CN109002068B - Temperature optimization control method for flame path - Google Patents

Temperature optimization control method for flame path Download PDF

Info

Publication number
CN109002068B
CN109002068B CN201811042776.4A CN201811042776A CN109002068B CN 109002068 B CN109002068 B CN 109002068B CN 201811042776 A CN201811042776 A CN 201811042776A CN 109002068 B CN109002068 B CN 109002068B
Authority
CN
China
Prior art keywords
flame path
temperature
improved
pls
fuzzy
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.)
Expired - Fee Related
Application number
CN201811042776.4A
Other languages
Chinese (zh)
Other versions
CN109002068A (en
Inventor
邓娜
曹红英
刘丽芳
陈艳红
王娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kaifeng University
Original Assignee
Kaifeng University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Kaifeng University filed Critical Kaifeng University
Priority to CN201811042776.4A priority Critical patent/CN109002068B/en
Publication of CN109002068A publication Critical patent/CN109002068A/en
Application granted granted Critical
Publication of CN109002068B publication Critical patent/CN109002068B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D23/00—Control of temperature
    • G05D23/19—Control of temperature characterised by the use of electric means
    • G05D23/20—Control of temperature characterised by the use of electric means with sensing elements having variation of electric or magnetic properties with change of temperature
    • C—CHEMISTRY; METALLURGY
    • C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
    • C10B—DESTRUCTIVE DISTILLATION OF CARBONACEOUS MATERIALS FOR PRODUCTION OF GAS, COKE, TAR, OR SIMILAR MATERIALS
    • C10B21/00—Heating of coke ovens with combustible gases
    • C10B21/10—Regulating and controlling the combustion

Landscapes

  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Combustion & Propulsion (AREA)
  • Materials Engineering (AREA)
  • Oil, Petroleum & Natural Gas (AREA)
  • Organic Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a flame path temperature optimization control method, which is characterized in that a flame path temperature detection model based on an improved PLS (partial least squares) is utilized to detect the problem of the flame path at the current moment, and the temperature of the flame path is controlled by adopting fuzzy PID (proportion integration differentiation) with variable universe of discourse according to a preset target value of the temperature of the flame path. The flame path temperature detection model based on the improved PLS realizes indirect measurement of the flame path temperature, reduces the cost, improves the efficiency, and simultaneously utilizes the fuzzy PID of the variable universe of discourse, thereby realizing the advantages of short temperature adjustment time, high steady-state precision, small steady-state error, small overshoot and the like.

Description

Temperature optimization control method for flame path
Technical Field
The invention relates to a temperature control method, in particular to a temperature optimization control method of a flame path.
Background
Coke ovens are complex industrial kilns in the metallurgical industry and are composed of a plurality of alternately arranged coking chambers and combustion chambers. The coal gas and the air enter the combustion chamber for diffusion and combustion after being preheated by the heat storage chamber, and the generated heat is transferred to the carbonization chamber through the furnace wall; the coal is subjected to high-temperature dry distillation in the carbonization chamber to form coke, waste gas generated by combustion is discharged after waste heat of the regenerator is recovered, and the flow directions of coal gas, air and the waste gas are exchanged once every 20 min. In the coking process, the flame path temperature is a crucial process parameter and is directly related to the coke quality and the service life of a furnace body, but the measurement of the flame path temperature is mainly carried out by a warm thermocouple at present, but the installation is difficult, and the coking temperature is too high and is easy to damage, and meanwhile, the control of the flame path temperature mostly depends on manual experience or a simple PID control algorithm, so that the precision is lower, and the adjustment time is too long.
Disclosure of Invention
In order to solve the problems that the direct measurement of the temperature of the flame path in the prior art is high in cost and easy to cause problems, and the control of the temperature of the flame path is not optimized enough, the invention aims to provide a temperature optimization control method of the flame path.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a temperature optimization control method of a flame path is characterized by comprising the following steps: (1) establishing a flame path temperature detection model based on the improved PLS according to the heating gas flow, the top temperature of the regenerator, the pressure of a gas main pipe, the gas flow of the gas, the temperature of a flue and historical data of the variables; (2) collecting the variables in real time, and calculating a flame path temperature deviation value according to a flame path detection value of the flame path temperature detection model based on the improved PLS and a preset flame path temperature target value; (3) and controlling the temperature of the flame path by using fuzzy PID (proportion integration differentiation) with variable discourse domains according to the deviation value of the temperature of the flame path.
Wherein the step (1) of establishing a flame path temperature detection model based on the improved PLS comprises the following steps:
(a) historical data of variables such as heating gas flow X1, regenerator top temperature X2, gas main pipe pressure X3, furnace gas flow X4 and flue temperature X5 are collected as input variables X which are sample data of the model, and flue temperature Y1 is used as output variables Y of the model;
(b) preprocessing the input variable X, firstly, carrying out standardized transformation on the input variable X to eliminate the influence of data dimension, and secondly, removing large error data;
(c) establishing a flame path temperature detection model based on the improved PLS according to the preprocessed input variables:
assuming that an input variable X is linearly related to an output variable Y, namely Y is XB + E, wherein B is a regression coefficient matrix, and E is a residual error matrix;
when the PLS modeling is improved, X and Y are decomposed to obtain
Figure BDA0001792528500000021
Wherein T, U is a score matrix, P, Q is a load matrix, and E1 and E2 are residual matrixes;
according to the formula II, the vectors are in relation with each other by linear regression: u ═ CT + R, where R is the residual matrix, C is the regression factor: c ═ UTT/(TTT), so that a regression coefficient matrix of
Figure BDA0001792528500000022
In order to ensure the prediction accuracy of the model, the model is verified by using the mean square error and the root mean square error.
When a new collected sample is received, fault judgment calculation needs to be carried out on the new sample, and input variables of the two standardized sample data are respectively set as Zi＝{zi1,zi1,,zim},Zj＝{zj1,zj1,,zjmThe failure judgment is related to two coefficients, wherein the coefficient 1 is:
Figure BDA0001792528500000023
Lij1 is 2 samples more consistent, Lij0 is the inconsistency of the two samples; the coefficient 2 is:
Figure BDA0001792528500000024
wherein EmaxIs a set value, Mij2 samples are the same for 1, MijTwo samples differ by 0. According to the two coefficients, the calculated value is: j ═ α Lij+(1-α)MijWhere α ∈ (0, 1). By calculating the calculated value, the flame path temperature detection based on the improved PLS is carried out after the calculated value meets the set threshold value, so that the accuracy of temperature prediction can be effectively ensured, error control caused by faults in a sensor is prevented, and the stable control of the system is improved.
The design of the fuzzy PID of the change domain comprises the following steps:
(A) the selection of the expansion factor of the input and output discourse domain, the basic discourse domain of the error e and the error change rate ec of the flame path temperature are respectively set as follows:
Figure BDA0001792528500000025
then the input scale factor is
Figure BDA0001792528500000026
Wherein lambda belongs to (0,1) and is the value range of the domain of discourseThe envelope coefficient reflects the precision of a control system, and the larger the lambda is, the more obvious the domain compression is; output scaling factor
Figure BDA0001792528500000031
(B) Fuzzification treatment: selecting 7 fuzzy subsets { NB, NM, NS, ZE, PS, PM, PB } to perform fuzzification processing;
(C) determining a membership function: selecting a triangular membership function at the zero point, and selecting a Gaussian membership function at the position close to the boundary of the fuzzy domain;
(D) establishing a fuzzy control rule table and fuzzy reasoning:
when e is larger, Kp is increased, response speed is increased, overshoot is avoided, and Kd is also increased properly; to reduce integral saturation, Ki is reduced;
when the Kp is moderate, the Kp is reduced, the overshoot is reduced, the Ki is increased, the stability is improved, and the value of the Kd is moderate;
when e is smaller, Ki needs to be increased to ensure stability, reduce Kp and improve steady-state precision, and when ec is larger, Kd is reduced, and when ec is smaller, Kd is increased;
(E) the integral quantity delta K of PID can be obtained by clarifying the fuzzy quantityp、ΔKi、ΔKdThen the PID parameters are:
Figure BDA0001792528500000032
wherein Kp0、Ki0、Kd0Setting PID initial parameters as follows: kp0＝7、Ki0＝5、Kd0＝2。
Drawings
Fig. 1 is a schematic diagram of the temperature-optimized control of a flame path.
Detailed Description
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
As shown in fig. 1, a method for controlling optimal temperature of a flue is characterized by comprising the following steps: (1) establishing a flame path temperature detection model based on the improved PLS according to the heating gas flow, the top temperature of the regenerator, the pressure of a gas main pipe, the gas flow of the gas, the temperature of a flue and historical data of the variables; (2) collecting the variables in real time, and calculating a flame path temperature deviation value according to a flame path detection value of the flame path temperature detection model based on the improved PLS and a preset flame path temperature target value; (3) and controlling the temperature of the flame path by using fuzzy PID (proportion integration differentiation) with variable discourse domains according to the deviation value of the temperature of the flame path.
Wherein the step (1) of establishing a flame path temperature detection model based on the improved PLS comprises the following steps:
(a) historical data of variables such as heating gas flow X1, regenerator top temperature X2, gas main pipe pressure X3, furnace gas flow X4 and flue temperature X5 are collected as input variables X which are sample data of the model, and flue temperature Y1 is used as output variables Y of the model;
(b) preprocessing the input variable X, firstly, carrying out standardized transformation on the input variable X to eliminate the influence of data dimension, and secondly, removing large error data;
(c) establishing a flame path temperature detection model based on the improved PLS according to the preprocessed input variables:
assuming that an input variable X is linearly related to an output variable Y, namely Y is XB + E, wherein B is a regression coefficient matrix, and E is a residual error matrix;
when the PLS modeling is improved, X and Y are decomposed to obtain
Figure BDA0001792528500000041
Wherein T, U is a score matrix, P, Q is a load matrix, and E1 and E2 are residual matrixes;
according to the formula II, the vectors are in relation with each other by linear regression: u ═ CT + R, where R is the residual matrix, C is the regression factor: c ═ UTT/(TTT), so that a regression coefficient matrix of
Figure BDA0001792528500000042
In order to ensure the prediction accuracy of the model, the model is verified by using the mean square error and the root mean square error.
When a new collected sample is received, fault judgment calculation needs to be carried out on the new sample, and input variables of the two standardized sample data are respectively set as Zi＝{zi1,zi1,,zim},Zj＝{zj1,zj1,,zjmThe failure judgment is related to two coefficients, wherein the coefficient 1 is:
Figure BDA0001792528500000043
Lij1 is 2 samples more consistent, Lij0 is the inconsistency of the two samples; the coefficient 2 is:
Figure BDA0001792528500000044
wherein EmaxIs a set value, Mij2 samples are the same for 1, MijTwo samples differ by 0. According to the two coefficients, the calculated value is: j ═ α Lij+(1-α)MijWhere α ∈ (0, 1). By calculating the calculated value, the flame path temperature detection based on the improved PLS is carried out after the calculated value meets the set threshold value, so that the accuracy of temperature prediction can be effectively ensured, error control caused by faults in a sensor is prevented, and the stable control of the system is improved.
The design of the fuzzy PID of the change domain comprises the following steps:
(A) the selection of the expansion factor of the input and output discourse domain, the basic discourse domain of the error e and the error change rate ec of the flame path temperature are respectively set as follows:
Figure BDA0001792528500000051
then the input scale factor is
Figure BDA0001792528500000052
Wherein, the lambda belongs to (0,1) and is a value range coefficient of the domain of discourse, which reflects the precision of the control system, and the larger the lambda is, the more obvious the domain of discourse compression is; output scaling factor
Figure BDA0001792528500000053
(B) Fuzzification treatment: selecting 7 fuzzy subsets { NB, NM, NS, ZE, PS, PM, PB } to perform fuzzification processing;
(C) determining a membership function: selecting a triangular membership function at the zero point, and selecting a Gaussian membership function at the position close to the boundary of the fuzzy domain;
(D) establishing a fuzzy control rule table and fuzzy reasoning: PID parameter tuning is achieved based on experience as follows,
when e is larger, Kp is increased, response speed is increased, overshoot is avoided, and Kd is also increased properly; to reduce integral saturation, Ki is reduced;
when the Kp is moderate, the Kp is reduced, the overshoot is reduced, the Ki is increased, the stability is improved, and the value of the Kd is moderate;
when e is smaller, Ki needs to be increased to ensure stability, reduce Kp and improve steady-state precision, and when ec is larger, Kd is reduced, and when ec is smaller, Kd is increased;
a fuzzy control rule table is obtained according to the above experience, as shown in table 1:
TABLE 1 fuzzy control rules Table
Figure BDA0001792528500000054
(E) The integral quantity delta K of PID can be obtained by clarifying the fuzzy quantityp、ΔKi、ΔKdThen the PID parameters are:
Figure BDA0001792528500000055
wherein Kp0、Ki0、Kd0Setting PID initial parameters as follows: kp0＝7、Ki0＝5、Kd0＝2。

Claims (2)

1. A temperature optimization control method of a flame path is characterized by comprising the following steps:
(1) according to historical data of variables such as heating gas flow, regenerator top temperature, gas main pipe pressure, gas flow and flue temperature, a flame path temperature detection model based on the improved PLS is established;
(2) collecting the variables in real time, and calculating a flame path temperature deviation value according to a flame path detection value of the flame path temperature detection model based on the improved PLS and a preset flame path temperature target value;
(3) controlling the temperature of the flame path by using fuzzy PID (proportion integration differentiation) with variable discourse domains according to the deviation value of the temperature of the flame path;
the step (1) of establishing a flame path temperature detection model based on the improved PLS comprises the following steps:
(a) historical data of variables such as heating gas flow X1, regenerator top temperature X2, gas main pipe pressure X3, furnace gas flow X4 and flue temperature X5 are collected as input variables X which are sample data of the model, and flue temperature Y1 is used as output variables Y of the model;
(b) preprocessing the input variable X, firstly, carrying out standardized transformation on the input variable X to eliminate the influence of data dimension, and secondly, removing large error data;
(c) establishing a flame path temperature detection model based on the improved PLS according to the preprocessed input variables:
assuming that an input variable X is linearly related to an output variable Y, namely Y is XB + E, wherein B is a regression coefficient matrix, and E is a residual error matrix;
when the PLS modeling is improved, X and Y are decomposed to obtain
Figure FDA0002805715120000011
Wherein T, U is a score matrix, P, Q is a load matrix, and E1 and E2 are residual matrixes;
according to the formula II, the vectors are in relation with each other by linear regression: u ═ CT + R, where R is the residual matrix, C is the regression factor: c ═ UTT/(TTT), so that a regression coefficient matrix of
Figure FDA0002805715120000012
And (3) controlling the temperature of the flame path by using a variable universe fuzzy PID according to the temperature deviation value of the flame path, wherein the design of the variable universe fuzzy PID comprises the following steps:
(A) the selection of the expansion factor of the input and output discourse domain, the basic discourse domain of the error e and the error change rate ec of the flame path temperature are respectively set as follows:
Figure FDA0002805715120000013
then the input scale factor is
Figure FDA0002805715120000014
Output scaling factor
Figure FDA0002805715120000015
(B) Fuzzification treatment: selecting 7 fuzzy subsets { NB, NM, NS, ZE, PS, PM, PB } to perform fuzzification processing;
(C) determining a membership function: selecting a triangular membership function at the zero point, and selecting a Gaussian membership function at the position close to the boundary of the fuzzy domain;
(D) establishing a fuzzy control rule table and fuzzy reasoning:
when e is larger, Kp is increased, response speed is increased, overshoot is avoided, and Kd is also increased properly; to reduce integral saturation, Ki is reduced;
when the Kp is moderate, the Kp is reduced, the overshoot is reduced, the Ki is increased, the stability is improved, and the value of the Kd is moderate;
when e is smaller, Ki needs to be increased to ensure stability, reduce Kp and improve steady-state precision, and when ec is larger, Kd is reduced, and when ec is smaller, Kd is increased;
(E) the integral quantity delta K of PID can be obtained by clarifying the fuzzy quantityp、ΔKi、ΔKdThen the PID parameters are:
Figure FDA0002805715120000021
wherein Kp0、Ki0、Kd0Is the initial value of PID parameter;
wherein upon receiving a new collected sample, it is requiredCarrying out fault judgment calculation on the new sample, and setting the input variables of the two standardized sample data as Zi＝{zi1,zi1,,zim},Zj＝{zj1,zj1,,zjmThe failure judgment is related to two coefficients, wherein the coefficient 1 is:
Figure FDA0002805715120000022
Lij1 is 2 samples more consistent, Lij0 is the inconsistency of the two samples; the coefficient 2 is:
Figure FDA0002805715120000023
wherein EmaxIs a set value, Mij2 samples are the same for 1, MijTwo samples differ for 0, and the calculated value is obtained from the two coefficients: j ═ α Lij+(1-α)MijAnd alpha belongs to (0,1), and the calculated value is calculated, and the flame path temperature detection based on the improved PLS is carried out after the calculated value meets the set threshold, so that the accuracy of temperature prediction can be effectively ensured, error control caused by the fault condition in the sensor is prevented, and the stable control of the system is improved.
2. The method for controlling the optimal temperature of the flame path according to claim 1, wherein the initial value of the PID parameter is as follows: kp0＝7、Ki0＝5、Kd0＝2。
CN201811042776.4A 2018-09-07 2018-09-07 Temperature optimization control method for flame path Expired - Fee Related CN109002068B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811042776.4A CN109002068B (en) 2018-09-07 2018-09-07 Temperature optimization control method for flame path

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811042776.4A CN109002068B (en) 2018-09-07 2018-09-07 Temperature optimization control method for flame path

Publications (2)

Publication Number Publication Date
CN109002068A CN109002068A (en) 2018-12-14
CN109002068B true CN109002068B (en) 2021-04-02

Family

ID=64591259

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811042776.4A Expired - Fee Related CN109002068B (en) 2018-09-07 2018-09-07 Temperature optimization control method for flame path

Country Status (1)

Country Link
CN (1) CN109002068B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110716593A (en) * 2019-10-31 2020-01-21 河北科技大学 Method and device for predicting and controlling temperature of reaction kettle and terminal equipment
CN111528532A (en) * 2020-04-24 2020-08-14 云南中烟工业有限责任公司 Temperature control system of device for heating non-combustible cigarettes and temperature control method thereof
CN111534308A (en) * 2020-05-08 2020-08-14 马鞍山钢铁股份有限公司 Method for controlling temperature of ultrahigh coke oven
CN112961709A (en) * 2021-02-02 2021-06-15 鹤岗市征楠煤化工有限公司 Tar removal process based on tail gas containing tar furnace
CN113110635B (en) * 2021-03-26 2023-08-18 北京北方华创微电子装备有限公司 Temperature control system, method and controller for semiconductor equipment and external ignition device
CN114510092B (en) * 2022-02-17 2023-02-10 太原理工大学 Transition packet internal temperature control system and method based on fuzzy PID (proportion integration differentiation) of prediction variable universe

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101067828A (en) * 2007-06-12 2007-11-07 中南大学 Coke oven fire path temperature integrated moulding and soft measuring method
CN101509812B (en) * 2008-12-18 2010-09-01 浙江大学 Soft measurement method for billet temperature distribution in smelting and heating-furnace
CN102213961B (en) * 2011-05-03 2013-06-12 东北大学 Method for predicting and controlling quality of hollow billet
CN103770832A (en) * 2014-02-21 2014-05-07 重庆邮电大学 Assisted power controlling method for electric power assisted steering system

Also Published As

Publication number Publication date
CN109002068A (en) 2018-12-14

Similar Documents

Publication Publication Date Title
CN109002068B (en) Temperature optimization control method for flame path
CN107023825B (en) Fluidized-bed combustion boiler control and combustion optimizing system
CN111562744B (en) Boiler combustion implicit generalized predictive control method based on PSO algorithm
CN110257577B (en) Burning process control method and system for ball type hot blast stove
CN104482525B (en) The control method of extra-supercritical unit reheat steam temperature and system
CN107515598A (en) Fired power generating unit distributed and coordinated control system based on multi-parameter dynamic matrix control
CN102494336A (en) Combustion process multivariable control method for CFBB (circulating fluidized bed boiler)
Prasad et al. Plant-wide predictive control for a thermal power plant based on a physical plant model
CN113110657B (en) Method, system and medium for controlling hearth pressure and exhaust gas temperature of heating furnace
CN113359425A (en) Thermal power plant boiler main steam temperature intelligent control system based on LSTM neural network PID optimization
CN101709863B (en) Hybrid control method for furnace pressure system of coal-fired boiler
WO2019085446A1 (en) Self-order-reducing multi-loop centralized prediction control system for reheat steam temperature of double reheat unit
CN111623369A (en) Control method for adjusting boiler fuel feeding quantity by using smoke oxygen content signal
CN110673482B (en) Power station coal-fired boiler intelligent control method and system based on neural network prediction
CN115685743A (en) Intelligent control coal-fired boiler and intelligent prediction regulation and control flue gas emission method thereof
CN113835342B (en) Disturbance rejection predictive control method for overheat steam temperature system
CN106468446B (en) Heating furnace control and combustion optimization method
CN111401652A (en) Boiler optimization method and system based on CO online detection
CN108488831B (en) Boiler combustion control system and method
CN105785761A (en) Method for controlling temperature intelligent switching of cement raw materials in the course of decomposition
CN113094896A (en) Power plant heat supply unit optimization control method
CN211554672U (en) Self-adjusting fuzzy Smith-PID temperature control system
CN113625556B (en) Self-adaptive control method of complex industrial system of circulating fluidized bed
CN111765446B (en) Boiler control method and system based on automatic optimization fuzzy three-level PID
CN1052698A (en) Expert systems is used in soaking furnace steel-smelting control

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20210402