Coal mill group control method and system adaptive to rapid load change

Abstract

The present application relates to the field of power plant boiler coal mill, provide a kind of coal mill group control method suitable for load rapid change, by obtaining the maximum output and economic output data of different coal used by each coal mill currently;The actual coal quantity of each coal mill under current load P, coal type and corresponding calorific value, load P' after T time are calculated and the judgment model of coal mill group start-stop control is established, and then the timing of controlling coal mill group start-stop is obtained;At the same time, the change of main indicators that have influence on boiler economy is quantified as the influence on unit power supply coal consumption, the calculation model of coal consumption change Δb is established, and the safety and environmental protection indicators are considered, the model is established by using machine learning algorithm, and the optimal calculation is carried out, to obtain the most economic coal quantity of each coal mill.The present application fully adapts to frequent and rapid change of load, improves production efficiency and equipment safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coal mill of power plant boiler, and particularly relates to a coal mill group control method suitable for rapid load change. BACKGROUND

[0002] At present, with the continuous advancement of China's power market reform, the power spot market is gradually opening up. It realizes the instant buying and selling of electricity through bilateral quotation or centralized bidding, has time sequence, flexibility and randomness, and can timely reflect the market supply and demand relationship and optimize the allocation of power resources. However, for the power generation side, it faces rapid fluctuations and unpredictability of load, which brings a series of problems to operation and equipment control. Among them, the coal mill of the coal pulverizing system of the coal-fired boiler faces rapid changes in load. The total coal quantity is positively correlated with the unit electric load, and each coal mill has upper and lower limits. Therefore, the operator needs to determine in real time whether to start a new coal mill or shut down a running coal mill. The start-up and shutdown procedures of each coal mill are relatively cumbersome, time-consuming, and require on-site inspection by the inspection personnel. Frequent start-stop brings great difficulties to the operator and the inspection personnel. At present, there is still no targeted technology to optimize the start-stop time of the coal mill under the condition of frequent load fluctuations and reduce the workload of the operator.

[0003] The Chinese invention patent application with the publication number CN112619871A "Coal mill start-stop automatic pre-selection module of thermal power unit" determines the coal mill start-stop object according to the analysis and decoupling of the boiler desuperheating water quantity, the furnace outlet flue gas temperature, and the furnace outlet nitrogen oxide content index, comprehensively considers the coal mill economy index and the operation time, and considers the safety, economy and environmental protection of the unit operation. The best combination mode of the coal mill start-stop is considered, but the start-stop time is not determined according to the rapid load change. The Chinese invention patent application with the publication number CN118002296A "Intelligent control method and device for coal mill suitable for deep load regulation of power station" combines the scheduling instructions of the coal-fired power unit, the coal type, the heat distribution of each heating surface of the boiler, and other boundary conditions, and incorporates intelligent operation analysis algorithms, and finally gives the start-stop equipment planning and start-stop time node of the coal pulverizing system, but does not consider the influence of the maximum output difference of each coal mill and the change of the coal type. SUMMARY

[0004] The technical problem to be solved by the present application is how to group control the coal mill group of the power plant boiler to obtain the start-stop time of the coal mill and the optimal coal quantity of each coal mill in order to adapt to the rapid load change.

[0005] The present application solves the above technical problems by the following technical means:

[0006] The application provides a coal mill group control method suitable for rapid load change, and the method comprises the following steps:

[0007] S1, obtaining maximum output and economic output data of each coal mill when different coal types are used;

[0008] S2, obtaining actual coal quantity, coal type and corresponding calorific value of each coal mill under current load P and load P' after T moment;

[0009] S3, calculating and establishing a judgment model for coal mill group start-stop control based on parameters obtained in steps S1 and S2;

[0010] S4, obtaining a time for controlling the start-stop of the coal mill group based on the calculation model for judging the start-stop control of the coal mill group obtained in step S3;

[0011] S5, quantifying changes of main indexes that have influence on the economy of the boiler into influences on the coal consumption of the unit, and establishing a calculation model for coal consumption change Δb;

[0012] S6, establishing a model by using a machine learning algorithm based on the calculation model for coal consumption change Δb obtained in step S5 and an energy balance formula, and taking into account safety and environmental protection indexes, and performing optimization calculation to obtain the most economic coal quantity of each coal mill.

[0013] Further, the maximum output and economic output data in step S1 have two obtaining modes, and specifically are as follows:

[0014] (1) periodically obtaining the maximum output and economic output data of each coal mill through performance test;

[0015] (2) obtaining historical data of each coal mill, at least including coal mill differential pressure, coal mill outlet air temperature, coal mill current, coal mill vibration parameter, coal mill power consumption, primary air pressure, primary air fan power consumption; and obtaining the maximum output and economic output data of each coal mill by using a machine learning algorithm model training and optimization.

[0016] The step S3 comprises the following steps:

[0017] S31, calculating the sum Q of heat input into the boiler by all running coal mills under current load P, the sum Q of heat input into the boiler by all running coal mills under maximum output, and the sum Q of heat input into the boiler by all running coal mills under economic output; now max eco

[0018] S32, calculating the difference ΔQ between the sum of input heat under current load P and the sum of input heat under load P' at T moment; ​​​

[0019] S33, judge the size relationship of P and P', if P'>P, ΔQ=P'-P, then execute step S34; if P'<P, ΔQ=P-P', then execute step S35;

[0020] S34, judge the size relationship of ΔQ and Q max -Q now , if ΔQ≤Q max -Q now , it represents that the load rising can be met by increasing the coal quantity, and the new coal mill does not need to be started; then skip step S4, and execute step S5; if ΔQ>Q max -Q now , it represents that the number of the coal mills currently running cannot meet the load rising, and the new coal mill needs to be started, then continue to execute step S4;

[0021] S35, judge the size relationship of ΔQ and Q now -Q eco , if ΔQ≤Q now -Q eco , it represents that the load descending can be met by reducing the coal quantity, and the coal mill can be kept in the economic output state, and the existing coal mill does not need to be stopped; then skip step S4, and execute step S5; if ΔQ>Q now -Q eco , it represents that part of the coal mills needs to be reduced to below the economic output state by reducing the coal quantity, which leads to the decrease of the stability of the coal mill, and the economy and safety of the unit are insufficient, and one coal mill needs to be stopped, then continue to execute step S4.

[0022] The step S4 is specifically:

[0023] When P'>P, the time point T 开启 of starting a new coal mill is T-T 暖磨 -T 启动 -T 升出力 ;

[0024] When P'<P, the time point T 关闭 of closing an existing coal mill is T-T 停运 -T 降出力 ;

[0025] Wherein, T 暖磨 is the time required for warming up the coal mill, T 启动 is the time required for starting the coal mill, T 升出力 is the time required for increasing the output of the coal mill, T 停运 is the time required for stopping the coal mill, and T 降出力 is the time required for reducing the output of the coal mill.

[0026] The step S5 is specifically:

[0027] The different increase or decrease of the coal quantity of each coal mill has different influences on the wall temperature of the boiler water cooling wall, the wall temperature of the superheater, the superheated steam temperature and pressure, the reheated steam temperature and pressure, the desuperheating water quantity, the nitrogen oxide emission quantity and the exhaust gas temperature, and further influences the boiler economy, safety and environmental protection indexes; the main indexes which have influences on the boiler economy include the superheated steam temperature t 过热 , the reheated steam temperature t 再热 , the superheated steam pressure p 过热 , the reheated steam pressure p 再热 , the superheated desuperheating water quantity m 过热 , the reheated desuperheating water quantity m 再热 , the exhaust gas temperature t 排烟 , and when the coal output of the coal mill is increased or decreased, the change amounts of the above parameters caused by the increase or decrease are Δt 过热 , Δt 再热 , Δp 过热 , Δp 再热 , Δm 过热 , Δm 再热 , Δt 排烟 .

[0028] The change of the unit parameter change amount of the above parameters on the unit causes the change of the unit economy to be quantified as the influence on the unit power supply coal consumption b, and the numerical values of the unit parameter change amount of the above parameters on the unit power supply coal consumption b are obtained by inquiring the experience values of the unit parameter change amount of the above parameters on the unit power supply coal consumption b of the different grade coal-fired units, which are b1, b2, b3, b4, b5, b6 and b7; b1-b7 are positive or negative values according to the parameter characteristics, and the change of the coal consumption caused by the overall parameter change after the increase or decrease of the coal quantity is Δb, and the calculation model is as follows:

[0029] Δb=Δt 过热 ·b1+Δt 再热 ·b2+Δp 过热 ·b3+Δp 再热 ·b4+Δm 过热 ·b5+Δm 再热 ·b6+Δt 排烟 ·b7(6)

[0030] Wherein, b is the power supply coal consumption, and the unit of the power supply coal consumption is kg / kwh.

[0031] Preferably, the step S5 considers the safety and environmental protection indexes including the wall temperature of the water cooling wall, the wall temperature of the superheater and the nitrogen oxide index.

[0032] The application also provides a coal mill group control system suitable for rapid load change, and the system comprises:

[0033] A parameter acquisition module is configured to acquire maximum output and economic output data of each coal mill when different coal types are used by the coal mill;

[0034] A working condition acquisition module is configured to acquire actual coal quantity, coal type and corresponding calorific value of each coal mill under current load P and load P' after time T;

[0035] A start-stop judgment model establishment module is configured to calculate and establish a judgment model for start-stop control of the coal mill group based on parameters obtained by the parameter acquisition module and the working condition acquisition module;

[0036] A start-stop control module is configured to obtain a timing for controlling start-stop of the coal mill group based on the calculation model for start-stop control of the coal mill group obtained by the start-stop judgment model establishment module;

[0037] An economic coal consumption model establishment module is configured to quantify changes in main indexes that have influence on boiler economy into influence on unit power supply coal consumption, and establish a calculation model for coal consumption change Δb;

[0038] A coal quantity control module is configured to establish a model by using a machine learning algorithm based on the calculation model for coal consumption change Δb obtained by the economic coal consumption model establishment module and an energy balance formula, and taking into account safety and environmental protection indexes, and to perform optimization calculation to obtain the most economic coal quantity of each coal mill.

[0039] Further, the parameter acquisition module acquires parameters in two ways, specifically as follows:

[0040] (1) Maximum output and economic output data of each coal mill are acquired through performance test at regular intervals;

[0041] (2) Historical data of each coal mill, at least including coal mill differential pressure, coal mill outlet air temperature, coal mill current, coal mill vibration parameter, coal mill power consumption, primary air pressure and primary air fan power consumption, are acquired, and a machine learning algorithm model is trained and optimized to obtain maximum output and economic output data of each coal mill.

[0042] The start-stop judgment model establishment module includes the following units:

[0043] A heat calculation unit is configured to calculate a sum Q of heat input into the boiler by all running coal mills under current load P, a sum Q of heat input into the boiler by all running coal mills under maximum output, and a sum Q of heat input into the boiler by all running coal mills under economic output; now max eco ;

[0044] A load calculation unit is configured to calculate a difference ΔQ between the sum of input heat under current load P and the sum of input heat under load P' at time T; ​​

[0045] The load increase and decrease judging unit is used to judge the size relation of P and P', if P'>P, ΔQ=P'-P, then the start control judging unit is executed; if P

[0046] The start control judging unit is used to judge the size relation of ΔQ and Q max -Q now , if ΔQ≤Q max -Q now , it means that the load increase can be met by increasing the coal quantity, and the new coal mill does not need to be started; the start-stop control module is skipped, and the economic coal consumption model establishing module is executed; if ΔQ>Q max -Q now , it means that the load increase cannot be met by the current number of running coal mills, and the new coal mill needs to be started, and the start-stop control module is continued to be executed.

[0047] The close control judging unit is used to judge the size relation of ΔQ and Q now -Q eco , if ΔQ≤Q now -Q eco , it means that the load decrease can be met by decreasing the coal quantity, and the coal mill can be kept in the economic output state, and the existing coal mill does not need to be stopped; the start-stop control module is skipped, and the economic coal consumption model establishing module is executed; if ΔQ>Q now -Q eco , it means that the load decrease needs to decrease the coal quantity of part of the coal mills to below the economic output state, which leads to the decrease of the stability of the coal mill, and the economy and safety of the unit are insufficient, and one coal mill needs to be stopped, and the start-stop control module is continued to be executed.

[0048] The specific execution mode of the economic coal consumption model establishing module is as follows:

[0049] The different increase or decrease of the coal quantity of each coal mill has different influences on the wall temperature of the boiler water cooling wall, the wall temperature of the superheater, the superheated steam temperature and pressure, the reheated steam temperature and pressure, the desuperheating water quantity, the nitrogen oxide emission quantity and the exhaust gas temperature, and further influences the economy, safety and environmental protection indexes of the boiler; the main indexes which have influences on the economy of the boiler include the superheated steam temperature t 过热 , the reheated steam temperature t 再热 , the superheated steam pressure p 过热 , the reheated steam pressure p 再热 , the superheated desuperheating water quantity m 过热 , the reheated desuperheating water quantity m 再热 and the exhaust gas temperature t 排烟 , and then when the coal output quantity of the coal mill is increased or decreased, the change amounts of the above parameters caused by the increase or decrease are Δt 过热 , Δt 再热, Δp 过热 , Δp 再热 , Δm 过热 , Δm 再热 , Δt 排烟 ;

[0050] The change of the above respective parameter unit change to the economy of the unit is quantified as the influence on the coal consumption b of the unit, and the numerical value of the parameter unit change of the unit influencing the coal consumption b is obtained by querying the empirical value of the parameter unit change of the unit influencing the coal consumption of the unit of different grades, which is b1, b2, b3, b4, b5, b6, b7; b1-b7 is positive or negative according to the parameter characteristics, then the coal consumption change caused by the overall parameter change after increasing or decreasing the coal quantity is Δb, and the calculation model is as follows:

[0051] Δb = Δt 过热 ·b1 + Δt 再热 ·b2 + Δp 过热 ·b3 + Δp 再热 ·b4 + Δm 过热 ·b5 + Δm 再热 ·b6 + Δt 排烟 ·b7 (6)

[0052] Wherein, b is the coal consumption, and the unit of the coal consumption is kg / kwh.

[0053] The present application is based on the actual load frequently changing in a short period of time in the future under the power spot transaction, and comprehensively considers the actual coal type of each coal mill, the output of the coal mill, the deterioration degree of the unit, the coal consumption and other data, and combines the corresponding boiler safety index, the emission environmental protection index, to perform group control on all coal mills of the power plant boiler, and accurately control the number of coal mill units starting to run and the output coal quantity of each coal mill. The group control method realizes the following beneficial effects:

[0054] (1) The optimal starting and stopping time of the coal mill unit is avoided, the equipment wear and tear caused by frequent starting and stopping of the equipment is avoided, the work load of the operation and maintenance personnel is increased, the safety and reliability of the equipment is improved, and the labor cost is reduced.

[0055] (2) The coal output of each coal mill is optimized, so that it always works above the economic output state, the stability and efficiency of the unit operation are improved, and the production cost is reduced.

[0056] (3) While adapting to the rapidly changing load and maximizing the production efficiency, the production safety and the influence of the emission on the environmental protection are taken into account, which meets the theme of environmental protection and sustainable development, and has positive social benefits. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1This is a schematic flowchart of a group control method for a power plant boiler coal mill according to Embodiment 1 of the present invention;

[0058] Figure 2 This is a cross-sectional schematic diagram of the layout of a typical power plant pulverized coal boiler pulverizing system in Embodiment 1 of the present invention;

[0059] Figure 3 This is a top view of the coal outlet of the F-mill in the typical power plant pulverized coal boiler pulverizing system layout of Embodiment 1 of the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] Example 1

[0062] like Figure 1 As shown, this embodiment, combined with a typical power plant pulverized coal boiler, further illustrates the coal mill group control method provided by the present invention that adapts to rapid load changes. The method includes:

[0063] S1. Obtain the maximum output and economic output data of each coal mill when using different types of coal;

[0064] S2. Obtain the actual coal quantity, coal type and corresponding calorific value of each coal mill under the current load P, and the load P' after time T;

[0065] S3. Based on the parameters obtained in steps S1 and S2, calculate and establish a judgment model for the start-up and shutdown control of the coal mill unit.

[0066] S4. Based on the calculation model for the start-up and shutdown control of the coal mill unit obtained in step S3, the timing for controlling the start-up and shutdown of the coal mill unit is obtained.

[0067] S5. Quantify the changes in the main indicators that affect the boiler's economy into their impact on the unit's coal consumption for power supply, and establish a calculation model for the coal consumption change Δb.

[0068] S6. Based on the calculation model and energy balance formula of coal consumption change Δb obtained in step S5, and taking into account safety and environmental protection indicators, a model is established using machine learning algorithms to perform optimization calculations and obtain the most economical coal quantity for each coal mill.

[0069] like Figure 2 and Figure 3As shown, 6 medium-speed coal mills are arranged in the tangential firing boiler of the embodiment, numbered A, B, C, D, E, and F. Due to the change of the coal market and the economic benefit decision of the thermal power plant, there are many types of coal entering the furnace, and the coal types are numbered 1, 2, 3,..., n in turn. In this embodiment, coal mills A, B, C, D, E, and F use coal types 1, 2, 3, 4, 5, and 6 respectively. Due to the change of the operating condition and the deterioration of the equipment, the maximum output m max and the economic output m eco of each coal mill are different.

[0070] The determination principle of the maximum output and the economic output of each coal mill refers to DLT469-2019 “Performance Test of Coal Mill and Pulverizing System for Power Station”. There are two ways. Way one: determine by performance test regularly, the time interval is half a year or one year. Way two: obtain the historical data of each coal mill, including the differential pressure of the coal mill, the outlet air temperature of the coal mill, the current of the coal mill, the vibration parameters of the coal mill, the power consumption of the coal mill, the primary air pressure, the power consumption of the primary air fan, etc., train and optimize the model by using the machine learning algorithm to obtain the maximum output and the economic output of each coal mill.

[0071] Under the background of spot trading, according to the quotation of the power plant, the actual load curve in the T period can be obtained after the power market is cleared, and the power grid AGC real-time load curve is sent to the generating unit by the power grid company. The unit generates electricity according to the load curve. Under the background of spot trading, the unpredictability of the market increases, and T is generally 15-60 min. The load after a longer time cannot be accurately predicted.

[0072] The power supply coal consumption b of each unit in a typical load section in a period of time is a fixed value. Based on the principle of energy conservation, the sum of the heat input into the boiler by all coal mills is equal to the sum of the input heat based on the power supply coal consumption and the power generation capacity.

[0073] m A1 ·q 1 +m B2 ·q 2 +m C3 ·q 3 +m D4 ·q 4 +m E5 ·q 5 +m F6 ·q 6 =Q now (1)

[0074] Where m A1 is the coal quantity of coal mill A under the working condition of coal type 1 (kg / h), q 1 is the low calorific value of coal type 1 (kJ / kg), Q now is the current total heat input into the boiler (kJ / h), and mB2 q represents the coal flow rate (kg / h) of mill B under coal type 2 operating conditions. 2 The lower heating value (kJ / kg) is for coal type 2, and so on.

[0075] Assuming the current unit load is P (kW) and the unit load after time T is P' (kW), then

[0076] b p ·P·Q net =Q now (2)

[0077] Among them, b p Let Q be the coal consumption for power generation under load P (kW). Generally, coal consumption for power generation increases as the unit load decreases. However, for a fixed load P, this value can be approximated as constant over a considerable period, meaning the coal consumption for power generation under a certain load is constant. This value is generally determined through periodic testing. net It is the calorific value of standard coal, 29308 kJ / kg, and is also a constant.

[0078] Combining formulas (1) and (2), we get:

[0079] m A1 ·q 1 +m B2 ·q 2 +m C3 ·q 3 +m D4 ·q 4 +m E5 ·q 5 +m F6 ·q 6 =b p ·P·Q net =α P ·P (3)

[0080] In the formula α P If the total heat of each coal mill is a constant that varies only with the load, then the total heat of each coal mill is proportional to the load.

[0081] In this embodiment, the current unit load is set as P (kW), and the unit load after time T is set as P' (kW). Currently, four coal mills (B, C, D, and E) are operating, each mill producing coal types 2, 3, 4, and 5, respectively. The coal flow rate (kg / h) of each mill is respectively m³. B2 m C3 m D4 m E5 The lower heating value (kJ / kg) of each type of coal in the mill is q 2 3 4 5

[0082] q, q, q.

[0083] In the spot market, the unpredictability of the market is increased, T is generally 15-60 min, the time is relatively short, based on the adjustment characteristics of the thermal power unit, so the load in the short time from the current time to T time is generally a monotonic function, that is, the load in the T period is monotonically increasing or monotonically decreasing.

[0084] When the load at T time increases, that is, P'>P, there are the following two cases:

[0085] (1) When the following relationship (4) is satisfied

[0086]

[0087] The current four coal mills can meet the load increase by increasing the coal quantity, without starting new coal mills, in the formula, B is the maximum output of the B coal mill under the condition of No. 2 coal, and the others are the maximum output under the corresponding coal.

[0088] At this time, the four coal mills increase the coal quantity (kg / h) m B , m C , m D , m E according to the load curve in T time, based on energy balance, the following relationship (5) is satisfied:

[0089] α P′ ·P′=(m B2 +m B )·q 2 +(m C3 +m C )·q 3 +(m D4 +m D )·q 4 +(m E5 +m E )·q 5 (5)

[0090] Where, α P’ is the corresponding constant derived from formula (3) when the load is P'.

[0091] Because the thermal load of the corresponding burner of different coal mills changes, that is, the different coal quantity of each coal mill affects the wall temperature of the boiler water cooling wall, the wall temperature of the reheater, the superheated steam temperature and pressure, the reheated steam temperature and pressure, the desuperheating water quantity, the nitrogen oxide emission quantity, and the exhaust gas temperature, which in turn affects the boiler economy, safety, and environmental protection indicators. The main indicators that affect the boiler economy include the superheated steam temperature t 过热 , the reheated steam temperature t 再热 , the superheated steam pressure p过热 , reheat steam pressure p 再热 , desuperheating water quantity m 过热 , desuperheating water quantity m 再热 , exhaust gas temperature t 排烟 , etc., m B , m C , m D , m E , the change amount of each parameter caused by the increase is Δt 过热 , Δt 再热 , Δp 过热 , Δp 再热 , Δm 过热 , Δm 再热 , Δt 排烟 .

[0092] The change of the unit change amount of each parameter in the above-mentioned to the economy of the unit can be quantified as the influence on the coal consumption of the unit for power supply, and the numerical value of the unit change amount of the parameter influencing the coal consumption of the unit for power supply (kg / kwh) is b1, b2, b3, b4, b5, b6, b7, respectively, b1-b7 can be positive or negative values according to the parameter characteristics, and there are experienced values for different grades of coal-fired units that can be consulted, so the change of the coal consumption caused by the overall change of each parameter after adding coal is Δb.

[0093] Δb = Δt 过热 ·b1 + Δt 再热 ·b2 + Δp 过热 ·b3 + Δp 再热 ·b4 + Δm 过热 ·b5 + Δm 再热 ·b6 + Δt 排烟 ·b7 (6)

[0094] In the process of increasing the load of the unit, on the basis of meeting the energy balance of formula (5), the coal quantity of the four coal mills is increased, and at the same time, the numerical value of the change of the coal consumption Δb in formula (6) is as small as possible, that is, the economic performance of the unit is the best, and at the same time, the water-cooled wall temperature, the over-reheater wall temperature, and the nitrogen oxide are not over-standard, a model is established by using a machine learning algorithm, and an optimization calculation is performed to optimize the best numerical value of m B , m C , m D , m E , so as to obtain the optimal coal quantity of the coal mills B, C, D and E.

[0095] (2) When the following relationship formula (7) is met

[0096]

[0097] Then the current 4 coal mills cannot meet the increasing load by increasing the coal quantity, and a new coal mill needs to be started to meet the load requirements of the power grid dispatching. At this time, coal mill A is started preferentially because starting a coal mill requires a warm-up time T 暖磨 , the starting time of the coal mill is T 启动 , the output increase time is T 升出力 , then at the time of T - T 暖磨 - T 启动 - T 升出力 , start coal mill A. Assume that coal mill A uses coal type 1, and the low calorific value (kJ / kg) of the coal type is q 1 . And the output reaches m A1 at time T. At this time, it should satisfy

[0098] α P′ ·P′ = m B2 ·q 2 + m C3 ·q 3 + m D4 ·q 4 + m E5 ·q 5 + m A1 ·q 1 (8)

[0099] On the basis that the coal quantity increased by 5 coal mills meets the energy balance of formula (8), while ensuring that the change Δb of coal consumption in formula (6) is as small as possible, that is, the economic performance of the unit is the best, and at the same time considering that the wall temperature of the water wall, the wall temperature of the reheater, and the nitrogen oxides do not exceed the standard, a model is established using the machine learning algorithm for optimization calculation to optimize m B2 , m C3 , m D4 , m E5 , m A1 The optimal value of the coal quantity, so as to obtain the optimal coal quantity of coal mills B, C, D, E, and A.

[0100] When the load rises at time T, that is, P’ < P, there are the following two situations:

[0101] (3) When the following relational formula (9) is satisfied

[0102]

[0103] Then the current 4 coal mills can meet the load reduction by reducing the coal quantity, and can ensure that the coal quantity of several coal mills remains above their respective economic outputs, and there is no need to stop the existing coal mills. In the formula

[0104] is the economic output of coal mill B under coal type 2, and the others can be known similarly as the economic outputs under the corresponding coal types. ​

[0105] At this time, the four coal mills reduce the coal quantity (kg / h) m³ according to the load curve at time T. B m C m D m E It satisfies the following relation (10):

[0106] α P′ ·P′=(m B2 -m B )·q 2 +(m C3 -m C )·q 3 +(m D4 -m D )·q 4 +(m E5 -m E )·q 5 (10)

[0107] The varying heat loads of burners corresponding to different coal mills, i.e., the different coal reduction rates for each mill, have varying impacts on boiler water-cooled wall temperature, superheater / reheater wall temperature, superheated steam temperature and pressure, reheated steam temperature and pressure, desuperheating water volume, nitrogen oxide emissions, and flue gas temperature. These differences consequently affect the boiler's economic efficiency, safety, and environmental performance. Among these, the main indicators affecting boiler economic efficiency include superheated steam temperature (t). 过热 Reheat steam temperature t 再热 Superheated steam pressure p 过热 Reheat steam pressure p 再热 Superheated desuperheating water volume (m) 过热 Reheat cooling water volume (m) 再热 , flue gas temperature t 排烟 If so, then m B m C m D m E The increase causes a change in each parameter by Δt. 过热 , Δt 再热 Δp 过热 Δp 再热 Δm 过热 Δm 再热 , Δt 排烟 .

[0108] The changes in the unit's economic efficiency caused by the unit changes in the aforementioned parameters can be quantified as the impact on the unit's coal consumption for power generation. The values ​​of the impact of the unit changes in the parameters on the unit's coal consumption for power generation (kg / kWh) are b1, b2, b3, b4, b5, b6, and b7, respectively. b1-b7 can be positive or negative depending on the parameter characteristics, and there are empirical values ​​available for different grades of coal-fired units. Therefore, the change in coal consumption caused by the overall changes in various parameters after adding coal is Δb.

[0109] Δb = Δt 过热 · b1 + Δt 再热 · b2 + Δp 过热 · b3 + Δp 再热 · b4 + Δm 过热 · b5 + Δm 再热 · b6 + Δt 排烟 · b7 (6)

[0110] In the process of unit load reduction, the coal quantity of the four coal mills is increased on the basis of satisfying the energy balance of formula (10), while ensuring that the coal consumption change Δb value in formula (6) is as small as possible, that is, the economic performance of the unit is best, while taking into account the water-cooled wall temperature, the over-reheater wall temperature, and the non-exceeding of nitrogen oxides, a model is established by using a machine learning algorithm to perform optimization calculation, and the optimal values of m B , m C , m D , m E are optimized, so as to obtain the optimal coal quantity of the coal mills B, C, D and E.

[0111] (4) When the following relationship (11) is satisfied

[0112]

[0113] The current four coal mills cannot meet the load reduction by reducing the coal quantity, and need to reduce the coal quantity of part of the coal mills to below the economic output, which leads to the decrease of the stability of the coal mills, the insufficient economic performance and safety of the unit, and the need to shut down one of the coal mills to meet the load required by the power grid dispatching. At this time, the E coal mill is preferentially shut down, because the shutdown of the coal mill needs a power reduction time T 降出力 , the shutdown time of the coal mill T 停运 , the E coal mill is prepared to be shut down at the moment of T-T 停运 -T 降出力 .

[0114] At this time, the following should be satisfied

[0115] α P′ · P' = m B2 · q 2 + m C3 · q 3 + m D4 · q 4 (12)

[0116] The coal quantity of the three coal mills is increased on the basis of satisfying the energy balance of formula (12), while ensuring that the coal consumption change Δb value in formula (6) is as small as possible, that is, the economic performance of the unit is best, while taking into account the water-cooled wall temperature, the over-reheater wall temperature, and the non-exceeding of nitrogen oxides, a model is established by using a machine learning algorithm to perform optimization calculation, and the optimal values of mB2 , m C3 , m D4 The optimal value of the coal quantity is obtained, so that the optimal coal quantity of the coal mills B, C and D is obtained.

[0117] Embodiment 2

[0118] It needs to be further explained that based on the same inventive concept, the application further provides a group control system of a coal mill of a power station boiler, which executes the method of embodiment 1 when running, and comprises:

[0119] A parameter acquisition module is configured to acquire the maximum output and economic output data of each coal mill when different coal types are used.

[0120] A working condition acquisition module is configured to acquire the actual coal quantity, coal type and corresponding calorific value of each coal mill under the current load P and the load P' after the T moment.

[0121] A start-stop judgment model establishment module is configured to calculate and establish a judgment model for start-stop control of the coal mill group based on the parameters acquired by the parameter acquisition module and the working condition acquisition module.

[0122] A start-stop control module is configured to obtain the timing for controlling the start-stop of the coal mill group based on the calculation model for start-stop control of the coal mill group acquired by the start-stop judgment model establishment module.

[0123] An economic coal consumption model establishment module is configured to quantify the changes of the main indexes that have an impact on the economy of the boiler into the impact on the coal consumption of the power generation unit, and establish a calculation model for the change Ab of the coal consumption.

[0124] A coal quantity control module is configured to establish a model by using a machine learning algorithm based on the calculation model for the change Ab of the coal consumption acquired by the economic coal consumption model establishment module and the energy balance formula, and taking into account the safety and environmental protection indexes, to perform optimization calculation and obtain the most economic coal quantity of each coal mill.

[0125] The parameter acquisition module has two ways of acquiring parameters, which are specifically:

[0126] (1) The maximum output and economic output data of each coal mill are acquired through performance tests at regular intervals;

[0127] (2) The maximum output and economic output data of each coal mill are acquired by acquiring historical data of each coal mill, at least including the differential pressure of the coal mill, the outlet air temperature of the coal mill, the current of the coal mill, the vibration parameters of the coal mill, the power consumption of the coal mill, the primary air pressure, and the power consumption of the primary air fan; the maximum output and economic output data of each coal mill are obtained by using a machine learning algorithm model to train and optimize.

[0128] The start-stop judgment model establishment module comprises the following units:

[0129] a heat calculation unit for calculating a sum Q of heat that all the running coal mills can input into the boiler at the current load P now a sum Q of heat that all the running coal mills can input into the boiler at maximum output max and a sum Q of heat that all the running coal mills can input into the boiler at economic output eco ;

[0130] a load calculation unit for calculating a difference AQ between the sum of input heat at the current load P and the sum of input heat at the load P' at the time T;

[0131] a load increase / decrease judgment unit for judging the size relation between P and P', if P'>P, AQ=P'-P, then executing the start control judgment unit; if P

[0132] the start control judgment unit for judging the size relation between AQ and Q max -Q now , if AQQ max -Q now , it means that the load increase can be met by increasing the coal quantity, and a new coal mill does not need to be started; then the start / stop control module is skipped, and the economic coal consumption model establishment module is executed; if AQ>Q max -Q now , it means that the load increase cannot be met by the number of currently running coal mills, and a new coal mill needs to be started; then the start / stop control module is continued to be executed.

[0133] the close control judgment unit for judging the size relation between AQ and Q now -Q eco , if AQQ now -Q eco , it means that the load decrease can be met by decreasing the coal quantity, and the coal mills can be kept running at economic output, and an existing coal mill does not need to be stopped; then the start / stop control module is skipped, and the economic coal consumption model establishment module is executed; if AQ>Q now -Q eco , it means that the load decrease needs to decrease the coal quantity of some coal mills to below the economic output, which leads to the decrease of the stability of the coal mills, and the economy and safety of the unit are insufficient, and a coal mill needs to be stopped; then the start / stop control module is continued to be executed.

[0134] The economic coal consumption model establishment module is specifically executed in the following manner:

[0135] The varying amounts of coal added or removed from each coal mill have different impacts on the boiler's water-cooled wall temperature, superheater / reheater wall temperature, superheated steam temperature and pressure, reheated steam temperature and pressure, desuperheating water volume, nitrogen oxide emissions, and flue gas temperature, thus affecting the boiler's economy, safety, and environmental performance. Key indicators affecting boiler economy include superheated steam temperature (t). 过热 Reheat steam temperature t 再热 Superheated steam pressure p 过热 Reheat steam pressure p 再热 Superheated desuperheating water volume (m) 过热 Reheat cooling water volume (m) 再热 , flue gas temperature t 排烟 Then, when the coal output of the coal mill increases or decreases, the change in the above parameters is Δt. 过热 Δt 再热 Δp 过热 Δp 再热 Δm 过热 Δm 再热 Δt 排烟 ;

[0136] The changes in the unit's economic efficiency caused by the unit changes in the aforementioned parameters are quantified as their impact on the unit's coal consumption for power generation, b. By consulting empirical values ​​for the impact of the unit changes in the aforementioned parameters on the unit's coal consumption for power generation for different grades of coal-fired units, the numerical values ​​of the impact of the unit changes in the aforementioned parameters on the unit's coal consumption for power generation are obtained as b1, b2, b3, b4, b5, b6, and b7, respectively. Since b1-b7 are either positive or negative based on the parameter characteristics, the calculation model for the change in coal consumption Δb caused by the overall changes in various parameters after increasing or decreasing the amount of coal is as follows:

[0137] Δb=Δt 过热 ·b1+Δt 再热 ·b2+Δp 过热 ·b3+Δp 再热 ·b4+Δm 过热 ·b5+Δm 再热 ·b6+Δt 排烟 ·b7(6)

[0138] Where b represents the coal consumption for power generation, and the unit of coal consumption for power generation is kg / kWh.

[0139] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling a group of coal mills to adapt to rapid load changes, characterized by, The method comprises the following steps: S1, obtaining the maximum output and economic output data of each coal mill when using different coal types; S2, obtaining the actual coal quantity, coal type and corresponding calorific value of each coal mill under the current load P and the load P' after time T; S3, calculating and establishing a judgment model for the start-stop control of the coal mill group based on the parameters obtained in steps S1 and S2; The method comprises the following steps: S31, calculate the sum of heat Q that all running coal mills can input into the boiler under the current load P now , the sum of heat Q that all running coal mills can input into the boiler when all the running coal mills are at maximum output max , and the sum of heat Q that all running coal mills can input into the boiler when all the running coal mills are at economic output eco ; S32, calculating the difference ΔQ between the sum of the input heat of the current load P and the sum of the input heat of the load P' after time T; S33, judging the size relationship between P and P', if P'>P, ΔQ=P'-P, then executing step S34; if P S34, judge the size relation of AQ and Q max -Q now , if AQ≤Q max -Q now , it represents that the load rising can be met by increasing the coal quantity, and the new coal mill does not need to be started; then step S4 is skipped, and step S5 is executed; if AQ>Q max -Q now , it represents that the number of currently running coal mills cannot meet the load rising, and the new coal mill needs to be started, and then step S4 is continuously executed; S35, judging the size relation between AQ and Q now -Q eco If AQ≤Q now -Q eco , it represents that the load reduction can be met by reducing the coal quantity, and the coal mill can be kept in the economic output state, and the existing coal mill does not need to be shut down; then step S4 is skipped, and step S5 is executed; if AQ>Q now -Q eco , it represents that part of the coal mill needs to be reduced to below the economic output state by reducing the coal quantity, which leads to the decrease of the stability of the coal mill, and the economy and safety of the unit are insufficient, so one coal mill needs to be shut down, and then step S4 is continuously executed. S4, obtaining the timing of controlling the start-stop of the coal mill group based on the calculation model for the start-stop control of the coal mill group obtained in step S3; S5, quantifying the changes of the main indexes affecting the economy of the boiler into the influence on the unit power supply coal consumption, and establishing a calculation model of the coal consumption change Δb; S6, based on the calculation model of the coal consumption change Δb obtained in step S5 and the energy balance formula, and taking into account the safety and environmental protection indexes, a model is established by using a machine learning algorithm for optimization calculation to obtain the most economical coal quantity of each coal mill; the safety and environmental protection indexes include: water wall temperature, superheater wall temperature, and nitrogen oxide index.

2. A method for controlling a group of coal mills to adapt to rapid load changes according to claim 1, characterized in that, The maximum output and economic output data in step S1 have two obtaining methods, specifically: (1) periodically obtaining the maximum output and economic output data of each coal mill through performance test; (2) obtaining the historical data of each coal mill, at least including the differential pressure of the coal mill, the outlet air temperature of the coal mill, the current of the coal mill, the vibration parameters of the coal mill, the power consumption of the coal mill, the primary air pressure, and the power consumption of the primary air fan; the maximum output and economic output data of the current each coal mill are obtained by using a machine learning algorithm model training and optimization.

3. A method for controlling a group of coal mills to adapt to rapid load changes according to claim 1, characterized in that, The step S4 is specifically: When P' > P, the time point T of starting a new coal mill 开启 = T - T 暖磨 - T 启动 - T 升出力 ; When P' < P, the time point T of closing one existing coal mill 关闭 = T - T 停运 - T 降出力 ; where T 暖磨 is the warm-up time of the coal mill required to start the coal mill, T 启动 is the time required to start the coal mill, T 升出力 is the time required to increase the output of the coal mill, T 停运 is the time required to shut down the coal mill, and T 降出力 is the time required to decrease the output of the coal mill.

4. The method of claim 1, wherein, The step S5 is specifically: The different increase or decrease of the coal quantity of each coal mill has different influences on the wall temperature of the boiler water cooling wall, the wall temperature of the reheater, the superheated steam temperature and pressure, the reheated steam temperature and pressure, the desuperheating water quantity, the nitrogen oxide emission quantity and the exhaust gas temperature, and further influences the boiler economy, safety and environmental protection indexes; the main indexes which have influences on the boiler economy include the superheated steam temperature t 过热 , the reheated steam temperature t 再热 , the superheated steam pressure p 过热 , the reheated steam pressure p 再热 , the superheated desuperheating water quantity m 过热 , the reheated desuperheating water quantity m 再热 , the exhaust gas temperature t 排烟 , and when the coal output of the coal mill increases or decreases, the change amounts of the above parameters caused are Δt 过热 , Δt 再热 , Δp 过热 , Δp 再热 , Δm 过热 , Δm 再热 , Δt 排烟 ; The changes of the above respective parameter unit changes in the unit of the unit economy are quantified into the influence on the unit power supply coal consumption b, and the values of the parameter unit changes influencing the unit power supply coal consumption are obtained by querying the experience values of the parameter unit changes influencing the unit power supply coal consumption of the coal-fired unit of different levels, which are b1, b2, b3, b4, b5, b6 and b7; b1-b7 are positive or negative values according to the parameter characteristics, so the coal consumption change Δb caused by the overall parameter changes after increasing or decreasing the coal quantity is calculated as follows: （6） Wherein, b is the power supply coal consumption, and the unit of the power supply coal consumption is kg / kwh.

5. A mill group control system that accommodates rapid load changes, characterized by, The system comprises: A parameter acquisition module for obtaining the maximum output and economic output data of each coal mill when using different coal types; A working condition acquisition module for obtaining the actual coal quantity, coal type and corresponding calorific value of each coal mill under the current load P and the load P' after time T; A start-stop judgment model establishment module for calculating and establishing a judgment model for the start-stop control of the coal mill group based on the parameters obtained by the parameter acquisition module and the working condition acquisition module; comprising the following units: a heat calculation unit for calculating the sum Q of the heat that all of the mills that are in operation can input into the boiler at the current load P now the sum Q of the heat that all of the mills that are in operation can input into the boiler at maximum output max and the sum Q of the heat that all of the mills that are in operation can input into the boiler at economic output eco ; A load calculation unit is configured to calculate a difference ΔQ between a sum of input heat of the current load P and a sum of input heat of the load P' at the time T; A load increase / decrease judgment unit is configured to judge the size relationship between P and P', if P'>P, ΔQ=P'-P, then execute the start control judgment unit; if P The start control judging unit is used for judging the size relation between ΔQ and Q max -Q now If ΔQ≤Q max -Q now , it represents that the load rising can be met by increasing the coal quantity, and the new coal mill does not need to be started; the start-stop control module is skipped, and the economic coal consumption model establishing module is executed; if ΔQ>Q max -Q now , it represents that the number of the currently running coal mills cannot meet the load rising, and the new coal mill needs to be started, and the start-stop control module is continuously executed. The closing control judging unit is used for judging the size relation between ΔQ and Q now -Q eco If ΔQ≤Q now -Q eco , it represents that the load drop can be met by reducing the coal quantity, and the coal mill can be kept in the economic output state, and the existing coal mill does not need to be stopped; the economic coal consumption model establishing module is executed by skipping the start-stop control module; if ΔQ>Q now -Q eco , it represents that the coal quantity of part of the coal mills needs to be reduced to below the economic output state by reducing the coal quantity, which leads to the decrease of the stability of the coal mill, the economy and safety of the unit are insufficient, and one coal mill needs to be stopped, and the start-stop control module is continuously executed. A start / stop control module is configured to obtain a judgment mill group start / stop control calculation model from the start / stop judgment model establishment module, and obtain the timing of controlling the mill group start / stop; An economic coal consumption model establishment module is configured to quantify the change of the main indexes that have influence on the boiler economy as the influence on the unit power supply coal consumption, and establish a calculation model of the coal consumption change Δb; A coal quantity control module is configured to obtain the calculation model of the coal consumption change Δb from the economic coal consumption model establishment module, and the energy balance formula, and take into account the safety and environmental protection indexes, and use the machine learning algorithm to establish the model, perform the optimization calculation, and obtain the most economic coal quantity of each mill.

6. A load following coal mill group control system according to claim 5, wherein, The parameter acquisition module acquires parameters in two ways, which are: (1) periodically obtaining the maximum output and economic output data of each mill through performance test; (2) obtaining the historical data of each mill, at least including the mill differential pressure, mill outlet air temperature, mill current, mill vibration parameter, mill power consumption, primary air pressure, primary air fan power consumption; using the machine learning algorithm model training and optimization to obtain the maximum output and economic output data of each mill.

7. A load following coal mill group control system according to claim 5, wherein, The economic coal consumption model establishment module is specifically executed in the following way: The different increase or decrease of the coal quantity of each coal mill has different influences on the wall temperature of the boiler water cooling wall, the wall temperature of the reheater, the superheated steam temperature and pressure, the reheated steam temperature and pressure, the desuperheating water quantity, the nitrogen oxide emission quantity and the exhaust gas temperature, and further influences the boiler economy, safety and environmental protection indexes; the main indexes which have influences on the boiler economy include the superheated steam temperature t 过热 , the reheated steam temperature t 再热 , the superheated steam pressure p 过热 , the reheated steam pressure p 再热 , the superheated desuperheating water quantity m 过热 , the reheated desuperheating water quantity m 再热 , the exhaust gas temperature t 排烟 , and when the coal output of the coal mill increases or decreases, the change amounts of the above parameters are Δt 过热 , Δt 再热 , Δp 过热 , Δp 再热 , Δm 过热 , Δm 再热 , Δt 排烟 ; Quantify the change of the above respective parameter unit change to the change of the unit economy to the influence on the unit power supply coal consumption b, and obtain the numerical value of the parameter unit change of the unit power supply coal consumption by querying the experience value of the parameter unit change of the unit power supply coal consumption of the different grade coal-fired unit, which are b1, b2, b3, b4, b5, b6, and b7; b1-b7 are positive or negative values according to the parameter characteristics, then the coal consumption change Δb caused by the overall parameter change after increasing or decreasing the coal quantity is calculated as follows: （6） Wherein, b is the power supply coal consumption, and the unit of the power supply coal consumption is kg / kwh.

Images