Method and system for intelligently adjusting boiler negative pressure based on induced draft fan AI

Abstract

The invention relates to a boiler negative pressure adjusting technology, and discloses a method and a system for intelligently adjusting boiler negative pressure based on induced draft fan AI, which are based on an induced draft fan adjusting data set, utilize a trained AI prediction model, and calculate the frequency of an induced draft fan at the next moment according to the current operating parameters and historical operating parameters and an induced draft fan AI intelligent adjusting control algorithm; according to the boiler negative pressure adjusting method designed by the invention, an induced draft fan frequency adjusting method of an AI prediction model is fused, the optimal operation frequency of the induced draft fan is calculated and predicted in real time according to the operation state of the boiler, and stable regulation and control of the hearth negative pressure are realized. The response speed and stability of the control system to the working condition change are improved, meanwhile, manual intervention is reduced, and the intelligentization and automation level of boiler system operation is improved.

Description

Technical Field

[0001] This invention relates to boiler negative pressure regulation technology, and more particularly to a method and system for intelligently regulating boiler negative pressure based on induced draft fan AI. Background Technology

[0002] During the operation of various industrial boilers, heating boilers, and combined heat and power units, the stability of furnace negative pressure is one of the key factors in ensuring the safe and economical operation of the boiler. A proper and stable furnace negative pressure not only helps maintain combustion efficiency and prevent flue gas backflow, but also effectively controls pollutant emission levels. Especially under conditions of high load fluctuations or frequent start-ups and shutdowns, furnace negative pressure is prone to fluctuations, thus affecting the safety and operational stability of the system.

[0003] Typically, boiler systems establish a negative pressure zone at the end of the flue by installing induced draft fans, maintaining the furnace interior at a pressure slightly below atmospheric pressure. The operating frequency (or damper opening) of the induced draft fans directly regulates flue gas flow and furnace pressure changes, making them a key actuator for achieving dynamic control of furnace negative pressure.

[0004] In traditional boiler systems, the operation and control of induced draft fans often rely on fixed-frequency operation or manual periodic adjustments, lacking the ability to provide real-time feedback and control based on actual furnace pressure changes. This can lead to excessively high or low furnace negative pressure. On the one hand, excessively high negative pressure may result in excessive air intake into the furnace, reducing thermal efficiency; on the other hand, insufficient negative pressure can easily cause flue gas leakage, increasing safety hazards.

[0005] As described in prior art 1, CN202411970565.2, a multi-objective combustion optimization method based on economic predictive control is provided. Given the frequent changes in power generation load and coal quality, conventional DCS control has shortcomings in areas such as airflow and oxygen optimization, balancing boiler efficiency and NOx emissions, and steam temperature control. Furthermore, some boiler parameters are difficult to measure accurately, limiting the applicability of existing combustion optimization technologies. This method first collects and preprocesses historical operating data, classifies and filters it according to steady-state conditions, and constructs an association model for preliminary combustion optimization. Next, based on unit characteristics and historical data, a load and coal quality change prediction model and a dynamic relationship model of the combustion process are established. Subsequently, in the online secondary optimization stage, initial parameter combinations are selected according to the current and predicted states, and the economic predictive control algorithm is used to calculate and apply the adjustment amounts of combustion parameters that satisfy multiple constraints for correction.

[0006] Existing technology 2, CN202510204325.X, discloses a method for simulating the state of unstable boiler combustion and overheating of the heating surface, including: screening of boiler sensitive parameters; boiler data acquisition: boiler data acquisition includes boiler combustion instability fault data acquisition, boiler overheating fault data acquisition, and boiler normal operation data acquisition; boiler data processing: based on the historical operating data of the unit, statistical and machine learning methods are used to process the collected operating parameters; establishing a state simulation function: the state simulation function integrates a state function, a probability function, and a prediction function; and evaluating the implementation effect.

[0007] Some boiler systems are equipped with furnace pressure sensors and induced draft fan frequency converters, and attempts are being made to use PID control to adjust the induced draft fan frequency based on the negative pressure deviation to achieve feedback regulation control. However, this traditional control method has the following main problems: Response lag: Classical control algorithms such as PID mainly rely on the current error for feedback adjustment, lacking predictive ability and making it difficult to cope with pressure fluctuations caused by rapid changes in boiler load; Low control precision: Control parameters are usually fixed and difficult to adapt to different operating conditions (such as start-up, low load, high load switching, etc.), resulting in insufficient control precision; The system is prone to oscillation: Under high-sensitivity adjustment, frequent adjustments or even "over-adjustment" are likely to occur, causing the system to oscillate and exacerbating negative pressure fluctuations. Reliance on manual intervention: In some complex working conditions, manual intervention by operators is still required, the control process is unstable, and the adjustment goals are not consistently achieved.

[0008] There are still problems in boiler negative pressure control, such as poor real-time performance, unstable regulation, and insufficient adaptability to operating conditions, making it difficult to achieve efficient coupling control between induced draft fan frequency and furnace pressure. Summary of the Invention

[0009] This invention addresses the problems of existing boiler negative pressure regulation methods, such as the inability to adjust the induced draft fan opening in a timely manner with multiple parameters and variables, and the complexity of precise calculation and operation of the induced draft fan. It provides a method and system for intelligent regulation of boiler negative pressure based on induced draft fan AI.

[0010] A method for intelligently regulating boiler negative pressure based on induced draft fan AI includes a boiler system, an induced draft fan regulating unit, an intelligent induced draft fan control cabinet, and an intelligent induced draft fan control system. The method includes: S1: Construct an intelligent induced draft fan control system database to collect and store the boiler system's operating parameters and the induced draft fan regulating unit's operating status data in real time; S2: Preprocess the data collected in step S1, and generate the induced draft fan adjustment dataset by correcting the collected data; S3: Based on the induced draft fan regulation dataset, using the trained AI prediction model, the frequency of the induced draft fan at the next moment is calculated according to the current operating parameters and historical operating parameters, based on the AI ​​intelligent regulation and control algorithm of the induced draft fan. S4: Send the frequency command of the induced draft fan calculated in step S3 to the intelligent control cabinet of the induced draft fan. S5: The induced draft fan regulating unit is controlled through the intelligent control cabinet of the induced draft fan to adjust the actual operating frequency of the induced draft fan.

[0011] As a preferred option, the operating parameters of the boiler system include main steam flow load data, blast furnace gas inlet valve opening data, and boiler negative pressure data. The operating status data of the induced draft fan regulating unit includes the induced draft fan set frequency point data, the induced draft fan feedback frequency point data, and the induced draft fan operating status point data.

[0012] Preferably, the data collected in step S1 is preprocessed to generate an induced draft fan adjustment dataset by correcting the collected data. The data correction process includes processing the input original time series dataset using the pre-filling principle.

[0013] Preferably, the input raw time series dataset includes the main steam flow load time series data x1_in of the boiler system, the blast furnace gas inlet valve opening time series data x2_in of the boiler system, the induced draft fan set frequency time series data s, and the boiler negative pressure time series data pa.

[0014] As a preferred method, based on the induced draft fan regulation dataset, the frequency of the induced draft fan at the next moment is calculated according to the current operating parameters and historical operating parameters using the AI ​​intelligent regulation and control algorithm of the induced draft fan. The AI ​​intelligent regulation and control algorithm is a linear regression algorithm based on machine learning.

[0015] As a preferred approach, the AI ​​intelligent regulation and control algorithm, which is a machine learning-based linear regression algorithm, establishes a linear mapping relationship between input and output variables; the linear regression model takes the form of: ; in: The predicted frequency of the induced draft fan at the next moment; Input variables for current or historical load, gas supply, and boiler negative pressure; These are the regression weight parameters.

[0016] To address the aforementioned technical problems, this invention also provides a system for intelligently regulating boiler negative pressure based on induced draft fan AI, comprising a boiler system, an induced draft fan regulating unit, an induced draft fan intelligent control cabinet, and an intelligent induced draft fan control system. The boiler system provides a negative pressure environment; the induced draft fan regulating unit, connected to the boiler system, adjusts the induced draft fan frequency; the intelligent induced draft fan control cabinet, connected to the induced draft fan regulating unit, receives frequency commands and controls the induced draft fan regulating unit; and the intelligent induced draft fan control system, connected to the intelligent induced draft fan control cabinet, processes data using an AI prediction model and calculates the induced draft fan frequency, then sends the calculated induced draft fan frequency command to the intelligent induced draft fan control cabinet.

[0017] To solve the above-mentioned technical problems, the present invention provides the following technical solution: The boiler negative pressure regulation method designed in this invention integrates an AI prediction model with an induced draft fan frequency regulation method. This method calculates and predicts the optimal operating frequency of the induced draft fan in real time based on the boiler's operating status, achieving stable control of the furnace negative pressure. This not only improves the response speed and stability of the control system to changes in operating conditions but also reduces manual intervention, enhancing the intelligence and automation level of the boiler system operation.

[0018] This invention relates to an automatic control algorithm for the induced draft fan process in boiler operation. It takes a multi-dimensional time series formed by multiple input operating indicators and control equipment regulation values, configures a decision model based on the control setpoint, loads the decision model into the algorithm, performs calculations, and then outputs the target regulation value of the control equipment.

[0019] This invention can determine the required induced draft fan frequency adjustment in real time based on the changing patterns and trends between the input operating indicators and equipment control values.

[0020] This invention can detect the frequency adjustment of the induced draft fan in real time, and it has good safety, low cost and simple operation. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention.

[0022] Figure 2 This is a rendering of the invention.

[0023] Figure 3 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0024] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0025] Example 1 A method for intelligently adjusting boiler negative pressure based on induced draft fan AI. Figure 1The system includes a boiler system, an induced draft fan regulating unit, an induced draft fan intelligent control cabinet, and an intelligent induced draft fan control system. The method includes: S1: Construct an intelligent induced draft fan control system database, collect and store the boiler system's operating parameters and the induced draft fan regulating unit's operating status data in real time; S2: Preprocess the data collected in step S1, and generate an induced draft fan regulating dataset by correcting the collected data; S3: Based on the induced draft fan regulating dataset, use a trained AI prediction model to calculate the frequency of the induced draft fan at the next moment according to the current operating parameters and historical operating parameters based on the induced draft fan AI intelligent regulating control algorithm; S4: Send the frequency command of the induced draft fan calculated in step S3 to the intelligent control cabinet of the induced draft fan; S5: Control the induced draft fan adjustment unit through the intelligent control cabinet of the induced draft fan to adjust the actual operating frequency of the induced draft fan.

[0026] The operating parameters of the boiler system include main steam flow load data, blast furnace gas inlet valve opening data, and boiler negative pressure data. The operating status data of the induced draft fan regulating unit includes the induced draft fan set frequency point data, the induced draft fan feedback frequency point data, and the induced draft fan operating status point data.

[0027] The data collected in step S1 is preprocessed, and the collected data is corrected to generate the induced draft fan adjustment dataset. The data correction process includes processing the input raw time series dataset according to the pre-order padding principle.

[0028] The input raw time series dataset includes the main steam flow load time series data of the boiler system x1_in, the blast furnace gas inlet valve opening time series data of the boiler system x2_in, the induced draft fan set frequency time series data s, and the boiler negative pressure time series data pa.

[0029] Based on the induced draft fan regulation dataset, the frequency of the induced draft fan at the next moment is calculated according to the current operating parameters and historical operating parameters using the AI ​​intelligent regulation and control algorithm for the induced draft fan. The AI ​​intelligent regulation and control algorithm is a linear regression algorithm based on machine learning.

[0030] For AI intelligent regulation and control algorithms, the linear regression algorithm, which is a machine learning algorithm, establishes a linear mapping relationship between input and output variables; the linear regression model takes the form of: ; in: The predicted frequency of the induced draft fan at the next moment; Input variables for current or historical load, gas supply, and boiler negative pressure; These are the regression weight parameters.

[0031] exist Figure 2 In the graph, the yellow line represents the furnace pressure curve under the algorithm control of the induced draft fan AI intelligent adjustment system; the blue line represents the furnace pressure curve under manual control. These two comparisons were conducted under the same control objective (controlling the furnace pressure to -150°C) and the same load. The results show that under AI control, furnace pressure fluctuations are reduced and converge within the target value, demonstrating better performance than manual control.

[0032] Example 2 Based on Example 1, this example is a system for intelligently regulating boiler negative pressure using an induced draft fan (AI). It includes a boiler system, an induced draft fan regulating unit, an intelligent induced draft fan control cabinet, and an intelligent induced draft fan control system. The boiler system provides a negative pressure environment. The induced draft fan regulating unit, connected to the boiler system, adjusts the induced draft fan frequency. The intelligent induced draft fan control cabinet, connected to the induced draft fan regulating unit, receives frequency commands and controls the regulating unit. The intelligent induced draft fan control system, connected to the intelligent induced draft fan control cabinet, processes data using an AI prediction model and calculates the induced draft fan frequency; then, it sends the calculated frequency command to the intelligent induced draft fan control cabinet.

[0033] Example 3 Based on the above embodiments, this embodiment... Figure 3 In the process, the boiler pressure is set by the operator (usually a slightly negative pressure, such as -100 Pa), and the actual boiler pressure is detected by the boiler pressure sensor. The DCS compares these two values, calculates the error, and generates a frequency setpoint using control algorithms such as PID (Proportional-Integral-Derivative) and sends it to the high-voltage frequency converter. The motor speed is controlled by adjusting the frequency of the output power supply, receiving control signals from the DCS. The frequency converter drives the motor, and the motor speed changes with the frequency of the frequency converter's output. The induced draft fan impeller is driven by the motor. The higher the speed, the greater the amount of flue gas extracted, and the greater the negative pressure (lower pressure) of the boiler. The frequency converter feeds back the actual operating status of the induced draft fan (such as actual operating frequency, output current, running / stopping status, fault codes, etc.) to the DCS system in real time, forming a closed-loop control that allows the DCS to accurately monitor the execution status. The temperature of critical components such as the induced draft fan bearings and motor windings is monitored. If the temperature exceeds the safety threshold, an alarm or even a cascading shutdown will be triggered to prevent equipment damage. The vibration amplitude of the induced draft fan and motor is also monitored. Excessive vibration may be a sign of mechanical failure (such as dust accumulation on the impeller, dynamic imbalance, or bearing wear), and the system will issue an alarm to prompt maintenance.

[0034] The DCS detected a significant deviation of -40 Pa between the actual pressure (-60 Pa) and the set pressure (-100 Pa). The PID algorithm immediately calculated the need for a substantial increase in induced draft and output a higher frequency command (48 Hz) to the inverter. Upon receiving the 48 Hz command, the inverter accelerated the drive motor, increasing the impeller speed and enhancing suction capacity. This stronger suction force extracted more flue gas from the furnace, gradually reducing the furnace pressure from -60 Pa back towards the target of -100 Pa. When the pressure returned to -100 Pa, the DCS error signal returned to zero, and the frequency output stabilized at a new, slightly higher value (e.g., 42 Hz) to accommodate the increased flue gas volume. The system entered a new steady-state.

Claims

1. A method for intelligently regulating boiler negative pressure based on induced draft fan AI, comprising a boiler system, an induced draft fan regulating unit, an induced draft fan intelligent control cabinet, and an intelligent induced draft fan control system, the method comprising: S1: Construct an intelligent induced draft fan control system database to collect and store the boiler system's operating parameters and the induced draft fan regulating unit's operating status data in real time; S2: Preprocess the data collected in step S1, and generate the induced draft fan adjustment dataset by correcting the collected data; S3: Based on the induced draft fan regulation dataset, using the trained AI prediction model, the frequency of the induced draft fan at the next moment is calculated according to the current operating parameters and historical operating parameters, based on the AI ​​intelligent regulation and control algorithm of the induced draft fan. S4: Send the frequency command of the induced draft fan calculated in step S3 to the intelligent control cabinet of the induced draft fan. S5: The induced draft fan regulating unit is controlled through the intelligent control cabinet of the induced draft fan to adjust the actual operating frequency of the induced draft fan.

2. The method for intelligently adjusting boiler negative pressure based on induced draft fan AI according to claim 1, characterized in that, The operating parameters of the boiler system include main steam flow load data, blast furnace gas inlet valve opening data, and boiler negative pressure data. The operating status data of the induced draft fan regulating unit includes the induced draft fan set frequency point data, the induced draft fan feedback frequency point data, and the induced draft fan operating status point data.

3. The method for intelligently adjusting boiler negative pressure based on induced draft fan AI according to claim 1, characterized in that, The data collected in step S1 is preprocessed, and the collected data is corrected to generate the induced draft fan adjustment dataset. The data correction process includes processing the input raw time series dataset according to the pre-order padding principle.

4. The method for intelligently adjusting boiler negative pressure based on induced draft fan AI according to claim 3, characterized in that, The input raw time series dataset includes the main steam flow load time series data of the boiler system x1_in, the blast furnace gas inlet valve opening time series data of the boiler system x2_in, the induced draft fan set frequency time series data s, and the boiler negative pressure time series data pa.

5. The method for intelligently adjusting boiler negative pressure based on induced draft fan AI according to claim 1, characterized in that, Based on the induced draft fan regulation dataset, the frequency of the induced draft fan at the next moment is calculated according to the current operating parameters and historical operating parameters using the AI ​​intelligent regulation and control algorithm for the induced draft fan. The AI ​​intelligent regulation and control algorithm is a linear regression algorithm based on machine learning.

6. The method for intelligently adjusting boiler negative pressure based on induced draft fan AI according to claim 1, characterized in that, For AI intelligent regulation and control algorithms, which are machine learning-based linear regression algorithms, a linear mapping relationship is established between input and output variables. The linear regression model takes the form of: ； in: The predicted frequency of the induced draft fan at the next moment; Input variables for current or historical load, gas supply, and boiler negative pressure; These are the regression weight parameters.

7. A system for intelligently regulating boiler negative pressure based on induced draft fan AI, comprising a boiler system, an induced draft fan regulating unit, an induced draft fan intelligent control cabinet, and an intelligent induced draft fan control system, characterized in that, Boiler systems are used to provide a negative pressure environment; The induced draft fan regulating unit is connected to the boiler system and is used to adjust the frequency of the induced draft fan. The induced draft fan intelligent control cabinet is connected to the induced draft fan regulating unit and is used to receive frequency commands and control the induced draft fan regulating unit. The intelligent induced draft fan control system is connected to the induced draft fan intelligent control cabinet, uses an AI prediction model to process data and calculate the frequency of the induced draft fan, and sends the calculated frequency command to the induced draft fan intelligent control cabinet.

Images