CN116910472A - 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method - Google Patents
660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method Download PDFInfo
- Publication number
- CN116910472A CN116910472A CN202310789671.XA CN202310789671A CN116910472A CN 116910472 A CN116910472 A CN 116910472A CN 202310789671 A CN202310789671 A CN 202310789671A CN 116910472 A CN116910472 A CN 116910472A
- Authority
- CN
- China
- Prior art keywords
- circulating fluidized
- fluidized bed
- data
- temperature
- ultra
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000005457 optimization Methods 0.000 title claims abstract description 31
- 230000008878 coupling Effects 0.000 claims abstract description 18
- 238000010168 coupling process Methods 0.000 claims abstract description 18
- 238000005859 coupling reaction Methods 0.000 claims abstract description 18
- 239000003245 coal Substances 0.000 claims abstract description 16
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 15
- 230000004044 response Effects 0.000 claims abstract description 13
- 238000002922 simulated annealing Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 8
- 230000007246 mechanism Effects 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000000137 annealing Methods 0.000 claims description 4
- 239000002893 slag Substances 0.000 claims description 4
- 235000019738 Limestone Nutrition 0.000 claims description 3
- 238000007599 discharging Methods 0.000 claims description 3
- 239000006028 limestone Substances 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000002277 temperature effect Effects 0.000 claims description 3
- 238000013459 approach Methods 0.000 claims description 2
- 238000001816 cooling Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 239000000446 fuel Substances 0.000 description 3
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 description 3
- 238000002485 combustion reaction Methods 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000000571 coke Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006477 desulfuration reaction Methods 0.000 description 1
- 230000023556 desulfurization Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 239000003546 flue gas Substances 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000005514 two-phase flow Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The invention relates to a 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method, which comprises the following steps: selecting relevant characteristic variable parameters affecting the bed temperature and the main steam pressure of the circulating fluidized bed unit; carrying out comprehensive pretreatment on related characteristic variable parameter data; selecting coal feeding quantity, water feeding quantity and primary air quantity as input characteristics; selecting a bed temperature value and a main steam pressure value at the first several moments as input characteristics; dividing the operation conditions of the circulating fluidized bed unit according to the load of the unit; establishing a circulating fluidized bed unit bed temperature and main steam pressure coupling prediction model under different working conditions, and carrying out parameter optimization by using a simulated annealing algorithm to determine dynamic response time steps of the circulating fluidized bed unit bed temperature and the main steam pressure under different working conditions; and a multi-working condition model generalized predictive control algorithm is adopted to predict and control and optimize the 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature. The invention can realize 660MW ultra-supercritical circulating fluidized bed boiler hearth high-quality prediction and control optimization.
Description
Technical Field
The invention belongs to the field of operation optimization of 660MW ultra-supercritical circulating fluidized bed boilers, and particularly relates to a 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method.
Background
The 660MW ultra-supercritical circulating fluidized bed boiler is one of ways for realizing clean and efficient utilization of coal under the 'double carbon' target, has the obvious advantages of multiple combustible fuel types, high operation efficiency, low emission of nitrogen oxides and the like, and has wide development potential. The bed temperature is 660MW ultra supercritical circulating fluidized bed boiler and important state parameters thereof, and has direct influence on the safe and stable operation of the boiler. The desulfurization efficiency of the boiler is easy to be reduced due to the overhigh bed temperature, the emission content of NOx in the flue gas is increased, and the hearth bed material is easy to coke; the combustion efficiency of the boiler is reduced due to the too low bed temperature, and the boiler is unstable in operation and even has a fire extinguishing phenomenon.
The factors influencing the bed temperature are many, the serious coupling among the factors, the delay effect and the gas-solid two-phase flow characteristic in the combustion process make the established mechanism model very complex, and are not suitable for the design and optimization of a control system of a power plant unit. At present, due to the tasks of new energy consumption, frequency modulation and the like of wind power, photovoltaic and the like, the dynamic process of the 660MW ultra-supercritical circulating fluidized bed boiler is more frequent, and the bed temperature performance and control difficulty in the dynamic process are increased. In the running process of the unit, the bed temperature and the main steam pressure in a 660MW ultra-supercritical circulating fluidized bed boiler closed-loop system model under the action of primary air quantity are seriously coupled, and a single-input single-output model and a multi-input single-output model established by a plurality of expert students are difficult to meet the current high-quality prediction and control requirements.
Disclosure of Invention
The invention aims to provide a 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method for solving the technical problems.
The invention provides a 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method, which comprises the following steps:
step 1, according to the actual process structure arrangement and operation flow of a 660MW ultra-supercritical circulating fluidized bed boiler, selecting relevant characteristic variable parameters influencing the bed temperature and the main steam pressure of a circulating fluidized bed unit by combining a power plant measuring point table; the related characteristic variable parameters comprise coal feeding amount, water feeding amount, air introducing amount, primary air amount, secondary air amount, limestone amount, slag discharging amount and material returning amount;
step 2, collecting the relevant characteristic variable parameter data selected in the step 1 from a unit database for comprehensive pretreatment; the comprehensive pretreatment comprises data cleaning treatment and data filtering treatment;
step 3, analyzing the relation between the pretreated related characteristic variable parameter data, the output characteristic bed temperature and the main steam pressure from the mechanism angle based on the strong coupling relation between the main steam pressure and the bed temperature, and selecting the coal feeding amount, the water feeding amount and the primary air quantity as input characteristics;
step 4, based on the time delay characteristic of the temperature response of the 660MW ultra-supercritical circulating fluidized bed boiler, selecting the temperature value of the bed at the first several moments and the pressure value of main steam as input characteristics;
step 5, dividing the operation conditions of the circulating fluidized bed unit according to the load of the unit;
step 6, establishing a circulating fluidized bed unit bed temperature and main steam pressure coupling prediction model under different working conditions according to the working condition division of the step 5, optimizing parameters by using a simulated annealing algorithm, and determining dynamic response time steps of the circulating fluidized bed unit bed temperature and the main steam pressure under different working conditions; the prediction model is as follows:
wherein: t (T), P (T) respectively represent the output of the bed temperature and the main steam pressure at the current moment T; x is x 1 (t)、x 2 (t)、x 3 (t) respectively representing the input of the coal feeding quantity, the water feeding quantity and the primary air quantity at the current time t; m, n, a, b and c respectively represent dynamic response time steps of bed temperature, main steam pressure, coal feeding amount, water feeding amount and primary air quantity, the dynamic response time steps are related to data sampling frequency, the value range is 0-20, and positive integers are taken;
and 7, based on the multi-working condition prediction model established in the step 6, adopting a multi-working condition model generalized prediction control algorithm to replace PID control, and performing 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization.
Further, the data cleaning process and the data filtering process in the step 2 include:
outlier data were removed using the box plot method:
let q 1 、q 3 For the 1 st quartile and the 3 rd quartile of data, the boxplot method can be expressed as:
x max ＝q 3 +1.5×(q 3 -q 1 )
x min ＝q 3 -1.5×(q 3 -q 1 )
wherein x is max Representing outlier maxima in data, x min Is an abnormal minimum value in the data; if the data is smaller than the abnormal minimum value, determining the data larger than the abnormal maximum value as outlier data, and removing the outlier data;
and carrying out data filtering processing by adopting a Kalman filtering method, wherein a Kalman filtering model is shown as follows:
X(k)＝A*X(k-1)+B*U(k)+W(k)
Z(k)＝H*X(k)+V(k)
wherein X (k) is the system state at time k, and U (k) is the control amount of the system at time k; a and B are system parameters, for a multi-model system A and B are matrices; z (k) is a measured value at the moment k, H is a parameter of the measuring system, and H is a matrix for the multi-measuring system; w (k) and V (k) represent noise of the process and the measurement, respectively; and optimally estimating the system state through the system input and output observation data.
Further, in step 5, a clustering algorithm based on projection on the convex set is adopted to divide working conditions.
Further, the clustering algorithm flow is as follows:
algorithm input: a data set X and a cluster number k;
algorithm output: clustering represents a set Y;
1) Randomly and randomly selecting k sample objects from the whole data set X as initial cluster centers;
2) Computing each sample object x in the dataset m To cluster center c i Is a Euclidean distance of (2);
3) Find each sample object x m To cluster center c i And the sample object x is determined m Is classified as being with c i In the same cluster;
4) Calculating a weighted average value of objects in the same cluster, and updating the cluster center; wherein: the weighting coefficients for each object are calculated as follows:
5) Repeating the steps 2) to 4) until the cluster center is not changed any more.
Further, the step of performing parameter optimization by using the simulated annealing algorithm in the step 6 is as follows:
(1) Selecting the iteration number q and the initial control temperature T q Let q=0; an initial state x is selected from the feasible solution space 0 Calculate the objective function value f (x 0 )；
(2) Generating a random disturbance in the feasible solution space, generating a new state x 1 Calculate the objective function value f (x 1 )；
(3) Judging whether to accept according to the state acceptance function: if f (x 1 )<f(x 0 ) Then accept the new state x 1 For the current state, otherwise, deciding whether to accept x according to Metropolis criterion 1 If accepted, let the current state equal to x 1 If not, let the current state equal to x 0 ；
(4) According to the temperatureCooling scheme T q+1 ＝CT q C is E (0, 1), and the control temperature T is reduced q+1 ；
(5) Judging whether the annealing process is terminated, stopping the algorithm when ten continuous new solutions are not accepted or the iteration times q are reached, and turning to the step (6), otherwise turning to the step (2);
(6) And outputting the current solution as an optimal solution.
Further, the step 6 includes:
the established multiple typical working point models approach the characteristics of the whole operation interval of the controlled object, and corresponding controllers are designed for each sub-model;
in actual operation, the limited sub-controller output is mapped into the final control bed temperature effect by switching or weighting.
By means of the scheme, the method for predicting, controlling and optimizing the temperature of the 660MW ultra-supercritical circulating fluidized bed boiler has the following technical effects:
1) The invention realizes 660MW ultra supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization based on 660MW ultra supercritical circulating fluidized bed boiler operation mechanism, combining with intelligent methods such as feature selection, operation condition division based on projection on a convex set, bed temperature and main steam pressure coupling prediction, multiplexing Kuang Moxing generalized prediction control and the like.
2) The invention fully considers the serious coupling of the bed temperature and the main steam pressure in the 660MW ultra-supercritical circulating fluidized bed boiler closed-loop system model under the action of the fuel quantity and the primary air quantity in the running process of the unit (the single-input single-output bed temperature model and the multi-input single-output bed temperature model established by a plurality of expert students are difficult to meet the current control requirements), establishes a dynamic characteristic model of the coupling prediction of the bed temperature and the main steam pressure, and lays a foundation for the automatic control and optimization of the bed temperature of the 660MW ultra-supercritical circulating fluidized bed boiler.
3) The invention further aims at the characteristics of strong nonlinearity, delay and the like of the hearth temperature of the circulating fluidized bed boiler, overcomes the defect of PID post-control, adopts a combination of a multi-model control strategy and generalized predictive control, improves the temperature control quality of the hearth, and enhances the disturbance rejection capability and the self-adaption capability.
The foregoing description is only an overview of the present invention, and is intended to provide a better understanding of the present invention, as it is embodied in the following description, with reference to the preferred embodiments of the present invention and the accompanying drawings.
Drawings
FIG. 1 is a flow chart of a 660MW ultra-supercritical circulating fluidized bed boiler temperature prediction and control optimization method of the invention.
FIG. 2 is an exemplary diagram of a model of a 660MW ultra-supercritical circulating fluidized bed boiler bed and main steam pressure coupling prediction model of the present invention;
FIG. 3 is a generalized predictive control example of a 660MW ultra-supercritical circulating fluidized bed boiler bed Wen Duogong Kuang Moxing of the invention.
Detailed Description
The following describes in further detail the embodiments of the present invention with reference to the drawings and examples. The following examples are illustrative of the invention and are not intended to limit the scope of the invention.
Referring to fig. 1, the embodiment provides a 660MW ultra supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method, which comprises the following steps:
step 1, according to the actual process structure arrangement and operation flow of the 660MW ultra-supercritical circulating fluidized bed boiler, combining with a power plant measuring point table, analyzing from a mechanism angle, and selecting relevant characteristic variable parameters influencing the bed temperature and main steam pressure of the circulating fluidized bed unit, wherein the relevant characteristic variable parameters mainly comprise coal feeding quantity, water feeding quantity, air introducing quantity, primary air quantity, secondary air quantity, limestone quantity, slag discharging quantity, returning quantity and the like.
And 2, collecting the data selected in the step 1 from a unit database, and comprehensively preprocessing the collected data in consideration of distortion abnormal points and high-frequency noise of the collected measurement data.
And 3, analyzing the relation between the input characteristics (related characteristic variable parameters) in the step 2, the output characteristic bed temperature and the main steam pressure in consideration of the strong coupling relation between the main steam pressure and the bed temperature, deleting variables with smaller correlation, and selecting the coal feeding quantity, the water feeding quantity and the primary air quantity as input characteristics.
And 4, considering that the 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature response has a larger time delay, in order to establish an accurate circulating fluidized bed unit bed temperature and main steam pressure coupling prediction model, the bed temperature value and the main steam pressure value at the first few moments are also used as input characteristics.
And 5, the heat accumulation, inertia and dynamic performance of the boiler are greatly changed under the operating condition. And dividing the operation working conditions of the circulating fluidized bed unit according to the load of the unit.
And 6, establishing a bed temperature and main steam pressure coupling prediction model (shown in figure 2) of the circulating fluidized bed unit under different working conditions according to the working condition division in the step 5, and carrying out parameter optimization by using a simulated annealing algorithm to determine dynamic response time steps of output characteristics (bed temperature and main steam pressure) under different working conditions.
And 7, considering that the control effect of a Generalized Predictive Control (GPC) algorithm on a large hysteresis object is obviously better than that of PID, adopting a multi-working-condition model generalized predictive control algorithm based on the multi-working-condition model established in the step 6 to replace PID control, and realizing 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization. As shown in fig. 3.
In this embodiment, in step 2, in order to ensure accuracy of data, in consideration of distortion outliers and high-frequency noise in the acquired measurement data, the acquired data is comprehensively preprocessed, which mainly includes cleaning and filtering. The cleaned data can comprise null value data and outlier data, and further can also comprise data of shutdown working conditions, namely data corresponding to a period when the unit load is 0 or close to 0.
Null data are data with null values at one or more measuring points at a certain moment, and outlier data are data beyond a normal range. And writing a corresponding code to remove null data, and removing outlier data by using a box graph method. Let q 1 、q 3 For the 1 st quartile and the 3 rd quartile of data, the boxplot method can be expressed as:
xmax＝q 3 +1.5×(q 3 -q 1 )
xmin＝q 3 -1.5×(q 3 -q 1 )
wherein x is max Representing outlier maxima in data, x min Is an outlier in the data. If the data is smaller than the abnormal minimum value and larger than the abnormal maximum value, determining the data as outlier data, and removing the outlier data.
Then, the data is subjected to a filtering process.
The method for filtering data adopts a Kalman filtering method, and Kalman filtering (Kalman filtering) is a method for utilizing a linear system state equation, wherein a common model is shown in the following formula:
X(k)＝A*X(k-1)+B*U(k)+W(k)
Z(k)＝H*X(k)+V(k)
in the formula, X (k) is a system state at time k, and U (k) is a control amount of the system at time k. A and B are system parameters and for a multi-model system a and B are matrices. Z (k) is the measurement value at time k, H is a parameter of the measurement system, and H is a matrix for the multi-measurement system. W (k) and V (k) represent noise of the process and measurement, respectively. And optimally estimating the system state through the system input and output observation data. The optimal estimate is considered as a filtering process, since the observed data includes the effects of noise and interference in the system.
In this embodiment, the relationship between the input feature and the output feature in step 2, the bed temperature and the main vapor pressure are analyzed in step 3, and the variables with smaller correlation are deleted. The specific coupling relationships are shown in the following table.
| Variable(s) | Bed temperature | Main steam pressure |
| Amount of coal fed | Strong strength | Strong strength |
| Air intake | Weak and weak | Weak and weak |
| Water supply quantity | Strong strength | Strong strength |
| Primary air quantity | Strong strength | Strong strength |
| Secondary air volume | Weak and weak | In general |
| Slag discharge amount | In general | Weak and weak |
Therefore, the coal feeding amount, the water feeding amount and the primary air quantity are selected from the mechanism angle as input characteristics.
In this embodiment, the operation condition of the circulating fluidized bed unit is divided according to the unit load in step 5, and the operation condition is divided by adopting a clustering method based on projection (Projections onto Convex Sets, POCS) on the convex set. A convex set in mathematics refers to a set in which the line segment between any two points is within the set. While projection is an operation of mapping a point onto a subspace in another space. Given a convex set and a point, it is possible to operate by finding the projection of the point onto the convex set. The projection is the point within the convex set closest to the point, and may be calculated by minimizing the distance between the point and any other point within the convex set. The clustering operation is achieved by mapping features onto a convex set in another space by projection.
The principle of operation of this algorithm is similar to the classical K-Means algorithm, but differs in the way each data point is processed: the K-Means algorithm weights the importance of each data point the same, but the POCS-based clustering algorithm weights the importance of each data point differently, which is proportional to the distance of the data point from the cluster center. The overall algorithm flow is as follows:
algorithm input: data set X, cluster number k
Algorithm output: cluster representative set Y
Step 1: randomly and randomly selecting k sample objects from the whole data set X as initial cluster centers;
step 2: computing each sample object x in the dataset m To cluster center c i Is a Euclidean distance of (2);
step 3: find each sample object x m To cluster center c i And the sample object x is determined m Is classified as being with c i In the same cluster;
step 4, calculating a weighted average value of the objects in the same cluster, and updating the cluster center; wherein: the weighting coefficients for each object are calculated as follows:
and 5, repeating the steps 2 to 4 until the cluster center is not changed.
In this embodiment, the bed temperature and main steam pressure coupling prediction model of the circulating fluidized bed unit under different working conditions is established according to the working condition division in the step 6. The model is as follows:
wherein: t (T), P (T) respectively represent the current time TOutputting the bed temperature and the main steam pressure; x is x 1 (t)、x 2 (t)、x 3 (t) respectively representing the input of the coal feeding quantity, the water feeding quantity and the primary air quantity at the current time t; m, n, a, b and c represent dynamic response time steps of bed temperature, main steam pressure, coal feeding amount, water feeding amount and primary air quantity respectively, and are related to data sampling frequency, and are generally in the range of 0-20, and positive integers are taken.
And step 6, carrying out parameter optimization by using a simulated annealing algorithm, and determining dynamic response time steps with output characteristics (bed temperature and main steam pressure) under different working conditions. The simulated annealing algorithm is a heuristic random search method, not only introduces proper random factors, but also introduces a natural mechanism of the annealing process of the physical system. In the iterative process, the method not only receives the point which makes the objective function value become good, but also receives the point which makes the objective function value become bad with a certain probability, the receiving probability gradually decreases along with the decrease of the temperature, and the algorithm can jump out of the local optimal solution to obtain the global optimal solution, thereby being beneficial to improving the reliability of obtaining the global optimal solution. The solving steps are as follows:
step 1: selecting the iteration number q and the initial control temperature T q (let q=0) and; an initial state x is selected from the feasible solution space 0 Calculate the objective function value f (x 0 )；
Step 2: generating a random disturbance in the feasible solution space, generating a new state x 1 Calculate the objective function value f (x 1 )；
Step 3: judging whether to accept according to the state acceptance function: if f (x 1 )<f(x 0 ) Then accept the new state x 1 For the current state, otherwise, deciding whether to accept x according to Metropolis criterion 1 If accepted, let the current state equal to x 1 If not, let the current state equal to x 0 ；
Step 4: according to temperature cooling scheme T q+1 ＝CT q C is E (0, 1), and the control temperature T is reduced q+1 ；
Step 5: judging whether the annealing process is terminated, stopping the algorithm when ten continuous new solutions are not accepted or the iteration times q are reached, and turning to the step 6, otherwise turning to the step 2;
step 6: the current solution is output as the optimal solution.
In the embodiment, the multi-working-condition model generalized predictive control algorithm based on the multi-working-condition model in the step 7 is adopted to replace PID control, so as to realize 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization. The specific scheme is that characteristics of a plurality of typical working point models established in the step 6 approximate to the whole operation interval of a controlled object are adopted, and corresponding controllers are designed for each sub-model. In actual operation, these limited sub-controller outputs are mapped into the final control bed temperature effect by switching or weighting.
The 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method has the following technical effects:
1) Based on 660MW ultra-supercritical circulating fluidized bed boiler operation mechanism, by combining with feature selection, operation condition division based on projection on convex sets, bed temperature and main steam pressure coupling prediction, multiplexing Kuang Moxing generalized prediction control and other intelligent methods, 660MW ultra-supercritical circulating fluidized bed boiler bed temperature prediction and control optimization are realized.
2) In the running process of the unit, the severe coupling of the bed temperature and the main steam pressure in a 660MW ultra-supercritical circulating fluidized bed boiler closed-loop system model under the action of the fuel quantity and the primary air quantity (the single-input single-output bed temperature model and the multi-input single-output bed temperature model established by a plurality of expert students are difficult to meet the current control requirements) is fully considered, a dynamic characteristic model of the coupling prediction of the bed temperature and the main steam pressure is established, and a foundation is laid for the automatic control and optimization of the bed temperature of the 660MW ultra-supercritical circulating fluidized bed boiler.
3) Aiming at the characteristics of strong nonlinearity, large delay and the like of the hearth temperature of the circulating fluidized bed, the defect of PID post-control is overcome, the combination of a multi-model control strategy and generalized predictive control is adopted, the temperature control quality of the hearth is improved, and the disturbance rejection capability and the self-adaptive capability are enhanced.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, and it should be noted that it is possible for those skilled in the art to make several improvements and modifications without departing from the technical principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.
Claims (6)
1. The 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method is characterized by comprising the following steps of:
step 1, according to the actual process structure arrangement and operation flow of a 660MW ultra-supercritical circulating fluidized bed boiler, selecting relevant characteristic variable parameters influencing the bed temperature and the main steam pressure of a circulating fluidized bed unit by combining a power plant measuring point table; the related characteristic variable parameters comprise coal feeding amount, water feeding amount, air introducing amount, primary air amount, secondary air amount, limestone amount, slag discharging amount and material returning amount;
step 2, collecting the relevant characteristic variable parameter data selected in the step 1 from a unit database for comprehensive pretreatment; the comprehensive pretreatment comprises data cleaning treatment and data filtering treatment;
step 3, analyzing the relation between the pretreated related characteristic variable parameter data, the output characteristic bed temperature and the main steam pressure from the mechanism angle based on the strong coupling relation between the main steam pressure and the bed temperature, and selecting the coal feeding amount, the water feeding amount and the primary air quantity as input characteristics;
step 4, based on the time delay characteristic of the temperature response of the 660MW ultra-supercritical circulating fluidized bed boiler, selecting the temperature value of the bed at the first several moments and the pressure value of main steam as input characteristics;
step 5, dividing the operation conditions of the circulating fluidized bed unit according to the load of the unit;
step 6, establishing a circulating fluidized bed unit bed temperature and main steam pressure coupling prediction model under different working conditions according to the working condition division of the step 5, optimizing parameters by using a simulated annealing algorithm, and determining dynamic response time steps of the circulating fluidized bed unit bed temperature and the main steam pressure under different working conditions; the prediction model is as follows:
wherein: t (T), P (T) respectively represent the output of the bed temperature and the main steam pressure at the current moment T; x is x 1 (t)、x 2 (t)、x 3 (t) respectively representing the input of the coal feeding quantity, the water feeding quantity and the primary air quantity at the current time t; m, n, a, b and c respectively represent dynamic response time steps of bed temperature, main steam pressure, coal feeding amount, water feeding amount and primary air quantity, the dynamic response time steps are related to data sampling frequency, the value range is 0-20, and positive integers are taken;
and 7, based on the multi-working condition prediction model established in the step 6, adopting a multi-working condition model generalized prediction control algorithm to replace PID control, and performing 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization.
2. The 660MW ultra supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method according to claim 1, wherein the data cleaning process and data filtering process in step 2 comprise:
outlier data were removed using the box plot method:
let q 1 、q 3 For the 1 st quartile and the 3 rd quartile of data, the boxplot method can be expressed as:
x max ＝q 3 +1.5×(q 3 -q 1 )
x min ＝q 3 -1.5×(q 3 -q 1 )
wherein x is max Representing outlier maxima in data, x min Is an abnormal minimum value in the data; if the data is smaller than the abnormal minimum value, determining the data larger than the abnormal maximum value as outlier data, and removing the outlier data;
and carrying out data filtering processing by adopting a Kalman filtering method, wherein a Kalman filtering model is shown as follows:
X(k)＝A*X(k-1)+B*U(k)+W(k)
Z(k)＝H*X(k)+V(k)
wherein X (k) is the system state at time k, and U (k) is the control amount of the system at time k; a and B are system parameters, for a multi-model system A and B are matrices; z (k) is a measured value at the moment k, H is a parameter of the measuring system, and H is a matrix for the multi-measuring system; w (k) and V (k) represent noise of the process and the measurement, respectively; and optimally estimating the system state through the system input and output observation data.
3. The 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method according to claim 1, wherein in step 5, a clustering algorithm based on projection on a convex set is adopted for working condition division.
4. The 660MW ultra-supercritical circulating fluidized bed boiler temperature prediction and control optimization method according to claim 3, wherein the clustering algorithm flow is as follows:
algorithm input: a data set X and a cluster number k;
algorithm output: clustering represents a set Y;
1) Randomly and randomly selecting k sample objects from the whole data set X as initial cluster centers;
2) Computing each sample object x in the dataset m To cluster center c i Is a Euclidean distance of (2);
3) Find each sample object x m To cluster center c i And the sample object x is determined m Is classified as being with c i In the same cluster;
4) Calculating a weighted average value of objects in the same cluster, and updating the cluster center; wherein: the weighting coefficients for each object are calculated as follows:
5) Repeating the steps 2) to 4) until the cluster center is not changed any more.
5. The 660MW ultra supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method according to claim 1, wherein the step of performing parameter optimization by using a simulated annealing algorithm in step 6 is as follows:
(1) Selecting the iteration number q and the initial control temperature T q Let q=0; an initial state x is selected from the feasible solution space 0 Calculate the objective function value f (x 0 )；
(2) Generating a random disturbance in the feasible solution space, generating a new state x 1 Calculate the objective function value f (x 1 )；
(3) Judging whether to accept according to the state acceptance function: if f (x 1 )<f(x 0 ) Then accept the new state x 1 For the current state, otherwise, deciding whether to accept x according to Metropolis criterion 1 If accepted, let the current state equal to x 1 If not, let the current state equal to x 0 ；
(4) According to temperature cooling scheme T q+1 ＝CT q C is E (0, 1), and the control temperature T is reduced q+1 ；
(5) Judging whether the annealing process is terminated, stopping the algorithm when ten continuous new solutions are not accepted or the iteration times q are reached, and turning to the step (6), otherwise turning to the step (2);
(6) And outputting the current solution as an optimal solution.
6. The 660MW ultra supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method according to claim 1, wherein the step 6 comprises:
the established multiple typical working point models approach the characteristics of the whole operation interval of the controlled object, and corresponding controllers are designed for each sub-model;
in actual operation, the limited sub-controller output is mapped into the final control bed temperature effect by switching or weighting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310789671.XA CN116910472A (en) | 2023-06-29 | 2023-06-29 | 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310789671.XA CN116910472A (en) | 2023-06-29 | 2023-06-29 | 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116910472A true CN116910472A (en) | 2023-10-20 |
Family
ID=88367560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310789671.XA Pending CN116910472A (en) | 2023-06-29 | 2023-06-29 | 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116910472A (en) |
-
2023
- 2023-06-29 CN CN202310789671.XA patent/CN116910472A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111754030B (en) | Thermal power generating unit power supply coal consumption optimization method based on HAC and RF-GA | |
| CN102494336A (en) | Combustion process multivariable control method for CFBB (circulating fluidized bed boiler) | |
| Tang et al. | Adaptive nonlinear model predictive control of NOx emissions under load constraints in power plant boilers | |
| CN104534507A (en) | Optimal control method for combustion of boiler | |
| CN110263452B (en) | Flue gas time distribution characteristic analysis method and system in flue and denitration system | |
| CN112365045A (en) | Main steam temperature intelligent prediction method based on big data | |
| CN113440990A (en) | EMD-LSTM based outlet SO2Concentration prediction method | |
| CN115077640A (en) | Real-time prediction method for main steam flow in urban solid waste incineration process | |
| CN111401652A (en) | Boiler optimization method and system based on CO online detection | |
| CN114186476A (en) | Boiler combustion online optimization method based on historical working conditions | |
| CN110286593A (en) | It is a kind of based on multi-objective predictive control chemisorption burning after CO2Trapping system control method | |
| CN103593578A (en) | Flue suction force feedback setting method in coke oven heating combustion process | |
| CN116910472A (en) | 660MW ultra-supercritical circulating fluidized bed boiler hearth temperature prediction and control optimization method | |
| US6230495B1 (en) | Method for optimizing fossil-fueled power stations | |
| CN110888401B (en) | Combustion control optimization method and device for thermal generator set and readable storage medium | |
| CN114970766A (en) | Power station unit operation parameter reference value obtaining method based on linear fitting | |
| CN115145152A (en) | Boiler combustion and denitration process collaborative optimization control method | |
| Wang et al. | A hybrid model with combined feature selection based on optimized VMD and improved multi-objective coati optimization algorithm for short-term wind power prediction | |
| Zhou et al. | Prediction Model of Coal Consumption Based on Random Forest Variable Selection and Random-Grid Hyperparametric Optimization Algorithm | |
| CN111852596B (en) | Method for predicting operation parameters and relative power generation coal consumption rate of boiler steam turbine generator unit | |
| Yuxin et al. | Prediction of Boiler Control Parameters Based on LSTM Neural Network | |
| CN116300666A (en) | Power plant boiler operation control method based on XGBoost optimization | |
| CN113705935A (en) | Multi-objective intelligent optimization-based load economic distribution method for cogeneration unit | |
| CN113759714A (en) | Fuzzy clustering prediction control method for ultra-supercritical thermal power generating unit | |
| CN114740713A (en) | Multi-objective optimization control method for wet flue gas desulfurization process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |