A method for controlling pulverized coal economic fineness of a coal mill

Abstract

The application discloses a coal mill pulverized coal economic fineness control method, which is applied to a shell. The bottom sides of the shell are respectively provided with an electric motor and a stone coal box. The inside of the shell comprises a center cylinder, a cyclone separator, a deflection plate, a mill stand, a speed reducer, a scraper, a grinding disc, a grinding roller, a fixing frame, a hydraulic pull rod and a driver. The electric motor is directly connected with the speed reducer and is used for driving the equipment to run. In the application, eight coupling working condition parameters of the coal mill and a boiler side are synchronously collected, a double-layer constraint economic fineness optimization model is constructed, the optimal economic fineness is solved with the lowest comprehensive energy consumption, the rotating speed of the deflection plate of the separator is adaptively adjusted by using a fuzzy PID closed loop, feedforward compensation of coal quality and load disturbance and boundary interlocking protection are simultaneously increased, the pulverized coal fineness is controlled in multiple dimensions, the coal mill fineness can be matched with the current working condition to realize efficient combustion of the boiler, and the plugging failure caused by excessively fine pulverized coal can be avoided from the algorithm level. The whole process is automatically adjusted in a closed loop, and manual frequent intervention is not needed.

Description

Technical Field

[0001] This invention relates to the technical field of dynamic separator control in coal mills, specifically to a method for controlling the economic fineness of pulverized coal in a coal mill. Background Technology

[0002] Coal mills are core heavy-duty auxiliary machines in coal-fired power plants, metallurgical, cement, and chemical coal powder preparation systems. Their core function is to crush, grind, and dry large pieces of raw coal into fine coal powder with uniform size and low moisture content, which is then supplied to boilers and kilns for complete combustion. Only by grinding solid coal into ultrafine coal powder can the combustion speed and efficiency be significantly improved, and the pollutants such as fly ash and carbon monoxide caused by incomplete combustion be reduced. At the same time, the equipment has a built-in hot air system that can simultaneously dry high-moisture raw coal, ensuring stable coal powder flowability. It is one of the four key auxiliary machines in thermal power plants.

[0003] Currently, the dynamic separator in a power plant boiler's coal mill is a crucial means of regulating pulverized coal fineness, and its control directly impacts the safe and economical operation of the boiler unit. In most units, the dynamic separator in the coal mill is manually controlled by operators or controlled by a function based on the coal feed rate. This control method is relatively simple, often resulting in pulverized coal fineness that is either too coarse or too fine, leading to mill blockage and affecting the unit's economic efficiency. Therefore, there is an urgent need to design a method for controlling the economical fineness of pulverized coal in the coal mill to address these issues. Summary of the Invention

[0004] The purpose of this invention is to provide a method for controlling the economic fineness of pulverized coal in a coal mill, so as to overcome the above-mentioned shortcomings in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for controlling the economic fineness of pulverized coal in a coal mill is applied to the casing. An electric motor and a coal / stone box are respectively installed on both sides of the bottom of the casing. The interior of the casing includes a central cylinder, a cyclone separator, a deflector plate, a base, a reducer, scrapers, a grinding disc, grinding rollers, a fixed frame, a hydraulic tie rod, and a driver. The electric motor is directly connected to the reducer to drive the equipment. The method includes the following steps: S0, Operation mode switching judgment: Real-time reading of unit load commands, automatic switching to anti-clogging and mill-priority control sub-mode when the unit is in deep peak shaving and low load condition; S1. Real-time acquisition of multi-dimensional operating condition data: Synchronously acquire eight real-time operating condition parameters, including coal feed rate of coal mill, hydraulic loading pressure of grinding roller, rotation current of grinding disc, inlet and outlet air temperature of mill, primary air volume, heat loss of flue gas in boiler furnace, carbon content of fly ash, and slag discharge of stone coal box. S2. Establish a multi-constraint economic fineness target model: With the highest comprehensive thermal efficiency of the boiler as the optimization target, set two-layer constraints. The first layer of constraints is the upper limit threshold of fly ash carbon content and the upper limit threshold of stone coal discharge. The second layer of constraints is the coal mill motor current not exceeding the limit and the no-blockage risk range. Based on eight operating condition parameters, construct a multi-variable coupled mathematical model of coal powder economic fineness to solve the target value of the optimal economic coal powder fineness under the current operating conditions. S3. Dynamic separator speed closed-loop adaptive adjustment: The difference between the target value of the optimal economic coal powder fineness and the measured coal powder fineness corresponding to the real-time speed of the cyclone separator deflector plate is calculated, and the fuzzy PID control algorithm is used to output the adjustment amount to continuously and dynamically adjust the drive speed of the cyclone separator deflector plate. S4. Online correction and compensation for operating condition fluctuations: Real-time monitoring of coal mill feed rate, raw coal moisture, and raw coal hardness disturbance variables. When any disturbance variable exceeds the preset steady-state range, a feedforward compensation coefficient is introduced to correct the output value of the economic fineness target model and simultaneously fine-tunes the speed of the cyclone separator deflector plate to avoid instantaneous excessive fineness of coal powder causing coal accumulation and mill blockage, or instantaneous excessive coarseness causing excessive carbon content in fly ash. S5. Boundary safety interlock protection: If the coal powder is still too coarse when the deflector plate is adjusted to the upper limit of the speed, or if the coal powder is still too fine when the deflector plate is adjusted to the lower limit of the speed, the coal mill load linkage adjustment logic will be triggered to simultaneously finely adjust the primary air volume and the hydraulic loading pressure of the grinding roller to bring the coal powder fineness back to the economic range, while avoiding mill blockage and overheating faults. S6. Full-condition data storage and model self-updating: Continuously stores the optimal economic fineness, separator speed, and boiler thermal efficiency operation datasets under various loads and coal types.

[0006] By adopting the above-mentioned invention, this solution simultaneously collects eight coupled operating parameters from the coal mill and boiler sides, constructs a dual-layer constrained economic fineness optimization model, and solves for the optimal economic fineness with the lowest comprehensive energy consumption. It adopts fuzzy PID closed-loop adaptive adjustment of the separator deflector speed, and adds coal quality, load disturbance feedforward compensation and boundary interlock protection. Multi-dimensional collaborative control of coal powder fineness can not only ensure that the coal powder fineness matches the current operating conditions to achieve efficient boiler combustion, but also avoid mill blockage caused by excessively fine coal powder from the algorithm level. The whole process is automatic closed-loop adjustment without frequent manual intervention, which significantly improves the safety and economic benefits of long-term operation of thermal power units.

[0007] Preferably, in step S1, the acquisition cycle of the eight real-time operating parameters is uniformly set to around 200ms. The parameters are acquired synchronously and in parallel through the driver, motor, and stone and coal box on the outside of the coal mill casing. The acquired data is uploaded to the unit controller in real time to complete timing alignment and outlier removal preprocessing.

[0008] By adopting the above-mentioned invention, all operating parameters are collected in parallel at a high frequency of 200ms, the timing is synchronized and abnormal values ​​of sensor jumps are eliminated, ensuring that the data input to the model is real-time and accurate, avoiding lag and misadjustment of fine-grained adjustment caused by lagging and distorted data, and improving the response speed and adjustment accuracy of closed-loop control.

[0009] Preferably, in step S2, the multi-constraint economic fineness target model is constructed by using the minimum sum of flue gas heat loss, mechanical incomplete combustion loss, and coal pulverizing power consumption corresponding to different coal fineness as the evaluation function, and using particle swarm optimization algorithm to iteratively solve for the optimal economic fineness target value.

[0010] By adopting the above-mentioned invention, the optimal economic fineness is solved iteratively using the particle swarm optimization algorithm. It comprehensively balances the three core losses: flue gas heat loss, incomplete combustion heat loss, and pulverizing power consumption. Unlike the traditional logic that only controls the coarseness of coal powder, it seeks optimization from the perspective of the overall unit energy consumption. The output fineness value is the economically optimal solution under all operating conditions, rather than simply satisfying a single safety threshold.

[0011] Preferably, the threshold values ​​of the double-layer constraint conditions in step S2 are calibrated and stored offline based on the type of coal used by the unit and the rated load of the boiler. When the coal type is switched, the corresponding threshold value group is automatically retrieved. The threshold values ​​include fly ash carbon content ≤3%, stone coal unit slag discharge ≤0.8t / h, mill motor current ≤90% of rated current, and mill inlet and outlet air temperature difference ≤45℃.

[0012] By adopting the above-mentioned invention, safety constraint thresholds are set in layers to distinguish the dual boundaries of combustion indicators and coal mill equipment operation indicators. At the same time, thresholds are stored offline in groups according to coal type. When the coal type is switched, the corresponding constraints are automatically matched to avoid the failure of a single fixed threshold after the coal quality changes. This balances the compliance of boiler combustion indicators with the coal mill equipment not exceeding the rated operating range, thus avoiding faults such as mill blockage, overflow, and excessive fly ash from the source.

[0013] Preferably, the fuzzy PID control algorithm in step S3 includes a fuzzification module, a fuzzy rule base, and a defuzzification module. The fuzzy input is the fineness deviation and the rate of change of the fineness deviation, and the fuzzy output is the speed adjustment increment of the deflector plate. The fuzzy rule base is established in segments according to different load ranges of the coal mill. In the low load range, the integral adjustment weight is amplified to prevent the coal powder from being too fine and clogging the mill. In the high load range, the proportional adjustment weight is amplified to quickly correct the coarse powder deviation.

[0014] By adopting the above-mentioned invention, the segmented fuzzy PID rule base distinguishes between high and low load conditions, amplifies the integral weight under low load, and suppresses excessive fine coal powder accumulation under low air volume.

[0015] Preferably, in step S4, the feedforward compensation coefficient of the disturbance variable is calibrated through an offline coal quality disturbance test. When the moisture content and hardness of the raw coal increase, the feedforward compensation coefficient is positively corrected to appropriately amplify the economic fineness target value and prevent excessively fine coal powder from accumulating and clogging the mill. When the moisture content and hardness of the raw coal decrease, the feedforward compensation coefficient is negatively corrected to reduce the economic fineness target value and decrease the carbon content of fly ash.

[0016] By adopting the above-mentioned invention, feedforward compensation coefficients are set for the two core disturbances of raw coal moisture and hardness. When the coal quality deteriorates and coal is prone to accumulation, the fineness target is actively relaxed, and when the coal quality is excellent, the fineness target is tightened. This achieves early compensation for disturbances, eliminating the need to wait for the fineness of coal powder to exceed the standard before making a delayed adjustment. This significantly reduces the large deviation in fineness caused by coal quality fluctuations and improves control stability.

[0017] Preferably, the boundary safety interlock protection in step S5 specifically includes: when the deflector plate speed reaches the lower limit and the mill internal pressure difference exceeds the mill blockage warning value for 30 consecutive seconds, the interlock increases the primary air volume, reduces the hydraulic loading pressure of the grinding roller, and simultaneously raises the economic fineness target value; when the deflector plate speed reaches the upper limit and the fly ash carbon content exceeds the standard for 30 consecutive seconds, the interlock increases the grinding roller loading pressure, reduces the primary air volume, and simultaneously lowers the economic fineness target value.

[0018] By adopting the above-mentioned invention, the upper and lower limit boundary interlocking logic of the separator speed is set. When the adjustment mechanism reaches the limit and still cannot meet the fineness requirements, the primary air volume and the grinding roller loading pressure are coordinated and adjusted. Multiple actuators work together to correct the fineness of coal powder, breaking the control limitation of relying solely on the speed adjustment of the separator, expanding the fineness adjustment range, and avoiding the equipment from going out of control after reaching the adjustment limit.

[0019] Preferably, in step S6, an offline particle swarm iteration is automatically initiated at regular intervals to update the internal coupling coefficients of the multi-constraint economic fineness target model, adapting to long-term coal quality fluctuations.

[0020] By adopting the above-mentioned invention, the full-condition operation data is automatically stored and the model coupling coefficient is periodically updated offline. When the equipment is running for a long time and the coal quality fluctuates year by year, the model can automatically iterate and adapt without the need for manual remodeling and calibration.

[0021] Preferably, the cyclone separator is installed on the outside of the central cylinder at the top of the shell, and the deflector plates are evenly arranged around the circumference of the cyclone separator. The deflector plates are independently driven by the top driver, and the speed adjustment range is 0 to 1200 r / min.

[0022] By adopting the above-mentioned invention, the circumferentially evenly arranged deflector plates are speed-regulated by an independent driver with a wide adjustment range. This allows for continuous and smooth changes in the rotation intensity of the airflow within the separator, precisely controlling the coal powder separation and grading effect. The mechanical drive structure has a fast response, and combined with closed-loop control, it can achieve fine adjustment of fineness.

[0023] Preferably, in step S0, the unit switches to the high-efficiency combustion priority control sub-mode under rated high load conditions, tightening the fly ash carbon content constraint threshold.

[0024] By adopting the above-mentioned invention, two control sub-modes are automatically switched according to the unit load. Deep peak shaving prioritizes low load to prevent wear and tear, while rated high load prioritizes combustion efficiency. This adapts to two typical operating scenarios: deep peak shaving and full power generation, taking into account both the unit's peak shaving flexibility and power generation economy.

[0025] In the above technical solution, the present invention provides a method for controlling the economic fineness of pulverized coal in a coal mill, which has the following beneficial effects: This method for controlling the economic fineness of pulverized coal in a coal mill involves simultaneously collecting eight coupled operating parameters from both the coal mill and boiler sides. A dual-layer constraint economic fineness optimization model is constructed, using the lowest comprehensive energy consumption to determine the optimal economic fineness. Fuzzy PID closed-loop adaptive adjustment of the separator deflector speed is employed, along with feedforward compensation for coal quality and load disturbances, and boundary interlocking protection. This multi-dimensional collaborative control of pulverized coal fineness ensures that the fineness matches the current operating conditions for efficient boiler combustion, while also mitigating mill blockage caused by excessively fine pulverized coal at the algorithmic level. The entire process is automated with closed-loop control, requiring minimal manual intervention and significantly improving the long-term safety and economic benefits of thermal power units.

[0026] This method for controlling the economic fineness of pulverized coal in a coal mill uses a particle swarm optimization algorithm to iteratively solve for the optimal economic fineness. It comprehensively balances three core losses: flue gas heat loss, incomplete combustion heat loss, and pulverizing power consumption. Unlike traditional methods that only control the coarseness of pulverized coal, this method seeks optimization from the perspective of the overall unit energy consumption. The output fineness value is the economically optimal solution under all operating conditions, rather than simply satisfying a single safety threshold.

[0027] This method for controlling the economic fineness of coal powder in a coal mill sets feedforward compensation coefficients for the two core disturbances of raw coal moisture and hardness. When the coal quality deteriorates and coal is prone to accumulation, the fineness target is actively relaxed, and when the coal quality is excellent, the fineness target is tightened. This achieves early compensation for disturbances, eliminating the need to wait for the fineness of the coal powder to exceed the standard before making a delayed adjustment. This significantly reduces the large deviation in fineness caused by fluctuations in coal quality and improves control stability. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0029] Figure 1 This is a first three-dimensional schematic diagram of an embodiment of a method for controlling the economic fineness of coal powder in a coal mill according to the present invention.

[0030] Figure 2 This is a second schematic diagram of a three-dimensional structure provided for an embodiment of the method for controlling the economic fineness of coal powder in a coal mill according to the present invention.

[0031] Figure 3 This is a third schematic diagram of a three-dimensional structure provided for an embodiment of the method for controlling the economic fineness of coal powder in a coal mill according to the present invention.

[0032] Figure 4 This is a partial cross-sectional view of a three-dimensional structure provided in an embodiment of the method for controlling the economic fineness of coal powder in a coal mill according to the present invention.

[0033] Figure 5 This is a schematic diagram of the working steps provided in an embodiment of the method for controlling the economic fineness of coal powder in a coal mill according to the present invention.

[0034] Explanation of reference numerals in the attached figures: 1. Shell; 2. Central cylinder; 3. Cyclone separator; 4. Deflector plate; 5. Base; 6. Reducer; 7. Scraper; 8. Grinding disc; 9. Grinding roller; 10. Fixing frame; 11. Hydraulic tie rod; 12. Driver; 13. Electric motor; 14. Stone and coal box. Detailed Implementation

[0035] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0036] like Figure 1-5 As shown in the figure, the embodiment of the present invention provides a method for controlling the economic fineness of coal powder in a coal mill, which is applied to the shell 1. The bottom two sides of the shell 1 are respectively provided with a motor 13 and a stone coal box 14. The shell 1 contains a central cylinder 2, a cyclone separator 3, a deflector plate 4, a base 5, a reducer 6, a scraper 7, a grinding disc 8, a grinding roller 9, a fixed frame 10, a hydraulic tie rod 11, and a driver 12. The motor 13 is directly connected to the reducer 6 and is used to drive the operation of the equipment.

[0037] Specifically, in this embodiment, in step S1, the acquisition period of the eight real-time operating condition parameters is uniformly set to about 200ms, and the data are synchronously and parallelly collected by the sensing modules matching the driver 12, the motor 13 and the stone coal box 14 on the outer side of the coal mill shell 1. The collected data are uploaded to the unit controller in real time to complete the preprocessing of time sequence alignment and outlier elimination. In step S2, the multi-constraint economic fineness target model is constructed by taking the minimum sum of exhaust heat loss, mechanical incomplete combustion loss and coal milling power consumption corresponding to different pulverized coal fineness as the evaluation function, and using the particle swarm optimization algorithm to iteratively solve for the optimal economic fineness target value. The thresholds of the double-layer constraints in step S2 are calibrated and stored off-line according to the coal type burned by the unit and the rated load of the boiler. When the coal type is switched, the corresponding exclusive threshold group for the coal type is automatically retrieved. The thresholds include carbon content in fly ash ≤ 3%, unit slag discharge of stone coal ≤ 0.8t / h, coal mill motor current ≤ 90% of the rated current, and air temperature difference between the mill inlet and outlet ≤ 45°C. The fuzzy PID control algorithm in step S3 includes a fuzzification module, a fuzzy rule base and a defuzzification module. The fuzzy input quantities are the fineness deviation and the fineness deviation change rate, and the fuzzy output quantity is the rotation speed adjustment increment of the deflector plate 4. The fuzzy rule base is established in sections according to different load intervals of the coal mill: the integral adjustment weight is enlarged in the low load interval to prevent the pulverized coal from being too fine to cause mill blockage, and the proportional adjustment weight is enlarged in the high load interval to quickly correct the coarse pulverized coal deviation. In step S4, the feedforward compensation coefficient of the disturbance variable is calibrated through off-line coal quality disturbance tests. When the moisture of raw coal increases and the hardness of raw coal increases, the feedforward compensation coefficient is positively corrected, and the economic fineness target value is moderately enlarged to prevent the pulverized coal from being too fine to accumulate and block the mill; when the moisture of raw coal decreases and the hardness of raw coal decreases, the feedforward compensation coefficient is negatively corrected, and the economic fineness target value is reduced to lower the carbon content in fly ash. The boundary safety interlock protection in step S5 specifically includes: when the rotation speed of the deflector plate 4 reaches the lower limit and the differential pressure inside the mill continuously exceeds the mill blockage warning value for 30s, the interlock increases the primary air volume, reduces the hydraulic loading pressure of the grinding roller 9, and synchronously raises the economic fineness target value; when the rotation speed of the deflector plate 4 reaches the upper limit and the carbon content in fly ash continuously exceeds the standard for 30s, the interlock increases the loading pressure of the grinding roller 9, reduces the primary air volume, and synchronously lowers the economic fineness target value. In step S6, the off-line particle swarm iteration is automatically started at intervals to update the internal coupling coefficient of the multi-constraint economic fineness target model, so as to adapt to long-term coal quality fluctuation. The cyclone separator 3 is installed on the outer side of the central cylinder 2 at the top of the shell 1, the deflector plates 4 are evenly arranged along the circumferential direction of the cyclone separator 3, the deflector plates 4 are independently driven by the top driver 12, and the rotation speed adjustment range is 0 to 1200r / min. In step S0, the unit switches to the high-efficiency combustion priority control sub-mode under the rated high-load operating condition, and tightens the constraint threshold for carbon content in fly ash.

[0038] This invention provides a method for controlling the economic fineness of pulverized coal in a coal mill. This method is suitable for a vertical medium-speed coal mill system consisting of a shell 1, a top central cylinder 2, a cyclone separator 3, a deflector plate 4, a base 5, a reducer 6, a scraper 7, a grinding disc 8, a grinding roller 9, a fixed frame 10, a hydraulic tie rod 11, a driver 12, a motor 13, and a stone and coal box 14. It abandons the simple and crude single-variable control logic and constructs a multi-parameter coupled closed-loop control system. This method pre-sets the operating mode determination step S0 and automatically switches control according to the unit load command. Sub-mode: Deep peak shaving at low loads activates the anti-clogging mill priority strategy, while rated high loads activate the high-efficiency combustion priority strategy; Step S1: The sensor modules of the outer casing 1 (driver 12, motor 13, and stone and coal box 14) collect eight coupled operating condition parameters in parallel at a period of 200ms. After time alignment and outlier removal preprocessing, the parameters are sent to the controller; Step S2: With the optimization objective of minimizing exhaust heat loss, mechanical incomplete combustion loss, and total pulverizing power consumption, a particle swarm optimization algorithm is used to construct an economically detailed mathematical model with double-layer safety constraints, approximately... The threshold values ​​are stored offline in groups according to coal type and load, and the threshold can be automatically matched and adapted when switching coal types; Step S3 uses a piecewise fuzzy PID algorithm to perform closed-loop adaptive speed adjustment of the deflector plate 4 of the cyclone separator 3. The deflector plate 4 is evenly distributed around the circumference of the cyclone separator 3 and is independently driven by the driver 12, with a speed range of 0 to 1200 r / min. High and low loads are distinguished by weight to suppress overshoot; Step S4 introduces feedforward compensation coefficients for raw coal moisture and raw coal hardness to correct for operating disturbances in advance and prevent large instantaneous deviations in coal powder coarseness; Step S5 The boundary interlock protection logic is set up so that when the deflector plate 4 reaches the speed regulation limit and still cannot meet the standard, the primary air volume and the hydraulic loading pressure of the grinding roller 9 are coordinated and regulated to avoid mill blockage and excessive fly ash. Step S6 continuously stores the full-condition operation data and periodically updates the model coupling coefficient offline to realize the algorithm's long-term autonomous adaptation to coal quality fluctuations without continuous manual intervention. This not only stably maintains the economic fineness of coal powder and improves the economic efficiency of unit power generation, but also avoids two core problems from the control logic level: coal accumulation and mill blockage due to excessively fine coal powder and decreased combustion efficiency due to excessively coarse coal powder.

[0039] In another embodiment provided by the present invention, S0, operation mode switching determination: real-time reading of unit load command, and automatic switching to anti-clogging and mill-preferred control sub-mode when the unit is in deep peak shaving and low load condition; S1. Real-time acquisition of multi-dimensional working condition data: Synchronously acquire eight real-time working condition parameters, including coal feed rate of coal mill, hydraulic loading pressure of grinding roller 9, rotation current of grinding disc 8, inlet and outlet air temperature of mill, primary air volume, heat loss of boiler flue gas, carbon content of fly ash, and slag discharge of stone coal box 14. S2. Establish a multi-constraint economic fineness target model: With the highest comprehensive thermal efficiency of the boiler as the optimization target, set two-layer constraints. The first layer of constraints is the upper limit threshold of fly ash carbon content and the upper limit threshold of stone coal discharge. The second layer of constraints is the range where the pulverizer motor current does not exceed the limit and there is no risk of mill blockage. Based on eight operating condition parameters, construct a multi-variable coupled mathematical model of pulverized coal economic fineness to solve the target value of the optimal economic pulverized coal fineness under the current operating conditions. S3, Dynamic separator speed closed-loop adaptive adjustment: The difference between the target value of the optimal economic coal powder fineness and the measured coal powder fineness corresponding to the real-time speed of the deflector plate 4 of the cyclone separator 3 is calculated, and the fuzzy PID control algorithm is used to output the adjustment amount to continuously and dynamically adjust the driving speed of the deflector plate 4 of the cyclone separator 3. S4. Online correction and compensation for operating condition fluctuations: Real-time monitoring of coal mill feed rate, raw coal moisture, and raw coal hardness disturbance variables. When any disturbance variable exceeds the preset steady-state range, a feedforward compensation coefficient is introduced to correct the output value of the economic fineness target model and simultaneously fine-tunes the rotation speed of the cyclone separator 3 deflector 4 to avoid instantaneous excessive fineness of coal powder causing coal accumulation and mill blockage, or instantaneous excessive coarseness causing excessive carbon content in fly ash. S5. Boundary safety interlock protection: If the coal powder is still too coarse when the deflector plate 4 is adjusted to the upper limit of the speed, or if the coal powder is still too fine when the deflector plate 4 is adjusted to the lower limit of the speed, the coal mill load linkage adjustment logic will be triggered, and the primary air volume and the hydraulic loading pressure of the grinding roller 9 will be adjusted simultaneously to bring the coal powder fineness back to the economic range, while avoiding mill blockage and overheating faults. S6. Full-condition data storage and model self-updating: Continuously stores the optimal economic fineness, separator speed, and boiler thermal efficiency operation datasets under various loads and coal types.

[0040] Working principle: When the equipment is running, raw coal falls into the center of the grinding disc 8. The hydraulic tie rod 11 presses the grinding roller 9 to cooperate with the grinding disc 8 to grind the raw coal. The motor 13 drives the grinding disc 8 to rotate through the reducer 6. The scraper 7 lifts up the qualified coal powder. The hot air carries the coal powder upward through the top central cylinder 2 of the shell 1 and enters the cyclone separator 3. The deflector plate 4 rotates to form a rotating airflow to classify the coal powder. The coarse coal powder falls back to the grinding disc 8 for re-grinding, the fine coal powder is sent to the boiler for combustion, and the gravel coal falls into the bottom gravel coal box 14 for periodic discharge. After the control process starts, the first step, S0, is to determine the operating mode. The controller reads the unit load command. Under low-depth peak load, it automatically switches to the anti-clogging mill priority sub-mode and appropriately relaxes the upper limit of coal powder fineness. Under rated high load, it switches to the high-efficiency combustion priority sub-mode and tightens the fly ash carbon content constraint threshold. Then, in step S1, the sensors on the outer side of the housing 1 (driver 12, motor 13, and stone coal box 14) synchronously collect eight parameters at a 200ms cycle: coal feed rate, hydraulic loading pressure of grinding roller 9, current of grinding disc 8, air temperature, primary air volume, flue gas heat loss, fly ash carbon content, and slag discharge rate of stone coal box 14. This completes the timing alignment and jump variation. After constant value removal and data collection, step S2 is initiated. The controller retrieves the dual-layer constraint thresholds corresponding to the current coal type and load. The first layer of constraints controls the carbon content of fly ash and the amount of stone coal slag discharge to ensure combustion indicators. The second layer of constraints limits the current of motor 13 and the pressure difference inside the mill to avoid the risk of mill blockage. With the goal of minimizing the sum of the three types of losses, the optimal economic coal powder fineness target value for the current working condition is solved using the particle swarm optimization algorithm. In step S3, the target fineness is calculated as a deviation from the measured fineness corresponding to the real-time rotational speed of the deflector plate 4. The piecewise fuzzy PID algorithm outputs the speed adjustment increment based on the deviation and the rate of change of the deviation, continuously adjusting the output of the driver 12 and changing the cyclone separator 3. The rotation speed of the internal deflector plate 4 is used to correct the coal powder classification effect in real time. During operation, the disturbance compensation logic of step S4 is continuously executed, and the moisture and hardness of the raw coal are monitored in real time. Once the steady-state range is exceeded, the offline calibrated feedforward compensation coefficient is retrieved to correct the economic fineness target value, and the rotation speed of the deflector plate 4 is adjusted simultaneously to offset the fineness deviation caused by coal quality disturbance in advance. If the speed of the deflector plate 4 is adjusted to the limit range of 0-1200 r / min, and the fineness still exceeds the standard, the boundary safety interlock protection of step S5 is triggered: when the speed limit is lowered and the pressure difference exceeds the warning value, the primary air is increased, the hydraulic loading pressure of the grinding roller 9 is reduced, and the fineness target is increased to prevent grinding blockage; when the speed limit is uppered and the pressure difference exceeds the warning value, the primary air is increased, the hydraulic loading pressure of the grinding roller 9 is reduced, and the fineness target is increased to prevent grinding blockage; when the speed limit is higher and the pressure difference exceeds the warning value, the primary air is increased, the hydraulic loading pressure of the grinding roller 9 is reduced, and the fineness target is increased to prevent grinding blockage. When the carbon content in the ash exceeds the standard, the primary air is reduced, the loading pressure of the grinding roller 9 is increased, and the fineness target is lowered to reduce incomplete combustion losses. Simultaneously, step S6 is continuously executed to store the optimal economic fineness, deflector plate 4 speed, and boiler thermal efficiency datasets under various operating conditions in real time. Offline particle swarm iteration is initiated every fixed period to update the internal coupling coefficients of the multi-constraint economic fineness model, adapting to the changes in operating conditions caused by long-term coal quality and equipment wear. The entire process forms a fully automatic control closed loop of "data acquisition - model optimization - closed-loop speed regulation - disturbance compensation - safety interlock - model self-learning", which stably maintains the pulverized coal in the economic fineness range that balances boiler thermal efficiency and safe operation of the coal mill.

[0041] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for controlling the economic fineness of coal powder in a coal mill, applied to a shell (1), wherein a motor (13) and a stone coal box (14) are respectively arranged on both sides of the bottom of the shell (1), and the shell (1) contains a central cylinder (2), a cyclone separator (3), a deflector plate (4), a base (5), a reducer (6), a scraper (7), a grinding disc (8), a grinding roller (9), a fixed frame (10), a hydraulic tie rod (11), and a driver (12), wherein the motor (13) is directly connected to the reducer (6) for driving the equipment operation, characterized in that, Includes the following steps: S0, Operation mode switching judgment: Real-time reading of unit load commands, automatic switching to anti-clogging and mill-priority control sub-mode when the unit is in deep peak shaving and low load condition; S1. Real-time acquisition of multi-dimensional working condition data: synchronously acquire eight real-time working condition parameters, including coal feed rate of coal mill, hydraulic loading pressure of grinding roller (9), rotating current of grinding disc (8), inlet and outlet air temperature of mill, primary air volume, heat loss of boiler furnace flue gas, carbon content of fly ash, and slag discharge of stone coal box (14). S2. Establish a multi-constraint economic fineness target model: with the highest comprehensive thermal efficiency of the boiler as the optimization target, set a double-layer constraint condition. The first layer constraint is the upper limit threshold of fly ash carbon content and the upper limit threshold of stone coal discharge. The second layer constraint is the range where the current of the coal mill motor (13) does not exceed the limit and there is no risk of mill blockage. Based on the eight working condition parameters, construct a multi-variable coupled mathematical model of coal powder economic fineness and solve the target value of the optimal economic coal powder fineness under the current working condition. S3, Dynamic separator speed closed-loop adaptive adjustment: The difference between the target value of the optimal economic coal powder fineness and the measured coal powder fineness corresponding to the real-time speed of the deflector plate (4) of the cyclone separator (3) is calculated, and the fuzzy PID control algorithm is used to output the adjustment amount to continuously and dynamically adjust the driving speed of the deflector plate (4) of the cyclone separator (3). S4. Online correction and compensation for working condition fluctuations: Real-time monitoring of coal mill feed rate, raw coal moisture, and raw coal hardness disturbance variables. When any disturbance variable exceeds the preset steady state range, a feedforward compensation coefficient is introduced to correct the output value of the economic fineness target model and simultaneously fine-tunes the rotation speed of the cyclone separator (3) deflector plate (4) to avoid the instantaneous excessive fineness of coal powder causing coal accumulation and blockage in the mill, or the instantaneous excessive coarseness causing the fly ash carbon content to exceed the standard. S5, Boundary safety interlock protection: If the coal powder is still too coarse when the deflector plate (4) is adjusted to the upper limit of the speed, or if the coal powder is still too fine when the deflector plate (4) is adjusted to the lower limit of the speed, the coal mill load linkage adjustment logic will be triggered, and the primary air volume and the hydraulic loading pressure of the grinding roller (9) will be adjusted simultaneously to bring the coal powder fineness back to the economic range, while avoiding the failure of mill blockage and overheating. S6. Full-condition data storage and model self-updating: Continuously stores the optimal economic fineness, separator speed, and boiler thermal efficiency operation datasets under various loads and coal types.

2. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, In step S1, the acquisition cycle of the eight real-time operating parameters is uniformly set at around 200ms. The parameters are acquired synchronously and in parallel through the driver (12), motor (13), and stone coal box (14) on the outside of the coal mill shell (1). The acquired data is uploaded to the unit controller in real time to complete the timing alignment and outlier removal preprocessing.

3. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, In step S2, the multi-constraint economic fineness target model is constructed as follows: the evaluation function is the minimum sum of flue gas heat loss, mechanical incomplete combustion loss, and coal pulverizing power consumption corresponding to different coal fineness, and the optimal economic fineness target value is solved iteratively using the particle swarm optimization algorithm.

4. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, The threshold values ​​of the double-layer constraint conditions in step S2 are calibrated and stored offline based on the coal type used by the unit and the rated load of the boiler. When the coal type is switched, the corresponding exclusive threshold group for the coal type is automatically retrieved. The threshold values ​​include fly ash carbon content ≤3%, stone coal unit slag discharge ≤0.8t / h, mill motor current ≤90% of rated current, and mill inlet and outlet air temperature difference ≤45℃.

5. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, The fuzzy PID control algorithm in step S3 includes a fuzzification module, a fuzzy rule base, and a defuzzification module. The fuzzy input is the fineness deviation and the rate of change of fineness deviation. The fuzzy output is the speed adjustment increment of the deflector (4). The fuzzy rule base is established in segments according to different load ranges of the coal mill. In the low load range, the integral adjustment weight is amplified to prevent the coal powder from being too fine and clogging the mill. In the high load range, the proportional adjustment weight is amplified to quickly correct the coarse powder deviation.

6. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, In step S4, the feedforward compensation coefficient for the disturbance variable is calibrated through offline coal quality disturbance tests. When the moisture content and hardness of the raw coal increase, the feedforward compensation coefficient is positively corrected to appropriately amplify the economic fineness target value and prevent excessively fine coal powder from accumulating and clogging the mill. When the moisture content and hardness of the raw coal decrease, the feedforward compensation coefficient is negatively corrected to reduce the economic fineness target value and decrease the carbon content of fly ash.

7. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, The specific steps of the boundary safety interlock protection in step S5 include: when the speed of the deflector plate (4) reaches the lower limit and the internal pressure difference exceeds the mill blockage warning value for 30 consecutive seconds, the primary air volume is increased and the hydraulic loading pressure of the grinding roller (9) is reduced, and the economic fineness target value is adjusted upwards. When the speed of the deflector plate (4) reaches the upper limit and the carbon content of fly ash exceeds the standard for 30 consecutive seconds, the loading pressure of the grinding roller (9) is increased and the primary air volume is reduced, and the economic fineness target value is adjusted downwards.

8. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, In step S6, an offline particle swarm iteration is automatically initiated at regular intervals to update the internal coupling coefficients of the multi-constraint economic fineness target model and adapt to long-term coal quality fluctuations.

9. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, The cyclone separator (3) is installed on the outside of the top center cylinder (2) of the housing (1). The deflector plate (4) is evenly arranged around the cyclone separator (3). The deflector plate (4) is driven independently by the top driver (12). The speed adjustment range is 0 to 1200 r / min.

10. The method for controlling the economic fineness of pulverized coal in a coal mill according to claim 1, characterized in that, In step S0, the unit switches to the high-efficiency combustion priority control sub-mode under rated high load conditions, tightening the fly ash carbon content constraint threshold.

Images