of flue gas conditioning is a dimensionless constant greater than zero; the fusion weighting index representing the influence of coal dust removal... The fusion weight index of the influence of flue gas conditioning The determination method is as follows: A large amount of historical operating data covering different coal qualities and conditioning combinations is collected. For each historical operating condition, the state level of dust resistivity under that condition is comprehensively determined by analyzing the shape of its electric field's volt-ampere characteristic curve, the frequency of back corona phenomena, and the change in outlet dust concentration, thus forming a state label. Then, using the normalized coal quality factor and normalized conditioning factor corresponding to that operating condition as input features, and the state label as the learning objective, a gradient boosting decision tree model is used for supervised learning training. After training, the importance scores of these two input features in the gradient boosting decision tree model are extracted. This gradient boosting decision tree score quantifies the contribution of the feature to reducing the overall prediction error when the decision tree splits into nodes. Finally, the importance scores of the two features are normalized respectively, and the resulting proportions are determined as follows: and The value of λ represents the time inertia coefficient of the change in operating conditions, which is a very small positive dimensionless constant. This time inertia coefficient is used to characterize the fact that when the coal quality or conditioning conditions change abruptly, the effect on the actual resistivity of dust is not instantaneous, but rather there is a physical process that gradually reaches a new steady state over time. In this embodiment, the method for determining λ is as follows: multiple events in which the coal quality and sulfur trioxide injection amount undergo significant step changes are selected from historical data. For each event, the evolution sequence of the operating condition evaluation index is recorded for a period of time after the change occurs. The evolution sequence of each event is fitted to an exponential decay function using a nonlinear fitting method. The reciprocal of the decay time constant of the exponential decay function is an estimated value of λ for that event. Finally, the average of the estimated values ​​of all events is taken as the value of λ used. The time difference between the current time and the moment of the most recent significant change in operating conditions is obtained in this embodiment as follows: Coal quality data and conditioning setpoints are continuously monitored, and the moments of the most recent significant changes in coal quality and conditioning setpoints are recorded respectively; then, the time difference between the current time and these two moments is calculated, and finally, the minimum of these two time differences is taken as the time difference. ; This represents the time normalization constant, with a value of 1 hour. Its function is to normalize the time... It is converted into a dimensionless value; exp represents the natural exponential function, which is multiplied here by the negative inertia coefficient and the normalized time difference to form a factor that gradually decays from 1, used to simulate the physical inertia effect that gradually and fully manifests the influence of the new working condition over time. The entire formula integrates the relative influence of coal quality and conditioning through the power operation of two ratios, and introduces a smooth transition in the time dimension through the exponential decay term, and finally outputs a dimensionless working condition evaluation index that can stably, continuously and accurately characterize the state of dust comprehensive resistivity in dynamic changes. S420: Based on the pre-established mapping relationship between the operating condition evaluation index and the benchmark expected current, the corresponding benchmark expected current is obtained according to the operating condition evaluation index; To transform the abstract operating condition evaluation index, which characterizes the operating condition, into a specific current value that can be directly used for control, it is necessary to establish a quantitative mapping relationship from the operating condition evaluation index to the corona current benchmark value. This quantitative mapping relationship is achieved through a machine learning model trained based on historical best operating data. The construction method of this machine learning model is as follows: extract a large amount of time period data from the historical long-cycle operating database where the boiler load, coal quality, and conditioning are all in a stable state; for each stable time period, calculate the corresponding operating condition evaluation index, and use the statistical median value of the secondary current time series when the target electric field of the electrostatic precipitator maintains high dust removal efficiency and does not experience abnormal states such as back corona or frequent flashover during that time period as the expected current label under the optimal operating condition. This results in a paired sample dataset containing operating condition assessment indices and expected current labels. Using this paired sample dataset, a radial basis function neural network (RBN) with a single-input, single-output structure is trained. This RBN learns the complex nonlinear correspondence between the operating condition assessment indices and expected current labels by adjusting the center and width of the hidden layer basis functions and the weights of the output layer. The trained RBN model serves as a knowledge base for querying the benchmark expected current from the operating condition assessment indices. In the current operation, after the operating condition assessment indices are calculated, they are input into the trained RBN model, and the current value output by the RBN model is used as the benchmark expected current under the current comprehensive operating condition. S430: Based on the reference desired current, the desired corona current range including the upper and lower limits is obtained by expanding it. Considering the slight fluctuations in equipment performance, measurement noise, and response characteristics of the control system during actual operation, extending the single reference expected current value into a safe operating range with a certain width can enhance the robustness and stability of the control system and avoid oscillations caused by frequent fine-tuning; it is necessary to calculate the lower and upper limits of the expected current corona range. The formula for calculating the lower limit coefficient of the desired corona current range is: ; In the formula, The coefficient representing the lower limit of the desired corona current range; This represents the baseline expected current value, which is obtained by querying the radial basis function neural network model. σ represents the maximum secondary current limit allowed for long-term operation of the high-voltage power supply equipment in the target electric field under safety regulations, and this value is determined by the equipment technical specifications; σ represents the basic relative fluctuation margin, which is a dimensionless constant between 0 and 0.2; in this embodiment, the method for setting the basic relative fluctuation margin σ is as follows: collect secondary current data during stable operating periods in long-term historical operation, calculate the standard deviation of the current data in each period, and divide the standard deviation of the current data in each period by the average value of the current in that period to obtain a series of relative fluctuation quantities; finally, take the percentile of the statistical distribution of all relative fluctuation quantities as the value of the basic relative fluctuation margin σ. The current adaptive range expansion coefficient is a very small positive dimensionless constant; in this embodiment, the current adaptive range expansion coefficient... The determination method is as follows: Divide the historical operating data into several intervals according to the average current level; for each current interval, statistically analyze all stable operating records that have not triggered the electric field protection action, and calculate the maximum relative range of current fluctuation in these stable operating records; analyze the relationship between the maximum relative fluctuation range and the current level, and obtain the expansion coefficient and the slope of the normalized current through linear regression fitting. This slope is the current adaptive interval expansion coefficient. The value; After calculating the lower limit coefficient of the desired corona current range, the lower limit coefficient of the desired corona current range is multiplied by the reference desired current value to obtain the lower limit value of the desired corona current range. The formula for calculating the upper limit coefficient of the desired corona current range is: ; In the formula, A coefficient representing the upper limit of the desired corona current range; Indicates the reference expected current value; σ represents the maximum secondary current limit that the high-voltage power supply equipment of the target electric field is allowed to operate for a long time under safety regulations; σ represents the relative fluctuation margin of the foundation. The range expansion coefficient represents the adaptive current range; the parameters in the formula , σ and Definition and determination method and calculation of the lower limit of the desired corona current range The timing is completely consistent; After calculating the upper limit coefficient of the desired corona current range, multiply the upper limit coefficient of the desired corona current range by the reference desired current value to obtain the upper limit of the desired corona current range. Ultimately, the closed interval defined by the lower limit and the upper limit of the desired corona current range is the desired corona current range of the target electric field under the current operating conditions. This desired corona current range will serve as the core basis for subsequent voltage adjustment decisions. S5: Extract the characteristic current value based on the secondary current time series data, calculate the electric field control margin based on the secondary voltage time series data, compare the characteristic current value with the expected corona current range, and determine the target voltage adjustment amount in combination with the electric field control margin. In this embodiment, step S5 includes the following specific details, which can be found in the flowchart below. Figure 6 , Figure 6 Here is a detailed diagram of the intelligent decision-making process provided in the embodiments of this application: S510: Perform statistical distribution analysis on the secondary current time series data within a set period, and determine the median of the statistical distribution as the characteristic current value; To accurately characterize the stable discharge level of the current electric field, it is necessary to eliminate the influence of abnormal disturbances such as instantaneous spark discharges. In this embodiment, all secondary current time-series data points within a five-minute period from the most recent complete operating cycle are selected to form a current data sample set. First, each original current measurement value in the set is divided by the rated secondary current value of the current electric field to obtain a dimensionless current ratio sequence. The rated secondary current value is a fixed constant determined based on the design parameters of the electrostatic precipitator and the capacity of the high-voltage power supply. Then, this current ratio sequence is statistically analyzed. To mitigate extreme high or low values ​​caused by instantaneous flashovers or signal interference, the median of the current ratio sequence is calculated instead of the arithmetic mean. The median more robustly reflects the stable corona discharge center trend within that time period. Finally, this calculated dimensionless median value is determined as the characteristic current value characterizing the current stable discharge intensity. S520: Calculate the average operating voltage based on the secondary voltage timing data, and determine the electric field control margin by comparing the average operating voltage with the preset breakdown voltage safety threshold. Assessing the safety margin between the current operating voltage and the breakdown voltage that could trigger a flashover is crucial for ensuring the safety of any voltage increase operation. First, the arithmetic mean of all original secondary voltage samples within the same time window as the characteristic current value extraction is calculated to obtain the current average operating voltage. Furthermore, to determine the safe voltage upper limit, the breakdown voltage safety threshold under the current operating condition needs to be dynamically calculated. This breakdown voltage safety threshold is calculated based on Paschen's law for gas discharge, with engineering corrections made for the actual flue gas composition and state. In this embodiment, the specific calculation process is as follows: The absolute temperature, absolute pressure, and water vapor partial pressure of the flue gas are collected in real time. The water vapor partial pressure is obtained by converting the measured absolute humidity using the ideal gas law. The breakdown voltage safety threshold is calculated to be directly proportional to the absolute pressure of the flue gas and the distance between the electrostatic precipitator plates, while also being significantly affected by flue gas temperature and humidity. To accurately quantify the influence of temperature and humidity, a temperature and humidity correction coefficient calibrated based on extensive field test data is used. This correction coefficient is determined by measuring the actual breakdown voltage values ​​under different combinations of flue gas temperature and humidity conditions in the laboratory and power plant. These measured values ​​are then compared with theoretical calculations based on standard Paschen's law. A multivariate nonlinear regression analysis method is used, with temperature and humidity as independent variables and the ratio of the measured breakdown voltage to the theoretical calculation value as the dependent variable, to fit a continuous function expression that can accurately predict temperature and humidity changes. The influence of temperature and humidity changes on breakdown voltage is assessed. Substituting real-time collected temperature and humidity data into this function yields the temperature and humidity correction coefficients under current conditions. Multiplying the correction coefficients, flue gas pressure, electrode spacing, and empirical constants for flue gas composition gives the dynamic breakdown voltage safety threshold. The empirical constants for flue gas composition are determined as follows: under optimal baseline operating conditions where the boiler burns the designed coal type and combustion is adjusted to its best state, the flue gas composition is stable and known. The breakdown voltage at this point is measured. The measured breakdown voltage is divided by the theoretical voltage value calculated considering only pressure, electrode spacing, and the current temperature and humidity correction coefficients; the quotient is the flue gas composition correction constant under these baseline operating conditions. This dynamic breakdown voltage safety threshold varies with the physical state of the flue gas. The system updates in real time. After obtaining the average operating voltage and the dynamic breakdown voltage safety threshold, the electric field control margin is calculated. The electric field control margin is defined as the relative proportion of the remaining available safe voltage space to the total safe space under the current operating voltage level. It is calculated by subtracting the ratio of the average operating voltage to the dynamic breakdown voltage safety threshold from a value of one. The result is a dimensionless scalar between zero and one. The closer the dimensionless scalar value is to one, the more abundant the safe space for voltage adjustment is, indicating that the current operating voltage is far below the breakdown threshold. The closer the dimensionless scalar value is to zero, the more the operating voltage is approaching the danger threshold, indicating that the voltage space for voltage adjustment is extremely limited, or even that the voltage is too high and voltage reduction needs to be considered to restore the safety margin. S530: Determine whether the characteristic current value is within the desired corona current range; if it is, set the target voltage adjustment to zero; if it is not, calculate the required adjustment based on the deviation between the characteristic current value and the median of the desired corona current range, and use the electric field control margin to constrain the required adjustment to generate the target voltage adjustment. First, determine whether the characteristic current value obtained by the characteristic current value extraction step falls within the expected corona current range determined by the expected corona current range determination step; if the characteristic current value is within the expected corona current range, it is determined that the discharge intensity of the current electric field has met the optimal target under the current operating conditions, and the target voltage adjustment is set to zero, that is, the current voltage operation is maintained without intervention. If the characteristic current value is not within the specified range, the voltage adjustment calculation is initiated. First, the arithmetic mean of the desired corona current range is calculated as the target median. Then, the deviation between the target median and the current characteristic current value is calculated. Based on this deviation, a preliminary voltage adjustment is calculated. In this embodiment, the specific calculation process is as follows: the current deviation value is divided by a preset scaling factor, the resulting quotient is input into a hyperbolic tangent function for calculation, the output value of the hyperbolic tangent function is multiplied by a preset voltage adjustment gain coefficient, and finally multiplied by the rated secondary voltage value of the electric field to obtain the preliminary voltage adjustment. The scaling factor is a small positive number. In this embodiment, its specific value is determined as follows: all records of voltage adjustment events marked as successful are extracted from the historical database. Each record contains the normalized current deviation value before adjustment and the actual executed and verified effective voltage adjustment. Voltage adjustment amount; using these historical current deviation values ​​as input, and the ratio of the corresponding voltage adjustment amount to the product of the rated voltage and the gain coefficient as the desired output, a nonlinear least squares fitting algorithm is used to fit the hyperbolic tangent function to these data points, minimizing the sum of squared errors between the function output and the historical actual values. The function input scaling parameter obtained by the final optimization process is determined as the value of this scaling factor; the voltage adjustment gain coefficient is a dimensionless constant. The method for determining the voltage adjustment gain coefficient is as follows: select multiple periods in the historical data where the electric field conditions are stable and the dust removal efficiency is consistently high, calculate the natural fluctuation standard deviation of the current during these periods; multiply the natural fluctuation standard deviation by an empirical multiple, and then convert it into an adjustment ratio for the rated voltage. This adjustment ratio is used as the reference value of the gain coefficient, and is finally determined after fine-tuning through a small number of on-site closed-loop tests; Next, the maximum allowable safe voltage adjustment amount is calculated. This calculation requires a safety margin consumption factor, which is a constant between zero and one. Its value is based on the voltage regulation response time constant explicitly specified in the technical specifications provided by the high-voltage power equipment manufacturer, and the maximum safety risk exposure ratio for a single operation as explicitly stipulated in the power plant operation safety regulations. Specifically, the voltage regulation time constant is multiplied by the risk exposure ratio specified in the regulations, and then divided by a constant determined based on the maximum allowable number of consecutive adjustments under typical operating conditions. The final result is normalized to the range of zero to one, yielding the safety margin consumption factor. Then, the calculated electric field control margin, the aforementioned safety margin consumption factor, and the dynamic breakdown voltage safety threshold are multiplied together; the product is the maximum allowable safe voltage adjustment amount for this operation. Finally, the absolute value of the calculated preliminary voltage demand adjustment is compared with the maximum allowable safe voltage adjustment. The smaller of the two values ​​is taken as the actual magnitude of the voltage adjustment. The adjustment direction is determined by the sign of the current deviation value: if the characteristic current value is lower than the target median, a voltage boost adjustment is performed; if the characteristic current value is higher than the target median, a voltage reduction adjustment is performed. This generates the final target voltage adjustment that combines target tracking and operational safety. To more intuitively illustrate the spatial relationship and control concept that the characteristic current value is the controlled variable, and the target is not a fixed point but a dynamic range defined by both safety and efficiency, please refer to [link to relevant documentation]. Figure 7 ; Figure 7 This is a schematic diagram illustrating the relationship between the characteristic current value and the desired corona current range provided in this embodiment. The diagram uses three unlabeled concentric rings to represent different control target areas. The innermost ring represents the theoretically optimal current point under ideal conditions, the middle ring represents the calculated desired corona current range, i.e., the optimal operating range that combines safety and efficiency, and the outermost ring represents the maximum operating boundary allowed by safety regulations. The arrows in the diagram represent the characteristic current values ​​extracted from real-time secondary current data. The arrows point to and fall within the middle ring, visually demonstrating that the core objective of the control method in this embodiment is to precisely guide and stabilize the actual operating characteristic current value within the dynamically optimized desired corona current range through closed-loop feedback adjustment. This embodiment uses closed-loop feedback control to drive the arrows representing the actual operating state to precisely approach and stabilize within the desired optimized range, thereby achieving the core objective of precise and stable optimized operation. S6: Adjust the operating voltage of the target electric field based on the target voltage adjustment amount, and re-collect data after adjustment to verify the adjustment effect; In this embodiment, step S6 includes the following specific details, which can be found in the flowchart below. Figure 8 , Figure 8 Here is a detailed diagram of the execution and feedback verification process provided in the embodiments of this application: S610: Based on the target voltage adjustment amount, issue a voltage adjustment command to the high-voltage power supply equipment of the target electric field; The target voltage adjustment is converted into a specific voltage setpoint instruction that the high-voltage power supply equipment controller can recognize. This specific voltage setpoint instruction is sent to the high-frequency power supply controller corresponding to the target electric field through the fieldbus communication network. After receiving the instruction, the high-frequency power supply controller smoothly adjusts the set reference value of the output voltage according to the internally preset smooth rate and ramp time, drives the high-voltage circuit, and makes the actual operating voltage of the electric field transition to the new target value without overshoot or oscillation, thereby changing the intensity of corona discharge. S620: After the voltage adjustment command is executed, the secondary voltage timing data and secondary current timing data of a set period are re-acquired; After the voltage adjustment command is issued, wait for the electric field to stabilize under the new voltage setting for a preset period of time to ensure that the transient process ends and a new steady state is entered. After the waiting period ends, restart a complete data acquisition cycle that is the same length as the decision cycle, and synchronously collect the original time-series data of the secondary voltage and secondary current of the target electric field under the new operating state to provide a data basis for effect verification. S630: Calculates new characteristic current values ​​based on the reacquired secondary current time-series data, and calculates new electric field control margin based on the reacquired secondary voltage time-series data. Based on the newly acquired secondary current time series data, the characteristic current value under the current operating condition is recalculated using the same method described in the characteristic current value extraction step, and recorded as the new characteristic current value. At the same time, based on the newly acquired secondary voltage time series data and the latest flue gas state parameters, the control margin of the current electric field is recalculated using the same method described in the electric field control margin calculation step, and recorded as the new electric field control margin. S640: If the new characteristic current value falls within the expected corona current range and the new electric field control margin is not lower than the preset threshold, the regulation is deemed effective and the current regulation parameters are maintained; otherwise, the regulation is deemed not to have met expectations and a new round of regulation process is triggered. The control effect is verified in a closed loop. The verification logic is based on two criteria: first, whether the newly calculated characteristic current value falls within the expected corona current range; second, whether the newly calculated electric field control margin is not lower than the preset safety threshold. The method for setting the safety threshold is as follows: extract data from the historical long-term operation database for all periods when the electric field is operating stably and no flashover faults have occurred; for each period, calculate the corresponding electric field control margin to form a control margin sample set; perform statistical analysis on the control margin sample set to calculate a specific low quantile, and determine this low quantile value as the preset safety threshold; this safety threshold represents the minimum safety buffer space that must be guaranteed under historical safe operating conditions; the safety threshold determined by this method is a specific value greater than zero, used in subsequent verification to determine whether the electric field after control is still in a state with the most basic safety redundancy; If both of the above conditions are met simultaneously, the voltage adjustment is deemed successful and effective, and the electric field operating state has been guided to an ideal range that is both efficient and safe. The current voltage parameters are maintained, and the complete data pair of this successful "operating condition-adjustment command-verification result" is stored in the historical knowledge base for future periodic optimization and updates of core models such as the benchmark expected current mapping model. If either of the above two conditions is not met, the adjustment is deemed to have failed to achieve the expected effect. The latest collected data is immediately used as the starting point for a new round of analysis, and the complete intelligent control process starting from data correction is re-triggered for rapid re-evaluation and re-decision, thereby forming an adaptive and robust closed-loop control process until the electric field operating state is stably controlled to the optimal target range. Please see Figure 9 , Figure 9 This is a schematic diagram of the structure of the intelligent control system for dust removal system based on Internet of Things data acquisition provided in the embodiments of this application; This embodiment demonstrates the overall system architecture for implementing the above method; the system constructs a complete technology chain from multi-source IoT data acquisition and intelligent condition assessment to closed-loop control execution through the collaborative work of six core functional modules; The data acquisition module serves as the system's data input interface, responsible for real-time acquisition of coal quality data entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide from the flue gas conditioning system, as well as secondary voltage time-series data and secondary current time-series data of the target electric field of the electrostatic precipitator. The data correction module, as a data quality assurance unit, is responsible for performing drift correction on the real-time sulfur trioxide concentration data based on historical standard data of sulfur trioxide, so as to obtain the effective concentration of sulfur trioxide. The impact factor calculation module, as the operating condition quantification unit, is responsible for calculating the coal quality dust removal impact factor based on the influence of ash and sulfur content in the coal quality data on dust resistivity; and calculating the flue gas conditioning impact factor based on the effect of effective sulfur trioxide concentration on dust resistivity. The working condition assessment module, as the state fusion and target generation unit, is responsible for fusing coal dust removal influencing factors and flue gas conditioning influencing factors, generating a working condition assessment index that characterizes the dust resistivity state, and determining the corresponding expected corona current range based on the working condition assessment index. The intelligent decision-making module, as the control command generation unit, is responsible for extracting the characteristic current value based on the secondary current time series data, calculating the electric field control margin based on the secondary voltage time series data, comparing the characteristic current value with the expected corona current range, and determining the target voltage adjustment amount in combination with the electric field control margin. The execution feedback module, as a closed-loop control unit, is responsible for regulating the operating voltage of the target electric field based on the target voltage adjustment amount, and re-collecting data after regulation to verify the regulation effect; Embodiments of the present invention also provide an electronic device, including a memory, a processor, and a communication bus; the memory and the processor are connected via the communication bus. The memory stores a dust removal system intelligent control method based on Internet of Things data acquisition, which can be loaded by the processor and executed as provided in the above embodiments.

[0020] The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the intelligent control method for a dust removal system based on IoT data acquisition provided in the above embodiments. The data storage area may store data involved in the intelligent control method for a dust removal system based on IoT data acquisition provided in the above embodiments.

[0021] The processor may include one or more processing cores. The processor executes instructions, programs, code sets, or instruction sets stored in memory, and calls data stored in memory to perform various functions and process data as described in this application. The processor may be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller, and microprocessor. It is understood that, for different devices, the electronic devices used to implement the above-described processor functions may also be other types, and the embodiments of this application do not specifically limit this.

[0022] A communication bus can include a pathway for transmitting information between the aforementioned components. The communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Communication buses can be categorized as address buses, data buses, control buses, etc.

[0023] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments, which is a method for intelligent control of a dust removal system based on Internet of Things data acquisition.

[0024] In this embodiment, a computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. A computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, a computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), spoofing random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0025] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0026] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A method for intelligent control of a dust removal system based on Internet of Things (IoT) data acquisition, characterized in that, Includes the following steps: S1: Real-time acquisition of coal quality data entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide in flue gas conditioning system, as well as secondary voltage time series data and secondary current time series data of target electric field of electrostatic precipitator. S2: Based on historical standard data of sulfur trioxide, drift correction is performed on the real-time concentration data of sulfur trioxide to obtain the effective concentration of sulfur trioxide; S3: Based on the influence of ash and sulfur content in the coal quality data on dust resistivity, the coal quality dust removal influence factor is calculated; based on the effect of effective sulfur trioxide concentration on dust resistivity, the flue gas conditioning influence factor is calculated. S4: Integrate the influencing factors of coal dust removal and flue gas conditioning to generate a working condition assessment index that characterizes the resistivity of dust, and determine the corresponding expected corona current range based on the working condition assessment index. S5: Extract the characteristic current value based on the secondary current time series data, calculate the electric field control margin based on the secondary voltage time series data, compare the characteristic current value with the expected corona current range, and determine the target voltage adjustment amount in combination with the electric field control margin. S6: Adjust the operating voltage of the target electric field based on the target voltage adjustment amount, and re-collect data after adjustment to verify the adjustment effect.

2. The intelligent control method for a dust removal system based on Internet of Things data acquisition according to claim 1, characterized in that, The real-time acquisition of coal quality data entering the boiler from the thermal power plant, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide from the flue gas conditioning system, and secondary voltage and secondary current time series data of the target electric field of the electrostatic precipitator includes: S110: Real-time acquisition of coal quality data entering the furnace of thermal power plants, including ash content and sulfur content; S120: Real-time acquisition of sulfur trioxide concentration data output by the sulfur trioxide concentration sensor of the flue gas conditioning system; S130: Obtain pre-calibrated historical standard data for sulfur trioxide; S140: Real-time acquisition of secondary voltage and secondary current timing data of the target electric field of the electrostatic precipitator.

3. The intelligent control method for a dust removal system based on IoT data acquisition according to claim 2, characterized in that, The process of applying drift correction to real-time sulfur trioxide concentration data based on historical standard data to obtain the effective concentration of sulfur trioxide includes: S210: Calculate the concentration measurement drift at the current moment based on the real-time sulfur trioxide concentration data and the historical standard data of sulfur trioxide; S220: The real-time concentration data of sulfur trioxide is compensated and corrected using the concentration measurement drift; S230: Outputs the compensated and corrected concentration value as the effective concentration of sulfur trioxide.

4. The intelligent control method for a dust removal system based on IoT data acquisition according to claim 3, characterized in that, The influence factor of coal quality dust removal is calculated based on the influence of ash content and sulfur content in the coal quality data of the furnace feed on dust resistivity. Based on the effect of effective sulfur trioxide concentration on dust resistivity, the flue gas conditioning influencing factors were calculated, including: S310: Based on the influence relationship between the ash content and the sulfur content on the resistivity of dust, the coal dust removal influence factor is calculated; S320: Based on the adjustment relationship between the effective concentration of sulfur trioxide and the resistivity of dust, the flue gas conditioning influence factor is calculated; S330: Output the coal dust removal influencing factor and the flue gas conditioning influencing factor.

5. The intelligent control method for a dust removal system based on Internet of Things data acquisition according to claim 4, characterized in that, The fusion of coal dust removal influencing factors and flue gas conditioning influencing factors generates a condition assessment index characterizing the resistivity of dust, and determines the corresponding expected corona current range based on the condition assessment index, including: S410: The working condition evaluation index is calculated based on the influence of the coal dust removal influencing factor and the flue gas conditioning influencing factor on the dust resistivity. S420: Based on the pre-established mapping relationship between the operating condition evaluation index and the benchmark expected current, the corresponding benchmark expected current is obtained according to the operating condition evaluation index; S430: Based on the reference desired current, the desired corona current range including the upper and lower limits is obtained by expanding it.

6. The intelligent control method for a dust removal system based on Internet of Things data acquisition according to claim 5, characterized in that, The process of extracting characteristic current values ​​based on secondary current time-series data, calculating the electric field control margin based on secondary voltage time-series data, comparing the characteristic current values ​​with the desired corona current range, and determining the target voltage adjustment amount in conjunction with the electric field control margin includes: S510: Perform statistical distribution analysis on the secondary current time-series data within a set period, and determine the median of the statistical distribution as the characteristic current value; S520: Calculate the average operating voltage based on the secondary voltage timing data, and determine the electric field control margin by comparing the average operating voltage with a preset breakdown voltage safety threshold. S530: Determine whether the characteristic current value is within the desired corona current range; if it is, set the target voltage adjustment amount to zero; if it is not, calculate the required adjustment amount based on the deviation between the characteristic current value and the median of the desired corona current range, and use the electric field control margin to constrain the required adjustment amount to generate the target voltage adjustment amount.

7. The intelligent control method for a dust removal system based on Internet of Things data acquisition according to claim 6, characterized in that, The process of regulating the operating voltage of the target electric field based on the target voltage adjustment amount, and then re-collecting data after regulation to verify the regulation effect, includes: S610: Based on the target voltage adjustment amount, issue a voltage adjustment command to the high-voltage power supply equipment of the target electric field; S620: After the voltage adjustment command is executed, secondary voltage timing data and secondary current timing data for a set period are re-acquired; S630: Calculates new characteristic current values ​​based on the reacquired secondary current time-series data, and calculates new electric field control margin based on the reacquired secondary voltage time-series data. S640: If the new characteristic current value falls within the expected corona current range and the new electric field control margin is not lower than the preset threshold, the regulation is deemed effective and the current regulation parameters are maintained; otherwise, the regulation is deemed not to have met expectations and a new round of regulation process is triggered.

8. An intelligent control system for a dust removal system based on Internet of Things (IoT) data acquisition, used to implement the intelligent control method for a dust removal system based on IoT data acquisition as described in any one of claims 1 to 7, characterized in that, include: The data acquisition module is used to collect real-time data on the coal quality entering the furnace of thermal power plants, real-time sulfur trioxide concentration data and historical standard data of sulfur trioxide from the flue gas conditioning system, as well as the secondary voltage time series data and secondary current time series data of the target electric field of the electrostatic precipitator. The data correction module is used to perform drift correction on the real-time concentration data of sulfur trioxide based on historical standard data of sulfur trioxide, so as to obtain the effective concentration of sulfur trioxide. The influencing factor calculation module is used to calculate the coal quality dust removal influencing factor based on the influence of ash and sulfur content in the coal quality data on dust resistivity; and to calculate the flue gas conditioning influencing factor based on the effect of effective sulfur trioxide concentration on dust resistivity. The operating condition assessment module is used to integrate the influencing factors of coal dust removal and the influencing factors of flue gas conditioning to generate an operating condition assessment index that characterizes the resistivity of dust, and to determine the corresponding expected corona current range based on the operating condition assessment index. The intelligent decision-making module is used to extract characteristic current values ​​based on secondary current time-series data, calculate electric field control margin based on secondary voltage time-series data, compare the characteristic current values ​​with the desired corona current range, and determine the target voltage adjustment amount in combination with the electric field control margin. The execution feedback module is used to regulate the operating voltage of the target electric field based on the target voltage adjustment amount, and to re-collect data after regulation to verify the regulation effect.

9. An electronic device comprising a processor and a memory, wherein, The memory stores a computer program that can be called by the processor; the processor executes the intelligent control method for dust removal system based on Internet of Things data acquisition as described in any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the intelligent control method for a dust removal system based on Internet of Things data acquisition as described in any one of claims 1 to 7.

Images