A method for predicting output power of subcritical circulating fluidized bed units under deep peak regulation

Abstract

The present invention provides a method for predicting the output power of a subcritical circulating fluidized bed unit under deep peak regulation, which uses the coal quality parameters of the subcritical CFB unit and the real-time operation data of deep peak regulation to predict the output power of the unit in real time, including the following steps: establishing a boiler system model under deep peak regulation of the subcritical CFB unit and determining the steady-state parameters; establishing a steam turbine system model and determining the steady-state parameters; fitting the functional relationship between the steady-state parameters; establishing an output power prediction model, using a particle swarm algorithm to identify the dynamic parameters in the model, and predicting the unit output power during deep peak regulation operation according to the real-time operation data of the subcritical CFB unit and the coal quality parameters. The present invention helps to guide the early adjustment of the unit operation control and reduce the fluctuation of important parameters.

Description

Technical Field

[0001] The invention belongs to the field of coordinated control of thermal power units and relates to an output power prediction method of a subcritical circulating fluidized bed unit under deep peak regulation. Background Art

[0002] In recent years, the growth rate of my country's renewable energy power generation installed capacity has continued to increase, and it is gradually completing the transition to the main power source. Large-scale renewable energy grid connection has reduced the stability of the power grid, and the use of thermal power units for peak and frequency regulation has become the most commonly used means. After a simple flexibility transformation, the circulating fluidized bed (CFB) unit has enhanced its peak-shaving capacity, changing from the traditional 50% to 100% rated output operation to a deep peak-shaving operation of 30% to 100% rated output. During deep peak-shaving operation, the grid load command changes frequently, and the CFB unit needs to adjust the unit's output power in time to respond to changes in the grid load, which is easy to cause large fluctuations in important operating parameters such as main steam pressure, main steam temperature, and bed temperature, affecting the safety of unit operation. If the prediction of the unit output power during deep peak-shaving operation of the CFB unit can be achieved, it will help the design of coordinated control strategies for the CFB unit, improve the unit's operation control level, ensure the unit's variable load rate, and reduce the fluctuation of important operating parameters.

[0003] Purpose of the Invention

[0004] The purpose of the present invention is to solve the problem in the prior art that during deep peak regulation operation, the grid load command changes frequently, and the CFB unit needs to adjust the output power of the unit in time to respond to the change of the grid load, which easily causes the main steam pressure, main steam temperature, bed temperature and other important operating parameters to fluctuate greatly, affecting the safe operation of the unit. A method for predicting the output power of a subcritical circulating fluidized bed unit under deep peak regulation is provided. The method uses the coal quality parameters of the subcritical CFB unit and the real-time operation data of deep peak regulation to predict the output power of the unit in real time, which is helpful to guide the early adjustment of the unit operation control and reduce the fluctuation of important parameters. Summary of the invention

[0005] The present invention provides a method for predicting output power of a subcritical circulating fluidized bed unit under deep peak regulation, comprising the following steps:

[0006] Step 1: According to the combustion and heat transfer mechanism of the CFB unit, a boiler system model of the subcritical CFB unit under deep peak regulation is established, and the steady-state parameters in the model are determined according to the historical operation data under steady-state conditions; wherein the boiler system model includes three sub-models, namely, a combustion sub-model, a heat transfer sub-model, and a steam-water sub-model; when establishing the boiler system model, the dynamic combustion process of the volatile matter in the coal fed into the furnace is ignored;

[0007] Step 2: According to the steam-water working medium characteristics and the energy conservation principle, a steam turbine system model under deep peak regulation of the subcritical CFB unit is established, and the steady-state parameters in the model are determined; when establishing the steam turbine system model under deep peak regulation of the subcritical CFB unit, the operating characteristics of the reheater are considered;

[0008] Step 3, using the dynamic historical operation data under deep peak regulation to fit the functional relationship between the steady-state parameters in the boiler system model in step 1 and the steady-state parameters in the steam turbine system model in step 2; further, after determining the functional form between the steady-state parameters based on mechanism analysis, fitting is performed using the nonlinear least squares method, with the minimum sum of squares of errors as the criterion.

[0009] Step 4: Use historical operating data and dynamic parameters in the coal quality parameter identification model to establish an output power prediction model for subcritical CFB units under deep peak regulation, wherein a particle swarm algorithm is used to identify the dynamic parameters in the model; and predict the unit output power during deep peak regulation based on the real-time operating data of the subcritical CFB unit and coal quality parameters.

[0010] Preferably, step 1 further comprises the following sub-steps:

[0011] Sub-step S11, establishing the combustion sub-model, including: defining the CFB unit as adopting a fluidized combustion mode, with a large amount of burning carbon residue in the furnace;

[0012] The mass of residual carbon m c It is expressed as shown in formula (1):

[0013]

[0014] In formula (1), W c is the coal feeding amount, in kg / s; X c R is the fixed carbon mass fraction on the received basis, expressed as % content; c is the burning rate of residual carbon, in kg / s; W PZ is the slag discharge rate, in kg / s; X c,p is the carbon content of the slag, expressed as % content; W FL is the fly ash flow rate, in kg / s; X c,f is the carbon content of fly ash, expressed as % content;

[0015] Burning rate R c It is related to the carbon residue and oxygen concentration and can be expressed as shown in formula (2):

[0016]

[0017] In formula (2), k 1 is the combustion coefficient; is the average oxygen concentration in mol / m 3 ;

[0018] The lumped parameter model of average oxygen concentration is expressed as shown in formula (3):

[0019]

[0020] In formula (3), k a is the volume-to-molar conversion coefficient; u Air is the total air volume, in Nm 3 / s;k v is the conversion factor from mass to mole; X o is the mass fraction of oxygen released after the combustion of volatile matter, expressed as % content; X v is the mass fraction of volatile matter in coal, expressed as % content; k f is the flue gas flow conversion coefficient; V is the furnace volume, in m 3 ;

[0021] The real-time combustion heat value is expressed as shown in formula (4):

[0022] Q a =Q c +Q v =R c H c +W c ·∑X i H i (4)

[0023] In formula (4), H c is the calorific value of the residual carbon, in kJ / kg; X i is the mass fraction of the i-th volatile component in coal, expressed as % content; H i is the combustion calorific value of the i-th volatile component, in kJ / kg; the volatile components in the coal are CH 4 , H 2 , CO, CH 0.689 O 0.014 , and the corresponding combustion heat values ​​are 50.02MJ / kg, 124.24MJ / kg, 10.08MJ / kg, and 37.0MJ / kg respectively;

[0024] The CH 4 , H 2 , CO, CH 0.689 O 0.014 The mass fraction of is expressed as shown in formula (5):

[0025]

[0026] Sub-step S12, establishing the heat transfer sub-model, including: setting the bed temperature T b The lumped parameter model of is expressed as shown in formula (6):

[0027]

[0028] In formula (6), C s is the heat capacity of the bed material, in kJ / ℃; Q s is the physical sensible heat of coal, in kJ / s; Δq air is the change in the calorific value of the wind, in kJ / m 3 ;h pz is the enthalpy of slag discharge, in kJ / kg; h FL is the enthalpy of fly ash, in kJ / kg; Q 1 is the heat absorbed by the steam-water medium and the wall, in kJ / s;

[0029] The amount of heat absorbed Q 1 It is expressed as shown in formula (7):

[0030]

[0031] In formula (7), k 2 , n is the model coefficient; D fw is the water flow rate, in kg / s; T st is the main steam temperature, in °C;

[0032] Sub-step S13, establishing the soda sub-model, comprises:

[0033] The lumped parameter model of the metal wall and the steam-water medium is expressed as shown in formula (8):

[0034]

[0035] In formula (8), p b is the drum pressure, in MPa; C b is the total heat storage coefficient of soda and metal, in kJ / MPa; h fw is the feed water enthalpy value, in kJ / kg; D sw is the cooling water flow rate, in kg / s; D st Main steam flow rate, unit is kg / s; h st is the main steam enthalpy, in kJ / kg;

[0036] The feed water enthalpy is related to the output power of the CFB unit. The feed water enthalpy is expressed as shown in formula (9):

[0037] hfw =f(N e ) (9),

[0038] In formula (9), N e is the unit output power, in MW; f(·) is N e Function of

[0039] The main steam pressure is determined by the drum pressure and the superheater pressure loss. The superheater pressure loss is expressed as shown in formula (10):

[0040] △p=p b -p st = p(N e ) (10),

[0041] In formula (10), Δp is the superheater pressure loss, in MPa; p st is the main steam pressure, in MPa; p(·) is about N e Function of

[0042] The main steam enthalpy of the subcritical CFB unit during deep peak regulation operation is expressed as shown in formula (11):

[0043] h st =h(p st ,N e ) (11),

[0044] In formula (11, h st is the main steam enthalpy, in kJ / kg; h(·) is about p st 、N e Function of

[0045] The main steam temperature is related to the main steam pressure and enthalpy value. The main steam temperature is expressed as shown in formula (12):

[0046] T st =T(p st ,h st ) (12),

[0047] In formula (12: T(·) is about p st 、h st function.

[0048] Preferably, step 2 further comprises: expressing the main steam flow rate as shown in formula (13):

[0049] D st =u t D(p st ,h st ) (13),

[0050] In formula (13): u t is the main steam valve opening, expressed as % content; D(·) is about p st 、h st Function of

[0051] The output power N of the subcritical CFB unit e It is expressed as shown in formula (14)-(15):

[0052]

[0053] k 3 =k(N e ) (15),

[0054] In formula (14) and (15), C n is the inertia time of the steam turbine, in seconds; k(·) is the unit of N e function.

[0055] Preferably, in step 3, f(·), h(·), p(·), D(·), k(·), Q 1 Fitting, the f(·), p(·), D(·), k(·), Q 1 The functional form of is shown in formula (16):

[0056]

[0057] In formula (16): a 1 、a 2 、a 3 、a 4 、a 5 、a 6 、b 1 、b 2 、b 3 、b 4 、b 5 、b 6 、c 1 、c 4 、c 6 d 1 d 4 d 6 are the parameters to be fitted.

[0058] Preferably, in step 4, the dynamic parameter C in the model is identified by using the historical operation data and coal quality parameters using a particle swarm algorithm. s , C b , C n; The coal quality, coal feed, total air volume, slag discharge, feed water flow, fly ash, main steam valve opening, and desuperheating water flow values ​​in the historical operation data are used as the input of the boiler system model under the deep peak regulation of the subcritical CFB unit to form an initial particle group, and each particle of the group is used as a dynamic parameter value to calculate the output of the boiler system model, and select p st 、N e The sum of the absolute values ​​of the relative errors between the model output value and the actual value is taken as the objective function, which is expressed as shown in formula (17):

[0059]

[0060] In formula (17), Δp st,i , ΔN e,i They represent the deviations of the main steam pressure and output power output values ​​of the model at the i-th sampling point from the actual data;

[0061] After updating the particle positions and velocities, the particle group continuously obtains smaller objective function values, and finally obtains optimized dynamic parameters;

[0062] The real-time operation data is read from the distributed control system DCS of the deep peak-shaving CFB unit, and the output power prediction model is used in combination with coal quality parameters to predict the output power of the subcritical CFB unit under deep peak-shaving. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flowchart of an output power prediction method of a subcritical circulating fluidized bed unit under deep peak regulation described in the present invention. DETAILED DESCRIPTION

[0064] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings. It should be noted that those skilled in the art should understand that the embodiments herein are only used to exemplarily describe the present invention and should not be regarded as limiting the scope of protection of the present invention. Any modification or equivalent substitution of the technical solution of the present invention that does not depart from the idea and purpose of the present invention shall fall within the scope of the present invention.

[0065] Figure 1 This is a flowchart of a method for predicting output power of a subcritical circulating fluidized bed unit under deep peak regulation according to the present invention. Figure 1 As shown, the prediction method of the present invention comprises the following steps:

[0066] Step 1: According to the combustion and heat transfer mechanism of the CFB unit, a boiler system model of the subcritical CFB unit under deep peak regulation is established, and the steady-state parameters in the model are determined according to the historical operation data under steady-state conditions; more specifically, the boiler system model includes three sub-models, namely, a combustion sub-model, a heat transfer sub-model, and a steam-water sub-model; when establishing the boiler system model, the dynamic combustion process of the volatile matter in the coal fed into the furnace is ignored;

[0067] Step 2: According to the steam-water working medium characteristics and the energy conservation principle, a steam turbine system model under deep peak regulation of the subcritical CFB unit is established, and the steady-state parameters in the model are determined; specifically, when establishing the steam turbine system model under deep peak regulation of the subcritical CFB unit, the operating characteristics of the reheater are considered;

[0068] Step 3, using the dynamic historical operation data under deep peak regulation to fit the functional relationship between the steady-state parameters in the boiler system model in step 1 and the steady-state parameters in the steam turbine system model in step 2; specifically, after determining the functional form between the steady-state parameters based on mechanism analysis, fitting is performed using the nonlinear least squares method, with the minimum sum of squares of errors as the criterion.

[0069] Step 4: Use historical operating data and the dynamic parameters in the coal quality parameter identification model to establish an output power prediction model for the subcritical CFB unit under deep peak regulation. Specifically, use a particle swarm algorithm to identify the dynamic parameters in the model; predict the unit output power during deep peak regulation based on the real-time operating data of the subcritical CFB unit and the coal quality parameters.

[0070] The specific implementation process of the above prediction method is described in detail below through an embodiment.

[0071] Example

[0072] 1. Establish a boiler system model for subcritical CFB units under deep peak regulation

[0073] 1.1 Combustion sub-model

[0074] The CFB unit adopts fluidized combustion. There is a large amount of burning carbon residue in the furnace. The mass of carbon residue is m c It can be expressed as

[0075]

[0076] Where: W c is the coal feeding amount, kg / s; X c is the fixed carbon mass fraction received, %; R c is the burning rate of residual carbon, kg / s; W PZ is the slag discharge rate, kg / s; X c,p is the carbon content of slag, %;

[0077] W FL is the fly ash flow rate, kg / s; X c,f is the carbon content of fly ash, %.

[0078] Burning rate R c It is related to the carbon residue and oxygen concentration and can be expressed as

[0079]

[0080] Where: k 1 is the combustion coefficient; is the average oxygen concentration, mol / m 3 .

[0081] The lumped parameter model of average oxygen concentration can be expressed as

[0082]

[0083] Where: k a is the volume-to-molar conversion coefficient; u Air is the total air volume, Nm 3 / s;k v is the conversion factor from mass to mole; X o is the mass fraction of oxygen released after the combustion of volatile matter, %; X v is the mass fraction of volatile matter in coal, %; k f is the flue gas flow conversion coefficient; V is the furnace volume, m 3 .

[0084] The real-time combustion heat can be expressed as

[0085] Q a =Q c +Q v =R c H c +W c ·∑X i H i (4)

[0086] Where: H c is the calorific value of the residual carbon, kJ / kg; X i is the mass fraction of the i-th volatile component in coal, %; H i is the calorific value of the volatile component i, kJ / kg. The volatile components in coal are CH 4 , H 2 , CO, CH 0.689 O 0.014 The calorific values ​​of combustion are 50.02MJ / kg, 124.24MJ / kg, 10.08MJ / kg and 37.0MJ / kg respectively.

[0087] CH 4 , H 2 , CO, CH 0.689 O 0.014 The mass fraction can be expressed as

[0088]

[0089] 1.2 Heat transfer sub-model

[0090] When the subcritical CFB unit is in deep peak load operation, the bed temperature changes greatly, which has a great impact on heat transfer. b The lumped parameter model can be expressed as

[0091]

[0092] Where: C s is the heat capacity of the bed material, kJ / ℃; Q s is the physical sensible heat of coal, kJ / s; Δq air is the change in the calorific value of the wind, kJ / m 3 ;h pz is the enthalpy of slag discharge, kJ / kg; h FL is the enthalpy of fly ash, kJ / kg; Q 1 is the amount of heat absorbed by the steam-water medium and the wall, kJ / s.

[0093] Heat absorption Q 1 It can be expressed as

[0094]

[0095] Where: k 2 , n is the model coefficient; D fw is the water flow rate, kg / s; T st is the main steam temperature, ℃.

[0096] 1.3 Soda sub-model

[0097] Based on the energy balance principle, the lumped parameter model of the metal wall and the steam-water medium can be expressed as

[0098]

[0099] Where: p b is the drum pressure, MPa; C b is the total heat storage coefficient of soda and metal, kJ / MPa; h fw is the feed water enthalpy, kJ / kg; D sw is the cooling water flow rate, kg / s; D st Main steam flow, kg / s; h st is the main steam enthalpy, kJ / kg.

[0100] The feed water enthalpy is related to the output power of the CFB unit and can be expressed as

[0101] h fw =f(N e ) (9)

[0102] Where: N e is the unit output power, MW; f(·) is N e function.

[0103] The main steam pressure is determined by the drum pressure and superheater pressure loss. The superheater pressure loss can be expressed as

[0104] △p=p b -p st = p(N e ) (10)

[0105] Where: Δp is the superheater pressure loss, MPa; p st is the main steam pressure, MPa; p(·) is about N e function.

[0106] When the subcritical CFB unit is in deep peak regulation operation, the main steam enthalpy changes greatly, which can be expressed as

[0107] h st =h(p st ,N e ) (11)

[0108] Where: h st is the main steam enthalpy, kJ / kg; h(·) is about p st 、N e function.

[0109] The main steam temperature is related to the main steam pressure and enthalpy, which can be expressed as

[0110] T st =T(p st ,h st ) (12)

[0111] Where: T(·) is about p st 、h st function.

[0112] 2. Establish a turbine system model for deep peak regulation of subcritical CFB units

[0113] The output power of the unit is mainly affected by the main steam flow rate, which can be expressed as

[0114] D st =u t D(pst ,h st ) (13)

[0115] Where: u t is the main steam valve opening, %; D(·) is about p st 、h st function.

[0116] According to the energy balance principle, considering the operating characteristics of the reheater, the unit output power N e It can be expressed as

[0117]

[0118] k 3 =k(N e ) (15)

[0119] Where: C n is the inertia time of the steam turbine, s; k(·) is the e function.

[0120] 3. Functional relationship fitting between key parameters

[0121] Using the dynamic historical operation data of subcritical CFB units under deep peak regulation, the minimum sum of squares of errors is taken as the criterion, and the nonlinear least squares method is used to calculate f(·), h(·), p(·), D(·), k(·), Q 1 Fitting, f(·), p(·), D(·), k(·), Q 1 The functional form of can be expressed as

[0122]

[0123] Where: a 1 、a 2 、a 3 、a 4 、a 5 、a 6 、b 1 、b 2 、b 3 、b 4 、b 5 、b 6 、c 1 、c 4 、c 6 ,d 1 ,d 4 ,d 6 are the parameters to be fitted.

[0124] 4. Dynamic parameter identification

[0125] Using historical operation data and coal quality parameters, the particle swarm algorithm is used to identify the dynamic parameter C in the model. s , C b , C n The coal quality, coal feed, total air volume, slag discharge, water feed flow, fly ash, main steam valve opening, and cooling water flow in the historical operation data are used as the input of the model to form an initial particle group. Each particle in the group is used as a dynamic parameter value to calculate the output of the model. Select p st 、N e The sum of the absolute values ​​of the relative errors between the model output value and the actual value is taken as the objective function. The objective function can be expressed as

[0126]

[0127] Where: Δp st,i , ΔN e,i They respectively represent the deviation between the main steam pressure and output power output value of the model at the i-th sampling point and the actual data.

[0128] After updating the position and velocity of the particle group, a smaller objective function value can be obtained, and finally the optimized dynamic parameters can be obtained.

[0129] After the above model is established, the real-time operation data is read from the distributed control system (DCS) of the deep peak-shaving CFB unit. The output power prediction model is used in combination with coal quality parameters to realize the output power prediction of the subcritical CFB unit under deep peak-shaving.

[0130] The present invention has the following beneficial effects:

[0131] (1) The operating characteristics of the CFB unit under medium and low load conditions were considered, and the boiler system model and steam turbine system model under deep peak regulation of the subcritical CFB unit were established.

[0132] (2) The historical operating data of a wide range of variable load conditions is used to fit the functional relationship of key parameters using the nonlinear least squares method, which improves the applicability of the functional relationship.

[0133] (3) The particle swarm optimization algorithm is used to identify and optimize the dynamic parameters in the output power prediction model, thereby improving the prediction accuracy of the model for the unit output power, which can provide guidance for the deep peak load control optimization and flexible operation of the CFB unit.

Claims

1. A method for predicting output power of a subcritical circulating fluidized bed unit under deep peak regulation, characterized in that: The steps include: Step 1: According to the combustion and heat transfer mechanism of the CFB unit, a boiler system model of the subcritical CFB unit under deep peak regulation is established, and the steady-state parameters in the model are determined according to the historical operation data under steady-state conditions; wherein the boiler system model includes three sub-models, namely, a combustion sub-model, a heat transfer sub-model, and a steam-water sub-model; when establishing the boiler system model, the dynamic combustion process of the volatile matter in the coal fed into the furnace is ignored; Step 2: According to the steam-water working medium characteristics and the energy conservation principle, a steam turbine system model under deep peak regulation of the subcritical CFB unit is established, and the steady-state parameters in the model are determined; when establishing the steam turbine system model under deep peak regulation of the subcritical CFB unit, the operating characteristics of the reheater are considered; Step 3, using the dynamic historical operation data under deep peak regulation to fit the functional relationship between the steady-state parameters in the boiler system model in step 1 and the steady-state parameters in the steam turbine system model in step 2; further, after determining the functional form between the steady-state parameters according to the mechanism analysis, fitting is performed by nonlinear least squares method, with the minimum sum of squares of errors as the criterion; Step 4: Using historical operation data and dynamic parameters in the coal quality parameter identification model, establish an output power prediction model for subcritical CFB units under deep peak regulation, wherein the dynamic parameters in the model are identified using a particle swarm algorithm; predict the unit output power during deep peak regulation operation based on the real-time operation data of the subcritical CFB unit and coal quality parameters; Step 1 further includes the following sub-steps: Sub-step S11, establishing the combustion sub-model, including: defining the CFB unit as adopting a fluidized combustion mode, with a large amount of burning carbon residue in the furnace; The mass of residual carbon m c It is expressed as shown in formula (1): In formula (1), W c is the coal feeding amount, in kg / s; X c R is the fixed carbon mass fraction on the received basis, expressed as % content; c is the burning rate of residual carbon, in kg / s; W PZ is the slag discharge rate, in kg / s; X c,p is the carbon content of the slag, expressed as % content; W FL is the fly ash flow rate, in kg / s; X c,f is the carbon content of fly ash, expressed as % content; Burning rate R c It is related to the carbon residue and oxygen concentration and can be expressed as shown in formula (2): In formula (2), k1 is the combustion coefficient; is the average oxygen concentration in mol / m 3 ; The lumped parameter model of average oxygen concentration is expressed as shown in formula (3): In formula (3), k a is the volume-to-molar conversion coefficient; u Air is the total air volume, in Nm 3 / s;k v is the conversion factor from mass to mole; X o is the mass fraction of oxygen released after the combustion of volatile matter, expressed as % content; X v is the mass fraction of volatile matter in coal, expressed as % content; k f is the flue gas flow conversion coefficient; V is the furnace volume, in m 3 ; The real-time combustion heat value is expressed as shown in formula (4): Q a =Q c +Q v =R c H c +W c ·∑X i H i (4), In formula (4), H c is the calorific value of the residual carbon, in kJ / kg; X i is the mass fraction of the i-th volatile component in coal, expressed as % content; H i is the combustion calorific value of the i-th volatile component, in kJ / kg; the volatile components in the coal are CH4, H2, CO, CH 0.689 O 0.014 , and the corresponding combustion heat values ​​are 50.02MJ / kg, 124.24MJ / kg, 10.08MJ / kg, and 37.0MJ / kg respectively; The CH4, H2, CO, CH 0.689 O 0.014 The mass fraction of is expressed as shown in formula (5): Sub-step S12, establishing the heat transfer sub-model, including: setting the bed temperature T b The lumped parameter model of is expressed as shown in formula (6): In formula (6), C s is the heat capacity of the bed material, in kJ / ℃; Q s is the physical sensible heat of coal, in kJ / s; Δq air is the change in the calorific value of the wind, in kJ / m 3 ;h pz is the enthalpy of slag discharge, in kJ / kg; h FL is the enthalpy value of fly ash, in kJ / kg; Q1 is the heat absorption between the steam-water medium and the wall, in kJ / s; The heat absorption Q1 is expressed as shown in formula (7): In formula (7), k2 and n are model coefficients; D fw is the water flow rate, in kg / s; T st is the main steam temperature, in °C; Sub-step S13, establishing the soda sub-model, comprises: The lumped parameter model of the metal wall and the steam-water medium is expressed as shown in formula (8): In formula (8), p b is the drum pressure, in MPa; C b is the total heat storage coefficient of soda and metal, in kJ / MPa; h fw is the feed water enthalpy value, in kJ / kg; D sw is the cooling water flow rate, in kg / s; D st Main steam flow rate, unit is kg / s; h st is the main steam enthalpy, in kJ / kg; The feed water enthalpy is related to the output power of the CFB unit. The feed water enthalpy is expressed as shown in formula (9): h fw =f(N e ) (9), In formula (9), N e is the unit output power, in MW; f(·) is N e Function of The main steam pressure is determined by the drum pressure and the superheater pressure loss. The superheater pressure loss is expressed as shown in formula (10): Δp=p b -p st =p(N e ) (10), In formula (10), Δp is the superheater pressure loss, in MPa; p st is the main steam pressure, in MPa; p(·) is about N e Function of The main steam enthalpy of the subcritical CFB unit during deep peak regulation operation is expressed as shown in formula (11): h st =h(p st ,N e ) (11), In formula (11, h st is the main steam enthalpy, in kJ / kg; h(·) is about p st 、N e Function of The main steam temperature is related to the main steam pressure and enthalpy value. The main steam temperature is expressed as shown in formula (12): T st =T(p st ,h st ) (12), In formula (12): T(·) is the st 、h st Function of Step 2 further includes: expressing the main steam flow rate as shown in formula (13): D st =u t D(p st ,h st ) (13), In formula (13): u t is the main steam valve opening, expressed as % content; D(·) is about p st 、h st Function of The output power N of the subcritical CFB unit e It is expressed as shown in formula (14)-(15): <h2 style=";text-align:left;direction:ltr">k3 = k(N<h2 style=";text-align:left;direction:ltr"> e <h2 style=";text-align:left;direction:ltr"> (15) In formula (14) and (15), C n is the inertia time of the steam turbine, in seconds; k(·) is the unit of N e function.

2. The method for predicting output power of a subcritical circulating fluidized bed unit under deep peak regulation according to claim 1, characterized in that: In step 3, f(·), h(·), p(·), D(·), k(·), and Q1 are fitted by nonlinear least squares method, and the functional form of f(·), p(·), D(·), k(·), and Q1 is expressed as shown in formula (16): In formula (16), a1, a2, a3, a4, a5, a6, b1, b2, b3, b4, b5, b6, c1, c4, c6, d1, d4, and d6 are the parameters to be fitted.

3. The method for predicting output power of a subcritical circulating fluidized bed unit under deep peak regulation according to claim 2, characterized in that: In step 4, the dynamic parameter C in the model is identified by using the historical operation data and coal quality parameters using the particle swarm algorithm. s , C b , C n ; The coal quality, coal feed, total air volume, slag discharge, feed water flow, fly ash, main steam valve opening, and desuperheating water flow values ​​in the historical operation data are used as the input of the boiler system model under the deep peak regulation of the subcritical CFB unit to form an initial particle group, and each particle of the group is used as a dynamic parameter value to calculate the output of the boiler system model, and select p st 、N e The sum of the absolute values ​​of the relative errors between the model output value and the actual value is taken as the objective function, which is expressed as shown in formula (17): In formula (17), Δp st,i , ΔN e,i They represent the deviations of the main steam pressure and output power output values ​​of the model at the i-th sampling point from the actual data; After updating the particle positions and velocities, the particle group continuously obtains smaller objective function values, and finally obtains optimized dynamic parameters; The real-time operation data is read from the distributed control system DCS of the deep peak-shaving CFB unit, and the output power prediction model is used in combination with coal quality parameters to predict the output power of the subcritical CFB unit under deep peak-shaving.

Images