Unit coordinated optimization method under energy spillage signal calculation of superheated steam system

Abstract

The application discloses a unit coordination optimization method under energy overflow signal calculation of superheated steam system, which unifies the main steam temperature control system and the unit coordination system, analyzes the heat transfer process of the superheated steam in stages, fits the energy overflow value of the heat transfer process of each stage according to the input amount of the desuperheating water and the deviation amount of the main steam temperature, calculates the equivalent fuel amount, and finally applies the equivalent fuel amount to the fuel given as a feedforward signal, so as to complete the accurate regulation and control of the load, pressure and temperature. The implementation of the method not only satisfies the stable control of the main steam pressure, the main steam temperature and the load, but also synchronously improves the unit AGC regulation performance of the unit by taking the energy overflow signal of the superheated steam as an optimization target.

Description

Technical Field

[0001] This invention belongs to the field of thermal power plant control technology, and relates to a design method for rapid and stable adjustment of thermal power unit load, specifically a unit coordination optimization method based on the calculation of superheated steam system energy overflow signal. Background Technology

[0002] The large-scale development of new energy sources has spurred research into the flexibility of thermal power unit regulation, aiming to achieve rapid frequency and peak shaving to adapt to the grid connection of new energy sources. Traditional regulation methods are mostly based on existing measurement data, researching new control strategies and applying them to traditional unit coordination models to optimize the regulation speed and accuracy for target parameters. However, with in-depth research into the dynamic performance of thermal power units, it has been found that the superheated steam system is crucial for rapid load adjustment, as the heat absorption process of the entire superheated steam system can directly and quickly reflect changes in boiler combustion, flue gas, and heat transfer. Therefore, compared to existing coordinated control systems that only use load and pressure deviations as adjustment quantities, quantifying the energy transfer in the superheated system's heat transfer process in a timely manner and incorporating it as a feedforward into the traditional coordinated control system will significantly improve the speed and accuracy of load and pressure regulation. Furthermore, using superheated steam energy as a feedforward signal avoids the problem of large fluctuations in main steam temperature at different stages during load adjustment.

[0003] The changes in superheated steam energy in thermal power units reflect the unit's ability to quickly adjust load. However, to reduce system coupling, current research on the main steam temperature system and the unit coordination system is mostly conducted separately. In reality, the control of main steam temperature is not only controlled by desuperheating water, but combustion and flue gas are also fundamental factors affecting the main steam temperature. The reason why the current research approach does not include the above factors in the control research of main steam temperature is that once they are involved in regulating the air temperature, they will cause significant interference to the control of the main output variables of the coordination system, namely load and pressure. Summary of the Invention

[0004] To address the aforementioned contradictions, this invention provides a unit coordination optimization method based on superheated steam system energy overflow signal calculation. By introducing the concept of superheated steam energy overflow, the combustion system, flue gas system, and feedwater system are treated as independent variables, while superheated steam energy overflow is treated as the dependent variable. Thus, the superheated steam energy overflow signal acts as a bridge to achieve unified control of the main steam system and the coordination system. The implementation of this method not only satisfies the stable control of main steam pressure, main steam temperature, and load, but also simultaneously improves the unit's AGC (Automatic Guided Control) performance by using the superheated steam energy overflow signal as the optimization target.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A method for unit coordination optimization based on superheated steam system energy overflow signal calculation includes the following steps:

[0007] Step 1: Taking the superheated steam system as the research object, based on the on-site boiler design structure, the heat transfer process of the entire superheated steam system is divided into: Section I low-temperature superheater, Section II screen-type superheater and Section III high-temperature superheater.

[0008] Step 2: Collect data of the thermal power unit under stable operating condition N. This data includes relevant signals that describe the combustion system, flue gas system and steam-water system of the unit at this moment. The fuzzy parameters of the whole domain are obtained by using the fuzzy neural network modeling method. Based on this, the above deviation data after the load deviation ΔN at any load point can be obtained. The difference is calculated to obtain a set of parameter deviations (after the operating condition N is adjusted to N+ΔN).

[0009] Step 3: Using neural network modeling, construct the energy overflow model for the three overheating segments mentioned in Step 1. After completing the full-condition model construction, a comprehensive energy overflow signal will be provided before each load adjustment action is executed.

[0010] Step 4: Based on the above comprehensive energy overflow signal, and using the principle of energy conservation, obtain the equivalent amount of fuel;

[0011] Step 5: Introduce the above fuel quantity as feedforward into the coordination system to achieve precise fuel allocation ahead of time under load command deviation ΔN.

[0012] Compared with the prior art, the present invention has the following advantages:

[0013] (1) Combine the research on the superheated steam system and the coordinated control system of the thermal power unit to reduce the fluctuation of the main steam temperature on the basis of completing the rapid adjustment of the unit load and pressure;

[0014] (2) Using the desuperheating water spray volume as the standard, the energy transfer overflow of the low-temperature superheated steam and the screen-type superheater is obtained respectively; the energy overflow of the high-temperature superheater section is characterized by the main steam increment.

[0015] (3) Quantify the energy overflow of the entire superheated system and add it as a feedforward to the traditional coordination system to optimize the coordination system’s regulation of load, pressure and main steam temperature. Attached Figure Description

[0016] Figure 1 This diagram illustrates the principle of superheated steam energy overflow participating in the coordination system regulation. Detailed Implementation

[0017] The technical solution of the present invention will be further described below with reference to the accompanying drawings, but it is not limited thereto. Any modifications or equivalent substitutions to the technical solution of the present invention that do not depart from the spirit and scope of the technical solution of the present invention should be covered within the protection scope of the present invention.

[0018] This invention provides a unit coordination optimization method based on superheated steam system energy overflow signal calculation. The method integrates the main steam temperature control system and the unit coordination system for unified study. By analyzing the heat transfer process of superheated steam in stages, and using the amount of desuperheating water input and the deviation of the main steam temperature, the superheated steam system energy overflow value for each stage of the heat transfer process is fitted, and the equivalent fuel quantity is calculated. Finally, this is used as a feedforward signal to the fuel setpoint, thereby achieving precise control of load, pressure, and temperature. Figure 1 As shown, the specific steps include the following:

[0019] Step 1: Taking the superheated steam system as the research object, based on the on-site boiler design structure, the heat transfer process of the entire superheated steam system is divided into: Section I low-temperature superheater, Section II screen-type superheater and Section III high-temperature superheater.

[0020] Step 2: Collect data from the thermal power unit under stable operating condition N. This data includes relevant signals describing the unit's combustion system, flue gas system, and steam-water system at this moment. A fuzzy neural network modeling method is used to obtain global fuzzy parameters. Based on this, the above deviation data after the load deviation ΔN at any load point can be obtained. The difference is calculated to obtain a set of parameter deviations (after adjusting operating condition N to N+ΔN), where:

[0021] N is the load value of the unit, which is the output load of the unit under steady state. Since it is a steady state, the output load is equal to the given load N.

[0022] The global fuzzy parameters consist of the following: Describing the parameters of the low-temperature superheater in segment I: load, flue gas volume, and flue gas temperature; Describing the parameters of the screen-type superheater in segment II: load, combustion intensity, and total air volume; Describing the parameters of the high-temperature superheater in segment III: load and total air volume.

[0023] Step 3: Using neural network modeling, construct the energy overflow models for the three overheating segments mentioned in Step 1. After completing the full-condition model construction, a comprehensive energy overflow signal will be provided before each load adjustment action is executed.

[0024] Step 4: Based on the above comprehensive energy overflow signal, and using the principle of energy conservation, obtain the equivalent amount of fuel;

[0025] Step 5: Introduce the above fuel quantity as a feedforward into the coordination system to achieve precise fuel allocation under load deviation ΔN in advance.

[0026] The core innovation of the above method lies in the following two points:

[0027] (1) The primary and secondary desuperheating water volumes are used as substitute signals for the heat absorption and overflow of different superheating sections to characterize the deviation of the superheating system in absorbing the total heat from the boiler. The reason why this reverse construction is possible is that after obtaining the field operation data, by analyzing the changes in the combustion system and flue gas system during the load increase and decrease process, it was found that the unit's desuperheating system can ensure that the main steam temperature is maintained between 535 and 545℃. Therefore, it is assumed that from the operating condition at time t1 to the operating condition at time t2, the heat change of the superheated steam system is analyzed separately as follows:

[0028] ΔQ=Q t2 -Q t1 (1)

[0029] Where ΔQ is the total energy change of the steam system, Q t1 With Q t2 These are the workable quantities of superheated steam under two different operating conditions. The energy changes are mainly reflected in the changes in steam flow rate Δq and temperature ΔT. In order to construct the influence of combustion and flue gas on the heat absorption of the superheated system, it is necessary to find the overflow heat caused by the heat absorption of the superheated system and the change in desuperheating water. This heat signal is the link between the systems, namely the superheated steam overflow energy signal.

[0030] (2) Sub-models were established using neural network modeling, namely “combustion parameter deviation - desuperheating water flow” and “combustion parameter - main steam temperature fluctuation deviation” neural network models.

[0031] Example:

[0032] Step 1: Taking the superheated steam system as the research object, based on the on-site boiler design structure, the heat transfer process of the entire superheated steam system is divided into three sections: Section I low-temperature superheater, Section II screen-type superheater, and Section III high-temperature superheater.

[0033] Step Two: Based on the above theoretical analysis, data collection and processing are performed, including collecting data on load, flue gas volume, flue gas temperature, combustion intensity, total air volume, primary desuperheating water volume, secondary desuperheating water volume, and main steam temperature. A fuzzy neural network model is then constructed, so that for each load, a set of specific parameters related to flue gas volume, flue gas temperature, combustion intensity, total air volume, primary desuperheating water volume, secondary desuperheating water volume, and main steam temperature are derived under the mapping of this fuzzy neural network. Through a large amount of historical data, relevant data (load, flue gas volume, flue gas temperature, combustion intensity, total air volume, primary desuperheating water volume, secondary desuperheating water volume, and main steam temperature) of the thermal power unit under each stable operating condition N can be collected. This data includes relevant signals describing the unit's combustion system, flue gas system, and steam-water system at that moment, resulting in a data set U. i Simultaneously, the set of parameter changes ΔU under the load deviation ΔN is collected. i (i is the number of samples), construct the extended dataset (N, (U i ),ΔN1,ΔU i The method employs a fuzzy neural network to learn and obtain global fuzzy parameters. The learning process consists of two steps: first, obtaining (N, U1), where N is the input and U1 is the output. i For the output, (U1,ΔN1,ΔU1) is obtained a second time, with U1 and ΔN1 as inputs and ΔU1 as output. After obtaining this set of parameters, fuzzy neural network modeling is performed again with ΔU1 (the changes in flue gas volume, flue gas temperature, combustion intensity, and total air volume) as input and the changes in desuperheating water volume and main steam temperature as outputs. The neural network model thus formed includes a data model with load N and ΔN1 as the original inputs and the changes in desuperheating water flow and main steam temperature as the outputs. Based on this, the above deviation data after the load deviation ΔN at any load point can be obtained. The difference is calculated to obtain a set of parameter deviations (after adjusting operating condition N to N+ΔN).

[0034] In this step, the fuzzy method is as follows: training is performed once for every 5% increase in load range between 50% and 100%, and the intermediate load points are distributed using the TS fuzzy method.

[0035] Step 3: Using neural network modeling, construct the energy overflow models for the three overheating segments mentioned in Step 1. After completing the full-condition model construction, a comprehensive energy overflow signal will be provided before each load adjustment action is executed.

[0036] For the boiler structure described above, the heat transfer process of the superheated steam system is analyzed separately. The process is as follows: First, the heat transfer mode of the primary superheater is constructed, then the heat transfer mode of the screen-type superheater is constructed, and finally the heat transfer mode of the secondary superheater is constructed. The analysis method for the above three superheaters uses a neural network analysis method. First, relevant data is collected, with fuel, air volume, and superheated steam temperature as input signals and desuperheating water or main steam temperature as output signals. The signal deviation of the heat transfer is obtained in stages, and then integrated to obtain a comprehensive signal. The data processing and fitting method first obtains the deviation of relevant data in each segment when the load N→N+ΔN, where N is a stable operating condition, and ΔN is the load adjustment amount under the stable operating condition. The corresponding relationship of each segment is shown in Table 1.

[0037] Table 1

[0038]

[0039] Step 4: Based on the above comprehensive energy overflow signal, the equivalent amount of fuel is obtained according to the principle of energy conservation.

[0040] The calculation of the spillover energy Q consists of the following three parts:

[0041] Q = Q s1 +Q s2 +Q s3 (2)

[0042] Among them, Q s1 For the energy overflow from the low-temperature superheated section I; Q s2 For the overflow energy of the segmented II-panel overheating section; Q s3 This refers to the energy overflow from the high-temperature superheated section III; here Q s1 Q s2 And Q s3 The calculation is as follows:

[0043] Q s1 =q1*(T1-T0) (3)

[0044] Where q1 is the flow rate of the first-stage desuperheating water, T1 is the outlet temperature of the low-temperature superheater, and T0 is the feed water temperature. Here, the feed water temperature is assumed to be the outlet temperature of the desuperheating water.

[0045] Q s2 =q2*(T2-T0) (4)

[0046] Where q2 is the flow rate of the secondary desuperheating water, and T2 is the outlet temperature of the screen-type superheater;

[0047] Q s3 The calculation involves the enthalpy-entropy change of steam, which will now be elaborated in detail, when the unit load changes from N... iChange to N s At that time, its main steam flow rate q i The change is q s Changes:

[0048] Δq=q s -q i (5)

[0049] Where, N i This refers to a specific load point at which the unit begins adjustment, N. s This is the target load under this adjustment action.

[0050] To calculate the change in steam energy, the energy per kilogram of steam before the engine is now used to characterize the energy level. si (N i ,P i The effective energy of superheated steam is calculated as follows:

[0051] E si (N i ,P i )=(H i -H0)-T0(S i -S0) (6)

[0052] Among them, H i H is the enthalpy in the initial state, kJ / kg; H0 is the ground state enthalpy, kJ / kg; S i Let H0 be the initial entropy value, kJ / (kg·K); S0 be the ground state entropy value, kJ / (kg·K); and T0 be the ground state temperature, K. The initial enthalpy is taken from the vapor enthalpy and entropy value at (450℃, 10MPa), and calculated to be H0 = 192.4346 kJ / (kg·K) and S0 = 0.3806 kJ / (kg·K).

[0053] The above enthalpy value can be obtained by the calculation method of steam parameters. That is, when the easily measurable pressure p and temperature t are selected as independent variables, the following formula (7) is given. The calculation method can be found in the enthalpy-entropy calculation table (industry consensus experimental data) in the reference:

[0054] H=f0(p,t) (7)

[0055] Similarly, the entropy of superheated steam at the above temperatures and pressures is calculated using the entropy calculation model adopted in the existing enthalpy-entropy calculation table (industry consensus experimental data) as follows:

[0056]

[0057] Where T is Kelvin temperature (K), P is standard atmospheric pressure (atm), and S is in kJ / (kg·K). The conversion relationship is as follows:

[0058] 1*P=101.325Kpa (9)

[0059] The method of calculating the enthalpy of superheated steam by dividing the region into segments, as used in the reference (Tang Qingsheng. Practical formula for calculating the enthalpy of superheated steam [J]. Electric Power Technology, 1983(03):40-42.), is as follows:

[0060]

[0061] The above steam pressure range is 80–180 atmospheres, and the temperature range is 450–570℃. The unit of H is KJ / KG. E in formula (6) can be calculated using formulas (7), (8), (9), and (10). si (N i ,P i ), combined with load point N i Traffic q i Steam can be used to generate workable energy Q. ssi Similarly, load point N s Combining the above formula, the following calculation data model for instantaneous energy storage of superheated steam in front of the turbine can be obtained:

[0062] Q ssi =q i ·E si (N i ,P i (11)

[0063] The instantaneous energy storage change of superheated steam between two points can be calculated using the formula (5) Δq. SS (N i N s )for:

[0064] ΔQ ss (N i N s )=Q sss -Q ssi (11)

[0065] Where, ΔQ SS (N i N s Q represents the instantaneous change in the stored energy of superheated steam between two points. ssi For load point N i Steam can be used as work energy, Q sss For load point N s Steam can be used as work energy, here ΔQ SS (Ni N s Q is the equivalent of Q. s3 .

[0066] Step 5: Introduce the above fuel quantity as feedforward into the coordination system to achieve precise fuel allocation ahead of time under load command deviation ΔN. Convert the above instantaneous energy storage deviation into the unit's equivalent hourly power (equivalent load deviation):

[0067] ΔN ss =ΔQ ss (N i N s ) / q0 (12)

[0068] Where, q0 = 3.6 × 10 6 J (joules), ΔN ss The instantaneous equivalent load deviation is used as a feedforward to incorporate into the traditional coordinated control system, thereby optimizing fuel supply.

Claims

1. A unit coordination optimization method based on superheated steam system energy overflow signal calculation, characterized in that... The method includes the following steps: Step 1: Taking the superheated steam system as the research object, based on the on-site boiler design structure, the heat transfer process of the entire superheated steam system is divided into: segments. Low-temperature superheater, segmented II-panel superheater and segmented High-temperature superheater; Step Two: Collect data from the thermal power unit under stable operating condition N. This data includes relevant data describing the unit's combustion system, flue gas system, and steam-water system at this moment. The relevant data includes load, flue gas volume, flue gas temperature, combustion intensity, total air volume, primary desuperheating water volume, secondary desuperheating water volume, and main steam temperature. Construct a fuzzy neural network model so that each load, under the mapping of this fuzzy neural network, yields a set of parameters related to flue gas volume, flue gas temperature, combustion intensity, total air volume, primary desuperheating water volume, secondary desuperheating water volume, and main steam temperature. Filter the collected data from a large amount of historical data under each stable operating condition N to obtain a data set U. i Simultaneously, the set of parameter changes ΔU under the load deviation ΔN is collected. i Let i be the number of samples, and construct an extended dataset (N, U). i , ΔN1,ΔU i The fuzzy neural network method is used to learn the global fuzzy parameters. The learning process is divided into two steps: the first step is to obtain (N, U1), where N is the input and U1 is the output. i For the output, (U1, ΔN1, ΔU1) is obtained a second time. U1 and ΔN1 are the inputs, and ΔU1 is the output. After obtaining this set of parameters, fuzzy neural network modeling is performed again, with ΔU1 as the input. ΔU1 includes the changes in flue gas volume, flue gas temperature, combustion intensity, and total air volume. The changes in desuperheating water volume and main steam temperature are used as the outputs for modeling. The neural network model thus formed includes a data model with load N and ΔN1 as the original inputs and the changes in desuperheating water flow and main steam temperature as the outputs. Based on this, the output data of the data model after superimposing the load deviation ΔN at any load point is obtained. The difference between the output data and the collected data corresponding to the operating condition N is calculated to obtain a set of parameter deviations after the operating condition N is adjusted to N+ΔN. Step 3: Using neural network modeling, construct the energy overflow models for the three overheating segments mentioned in Step 1. After completing the full-condition model construction, a comprehensive energy overflow signal will be provided before each load adjustment action is executed. The comprehensive energy overflow signal is segmented... Low-temperature superheated section overflow energy, segmented II-screen superheated section overflow energy, and segmented The sum of the energy overflowing from the high-temperature overheating section, the segmented... The energy overflowing from the high-temperature superheated section is the instantaneous change in the stored energy of the superheated steam between two points, determined by the unit load from N. i Change to N s At that time, its main steam flow rate q i The change is q s The change Δq is calculated to obtain N. i This refers to a specific load point at which the unit begins adjustment, N. s This is the target load under this adjustment action; Step 4: Based on the above comprehensive energy overflow signal, and using the principle of energy conservation, obtain the equivalent amount of fuel; Step 5: Introduce the above fuel quantity as feedforward into the coordination system to achieve precise fuel allocation under load deviation ΔN.

2. The unit coordination optimization method based on superheated steam system energy overflow signal calculation according to claim 1, characterized in that... In step two, the global fuzzy parameters consist of the following: parameters describing segment I low-temperature superheater: load, flue gas volume, and flue gas temperature; parameters describing segment II screen-type superheater: load, combustion intensity, and total air volume; and parameters describing segment III high-temperature superheater: load and total air volume.

3. The unit coordination optimization method based on superheated steam system energy overflow signal calculation according to claim 1, characterized in that... In step four, the calculation of the overall overflow energy Q consists of the following three parts: Among them, Q s1 For the energy overflow from the low-temperature superheated section I; Q s2 For the overflow energy of the segmented II-panel overheating section; Q s3 This refers to the overflow energy from the high-temperature overheating section III.

4. The unit coordination optimization method based on superheated steam system energy overflow signal calculation according to claim 3, characterized in that... The Q s1 The calculation method is as follows: Where q1 is the flow rate of the first-stage desuperheating water, T1 is the outlet temperature of the low-temperature superheater, and T0 is the feedwater temperature.

5. The unit coordination optimization method based on superheated steam system energy overflow signal calculation according to claim 3, characterized in that... The Q s2 The calculation method is as follows: Where q2 is the flow rate of the secondary desuperheating water, T2 is the outlet temperature of the screen-type superheater, and T0 is the feedwater temperature.

6. The unit coordination optimization method based on superheated steam system energy overflow signal calculation according to claim 3, characterized in that... The Q s3 The calculation method is as follows: Where, ΔQ SS (N i N s Q represents the instantaneous change in the stored energy of superheated steam between two points. ssi For load point N i Steam can be used as work energy, Q sss For load point N s Steam can be used as work energy, here ΔQ SS (N i N s Q is the equivalent of Q. s3 .

Images