A dynamic intelligent speed regulation control method for an electric furnace steelmaking dust removal fan

Abstract

The present application belongs to the technical field of industrial process automation control, and provides a dynamic intelligent speed regulation control method for an electric furnace steelmaking dust removal fan, which comprises the following steps: collecting fourth hole outlet temperature of an electric furnace, waste heat recovery inlet temperature, flue gas volume flow, PM2.5 concentration and electric furnace smelting parameters and other multi-parameter data through deployment of sensors; correcting the temperature data in time and space by using flue gas volume flow transmission delay, and predicting the waste heat recovery inlet temperature change by using an LSTM network; dividing low-temperature, waste heat optimal, high-temperature and other regulation and control scenes according to the predicted temperature, combining the PM2.5 concentration, dynamically adjusting the fan wind speed by using a PID controller, and balancing energy saving, waste heat recovery and dust suppression effect; finally, calculating the actual transmission delay by using a cross-correlation function, optimizing the LSTM parameters, and optimizing the wind speed again, so that the present application realizes intelligent speed regulation of the fan, improves the waste heat recovery efficiency, reduces energy consumption, and solves the contradiction between smoke dust emission, waste heat utilization and energy saving.

Description

Technical Field

[0001] This invention belongs to the field of industrial process automation control technology, specifically a dynamic intelligent speed regulation control method for dust removal fans in electric arc furnace steelmaking. Background Technology

[0002] The electric arc furnace smelting process generates a large amount of high-temperature dusty flue gas. Due to production conditions such as opening the furnace cover and disconnecting the fourth flue of the electric arc furnace during the charging and tapping stages, the flue gas temperature will deviate significantly with the changes in the electric arc furnace smelting conditions.

[0003] Conventional dust removal fans need to operate at a constant high speed to ensure dust removal and purification effects. However, when an electric furnace waste heat recovery system is configured, it is necessary to use the heat energy of high-temperature flue gas to generate steam. The boiler heat exchanger of the waste heat recovery system has an optimal heat exchange temperature range. When the flue gas temperature at the waste heat recovery inlet is lower than the heat exchange temperature matched by the system, it will lead to low heat exchange efficiency and low steam production of the waste heat recovery system.

[0004] Traditional control only responds to single variables and cannot match complex operating conditions in real time, resulting in high fan energy consumption and low waste heat utilization. It is difficult to solve the contradiction of optimal matching of dust emission, waste heat utilization and energy saving. For example, although low wind speed in the high temperature flue gas stage is conducive to waste heat recovery and energy saving, the dust emission suppression effect is poor. In the low temperature flue gas stage, high wind speed is required in the waste heat recovery section to improve heat exchange efficiency, but if the flue gas temperature does not match the boiler's optimal heat exchange temperature, it will have the opposite effect.

[0005] Therefore, there is an urgent need for a control method that can dynamically adjust the fan speed according to the changes in flue gas temperature under different operating conditions of the electric furnace, so as to balance the trade-offs among the three factors and achieve the goal of optimal matching and high efficiency and energy saving.

[0006] Therefore, the present invention provides a dynamic intelligent speed regulation control method for dust removal fans in electric arc furnace steelmaking. Summary of the Invention

[0007] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0008] The technical solution adopted by this invention to solve its technical problem is:

[0009] In a first aspect, the present invention provides a dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking, comprising:

[0010] S1: Deploy sensors to collect multi-parameter data;

[0011] S2: Use transmission delay to perform spatiotemporal correction on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data, and use the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the data to predict the change of the inlet temperature of the waste heat recovery in the data.

[0012] S3: Based on the predicted changes in the waste heat recovery inlet temperature, control scenarios are divided, and the fan speed is dynamically adjusted according to the divided control scenarios and the PM2.5 concentration.

[0013] S4: Based on the control scenario, further optimize the fan speed after dynamic control.

[0014] In this invention, as a further improvement, the specific process of deploying the sensor is as follows:

[0015] A temperature sensor is installed at the fourth hole of the electric furnace to collect temperature information inside the furnace. A temperature sensor is also installed at the boiler waste heat recovery inlet to reflect the recoverable temperature. A flue gas flow meter is installed at the boiler waste heat recovery inlet to record the actual flue gas volume of the waste heat recovery system. PM2.5 monitors are deployed on the top and outside of the plant to record PM2.5 concentration data.

[0016] In this invention, as a further improvement, the multi-parameter data specifically includes:

[0017] Temperature data recorded by the temperature sensor installed at the fourth hole of the electric furnace is recorded as the outlet temperature of the fourth hole of the electric furnace; temperature data recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery is recorded as the inlet temperature of the waste heat recovery; flue gas volume flow rate data recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery; current and voltage data and electrode position change rate data obtained from the electric furnace control system via the industrial bus; and PM2.5 concentration data recorded by the PM2.5 monitor.

[0018] In this invention, as a further improvement, the specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in the multi-parameter data includes:

[0019] Process the collected multi-parameter data and remove abnormal data;

[0020] The temperature at the outlet of the fourth hole of the electric furnace, recorded by the temperature sensor installed at the fourth hole, is denoted as T1. The temperature at the inlet of the waste heat recovery system, recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery system, is denoted as T2. The flue gas volumetric flow rate recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery system is denoted as V. 余 The current, voltage, and electrode position data obtained from the electric furnace control system via the industrial bus are denoted as I. 电 U 电 R 电 The monitoring data recorded by the PM2.5 monitor is denoted as ug;

[0021] For any parameter including: T1, T2, V 余I 电 U 电 R 电 ,ug; calculate the mean μ and standard deviation σ of the parameter in real time over the past A minutes; compare the newly collected parameter with the mean μ and standard deviation σ of the parameter to determine whether the newly collected parameter is abnormal data;

[0022] If the newly collected data X satisfies |X-μ|>3σ, then the new data is determined to be abnormal data, and it is removed and replaced with the valid value from the previous time step.

[0023] If the newly collected data X satisfies |X-μ|<3σ, then the new data is determined to be normal data.

[0024] In this invention, as a further improvement, the specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in the multi-parameter data further includes:

[0025] Spatiotemporal correction was performed on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data after removing abnormal data. The transmission delay of the flue gas volume flow rate was calculated based on the outlet temperature T1 of the fourth hole of the electric furnace recorded by the temperature sensor installed in the fourth hole of the electric furnace, and spatiotemporal correction was performed based on the transmission delay of the flue gas volume flow rate.

[0026] Obtain the cross-sectional area S and length L of the flue, and then calculate the flue gas volumetric flow rate V. 余 First, calculate the transmission delay of the flue gas volume flow rate, and then perform spatiotemporal correction based on the transmission delay of the flue gas volume flow rate.

[0027] Calculate the flue gas volumetric flow rate per second: Among them, V 余 Let be the flue gas volumetric flow rate, then the flue gas volumetric flow rate transmission delay is:

[0028] The time series data of the fourth outlet temperature of the electric furnace is delayed by Δt and then spatiotemporally matched with the time series data of the waste heat recovery inlet temperature to ensure that the causal relationship between the rise of the fourth outlet temperature of the electric furnace and the rise of the flue gas temperature at the waste heat recovery inlet is synchronized.

[0029] In this invention, as a further improvement, the specific process of predicting the change in the waste heat recovery inlet temperature using the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is as follows:

[0030] The system selects the furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, and average current from the previous N time points, and the electrode position change rate from the previous N / 2 time points. It adopts a network structure with two layers of Long Short-Term Memory (LSTM) network and one fully connected layer, with 64 neurons in each layer, and outputs the predicted value of waste heat recovery inlet temperature for a future period. It obtains normal operating condition data, trains the model with mean absolute error as the loss function, and inputs the latest N time points of furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, average current, and N / 2 electrode position change rate to predict the change of waste heat recovery inlet temperature data.

[0031] In this invention, as a further improvement, the specific process of dynamically regulating the fan speed includes:

[0032] When the outlet temperature of the fourth hole of the electric furnace is lower than the low-temperature section threshold of the electric furnace smelting process and the predicted change in the waste heat recovery inlet temperature is less than the low-temperature section recovery threshold, the frequency converter of the fan is controlled by the PID controller to gradually reduce the wind speed of the primary dust removal fan and the secondary dust removal fan to the energy-saving mode and maintain the minimum safe wind speed.

[0033] In this invention, as a further improvement, the specific process of dynamically regulating the fan speed further includes:

[0034] When the waste heat recovery inlet temperature is within the optimal temperature range for waste heat recovery, the frequency converter of the fan is controlled by the PID controller to switch the fan speed to energy efficiency mode, and the fan speed is adjusted according to V. 机 =k·T2+b gradually increases the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient and b≈30 is the base speed drift;

[0035] PM2.5 concentration data is obtained through PM2.5 monitoring stations. When the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to accelerate to the rated maximum speed. Activate dust suppression mode;

[0036] When the PM2.5 concentration does not exceed the standard, the fan speed is set according to V. 机 =k·T2+b gradually increases the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient and b≈30 is the base speed drift.

[0037] In this invention, as a further improvement, the specific process of dynamically regulating the fan speed further includes:

[0038] When the waste heat recovery inlet temperature rises above the system's safe temperature, the fan is forced to slow down to the minimum safe speed, and an over-temperature alarm is issued. Once the temperature returns to the optimal waste heat recovery temperature range, the PID controller controls the fan's frequency converter to switch the fan speed to energy efficiency mode, with the fan speed following the V... 机 =k·T2+b Gradually increase the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient, b≈30 is the base speed drift, PM2.5 concentration data is obtained through PM2.5 monitoring points, and when the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to speed up to the rated speed or the set maximum speed. Activate dust suppression mode.

[0039] In this invention, as a further improvement, the specific process of optimizing the dynamically adjusted fan speed is as follows:

[0040] The real-time outlet temperature of the fourth hole of the electric furnace in group Y and the inlet temperature of the waste heat recovery are obtained. The cross-correlation function between the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is calculated. The peak value τ is the actual transmission delay. The actual transmission delay is updated to the parameters of the Long Short-Term Memory (LSTM) network to predict the accurate waste heat recovery inlet temperature. Based on the predicted accurate waste heat recovery inlet temperature, the fan speed is optimized again.

[0041] Secondly, the present invention provides a dynamic intelligent speed control system for dust removal fans in electric arc furnace steelmaking, comprising:

[0042] Data acquisition module: Deploys sensors to collect multi-parameter data;

[0043] Data spatiotemporal correction module: Uses transmission delay to perform spatiotemporal correction on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data, and uses the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the module to predict the change of the inlet temperature of the waste heat recovery in the module.

[0044] Fan speed control module: Based on the predicted changes in waste heat recovery inlet temperature, control scenarios are divided, and the fan speed is dynamically controlled according to the divided control scenarios and PM2.5 concentration.

[0045] Fan speed optimization module: Based on the control scenario, the fan speed after dynamic control is further optimized.

[0046] The beneficial effects of this invention are as follows:

[0047] Achieving multi-parameter coordinated control, by collecting data such as the outlet temperature of the fourth hole of the electric furnace, the inlet temperature of the waste heat recovery, the volumetric flow rate of flue gas, the PM2.5 concentration, and the smelting parameters of the electric furnace, and combining transmission delay spatiotemporal correction and LSTM network prediction, it accurately matches the complex operating conditions of the electric furnace and solves the limitations of traditional single-variable control.

[0048] By dynamically balancing multiple objectives and dividing control scenarios such as low temperature, optimal waste heat, and high temperature, the system can improve waste heat recovery efficiency and significantly reduce fan energy consumption while ensuring that dust emissions meet standards, thus resolving the contradiction between dust emissions, waste heat utilization, and energy conservation.

[0049] To improve control accuracy and stability, the system optimizes transmission delay in real time through cross-correlation functions, updates LSTM network parameters, and further optimizes the dynamically adjusted wind speed, reducing drastic fluctuations in fan speed, extending equipment life, enhancing system robustness, and adapting to industrial scenarios with flexibility. Thresholds and parameters can be set according to actual production conditions, adapting to different electric furnace scales and operating conditions, and has strong practicality and promotional value. Attached Figure Description

[0050] The invention will now be further described with reference to the accompanying drawings.

[0051] Figure 1 This is a flowchart of the steps of a dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to the present invention;

[0052] Figure 2 This is a system module diagram of a dynamic intelligent speed regulation and control system for dust removal fans in electric furnace steelmaking according to the present invention. Detailed Implementation

[0053] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0054] Example 1

[0055] like Figure 1 As shown, the present invention provides a dynamic intelligent speed regulation control method for dust removal fans in electric arc furnace steelmaking, comprising:

[0056] S1: Deploy sensors to collect multi-parameter data;

[0057] In S1, firstly, multiple sensors are deployed. A temperature sensor is installed at the fourth hole of the electric furnace to collect temperature information inside the furnace and reflect the peak smelting temperature. A temperature sensor is arranged at the boiler waste heat recovery inlet to reflect the recoverable temperature. At the same time, a flue gas flow meter is installed at the boiler waste heat recovery inlet to record the actual flue gas volume of the waste heat recovery system. PM2.5 monitoring points are deployed on the top of the plant and outside the plant to reflect the collection effect of electric furnace dust.

[0058] In addition, real-time smelting parameters of the electric furnace are obtained from the electric furnace control system via an industrial bus.

[0059] For example, a flue gas flow meter is installed in the horizontal flue section downstream of the waste heat recovery inlet, positioned at a distance of one times the pipe diameter from the upstream bend and three times the pipe diameter from the downstream bend to avoid flow field distortion. An ultrasonic flow meter with a range of 0–300,000 m³ / h is used. 3 / h, with an accuracy of ±1%, the volumetric flow rate of flue gas is measured using the ultrasonic time-of-flight propagation method;

[0060] The system connects to the edge computing gateway via RS485 interface using Modbus-RTU protocol, with a sampling frequency of 1Hz, and provides real-time feedback on the actual flue gas volume of the waste heat recovery system; real-time smelting parameters of the electric furnace are extracted from the process data block of the electric furnace PLC control system to reflect the boiler load status.

[0061] It should be noted that the real-time smelting parameters of the electric furnace include, but are not limited to: current, voltage, and electrode position change rate.

[0062] The flue gas flow meter is installed at a distance of two times the pipe diameter from the upstream elbow and three times the pipe diameter from the downstream elbow, specifically in the horizontal flue section downstream of the waste heat recovery inlet. This facilitates accurate subsequent calculations and avoids measurement errors caused by flow field distortion.

[0063] In S1, the second specific multi-parameter data includes: temperature data recorded by the temperature sensor installed at the fourth hole of the electric furnace, denoted as the outlet temperature of the fourth hole of the electric furnace; temperature data recorded by the temperature sensor installed at the waste heat recovery inlet of the boiler, denoted as the waste heat recovery inlet temperature; flue gas volume flow rate data recorded by the flue gas flow meter installed at the waste heat recovery inlet of the boiler; current and voltage data and electrode position change rate data obtained from the electric furnace control system via the industrial bus; and PM2.5 concentration data recorded by the PM2.5 monitor.

[0064] Among them, flue gas volume flow rate refers to the volume of flue gas passing through a certain cross section of the flue per unit time, with the unit being cubic meters per hour. It is used to characterize the amount of flue gas entering the waste heat recovery system and is the core parameter for calculating flue gas transmission speed, waste heat recovery efficiency, and matching boiler load.

[0065] S2: Use transmission delay to perform spatiotemporal correction on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data, and use the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the data to predict the change of the inlet temperature of the waste heat recovery in the data.

[0066] In S2, the first specific step is to process the collected multi-parameter data and remove abnormal data.

[0067] The temperature at the outlet of the fourth hole of the electric furnace, recorded by the temperature sensor installed at the fourth hole, is denoted as T1. The temperature at the inlet of the waste heat recovery system, recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery system, is denoted as T2. The flue gas volumetric flow rate recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery system is denoted as V. 余 The current, voltage, and electrode position data obtained from the electric furnace control system via the industrial bus are denoted as I. 电 U 电 The monitoring data recorded by the PM2.5 monitor is denoted as ug;

[0068] For any parameter (T1, T2, V) 余 I 电 U 电 P 锅 V 产 The system calculates the mean μ and standard deviation σ of the parameters over the past A minutes in real time; it then compares the newly collected parameters with the mean μ and standard deviation σ corresponding to the parameters to determine whether the newly collected parameters are abnormal data.

[0069] If the newly collected data X satisfies |X-μ|>3σ, then the new data is determined to be abnormal data, and it is removed and replaced with the valid value from the previous time step.

[0070] If the newly collected data X satisfies |X-μ|<3σ, then the new data is determined to be normal data;

[0071] It should be noted that, for any parameter, the mean μ and standard deviation σ of the past A minutes are calculated in real time. In this invention, the past A minutes can be set to calculate the past 2 minutes, etc. The specific time length is set reasonably according to the actual sensor acquisition frequency and the range of the acquired samples. This invention does not limit this.

[0072] In S2, the second specific step is to perform spatiotemporal correction on the electric furnace fourth hole outlet temperature and waste heat recovery inlet temperature data in the multi-parameter data after removing abnormal data. The electric furnace fourth hole outlet temperature T1 recorded by the temperature sensor installed in the electric furnace fourth hole is used as the standard to calculate the transmission delay of flue gas volume flow rate and perform spatiotemporal correction based on the transmission delay of flue gas volume flow rate.

[0073] Furthermore, obtain the cross-sectional area S and length L of the flue, and then determine the flue gas volumetric flow rate V. 余 First, calculate the transmission delay of the flue gas volume flow rate, and then perform spatiotemporal correction based on the transmission delay of the flue gas volume flow rate.

[0074] Furthermore, calculate the flue gas volumetric flow rate per second: Among them, V 余Let be the flue gas volumetric flow rate, then the flue gas volumetric flow rate transmission delay is:

[0075] After delaying the time series data of the fourth outlet temperature of the electric furnace by Δt, it is spatiotemporally matched with the time series data of the waste heat recovery inlet temperature to ensure that the causal relationship between the rise of the fourth outlet temperature of the electric furnace and the rise of the flue gas temperature transferred to the waste heat recovery inlet temperature is synchronized, thus avoiding control lag.

[0076] There is a physical distance between the fourth outlet of the electric furnace and the waste heat recovery inlet. The flue gas flows from the fourth outlet of the electric furnace to the waste heat recovery inlet at a flow rate v, which inevitably results in a transmission delay. If the real-time temperature data of the fourth outlet of the electric furnace and the real-time temperature data of the waste heat recovery inlet are used for control, the control action will be misaligned with the actual operating conditions. Asynchronous data will cause the control algorithm to oscillate frequently. Synchronized data makes the control more logical, reduces the drastic fluctuations in the fan speed, extends the equipment life, and reduces the contradiction between energy consumption and emissions.

[0077] In S2, the third specific step is to use the spatiotemporally corrected data of the fourth outlet temperature of the electric furnace and the waste heat recovery inlet temperature to predict the changes in the waste heat recovery inlet temperature data.

[0078] In the processed multi-parameter data, the furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, and average current value are selected from the first N (N is an even number and N is greater than or equal to 2) time points. The electrode position change rate is selected from the first N / 2 time points. A network structure of 2 layers of long short-term memory network LSTM and 1 layer of fully connected layer is adopted, with 64 neurons in each layer. The predicted value of waste heat recovery inlet temperature for a future period is output. Normal operating condition data is obtained. The model is trained with mean absolute error as the loss function. The latest N time point electric furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, average current value, and N / 2 electrode position change rate are input to predict the change of waste heat recovery inlet temperature data.

[0079] S3: Based on the predicted changes in the waste heat recovery inlet temperature, control scenarios are divided, and the fan speed is dynamically adjusted according to the divided control scenarios and the PM2.5 concentration.

[0080] In S3, firstly, based on the predicted changes in the waste heat recovery inlet temperature data, control scenarios are divided. According to the predicted changes in the waste heat recovery inlet temperature data, and based on the range of changes in the waste heat recovery inlet temperature data, control scenarios are divided into waste heat recovery inlet temperature data, and the fan is dynamically controlled in conjunction with PM2.5 concentration data.

[0081] In the first control scenario, when the outlet temperature of the fourth hole of the electric furnace is lower than the low-temperature section threshold of the electric furnace smelting process and the predicted change in the waste heat recovery inlet temperature is less than the low-temperature section recovery threshold, the frequency converter of the fan is controlled by the PID controller to gradually reduce the wind speed of the primary dust removal fan and the secondary dust removal fan to the energy-saving mode and maintain the minimum safe wind speed.

[0082] The low-temperature threshold and the low-temperature recovery threshold mentioned above are reference values ​​set by technical personnel in this industry based on production experience. The low-temperature threshold can be set to 300°C and the low-temperature recovery threshold can be set to 200°C. It should be noted that due to the layout of the pipeline, heat loss will occur during the transmission of flue gas, and the longer the pipeline, the higher the heat loss. Therefore, the low-temperature threshold is greater than the low-temperature recovery threshold. In this invention, the low-temperature threshold and the low-temperature recovery threshold are not limited and can be set according to the actual production situation.

[0083] In the second control scenario, when the waste heat recovery inlet temperature is within the optimal temperature range for waste heat recovery, the PID controller controls the fan's frequency converter to adjust the fan speed, switching the fan speed to energy efficiency mode. The fan speed is adjusted according to V. 机 =k·T2+b gradually increases the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient and b≈30 is the base speed drift;

[0084] At this time, PM2.5 concentration data is obtained through PM2.5 monitoring points. When the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to speed up to the rated maximum speed. Activate dust suppression mode;

[0085] When the PM2.5 concentration does not exceed the standard, the fan speed is set according to V. 机 =k·T2+b Increase the fan speed, prioritizing energy saving;

[0086] It should be noted that in this invention, the optimal temperature range for waste heat recovery can be set according to the actual situation. This invention does not limit the optimal temperature range for waste heat recovery. For example, the optimal temperature range for waste heat recovery can be set to 400°C to 800°C.

[0087] In scenario three, when the waste heat recovery inlet temperature rises above the system's safe temperature, the fan is forced to slow down to the minimum safe speed, and a high-temperature alarm is issued. Once the temperature returns to the optimal waste heat recovery temperature range, the PID controller controls the fan's inverter to switch the fan speed to energy efficiency mode, with the fan speed following the V... 机 =k·T2+b Gradually increase the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient, b≈30 is the base speed drift, PM2.5 concentration data is obtained through PM2.5 monitoring points, and when the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to speed up to the rated speed or the set maximum speed. Activate dust suppression mode;

[0088] S4: Optimize the fan speed again after dynamic adjustment based on the control scenario;

[0089] In S4, the first specific step is to obtain the flue gas volume flow rate transmission delay. When the waste heat recovery inlet temperature is within the optimal temperature range for waste heat recovery, the flue gas volume flow rate transmission delay is recalculated.

[0090] The waste heat recovery inlet temperature is predicted by an LSTM network after the electric furnace fourth hole outlet temperature is spatiotemporally corrected. Since the flue gas volume flow rate increases rapidly, the flue gas volume flow rate transmission delay decreases. Therefore, the spatiotemporal correction of the electric furnace fourth hole outlet temperature and the waste heat recovery inlet temperature needs to be adjusted in real time.

[0091] Specifically, the real-time outlet temperature of the fourth hole of the electric furnace in group Y and the inlet temperature of the waste heat recovery are obtained. The cross-correlation function between the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is calculated. The peak value τ is the actual transmission delay. The actual transmission delay is updated to the parameters of the Long Short-Term Memory (LSTM) network to predict the accurate waste heat recovery inlet temperature. Based on the predicted accurate waste heat recovery inlet temperature, the fan speed is optimized again.

[0092] Fast Fourier Transform is used to accelerate cross-correlation calculation, reduce time complexity, and meet real-time requirements;

[0093] Example 2

[0094] like Figure 2 As shown in Example 1, the present invention provides a dynamic intelligent speed control system for dust removal fans in electric arc furnace steelmaking, comprising:

[0095] Data acquisition module: Deploys sensors to collect multi-parameter data;

[0096] The specific process for deploying the sensors is as follows:

[0097] A temperature sensor is installed at the fourth hole of the electric furnace to collect temperature information inside the furnace. A temperature sensor is also installed at the boiler waste heat recovery inlet to reflect the recoverable temperature. A flue gas flow meter is installed at the boiler waste heat recovery inlet to record the actual flue gas volume of the waste heat recovery system. PM2.5 monitors are deployed on the top and outside of the plant to record PM2.5 concentration data.

[0098] The specific multi-parameter data is as follows:

[0099] Temperature data recorded by the temperature sensor installed at the fourth hole of the electric furnace is recorded as the outlet temperature of the fourth hole of the electric furnace; temperature data recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery is recorded as the inlet temperature of the waste heat recovery; flue gas volume flow rate data recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery; current and voltage data and electrode position change rate data obtained from the electric furnace control system via the industrial bus; and PM2.5 concentration data recorded by the PM2.5 monitor.

[0100] Data spatiotemporal correction module: Uses transmission delay to perform spatiotemporal correction on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data, and uses the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the data to predict the change of the inlet temperature of the waste heat recovery in the data.

[0101] The specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in multi-parameter data includes:

[0102] Process the collected multi-parameter data and remove abnormal data;

[0103] The temperature at the outlet of the fourth hole of the electric furnace, recorded by the temperature sensor installed at the fourth hole, is denoted as T1. The temperature at the inlet of the waste heat recovery system, recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery system, is denoted as T2. The flue gas volumetric flow rate recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery system is denoted as V. 余 The current, voltage, and electrode position data obtained from the electric furnace control system via the industrial bus are denoted as I. 电 U 电 R 电 The monitoring data recorded by the PM2.5 monitor is denoted as ug;

[0104] For any parameter including: T1, T2, V 余 I 电 U 电 R 电 ,ug; calculate the mean μ and standard deviation σ of the parameter in real time over the past A minutes; compare the newly collected parameter with the mean μ and standard deviation σ of the parameter to determine whether the newly collected parameter is abnormal data;

[0105] If the newly collected data X satisfies |X-μ|>3σ, then the new data is determined to be abnormal data, and is removed and replaced with the valid value from the previous time step.

[0106] If the newly collected data X satisfies |X-μ|<3σ, then the new data is determined to be normal data.

[0107] The specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in the multi-parameter data also includes:

[0108] Spatiotemporal correction was performed on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data after removing abnormal data. The transmission delay of the flue gas volume flow rate was calculated based on the outlet temperature T1 of the fourth hole of the electric furnace recorded by the temperature sensor installed in the fourth hole of the electric furnace, and spatiotemporal correction was performed based on the transmission delay of the flue gas volume flow rate.

[0109] Obtain the cross-sectional area S and length L of the flue, and then calculate the flue gas volumetric flow rate V. 余 First, calculate the transmission delay of the flue gas volume flow rate, and then perform spatiotemporal correction based on the transmission delay of the flue gas volume flow rate.

[0110] Calculate the flue gas volumetric flow rate per second: Among them, V 余 Let be the flue gas volumetric flow rate, then the flue gas volumetric flow rate transmission delay is:

[0111] The time series data of the fourth outlet temperature of the electric furnace is delayed by Δt and then spatiotemporally matched with the time series data of the waste heat recovery inlet temperature to ensure that the causal relationship between the rise of the fourth outlet temperature of the electric furnace and the rise of the flue gas temperature at the waste heat recovery inlet is synchronized.

[0112] The specific process of predicting the change in the waste heat recovery inlet temperature using the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is as follows:

[0113] The system selects the furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, and average current from the previous N time points, and the electrode position change rate from the previous N / 2 time points. It adopts a network structure with two layers of Long Short-Term Memory (LSTM) network and one fully connected layer, with 64 neurons in each layer, and outputs the predicted value of waste heat recovery inlet temperature for a future period. It obtains normal operating condition data, trains the model with mean absolute error as the loss function, and inputs the latest N time points of furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, average current, and N / 2 electrode position change rate to predict the change of waste heat recovery inlet temperature data.

[0114] Fan speed control module: Based on the predicted changes in waste heat recovery inlet temperature, control scenarios are divided, and the fan speed is dynamically controlled according to the divided control scenarios and PM2.5 concentration.

[0115] The specific process of dynamically regulating the fan speed includes:

[0116] When the outlet temperature of the fourth hole of the electric furnace is lower than the low-temperature section threshold of the electric furnace smelting process and the predicted change in the waste heat recovery inlet temperature is less than the low-temperature section recovery threshold, the frequency converter of the fan is controlled by the PID controller to gradually reduce the wind speed of the primary dust removal fan and the secondary dust removal fan to the energy-saving mode and maintain the minimum safe wind speed.

[0117] The specific process of dynamically controlling the fan speed also includes:

[0118] When the waste heat recovery inlet temperature is within the optimal temperature range for waste heat recovery, the frequency converter of the fan is controlled by the PID controller to adjust the fan speed, switching the fan speed to energy efficiency mode. The fan speed is adjusted according to V. 机 =k·T2+b gradually increases the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient and b≈30 is the base speed drift;

[0119] PM2.5 concentration data is obtained through PM2.5 monitoring stations. When the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to accelerate to the rated speed or the maximum speed set at the specified point. Activate dust suppression mode;

[0120] When the PM2.5 concentration does not exceed the standard, the fan speed is set according to V. 机 =k·T2+b gradually increases the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient and b≈30 is the base speed drift.

[0121] The specific process of dynamically controlling the fan speed also includes:

[0122] When the waste heat recovery inlet temperature rises above the system's safe temperature, the fan is forced to slow down to the minimum safe speed, and an over-temperature alarm is issued. Once the temperature returns to the optimal waste heat recovery temperature range, the PID controller controls the fan's frequency converter to switch the fan speed to energy efficiency mode, with the fan speed following the V... 机 =k·T2+b Gradually increase the fan speed, where k≈4 is the temperature fan speed sensitivity coefficient, b≈30 is the base speed drift, PM2.5 concentration data is obtained through PM2.5 monitoring points, and when the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to speed up to the rated speed or the set maximum speed. Activate dust suppression mode.

[0123] Fan speed optimization module: Based on the control scenario, the fan speed after dynamic control is further optimized.

[0124] The specific process for further optimizing the dynamically adjusted fan speed is as follows:

[0125] The real-time outlet temperature of the fourth hole of the electric furnace in group Y and the inlet temperature of the waste heat recovery are obtained. The cross-correlation function between the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is calculated. The peak value τ is the actual transmission delay. The actual transmission delay is updated to the LSTM network parameters to predict the accurate waste heat recovery inlet temperature. Based on the predicted accurate waste heat recovery inlet temperature, the fan speed is optimized again.

[0126] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking, characterized in that: include: S1: Deploy sensors to collect multi-parameter data; S2: Use transmission delay to perform spatiotemporal correction on the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the multi-parameter data, and use the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery in the data to predict the change of the inlet temperature of the waste heat recovery in the data. The specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in the multi-parameter data also includes: The collected multi-parameter data were processed to remove outliers. Spatiotemporal correction was then performed on the outlier-removed data for the electric furnace fourth orifice outlet temperature and the waste heat recovery inlet temperature. The electric furnace fourth orifice outlet temperature was then recorded by the temperature sensor installed at the fourth orifice. Using the standard, the transmission delay of flue gas volume flow rate is calculated, and spatiotemporal correction is performed based on the transmission delay of flue gas volume flow rate. Obtain the cross-sectional area of ​​the flue flue length According to flue gas volume flow rate First, calculate the transmission delay of the flue gas volume flow rate, and then perform spatiotemporal correction based on the transmission delay of the flue gas volume flow rate. Calculate the flue gas volumetric flow rate per second: ,in, Let be the flue gas volumetric flow rate, then the flue gas volumetric flow rate transmission delay is: ; Delay the time series data of the fourth outlet temperature of the electric furnace Then, spatiotemporal matching was performed with the time series data of the waste heat recovery inlet temperature to ensure the synchronization of the causal relationship between the temperature rise of the fourth outlet of the electric furnace and the temperature rise of the flue gas transmitted to the waste heat recovery inlet. S3: Based on the predicted changes in the waste heat recovery inlet temperature, control scenarios are divided, and the fan speed is dynamically adjusted according to the divided control scenarios and the PM2.5 concentration. S4: Based on the control scenario, further optimize the fan speed after dynamic control.

2. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific process for deploying the sensors is as follows: A temperature sensor is installed at the fourth hole of the electric furnace to collect temperature information inside the furnace. A temperature sensor is also installed at the boiler waste heat recovery inlet to reflect the recoverable temperature. A flue gas flow meter is installed at the boiler waste heat recovery inlet to record the actual flue gas volume of the waste heat recovery system. PM2.5 monitors are deployed on the top and outside of the plant to record PM2.5 concentration data.

3. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific multi-parameter data is as follows: Temperature data recorded by the temperature sensor installed at the fourth hole of the electric furnace is recorded as the outlet temperature of the fourth hole of the electric furnace; temperature data recorded by the temperature sensor installed at the inlet of the boiler waste heat recovery is recorded as the inlet temperature of the waste heat recovery; flue gas volume flow rate data recorded by the flue gas flow meter installed at the inlet of the boiler waste heat recovery; current and voltage data and electrode position change rate data obtained from the electric furnace control system via the industrial bus; and PM2.5 concentration data recorded by the PM2.5 monitor.

4. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific process of using transmission delay to perform spatiotemporal correction on the electric furnace fourth orifice outlet temperature and waste heat recovery inlet temperature in multi-parameter data includes: Process the collected multi-parameter data and remove abnormal data; The outlet temperature of the fourth hole of the electric furnace, recorded by the temperature sensor installed at the fourth hole, is denoted as [missing information]. The temperature at the waste heat recovery inlet, recorded by the temperature sensor installed at the boiler waste heat recovery inlet, is denoted as... The flue gas volumetric flow rate data recorded by the flue gas flow meter installed at the boiler waste heat recovery inlet is denoted as... The current, voltage, and electrode position data obtained from the electric furnace control system via the industrial bus are respectively denoted as... , , The monitoring data recorded by the PM2.5 monitor is denoted as ; For any parameter, including: , , , , , , Real-time calculation of the average value of parameters over the past A minutes. and standard deviation ; Compare the newly collected parameters with the mean value corresponding to those parameters. and standard deviation Compare the data to determine whether the newly acquired parameters are abnormal. If newly collected data satisfy If the new data is deemed abnormal, it will be removed and replaced with a valid value from the previous time step. If newly collected data satisfy If so, the new data is determined to be normal data.

5. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific process of predicting the change in the waste heat recovery inlet temperature using the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is as follows: The system selects the furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, and average current from the previous N time points, and the electrode position change rate from the previous N / 2 time points. It adopts a network structure with two layers of Long Short-Term Memory (LSTM) network and one fully connected layer, with 64 neurons in each layer, and outputs the predicted value of waste heat recovery inlet temperature for a future period. It obtains normal operating condition data, trains the model with mean absolute error as the loss function, and inputs the latest N time points of furnace fourth hole outlet temperature data, waste heat recovery inlet temperature data, average current, and N / 2 electrode position change rate to predict the change of waste heat recovery inlet temperature data.

6. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific process of dynamically regulating the fan speed includes: When the outlet temperature of the fourth hole of the electric furnace is lower than the low-temperature section threshold of the electric furnace smelting process and the predicted change in the waste heat recovery inlet temperature is less than the low-temperature section recovery threshold, the frequency converter of the fan is controlled by the PID controller to gradually reduce the wind speed of the primary dust removal fan and the secondary dust removal fan to the energy-saving mode and maintain the minimum safe wind speed.

7. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 4, characterized in that: The specific process of dynamically controlling the fan speed also includes: When the waste heat recovery inlet temperature is within the optimal temperature range for waste heat recovery, the frequency converter of the fan is controlled by the PID controller to adjust the fan speed, switching the fan speed to energy efficiency mode. The fan speed is adjusted according to... Gradually increase the fan speed, among which, This is the temperature fan speed sensitivity coefficient. The base speed drift; PM2.5 concentration data is obtained through PM2.5 monitoring stations. When the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to accelerate to the rated speed or the maximum speed set at the specified point. Execute dust suppression mode; When the PM2.5 concentration does not exceed the standard, the fan speed is as follows: Gradually increase the fan speed, among which, This is the temperature fan speed sensitivity coefficient. This is the base speed drift.

8. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 4, characterized in that: The specific process of dynamically controlling the fan speed also includes: When the waste heat recovery inlet temperature rises above the system's safe temperature, the fan is forced to slow down to the minimum safe speed, and an over-temperature alarm is issued. Once the temperature returns to the optimal waste heat recovery temperature range, the PID controller controls the fan's frequency converter to switch the fan speed to energy efficiency mode, with the fan speed adjusted according to... Gradually increase the fan speed, among which, This is the temperature fan speed sensitivity coefficient. Based on the base speed drift, PM2.5 concentration data is obtained from PM2.5 monitoring points. When the PM2.5 concentration exceeds the standard, the secondary dust removal fan is forced to accelerate to the rated speed or the maximum set speed. Execute dust suppression mode.

9. The dynamic intelligent speed control method for dust removal fans in electric arc furnace steelmaking according to claim 1, characterized in that: The specific process for further optimizing the dynamically adjusted fan speed is as follows: The real-time outlet temperature of the fourth hole of the electric furnace in group Y and the inlet temperature of the waste heat recovery are obtained. The cross-correlation function between the outlet temperature of the fourth hole of the electric furnace and the inlet temperature of the waste heat recovery is calculated. The peak value τ is the actual transmission delay. The actual transmission delay is updated to the parameters of the Long Short-Term Memory (LSTM) network to predict the accurate waste heat recovery inlet temperature. Based on the predicted accurate waste heat recovery inlet temperature, the fan speed is optimized again.

Images