CN115326237A - Wall temperature measuring method and system considering influence of boiler superheater oxide skin - Google Patents
Wall temperature measuring method and system considering influence of boiler superheater oxide skin Download PDFInfo
- Publication number
- CN115326237A CN115326237A CN202210961018.2A CN202210961018A CN115326237A CN 115326237 A CN115326237 A CN 115326237A CN 202210961018 A CN202210961018 A CN 202210961018A CN 115326237 A CN115326237 A CN 115326237A
- Authority
- CN
- China
- Prior art keywords
- wall
- wall temperature
- superheater
- boiler
- boiler superheater
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F22—STEAM GENERATION
- F22B—METHODS OF STEAM GENERATION; STEAM BOILERS
- F22B37/00—Component parts or details of steam boilers
- F22B37/02—Component parts or details of steam boilers applicable to more than one kind or type of steam boiler
- F22B37/38—Determining or indicating operating conditions in steam boilers, e.g. monitoring direction or rate of water flow through water tubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physiology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Thermal Sciences (AREA)
- Mechanical Engineering (AREA)
- Measuring Temperature Or Quantity Of Heat (AREA)
Abstract
The invention discloses a wall temperature measuring method and system considering the influence of oxide skin of a boiler superheater, belonging to the technical field of boiler safety and comprising the following steps: obtaining influence parameters of the wall temperature of the inner wall of the boiler superheater based on an expert algorithm, and constructing a training set based on the influence parameter data and the corresponding wall temperature data of the inner wall of the boiler superheater; inputting the training set into a prediction model of the wall temperature of the inner wall of the superheater based on a machine learning algorithm to obtain a trained prediction model; predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment; measuring the thickness of the oxide layer on the inner wall of the boiler superheater based on an ultrasonic testing system to obtain the thickness of the oxide layer of the boiler superheater at the moment; and when the thickness of the oxide layer is larger than a first threshold value, correcting the wall temperature of the inner wall of the boiler superheater to obtain the corrected wall temperature of the inner wall of the boiler superheater, so that the wall temperature measurement of the boiler superheater becomes more reasonable, stable and reliable.
Description
Technical Field
The invention belongs to the technical field of boiler safety, and particularly relates to a wall temperature measuring method and system considering the influence of oxide skin of a boiler superheater.
Background
Nowadays, china already puts into operation ultra-supercritical units in many provinces, and the ultra-supercritical power generation technology further replaces the supercritical power generation technology and becomes the mainstream technology of thermal power generation. The ultra-supercritical unit reduces the coal consumption of power generation and pollutant discharge, and simultaneously, because the working medium parameters are high, the requirement on high temperature resistance is correspondingly improved, especially in the aspects of high temperature resistance, creep resistance and oxidation, if the material performance can not meet the requirement, the ultra-temperature pipe explosion accident is easy to occur in the operation process of the boiler, and the safe and economic operation of a power plant is influenced. During the operation of the boiler, because of the reasons of large heat load, high working medium parameters and the like, boiler accidents caused by four-pipe explosion leakage are more and more frequent, particularly high-temperature superheaters, the working environment of the high-temperature superheaters is more severe, and the high-temperature superheaters are simultaneously subjected to the effects of radiation heat exchange and convection heat exchange, so that the heat load of the pipelines is high, the wall temperature of the heated pipelines is close to the highest allowable temperature for a long time, and the overtemperature pipe explosion accidents are easy to occur at the moment, and the safe operation of the boiler is seriously influenced.
The research and prevention on the boiler superheater at home and abroad mainly measures the temperature through a thermocouple arranged at the outer pipe wall of an inlet and an outlet of the superheater, the temperature change trend of the outer wall of the boiler is close to the temperature of the inner wall, so that the temperature of the inner wall of the boiler is deduced, the temperature of the inner wall is determined through artificial intelligence or the thermal deviation theory of a pipeline, but in the operation process, the inner wall of the boiler superheater generates oxide skin, the temperature of the inner wall is increased due to overlarge thickness of the oxide skin, the thermal conductivity coefficient of the oxide skin is less than 1W/(m.K), the thermal conductivity coefficient of T92 steel is 27W/(m.K), when the artificial intelligence algorithm obtained through training without the oxide skin or the thermal deviation theory of the pipeline is adopted to carry out soft measurement on the temperature of the inner wall at the moment, the error at the moment is very large, the temperature of the inner wall at the moment cannot be truly reflected, and the temperature of the inner wall of the superheater cannot be accurately mastered at the moment seriously, so that the overtemperature tube explosion accident occurs.
Based on the technical problems, a wall temperature measuring method and system considering the influence of oxide skin of a boiler superheater need to be designed.
Disclosure of Invention
The invention aims to provide a wall temperature measuring method and system considering the influence of oxide skin of a boiler superheater.
In order to solve the technical problem, a first aspect of the present invention provides a wall temperature measuring method considering an influence of a scale of a boiler superheater, including:
s1, obtaining influence parameters of the wall temperature of the inner wall of the boiler superheater based on an expert algorithm, and constructing a training set based on the influence parameter data and the corresponding wall temperature data of the inner wall of the boiler superheater;
s2, inputting the training set into a superheater inner wall temperature prediction model based on a machine learning algorithm to obtain a trained prediction model;
s3, predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment;
s4, measuring the thickness of the oxide layer on the inner wall of the boiler superheater based on an ultrasonic testing system to obtain the thickness of the oxide layer of the boiler superheater at the moment;
and S5, when the thickness of the oxide layer is larger than a first threshold value, correcting the wall temperature of the inner wall of the boiler superheater to obtain the corrected wall temperature of the inner wall of the boiler superheater.
The influence parameters of the wall temperature of the inner wall of the boiler superheater are obtained based on an expert algorithm, the corresponding data of the wall temperature of the inner wall of the boiler superheater are obtained, the data of the influence parameters serve as input quantity, the data of the inner wall temperature of the superheater serve as output quantity, a prediction model based on a machine learning algorithm is trained, the prediction model is obtained on the basis, the wall temperature of the inner wall of the boiler superheater at the moment can be known through the prediction model, in order to enable the prediction result to be more accurate, a mathematical mechanism model of the thickness of the oxidation layer of the inner wall of the boiler superheater is built, the thickness of the oxidation layer of the boiler at the moment is obtained, the wall temperature of the inner wall of the boiler superheater is corrected after the thickness is larger than a first threshold value, the measurement of the wall temperature of the inner wall of the boiler at the moment is comprehensively considered on the basis of the thickness of the oxidation layer, and the accurate measurement of the wall temperature of the inner wall of the boiler is achieved.
The influence parameters of the wall temperature of the inner wall of the boiler superheater are extracted by adopting an expert algorithm, so that the obtained input data concentrated by training becomes more accurate, and the prediction efficiency of the wall temperature of the inner wall of the boiler superheater is improved. The superheater inner wall temperature prediction model based on the machine learning algorithm is built, and the superheater inner wall temperature at the moment is obtained through prediction, so that the prediction of the superheater inner wall temperature of the boiler becomes more accurate, and the overall precision is improved. Through the mathematical mechanism model of founding boiler superheater inner wall oxide layer thickness, thereby can calculate the boiler superheater inner wall oxide layer thickness that obtains this moment, and when being greater than first threshold value, it is right boiler superheater inner wall temperature revises, make the true condition of the reaction this moment that the boiler superheater inner wall temperature data that obtain this moment can be more accurate, because boiler inner wall oxide layer thickness is too big, the data difference when leading to boiler superheater inner wall temperature data this moment and no oxide layer is very big, adopt the correction method, revise boiler inner wall temperature data, make the error at this moment obtain further correction, thereby the condition of reaction this moment that can be accurate, avoid the appearance of the pipe explosion condition that leads to because boiler superheater inner wall temperature transfinites.
The further technical scheme is that the influence parameters of the wall temperature of the inner wall of the superheater comprise power of a generator, main steam pressure, main steam temperature, water supply flow, coal supply quantity, primary air quantity, secondary air quantity, superheated steam outlet pressure, superheated steam flow and superheated swallow steam outlet temperature.
The influence parameters of the wall temperature of the inner wall of the superheater include power of a generator, main steam pressure, main steam temperature, water supply flow, coal supply amount, primary air amount, secondary air amount, superheated steam outlet pressure, superheated steam flow and superheated swallow steam outlet temperature through an expert algorithm, and the problem that the original calculation speed is slow due to too much extraction amount or the calculation accuracy is low due to too little extraction amount is avoided.
The further technical scheme is that before a training set is constructed based on the influence parameter data and corresponding boiler superheater inner wall temperature data, the influence parameter data are subjected to data cleaning through a mean value filtering algorithm.
Before a training set is constructed based on the influence parameter data and the corresponding boiler superheater inner wall temperature data, the influence parameter data are subjected to data cleaning, some error data or error data are screened, so that the influence parameter data can be accurate, and the problem that the error is high or the convergence speed is slow due to inaccurate parameter data is solved.
The further technical scheme is that the data of the wall temperature of the inner wall of the boiler superheater comprises inlet temperature, middle section temperature and outlet temperature.
Because its temperature data of boiler superheater's different sections are also different at boiler superheater's data of boiler superheater wall temperature, come to characterize boiler superheater wall temperature data through the temperature data of sectional type for the characterization to boiler superheater wall temperature becomes more accurate this moment, thereby makes the more accurate cognition to boiler superheater wall temperature that the dispatch operation personnel can be more accurate.
The further technical scheme is that the machine learning algorithm adopts an FOA-GRNN algorithm, and the specific steps are as follows:
s1: initializing parameters, and initializing the group positions of the drosophila;
s2: giving the fruit fly individual a random distance and direction for searching food by using smell, and obtaining the random position of the fruit fly individual at the moment;
s3: calculating a judgment value of the odor concentration at the moment, substituting the judgment value into an odor concentration judgment function, obtaining the odor concentration of the fruit fly individual at the moment and marking the position of the fruit fly individual;
s4, finding out the odor concentration value and the position of the fruit fly with the highest taste concentration in the fruit fly group, wherein the fruit fly group flies to the position by using vision;
s5, performing iterative optimization, repeating S3 and S4, and ending the cycle when the iteration is performed until the target times or the odor concentration value reaches the maximum;
and S6, constructing a prediction model of the inner wall temperature of the superheater by taking the obtained odor concentration value as a smoothing factor of the GRNN algorithm at the moment.
The smoothing factor of the GRNN algorithm is optimized through the drosophila algorithm, so that the prediction efficiency of the GRNN algorithm is higher, the calculation speed of the wall temperature of the boiler superheater is higher, and the wall temperature of the boiler superheater at the moment can be accurately and quickly obtained.
A further technical solution consists in that the value of the odor concentration at the time when the value of the odor concentration reaches a maximum is no longer superior to the value of the odor concentration of the preceding iteration.
The further technical scheme is that the boiler superheater oxide layer thickness obtained by calculation based on the mechanism model at the moment comprises an inlet superheater oxide layer thickness, a middle superheater oxide layer thickness and an outlet superheater oxide layer thickness.
Because the temperature and the environmental condition of the boiler superheater are the same at different positions, the thickness of the boiler superheater oxide layer at the moment is calculated and obtained on the basis of the mechanism model and comprises the thickness of an inlet superheater oxide layer, the thickness of a middle section superheater oxide layer and the thickness of an outlet superheater oxide layer, so that the wall temperature is corrected more accurately according to the thickness of the superheater oxide layer at the moment.
The further technical scheme includes that the method further comprises a second threshold value, and when the thickness of the oxide layer is larger than the second threshold value, the machine learning algorithm-based superheater inner wall temperature prediction model is retrained based on real-time data at the moment.
When the thickness of the oxide layer is too thick, the superheater inner wall temperature prediction model based on the machine learning algorithm cannot correctly reflect the boiler water wall temperature at the moment, and is retrained by acquiring real-time data, so that the superheater inner wall temperature prediction model can better reflect the boiler water wall temperature at the moment, and the prediction precision is further improved
The further technical scheme is that the first threshold value and the second threshold value are determined according to the position of a superheater, the type and the capacity of a boiler.
The first threshold value and the second threshold value are adjusted according to the state of the boiler instead of being fixed, and the influence of the thickness of the oxide layer of the superheater of different boilers is different, so that the setting of the thickness threshold value of the oxide layer becomes more scientific.
The invention provides a wall temperature measuring system considering the influence of oxide skin of a boiler superheater, and the wall temperature measuring method considering the influence of the oxide skin of the boiler superheater comprises the following steps:
the system comprises a boiler superheater inner wall temperature prediction module, a boiler superheater oxide layer calculation module and a result output module, wherein the boiler superheater inner wall temperature prediction module is used for obtaining an influence parameter of the wall temperature of the boiler superheater inner wall based on an expert algorithm and constructing a training set based on the influence parameter data and the corresponding boiler superheater inner wall temperature data; inputting the training set into a prediction model of the wall temperature of the inner wall of the superheater based on a machine learning algorithm to obtain a trained prediction model; predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment; the boiler superheater oxidation layer calculation module is responsible for measuring the thickness of the boiler superheater inner wall oxidation layer based on an ultrasonic testing system to obtain the thickness of the boiler superheater oxidation layer at the moment; and the result output module is responsible for correcting the wall temperature of the inner wall of the boiler superheater when the thickness of the oxidation layer is larger than a first threshold value, so as to obtain the corrected wall temperature of the inner wall of the boiler superheater.
Additional features and advantages will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a wall temperature measurement method considering the influence of scale on a boiler superheater in example 1;
FIG. 2 is a flowchart of the FOA-GRNN algorithm-based prediction steps in example 1;
FIG. 3 is a schematic diagram of a wall temperature measuring system in accordance with example 2 in which the influence of scale on a superheater of the boiler is taken into consideration;
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Nowadays, china already puts into operation ultra-supercritical units in many provinces, and the ultra-supercritical power generation technology further replaces the supercritical power generation technology and becomes the mainstream technology of thermal power generation. The ultra-supercritical unit reduces the coal consumption of power generation and pollutant emission, and simultaneously, due to high working medium parameters, the requirement on high temperature resistance is correspondingly improved, especially in the aspects of high temperature resistance, creep resistance and oxidation, if the material performance cannot meet the requirement, an over-temperature pipe explosion accident is easy to occur in the operation process of a boiler, and the safe and economic operation of a power plant is influenced. In the operation of a boiler, due to the reasons of large heat load, high working medium parameters and the like, boiler accidents caused by four-pipe explosion leakage are more and more, particularly, a high-temperature superheater has a more severe working environment and is simultaneously subjected to the effects of radiation heat exchange and convection heat exchange, the heat load of a pipeline is high, the wall temperature of a heated pipeline is close to the maximum allowable temperature for a long time, and the overtemperature explosion accident easily occurs at the moment, so that the safe operation of the boiler is seriously influenced.
Research and prevention to the boiler over heater at home and abroad, mainly measure the temperature through the thermocouple of installing in the exit outer tube wall department of over heater, be close to the inner wall temperature because of boiler outer wall temperature variation trend, deduce boiler inner wall temperature with this, confirm the inner wall temperature through the thermal deviation theory of artificial intelligence or pipeline, but because at the operation in-process, boiler over heater inner wall can produce the cinder, this cinder can make the temperature of inner wall temperature rise, the artificial intelligence algorithm that the training when adopting the no cinder obtained or the thermal deviation theory of pipeline carry out soft measurement to the inner wall temperature this moment, the error that can lead to this moment is very big, inner wall temperature this moment of true reaction can not be, owing to can not accurately master over heater inner wall temperature during serious, thereby lead to over heater emergence overtemperature pipe explosion accident.
Example 1
FIG. 1 is a flow chart of a wall temperature measuring method considering the influence of oxide skin of a boiler superheater according to the invention, which comprises the following steps:
s1, obtaining influence parameters of the wall temperature of the inner wall of the boiler superheater based on an expert algorithm, and constructing a training set based on the influence parameter data and the corresponding wall temperature data of the inner wall of the boiler superheater;
s2, inputting the training set into a superheater inner wall temperature prediction model based on a machine learning algorithm to obtain a trained prediction model;
s3, predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment;
s4, measuring the thickness of the oxide layer on the inner wall of the boiler superheater based on an ultrasonic testing system to obtain the thickness of the oxide layer of the boiler superheater at the moment;
and S5, when the thickness of the oxide layer is larger than a first threshold value, correcting the wall temperature of the inner wall of the boiler superheater to obtain the corrected wall temperature of the inner wall of the boiler superheater.
The influence parameters of the wall temperature of the inner wall of the boiler superheater are obtained based on an expert algorithm, the corresponding data of the wall temperature of the inner wall of the boiler superheater are obtained, the data of the influence parameters serve as input quantity, the data of the inner wall temperature of the superheater serve as output quantity, a prediction model based on a machine learning algorithm is trained, the prediction model is obtained on the basis, the wall temperature of the inner wall of the boiler superheater at the moment can be known through the prediction model, in order to enable the prediction result to be more accurate, a mathematical mechanism model of the thickness of the oxidation layer of the inner wall of the boiler superheater is built, the thickness of the oxidation layer of the boiler at the moment is obtained, the wall temperature of the inner wall of the boiler superheater is corrected after the thickness is larger than a first threshold value, the measurement of the wall temperature of the inner wall of the boiler at the moment is comprehensively considered on the basis of the thickness of the oxidation layer, and the accurate measurement of the wall temperature of the inner wall of the boiler is achieved.
The influence parameters of the wall temperature of the inner wall of the boiler superheater are extracted by adopting an expert algorithm, so that the obtained input data concentrated by training becomes more accurate, and the prediction efficiency of the wall temperature of the inner wall of the boiler superheater is improved. The superheater inner wall temperature prediction model based on the machine learning algorithm is built, and the superheater inner wall temperature at the moment is obtained through prediction, so that the prediction of the boiler superheater inner wall temperature becomes more accurate, and the integral precision is improved. The method comprises the steps of establishing a mathematical mechanism model of the thickness of an oxidation layer on the inner wall of the boiler superheater, calculating the thickness of the oxidation layer on the inner wall of the boiler superheater at the moment, correcting the wall temperature of the inner wall of the boiler superheater when the thickness of the oxidation layer is larger than a first threshold value, enabling the wall temperature data of the inner wall of the boiler superheater obtained at the moment to reflect the real situation at the moment more accurately, adopting a correction method to correct the wall temperature data of the inner wall of the boiler, enabling the error at the moment to be further corrected, accurately reflecting the situation at the moment, and avoiding the occurrence of pipe explosion caused by the fact that the wall temperature of the inner wall of the boiler superheater exceeds the limit.
In another possible embodiment, the parameters influencing the wall temperature of the superheater include power of a generator, main steam pressure, main steam temperature, feed water flow, coal supply quantity, primary air quantity, secondary air quantity, superheated steam outlet pressure, superheated steam flow and superheated swallow steam outlet temperature.
The influence parameters of the wall temperature of the inner wall of the superheater include power of a generator, main steam pressure, main steam temperature, water supply flow, coal supply amount, primary air amount, secondary air amount, superheated steam outlet pressure, superheated steam flow and superheated swallow steam outlet temperature through an expert algorithm, and the problem that the original calculation speed is slow due to too much extraction amount or the calculation accuracy is low due to too little extraction amount is avoided.
In another possible embodiment, the influence parameter data is data-cleaned by a mean filtering algorithm before a training set is constructed based on the influence parameter data and corresponding boiler superheater inner wall temperature data.
Before a training set is constructed based on the influence parameter data and the corresponding boiler superheater inner wall temperature data, the influence parameter data are subjected to data cleaning, some error data or error data are screened, so that the data can be accurate, and the problem that the error is high or the convergence speed is low due to inaccurate parameter data is solved.
In another possible embodiment, the data of the wall temperature of the inner wall of the boiler superheater comprises an inlet temperature, a middle section temperature and an outlet temperature.
Because its temperature data of boiler superheater's different sections are also different at boiler superheater's data of boiler superheater wall temperature, come to characterize boiler superheater wall temperature data through the temperature data of sectional type for the characterization to boiler superheater wall temperature becomes more accurate this moment, thereby makes the more accurate cognition to boiler superheater wall temperature that the dispatch operation personnel can be more accurate.
In another possible embodiment, the machine learning algorithm is a FOA-GRNN algorithm, which includes the following specific steps:
s1: initializing parameters, and initializing the fruit fly colony position;
s2: giving the fruit fly individual a random distance and direction for searching food by using smell, and obtaining the random position of the fruit fly individual at the moment;
s3: calculating the odor concentration judgment value at the moment, bringing the judgment value into an odor concentration judgment function, obtaining the odor concentration of the fruit fly individual at the moment, and marking the position of the fruit fly individual;
s4, finding out the odor concentration value and the position of the fruit fly with the highest taste concentration in the fruit fly group, wherein the fruit fly group flies to the position by using vision;
s5, performing iterative optimization, repeating S3 and S4, and ending the cycle when the iteration is performed until the target times or the odor concentration value reaches the maximum;
and S6, constructing a prediction model of the inner wall temperature of the superheater by taking the obtained odor concentration value as a smoothing factor of the GRNN algorithm at the moment.
The smoothing factor of the GRNN algorithm is optimized through the drosophila algorithm, so that the prediction efficiency of the GRNN algorithm is higher, the calculation speed of the wall temperature of the boiler superheater is higher, and the wall temperature of the boiler superheater at the moment can be accurately and quickly obtained.
In a further possible embodiment, the value of the odor concentration at which the value of the odor concentration reaches a maximum is no longer superior to the value of the odor concentration of the previous iteration.
In another possible embodiment, the thicknesses of the boiler superheater oxide layers at the moment are calculated and obtained based on the mechanism model and comprise an inlet superheater oxide layer thickness, a middle superheater oxide layer thickness and an outlet superheater oxide layer thickness.
Because the temperature and the environmental condition of the boiler superheater are the same at different positions, the thickness of the oxidation layer of the boiler superheater at the moment comprises the thickness of the oxidation layer of an inlet superheater, the thickness of the oxidation layer of a middle section superheater and the thickness of the oxidation layer of an outlet superheater, which are calculated based on the mechanism model, so that the wall temperature can be corrected more accurately according to the thickness of the oxidation layer of the superheater at the moment.
In another possible embodiment, the method further comprises a second threshold value, and when the thickness of the oxide layer is larger than the second threshold value, the machine learning algorithm-based superheater inner wall temperature prediction model is retrained based on real-time data at the moment.
When the thickness of the oxide layer is too thick, the superheater inner wall temperature prediction model based on the machine learning algorithm cannot correctly reflect the boiler water wall temperature at the moment, and is retrained by acquiring real-time data, so that the superheater inner wall temperature prediction model can better reflect the boiler water wall temperature at the moment, and the prediction precision is further improved
In a further possible embodiment, the first and second threshold values are determined depending on the location of the superheater, the type and capacity of the boiler.
The first threshold value and the second threshold value are adjusted according to the state of the boiler instead of being fixed, and the influence of the thickness of the oxide layer of the superheater of different boilers is different, so that the setting of the thickness threshold value of the oxide layer becomes more scientific.
Example 2
As shown in fig. 3, the wall temperature measuring system considering the influence of the scale of the boiler superheater adopts the wall temperature measuring method considering the influence of the scale of the boiler superheater, and includes:
the system comprises a boiler superheater inner wall temperature prediction module, a boiler superheater oxidation layer calculation module and a result output module, wherein the boiler superheater inner wall temperature prediction module is responsible for obtaining influence parameters of the boiler superheater inner wall temperature based on an expert algorithm and constructing a training set based on the influence parameter data and the corresponding boiler superheater inner wall temperature data; inputting the training set into a superheater inner wall temperature prediction model based on a machine learning algorithm to obtain a trained prediction model; predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment; the boiler superheater oxidation layer calculation module is responsible for measuring the thickness of the boiler superheater inner wall oxidation layer based on an ultrasonic testing system to obtain the thickness of the boiler superheater oxidation layer at the moment; and the result output module is responsible for correcting the wall temperature of the inner wall of the boiler superheater when the thickness of the oxidation layer is larger than a first threshold value, so as to obtain the corrected wall temperature of the inner wall of the boiler superheater.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.
Claims (10)
1. A wall temperature measuring method considering influence of oxide skin of a boiler superheater comprises the following steps:
s1, obtaining influence parameters of the wall temperature of the inner wall of the boiler superheater based on an expert algorithm, and constructing a training set based on the influence parameter data and the corresponding wall temperature data of the inner wall of the boiler superheater;
s2, inputting the training set into a superheater inner wall temperature prediction model based on a machine learning algorithm to obtain a trained prediction model;
s3, predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment;
s4, measuring the thickness of the oxide layer on the inner wall of the boiler superheater based on an ultrasonic testing system to obtain the thickness of the oxide layer of the boiler superheater at the moment;
and S5, when the thickness of the oxide layer is larger than a first threshold value, correcting the wall temperature of the inner wall of the boiler superheater to obtain the corrected wall temperature of the inner wall of the boiler superheater.
2. The wall temperature measuring method according to claim 1, wherein the parameters affecting the wall temperature of the superheater inner wall include generator power, main steam pressure, main steam temperature, feed water flow, coal feed amount, primary air amount, secondary air amount, superheated steam outlet pressure, superheated steam flow, and superheated swallow steam outlet temperature.
3. The wall temperature measuring method according to claim 1, wherein the influence parameter data is subjected to data cleaning by a mean filtering algorithm before a training set is constructed based on the influence parameter data and corresponding boiler superheater inner wall temperature data.
4. The wall temperature measuring method according to claim 1, wherein the data of the wall temperature of the superheater of the boiler comprises an inlet temperature, a middle section temperature and an outlet temperature.
5. The wall temperature measuring method according to claim 1, wherein the machine learning algorithm is FOA-GRNN algorithm, and comprises the following specific steps:
s1: initializing parameters, and initializing the fruit fly colony position;
s2: giving the fruit fly individual a random distance and direction for searching food by using smell, and obtaining the random position of the fruit fly individual at the moment;
s3: calculating a judgment value of the odor concentration at the moment, substituting the judgment value into an odor concentration judgment function, obtaining the odor concentration of the fruit fly individual at the moment and marking the position of the fruit fly individual;
s4, finding out the odor concentration value and the position of the fruit fly with the highest odor concentration in the fruit fly colony, wherein the fruit fly colony flies to the position by using vision;
s5, performing iterative optimization, repeating S3 and S4, and ending the cycle when the iteration is performed until the target times or the odor concentration value reaches the maximum;
and S6, constructing a prediction model of the inner wall temperature of the superheater by taking the obtained odor concentration value as a smoothing factor of the GRNN algorithm at the moment.
6. Method according to claim 5, characterized in that the odor concentration value at which the odor concentration value reaches a maximum is no longer superior to the taste concentration of the previous iteration.
7. The wall temperature measuring method according to claim 1, wherein the calculated thickness of the boiler superheater oxide layer at the moment comprises an inlet superheater oxide layer thickness, a middle superheater oxide layer thickness and an outlet superheater oxide layer thickness based on the mechanism model.
8. The wall temperature measuring method according to claim 1, further comprising a second threshold value, and retraining the superheater inner wall temperature prediction model based on the machine learning algorithm based on real-time data when the thickness of the oxide layer is greater than the second threshold value.
9. The wall temperature measuring method according to claim 1, wherein the first threshold value and the second threshold value are determined according to a location of a superheater, a type and a capacity of a boiler.
10. A wall temperature measuring system considering the influence of oxide skin of a boiler superheater, which adopts the wall temperature measuring method considering the influence of oxide skin of the boiler superheater as claimed in any one of claims 1 to 9, and comprises:
the system comprises a boiler superheater inner wall temperature prediction module, a boiler superheater oxidation layer calculation module and a result output module, wherein the boiler superheater inner wall temperature prediction module is responsible for obtaining influence parameters of the boiler superheater inner wall temperature based on an expert algorithm and constructing a training set based on the influence parameter data and the corresponding boiler superheater inner wall temperature data; inputting the training set into a superheater inner wall temperature prediction model based on a machine learning algorithm to obtain a trained prediction model; predicting to obtain the wall temperature of the inner wall of the boiler superheater at the moment based on the influence parameter data at the moment; the boiler superheater oxidation layer calculation module is responsible for measuring the thickness of the boiler superheater inner wall oxidation layer based on an ultrasonic testing system to obtain the thickness of the boiler superheater oxidation layer at the moment; and the result output module is responsible for correcting the wall temperature of the inner wall of the boiler superheater when the thickness of the oxidation layer is larger than a first threshold value, so as to obtain the corrected wall temperature of the inner wall of the boiler superheater.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210961018.2A CN115326237A (en) | 2022-08-11 | 2022-08-11 | Wall temperature measuring method and system considering influence of boiler superheater oxide skin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210961018.2A CN115326237A (en) | 2022-08-11 | 2022-08-11 | Wall temperature measuring method and system considering influence of boiler superheater oxide skin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115326237A true CN115326237A (en) | 2022-11-11 |
Family
ID=83922061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210961018.2A Pending CN115326237A (en) | 2022-08-11 | 2022-08-11 | Wall temperature measuring method and system considering influence of boiler superheater oxide skin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115326237A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117495125A (en) * | 2023-11-03 | 2024-02-02 | 天津大学 | Wall temperature and oxide skin generation distribution prediction method for high-temperature heating surface of coal-fired boiler |
-
2022
- 2022-08-11 CN CN202210961018.2A patent/CN115326237A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117495125A (en) * | 2023-11-03 | 2024-02-02 | 天津大学 | Wall temperature and oxide skin generation distribution prediction method for high-temperature heating surface of coal-fired boiler |
| CN117495125B (en) * | 2023-11-03 | 2024-05-24 | 天津大学 | Wall temperature and oxide skin generation distribution prediction method for high-temperature heating surface of coal-fired boiler |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108509752B (en) | Superheater mechanism model dynamic parameter setting method combined with field data | |
| CN112818595B (en) | Correction method and system for digital twin model data of evaporation area of thermal power plant | |
| CN113011010A (en) | Boiler fault diagnosis method and diagnosis system based on structural mechanism and operation data | |
| CN110941186B (en) | Steam temperature control optimization method based on neural network and universal gravitation search algorithm | |
| CN107168392B (en) | Method for setting PID (proportion integration differentiation) parameter of boiler drum water level | |
| CN106936627A (en) | A kind of thermal power generating equipment performance monitoring method based on big data analysis mining | |
| CN113868966B (en) | Optimization design method for structural parameters of biomass boiler economizer | |
| CN115326237A (en) | Wall temperature measuring method and system considering influence of boiler superheater oxide skin | |
| CN114925529B (en) | State correction method, system, equipment and medium for digital twin model of condenser | |
| Shi et al. | Ash fouling monitoring and key variables analysis for coal fired power plant boiler | |
| JP2014181570A (en) | Steam turbine control device | |
| CN106289754B (en) | A kind of fired power generating unit component capabilities variation determination method based on Statistical Control Technique | |
| CN114429003A (en) | System of boiler four-tube service life prediction method | |
| CN115510904A (en) | Boiler heating surface ash deposition monitoring method based on time sequence prediction | |
| CN112396162B (en) | Neural network model for predicting wall temperature of screen type superheater of coal-fired unit | |
| CN103728055B (en) | A kind of real-time estimation method of thermal power unit boiler furnace outlet flue gas energy | |
| CN115290218A (en) | Soft measurement method and system for wall temperature of boiler water wall of thermal generator set | |
| CN103697958A (en) | Real-time measurement method of mass flow of saturated steam of drum outlet of coal-fired unit | |
| CN116186625A (en) | Overtemperature early warning method for heating surface of boiler | |
| CN107451304B (en) | Mechanism modeling calculation method of reheater transfer function model | |
| CN109404071B (en) | Identification method for pressurized water reactor steam generator model time constant | |
| CN115296335B (en) | Method and device for solving dynamic characteristics of thermal parameters of boiler in primary frequency modulation process | |
| CN113094992A (en) | Desulfurizing tower optimization method based on neural network discrimination | |
| CN110794679A (en) | Prediction control method and system for load regulation of industrial steam supply system | |
| CN118246164A (en) | Method and device for determining parameters of drum boiler model of primary frequency modulation of thermal power generating unit |
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 |