CN112255364B - Soft measurement method for real-time quantitative determination of sintering end point state - Google Patents

Soft measurement method for real-time quantitative determination of sintering end point state Download PDF

Info

Publication number
CN112255364B
CN112255364B CN202011127923.5A CN202011127923A CN112255364B CN 112255364 B CN112255364 B CN 112255364B CN 202011127923 A CN202011127923 A CN 202011127923A CN 112255364 B CN112255364 B CN 112255364B
Authority
CN
China
Prior art keywords
num
end point
sintering
tail
gas temperature
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.)
Active
Application number
CN202011127923.5A
Other languages
Chinese (zh)
Other versions
CN112255364A (en
Inventor
刘颂
赵亚迪
卜象平
杨秀伟
冯伟
张小松
赵志伟
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.)
Tangshan Shuyu Technology Co ltd
Tangshan University
Original Assignee
Tangshan Shuyu Technology Co ltd
Tangshan 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 Tangshan Shuyu Technology Co ltd, Tangshan University filed Critical Tangshan Shuyu Technology Co ltd
Priority to CN202011127923.5A priority Critical patent/CN112255364B/en
Publication of CN112255364A publication Critical patent/CN112255364A/en
Application granted granted Critical
Publication of CN112255364B publication Critical patent/CN112255364B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N31/00—Investigating or analysing non-biological materials by the use of the chemical methods specified in the subgroup; Apparatus specially adapted for such methods
    • G01N31/12—Investigating or analysing non-biological materials by the use of the chemical methods specified in the subgroup; Apparatus specially adapted for such methods using combustion
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N25/00—Investigating or analyzing materials by the use of thermal means
    • G01N25/02—Investigating or analyzing materials by the use of thermal means by investigating changes of state or changes of phase; by investigating sintering
    • G01N25/12—Investigating or analyzing materials by the use of thermal means by investigating changes of state or changes of phase; by investigating sintering of critical point; of other phase change
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N2021/8411—Application to online plant, process monitoring

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Molecular Biology (AREA)
  • Theoretical Computer Science (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Combustion & Propulsion (AREA)
  • Manufacture And Refinement Of Metals (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a soft measurement method for quantitatively judging the state of a sintering end point in real time, belonging to the field of sintering process control. According to the method, a secondary curve is fitted to a thermocouple insertion position and a waste gas temperature detection value on the basis of a waste gas temperature detection value of an air box in a sintering production line and an image shot by a tail CCD camera, and a quantitative end point position (MBTP) and a quantitative end point temperature (Tmax) are preliminarily obtained by calculating coordinates (X and Y) of a maximum value point of the curve. The air leakage problem of the sintering production line occasionally causes the detection distortion of the exhaust gas temperature of the air box, so that the secondary curve fitting method obtains wrong end point state, and the end point state obtained by the secondary curve fitting method is corrected by adopting a machine tail image recognition result. The terminal position and the temperature obtained by calculation by the method are more real and accurate, and the method can assist field operators in judging the terminal state and the change trend thereof quantitatively in real time.

Description

Soft measurement method for real-time quantitative determination of sintering end point state
Technical Field
The invention relates to a soft measurement method for quantitatively judging the state of a sintering end point in real time, belonging to the field of sintering process control.
Background
In the sintering production, the sintering end point is an important process parameter influencing the quality, the yield and the cost of the sintered ore, and the proper and stable end point state is the key point for ensuring that the blast furnace obtains the high-quality sintered ore. When the sintering end point is advanced, the production capacity of the sintering machine cannot be fully utilized, so that the yield of sintered ore is reduced; when the sintering end point is delayed, the mixture is not completely burnt through and the machine tail is unloaded, so that the percent of pass of the sinter is reduced, and the cost is increased.
The environment of the sintering machine tail is severe, and under the conditions of high temperature, high humidity, high dust and strong interference, instrument equipment for directly detecting the sintering end point is not available at present. No matter the end point position is estimated by a waste gas temperature method, a negative pressure method and a waste gas component judgment method, or a red layer is observed at the tail of the sintering machine, the end point state can be judged only qualitatively, and accurate control of the end point state and optimal adjustment of parameters of the sintering process are necessarily influenced. Therefore, the research on how to quantitatively and accurately detect the end point state has important guiding significance on the fine operation of the sintering production process.
A quadratic curve fitting method based on the exhaust gas temperature of the air box is a commonly adopted end point state soft measurement method at present. In the sintering production process, the problems of trolley material surface air blowby, machine tail air leakage and the like often occur, so that the exhaust gas temperature of the air box is not consistent with the actual real temperature, the end point state fitting result based on the exhaust gas temperature of the air box is jittered or deviated, and the output result cannot effectively guide an on-site operator to accurately judge the end point state. In order to obtain a more stable and accurate end point state detection result, the soft measurement method of the end point state is improved and optimized.
Disclosure of Invention
The invention provides a soft measurement method for quantitatively judging the sintering end point state in real time, which can stably detect the end point state in the sintering production process in real time and quantitatively, and has important guidance effect on timely mastering the end point state and carrying out fine operation for field operators.
Specifically, the soft measurement method for quantitatively judging the sintering end point state in real time comprises the following steps:
the method comprises the following steps: acquiring data, namely acquiring a waste gas temperature detection value of an air box in a sintering production line and an image shot by a tail CCD camera in real time;
step two: fitting data, namely fitting a secondary curve to the thermocouple insertion position and the waste gas temperature detection value to obtain coordinates (X, Y) of a curve maximum value point and preliminarily obtain a quantitative terminal point position (MBTP) and a terminal point temperature (Tmax);
step three: classifying and judging the machine tail images, namely classifying the machine tail images into 3 categories, namely under-burning, normal and over-burning, and classifying and judging the machine tail images by adopting a convolutional neural network model;
step four: and (4) correcting the end point position, namely establishing an expert rule by taking the recognized category of the tail image as a reference, correcting the end point state obtained by a quadratic curve fitting method, and obtaining real and accurate end point state information in real time.
The first step specifically comprises:
the method comprises the steps of acquiring a detection value of the exhaust gas temperature of the air box in the whole sintering production line and the specific insertion position of a thermocouple in the air box in real time, recording tail images shot by a CCD camera at the tail of the sintering machine in the same period, and taking the exhaust gas temperature of the air box and the detected images at the tail as basic data sources for real-time judgment of the end point state.
The second step specifically comprises:
and fitting a secondary curve according to the thermocouple mounting position and the exhaust gas temperature detection value, and preliminarily calculating the terminal point position (MBTP) and the terminal point temperature (Tmax) in the following manner.
2-1, when the exhaust gas temperature of the last windbox is maximum, the sintering end point (BTP) = the last windbox number (Num) — 0.5 (the thermocouple insertion position is generally in the middle of the windbox, so 0.5 is taken, and this coefficient can be adjusted according to the actual practice), the sintering end point position (MBTP) = the sintering end point (BTP) × the length of the windbox, and the sintering end point temperature (Tmax) is the exhaust gas temperature of the last windbox.
2-2, when the exhaust gas temperature of the penultimate air box is maximum, taking the exhaust gas temperature detection values (T) of the penultimate air box, the penultimate air box and the last air boxNum-2，TNum-1T) and thermocouple position (Num-2.5, Num-1.5, Num-0.5) to calculate the coordinates (X, Y) of the curve maximum point. Assume a quadratic form of a unitary equation of the form aX2+ bX + c = Y, and the thermocouple position is defined asAnd X, taking the exhaust gas temperature detection value as Y, and respectively substituting each group of detection data into equation solving coefficients a, b and c. Sintering end point (BTP) = -b/(2a), sintering end point position (MBTP) = sintering end point (BTP) × length of windbox, sintering end point temperature (Tmax) = a × BTP2+b*BTP+c。
And 2-3, counting the fluctuation ranges of the historical end points of the plurality of sintering machines, wherein the fluctuation range of the end point is generally positioned between (Num-6) bellows and (Num) bellows. And calculating the end point state between the (Num-5) bellows and the (Num-1) bellows by adopting the method in the step 2-2.
2-4, when the waste gas temperature of the (Num-6) windbox is maximum, the sintering end point (BTP) = windbox number (Num-6) -0.5 (thermocouple insertion position is generally in the middle of the windbox, so 0.5 is taken, the coefficient can be adjusted according to the actual situation), the sintering end point position (MBTP) = sintering end point (BTP) × length of the windbox, and the sintering end point temperature (Tmax) is the waste gas temperature of the (Num-6) windbox.
The third step specifically comprises:
according to the position difference of the red fire layer in the tail image, the images are divided into 3 categories, namely under-burning, normal and over-burning. Under-burning is that the red fire layer is positioned at the top of the machine tail image, and the lower part of the red fire layer is all raw materials; normally, the red fire layer is arranged at the bottom of the tail image and occupies about one third of the height of the whole section; the overburning is that only a narrow red fire layer is arranged at the bottom of a tail image, and the upper part of the red fire layer is provided with a sintering block. And (4) carrying out classification judgment on the tail image by adopting a Convolutional Neural Network (CNN) model.
The fourth step specifically comprises:
and correcting the end point state obtained by the quadratic curve fitting method by taking the category of the tail image identified by the Convolutional Neural Network (CNN) model as a reference, wherein the specific expert rule is as follows.
4-1, when the tail image is identified as normal by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num-2) to (Num-1) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-2, when the tail image is identified as normal by the Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num-2) to (Num-1) of the bellows, outputting the end position and the end temperature which are located between the range from (Num-2) to (Num-1) of the bellows at the last moment;
4-3, when the tail image is identified as under burning by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num-6) to (Num-2) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-4, when the tail image is identified as under-burning by the Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num-6) to (Num-2) of the bellows, outputting the end position and the end temperature of which the last time is located in the range from (Num-6) to (Num-3) of the bellows;
4-5, when the tail image is identified as overburning by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num) to (Num-1) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-6, when the tail image is identified as overfire by a Convolution Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num) to (Num-1) of the bellows, outputting an end position and an end temperature which are located between the range from (Num) to (Num-1) of the bellows at the last moment;
a soft measurement method for quantitatively judging the state of a sintering end point in real time comprises the following steps: a CCD camera, a storage device, a processor, etc.; the processor loads and executes data, images and rules in the storage device to realize the soft measurement method for quantitatively judging the sintering end point state in real time as claimed in claims 1-5.
Compared with the prior art, the invention has the advantages that:
(1) according to the soft measurement method for quantitatively judging the sintering end point state in real time, the sintering end point is multiplied by the length of the air box to obtain the distance between the position of the sintering end point and the starting point of the No. 1 air box, so that the judgment precision of the end point position is improved.
(2) The invention discloses a soft measurement method for quantitatively judging the sintering end point state in real time, which selects a plurality of data sources such as numerical values (air box waste gas temperature), images (tail images) and the like as signals for judging the end point state, and fully considers the input information related to the end point state.
(3) According to the soft measurement method for quantitatively judging the sintering end point state in real time, the end point state obtained by a quadratic curve fitting method is corrected by taking the type of the convolutional neural network for identifying the tail image as a reference, so that the stability of the detection of the end point state in the actual production process is improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a soft measurement method for real-time quantitative determination of sintering end point state according to the present invention.
Fig. 2 is a schematic diagram of the category of images at the tail of the aircraft.
Fig. 3 is a comparison graph of the correction of the quadratic curve fitting value using the machine tail image recognition result.
FIG. 4 is a graph showing the effect of the sintering end point state detection application of the present invention.
Detailed Description
In order to make the technical features, objects, and effects of the present invention more clearly understood, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
A soft measurement method for quantitatively judging the state of a sintering end point in real time is shown in figure 1 and specifically comprises the following steps:
specifically, the soft measurement method for quantitatively judging the sintering end point state in real time comprises the following steps:
the method comprises the following steps: acquiring data, namely acquiring a waste gas temperature detection value of an air box in a sintering production line and an image shot by a tail CCD camera in real time;
step two: fitting data, namely fitting a secondary curve to the thermocouple insertion position and the waste gas temperature detection value to obtain coordinates (X, Y) of a curve maximum value point and preliminarily obtain a quantitative terminal point position (MBTP) and a terminal point temperature (Tmax);
step three: classifying and judging the machine tail images, namely classifying the machine tail images into 3 categories, namely under-burning, normal and over-burning, and classifying and judging the machine tail images by adopting a convolutional neural network model;
step four: and (4) correcting the end point position, namely establishing an expert rule by taking the recognized category of the tail image as a reference, correcting the end point state obtained by a quadratic curve fitting method, and obtaining real and accurate end point state information in real time.
The first step specifically comprises:
the method comprises the steps of acquiring a detection value of the exhaust gas temperature of the air box in the whole sintering production line and the specific insertion position of a thermocouple in the air box in real time, recording tail images shot by a CCD camera at the tail of the sintering machine in the same period, and taking the exhaust gas temperature of the air box and the detected images at the tail as basic data sources for real-time judgment of the end point state.
The second step specifically comprises:
and fitting a secondary curve according to the thermocouple mounting position and the exhaust gas temperature detection value, and preliminarily calculating the terminal point position (MBTP) and the terminal point temperature (Tmax) in the following manner.
2-1, when the exhaust gas temperature of the last windbox is maximum, the sintering end point (BTP) = the last windbox number (Num) — 0.5 (the thermocouple insertion position is generally in the middle of the windbox, so 0.5 is taken, and this coefficient can be adjusted according to the actual practice), the sintering end point position (MBTP) = the sintering end point (BTP) × the length of the windbox, and the sintering end point temperature (Tmax) is the exhaust gas temperature of the last windbox.
2-2, when the exhaust gas temperature of the penultimate wind box is maximum, taking the exhaust gas temperature detection values (T) of the penultimate wind box, the penultimate wind box and the last wind boxNum-2，TNum-1T) and thermocouple position (Num-2.5, Num-1.5, Num-0.5) to calculate the coordinates (X, Y) of the curve maximum point.Assume a quadratic form of a unitary equation of the form aX2And + bX + c = Y, taking the thermocouple position as X, taking the exhaust gas temperature detection value as Y, and substituting each group of detection data into equation solving coefficients a, b and c respectively. Sintering end point (BTP) = -b/(2a), sintering end point position (MBTP) = sintering end point (BTP) × length of windbox, sintering end point temperature (Tmax) = a × BTP2+b*BTP+c。
And 2-3, counting the fluctuation ranges of the historical end points of the plurality of sintering machines, wherein the fluctuation range of the end point is generally positioned between (Num-6) air boxes and (Num) air boxes. And calculating the end point state between the (Num-5) bellows and the (Num-1) bellows by adopting the method in the step 2-2.
2-4, when the waste gas temperature of the (Num-6) windbox is maximum, the sintering end point (BTP) = windbox number (Num-6) -0.5 (thermocouple insertion position is generally in the middle of the windbox, so 0.5 is taken, the coefficient can be adjusted according to the actual situation), the sintering end point position (MBTP) = sintering end point (BTP) × length of the windbox, and the sintering end point temperature (Tmax) is the waste gas temperature of the (Num-6) windbox.
The third step specifically comprises:
according to the position difference of the red fire layer in the tail image, the images are divided into 3 categories, namely under-burning, normal and over-burning. Under-burning is that the red fire layer is positioned at the top of the machine tail image, and the lower part of the red fire layer is all raw materials; normally, the red fire layer is arranged at the bottom of the tail image and occupies about one third of the height of the whole section; the overburning is that only a narrow red fire layer is arranged at the bottom of a tail image, and the upper part of the red fire layer is provided with a sintering block. And (4) carrying out classification judgment on the tail image by adopting a Convolutional Neural Network (CNN) model.
The fourth step specifically comprises:
and correcting the end point state obtained by the quadratic curve fitting method by taking the category of the tail image identified by the Convolutional Neural Network (CNN) model as a reference, wherein the specific expert rules are as follows.
4-1, when the tail image is identified as normal by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num-2) to (Num-1) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-2, when the tail image is identified as normal by the Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num-2) to (Num-1) of the bellows, outputting the end position and the end temperature which are located between the range from (Num-2) to (Num-1) of the bellows at the last moment;
4-3, when the tail image is identified as under burning by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num-6) to (Num-2) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-4, when the tail image is identified as under-burning by the Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num-6) to (Num-2) of the bellows, outputting the end position and the end temperature of which the last time is located in the range from (Num-6) to (Num-3) of the bellows;
4-5, when the tail image is identified as overburning by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located between the range of (Num) to (Num-1) bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method;
4-6, when the tail image is identified as overfire by a Convolutional Neural Network (CNN) model, if the end position state obtained by the quadratic curve fitting method is located in the range from (Num) to (Num-1) of the bellows, outputting the end position and the end temperature which are located between the range from (Num) to (Num-1) of the bellows at the last moment;
according to the method, the exhaust gas temperature of the air box and the tail image are used as input signals, the sintering end point position and the sintering end point temperature are used as output results, the on-line and quantitative detection of the end point state is realized, and the method has an important application value for guiding sintering production by field operators.
The embodiment of the invention adopts one 360m sintering plant2Historical air box waste gas temperature detection of sintering machineAnd (4) establishing a soft measurement method for judging the sintering end point state according to the steps from the first step to the fourth step on the basis of the value and the CCD camera at the tail of the sintering machine, and finally testing by adopting the on-site actual production data of the sintering machine. The method comprises the following specific steps:
(1) data acquisition
One sintering plant in China is 360m2The sintering machine was used for example, and the sintering machine had an effective sintering area of 80 m and was equipped with 22 windboxes. And thermocouples are respectively arranged in the middle parts of the air boxes 1#, 2#, 3#, 5#, 7#, 9#, 11#, 13#, 15#, 16#, 18#, 20#, 21# and 22# and are used for carrying out second-level continuous temperature measurement on the temperature of the waste gas in the air boxes. The detection value of the historical exhaust gas temperature and the detection video of the CCD camera at the tail of the machine are obtained from the field automation system, the time span is 6 months in 2019 to 6 months in 2020, and the frequency of the exhaust gas temperature acquisition is 1 minute.
(2) Fitting of data
Based on the above-mentioned historical data, a secondary curve was fitted to thermocouple installation positions of 16#, 18#, 20#, 21# and 22# windboxes and the exhaust gas temperature detection values, and the terminal point position (MBTP) and the terminal point temperature (Tmax) were preliminarily calculated.
2-1, when the waste gas temperature of 22# windbox was at a maximum, sintering end point (BTP) =22 (Num) -0.5 (thermocouple was inserted in the middle of windbox, so 0.5) =21.5, sintering end point position (MBTP) =21.5 (BTP) × 3.636 (length of windbox) =78.174(m), and sintering end point temperature (Tmax) was the waste gas temperature detection value of 22# windbox.
2-2, when the exhaust gas temperature of the No. 21 windbox is maximum, taking the thermocouple positions (19.5, 20.5, 21.5) of the No. 20 windbox, the No. 21 windbox and the No. 22 windbox and the detected value (T) of the exhaust gas temperatureNum-2，TNum-1And T) fitting a quadratic curve. And (3) respectively obtaining three unary quadratic equations by adopting the 3 groups of points, and calculating to obtain coefficients of the equations: a = (T)Num-2+T-2*TNum-1)/2，b= 41*TNum-1-21*TNum-2-20*T，c=(4*TNum-2-1521 xa-78 xb)/4. Sintering end point (BTP) = -b/(2a), sintering end point position (MBTP) = sintering end point (BTP) × 3.636 (length of windbox), sintering end point temperature (Tmax) = a × BTP2+b*BTP+c。
2-3, when the exhaust gas temperature of the No. 20 wind box is maximum, taking the thermocouple positions (17.5, 19.5, 20.5) of the No. 18 wind box, the No. 20 wind box and the No. 21 wind box and the detected value (T) of the exhaust gas temperatureNum-4，TNum-2, TNum-1) And fitting a quadratic curve. And (3) respectively obtaining three unary quadratic equations by adopting the 3 groups of points, and calculating to obtain coefficients of the equations: a = (T)Num-4+2*TNum-1-3*TNum-2)/6，b=(57*TNum-2-20*TNum-4-37*TNum-1)/3，c=(4* TNum-4-1225 a-70 b)/4. Sintering end point (BTP) = -b/(2a), sintering end point position (MBTP) = sintering end point (BTP) × 3.636 (length of windbox), sintering end point temperature (Tmax) = a × BTP2+b*BTP+c。
2-4, when the exhaust gas temperature of the 18# wind box is maximum, taking the thermocouple positions (15.5, 17.5, 19.5) of the 16#, 18# and 20# wind boxes and the detected value (T) of the exhaust gas temperatureNum-6，TNum-4, TNum-2) And fitting a quadratic curve. And (3) respectively obtaining three unary quadratic equations by adopting the 3 groups of points, and calculating to obtain coefficients of the equations: a = (T)Num-6+ TNum-2-2* TNum-4)/8，b=(70* TNum-4-37* TNum-6-33* TNum-2)/8，c=(4* TNum-6-961-62 a)/4. Sintering end point (BTP) = -b/(2a), sintering end point position (MBTP) = sintering end point (BTP) × 3.636 (length of windbox), sintering end point temperature (Tmax) = a × BTP2+b*BTP+c。
2-5, when the waste gas temperature of the 16# windbox was the maximum, the sintering end point (BTP) =16 (Num-6) -0.5 (the thermocouple was inserted in the middle of the windbox, so 0.5) =15.5, the sintering end point position (MBTP) =15.5 (BTP) × 3.636 (length of windbox) =56.358(m), and the sintering end point temperature (Tmax) was the waste gas temperature detection value of the 16# windbox.
(3) Machine tail image classification judgment
And according to the collection frequency of the exhaust gas temperature detection values, classifying and judging the detection videos of the tail in the same time period by adopting a convolutional neural network model, and determining whether the position state of the terminal is under-burning, normal or over-burning.
(4) End point position correction
And correcting the end point state obtained by the quadratic curve fitting method according to the recognized type of the tail image as a reference.
The original end position state obtained by quadratic curve fitting is compared with the end position state corrected based on the machine tail image recognition, as shown in fig. 3. In the positions of the first step and the second step in the figure, the convolution neural network model is adopted to identify that the images of the tail of the machine are in the overburning state and the normal state respectively, so the end point position state of the fitting result of the quadratic curve is corrected according to the process rules of 4-6 and 4-2 in the fourth step of the soft measurement method of the sintering end point state.
(5) Application effects
The soft measurement model of the sintering end point state established by the method is applied to the field latest production data for testing, and the 360m is selected2Actual production data of the sintering machine in 2020 and 8 months intercept results of about 1000 sets of test samples for display, as shown in fig. 4. In fig. 4, the abscissa is the number of test samples, and the ordinate is the end point position and the end point temperature, and the original end point state of the quadratic curve fitting and the end point state after correction based on image recognition are plotted simultaneously and represented by curves of different labels. As can be seen from fig. 4, the endpoint status after the correction based on the image recognition not only can display the actual situation of the on-site sintering endpoint more truly and stably, but also has higher accuracy. Therefore, the soft measurement method for quantitatively judging the sintering end point state in real time provided by the invention has a good application effect, and can be well applied and popularized to automatic systems of other sintering machines.
While specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to the above-described specific embodiments, which are intended to be illustrative only and not limiting. Those skilled in the art, having the benefit of this disclosure, may effect numerous modifications thereto without departing from the scope and spirit of the invention as set forth in the claims and the following claims.

Claims (4)

1. A soft measurement method for quantitatively judging the state of a sintering end point in real time comprises the following steps:
the method comprises the following steps: acquiring data, namely acquiring a waste gas temperature detection value of an air box in a sintering production line and an image shot by a tail CCD camera in real time;
step two: fitting data, namely fitting a secondary curve to the thermocouple inserting position and the waste gas temperature detection value to obtain coordinates (X, Y) of a curve maximum value point, preliminarily obtaining a quantitative terminal point position MBTP and a terminal point temperature Tmax in the following way,
2-1, when the waste gas temperature of the last wind box is maximum, the sintering end point BTP is the number Num-0.5 of the last wind box, the thermocouple is generally arranged in the middle of the wind box, so the thermocouple is 0.5, the coefficient can be adjusted according to the actual situation, the sintering end point position MBTP is the length of the sintering end point BTP, the sintering end point temperature Tmax is the waste gas temperature of the last wind box,
2-2, when the exhaust gas temperature of the penultimate wind box is maximum, taking the exhaust gas temperature detection values (T) of the penultimate wind box, the penultimate wind box and the last wind boxNum-2，TNum-1T) and thermocouple position (Num-2.5, Num-1.5, Num-0.5) to calculate the coordinates (X, Y) of the curve maximum point, assuming the form of a unitary quadratic equation as aX2Using the thermocouple position as X and the exhaust gas temperature detection value as Y, respectively substituting each group of detection data into equation solving coefficients a, b and c, wherein the sintering end point BTP is-b/(2 a), the sintering end point position MBTP is the sintering end point BTP and the length of the wind box, and the sintering end point temperature Tmax is a BTP2+b*BTP+c，
2-3, counting the fluctuation range of the historical endpoints of the plurality of sintering machines, determining that the fluctuation range of the endpoint is generally positioned between a Num-6 air box and a Num air box, calculating the endpoint state between a Num-5 air box and a Num-1 air box by adopting the method of the 2-2 steps,
2-4, when the waste gas temperature of the Num-6 wind box is maximum, the sintering end point BTP is equal to the number Num-6-0.5 of the wind box, the thermocouple is generally arranged in the middle of the wind box, so that the value is 0.5, the coefficient can be adjusted according to the actual situation, the sintering end point position MBTP is equal to the length of the sintering end point BTP, and the sintering end point temperature Tmax is the waste gas temperature of the Num-6 wind box;
step three: classifying and judging the machine tail images, namely classifying the machine tail images into 3 categories, namely under-burning, normal and over-burning, and classifying and judging the machine tail images by adopting a convolutional neural network model;
step four: correcting the end point position, establishing an expert rule by taking the recognized category of the tail image as a reference, correcting the end point state obtained by a quadratic curve fitting method, and obtaining real and accurate end point state information in real time, wherein the expert rule is as follows,
4-1, when the tail image is identified as normal by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is located between the range of Num-2 to Num-1 bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method,
4-2, when the tail image is identified as normal by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is located in the range from Num-2 to Num-1 bellows, outputting the end position and the end temperature which are located between the range from Num-2 to Num-1 bellows at the last moment,
4-3, when the tail image is identified as under burning by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is between the range of Num-6 to Num-2 bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method,
4-4, when the tail image is identified as under burning by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is located in the range from Num-6 to Num-2 bellows, outputting the end position and the end temperature which are located between Num-6 and Num-3 bellows at the last moment,
4-5, when the tail image is identified as overburning by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is between the range of Num and Num-1 bellows, outputting the end position and the end temperature obtained by the quadratic curve fitting method,
and 4-6, when the tail image is identified as overfire by the convolutional neural network CNN model, if the end position state obtained by the quadratic curve fitting method is located in a range from Num to Num-1 bellows, outputting the end position and the end temperature which are located between Num and Num-) bellows at the last moment.
2. The soft measurement method for quantitatively determining the state of the sintering endpoint in real time according to claim 1, wherein the first step specifically comprises:
the method comprises the steps of acquiring a detection value of the exhaust gas temperature of the wind box in the whole sintering production line and the specific inserting position of a thermocouple in the wind box in real time, recording tail images shot by a CCD camera at the tail of the sintering machine in the same period, and taking the exhaust gas temperature of the wind box and the detected images of the tail as basic data sources for real-time judgment of the end point state.
3. The soft measurement method for quantitatively determining the state of the sintering endpoint in real time according to claim 1, wherein the third step specifically comprises:
according to the position difference of a red fire layer in the machine tail image, dividing the image into 3 categories, namely under-burning, normal and over-burning, wherein the under-burning is that the red fire layer is positioned at the top of the machine tail image, and the lower part of the red fire layer is all raw materials; normally, the red fire layer is at the bottom of the tail image and occupies one third of the whole section height; the overburning is that only a narrow red fire layer is arranged at the bottom of the machine tail image, all the upper parts of the red fire layers are sintered blocks, and the machine tail image is classified and judged by adopting a Convolutional Neural Network (CNN) model.
4. The soft measurement method for quantitatively determining the sintering end point state in real time according to claim 1, comprising: a CCD camera, a storage device and a processor; the processor loads and executes data, images and rules in the storage device to realize the soft measuring method for quantitatively judging the state of the sintering end point in real time as claimed in any one of claims 1 to 3.
CN202011127923.5A 2020-10-20 2020-10-20 Soft measurement method for real-time quantitative determination of sintering end point state Active CN112255364B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011127923.5A CN112255364B (en) 2020-10-20 2020-10-20 Soft measurement method for real-time quantitative determination of sintering end point state

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011127923.5A CN112255364B (en) 2020-10-20 2020-10-20 Soft measurement method for real-time quantitative determination of sintering end point state

Publications (2)

Publication Number Publication Date
CN112255364A CN112255364A (en) 2021-01-22
CN112255364B true CN112255364B (en) 2022-07-01

Family

ID=74244055

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011127923.5A Active CN112255364B (en) 2020-10-20 2020-10-20 Soft measurement method for real-time quantitative determination of sintering end point state

Country Status (1)

Country Link
CN (1) CN112255364B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113111571B (en) * 2021-03-10 2022-12-02 山东诺德能源科技有限公司 Method for predicting sintering end point according to air box temperature
CN113511885B (en) * 2021-08-03 2023-10-27 苏州楚翰真空科技有限公司 Method and system for sintering ceramic material
CN114216349B (en) * 2021-12-06 2022-09-16 浙江大学 Sintering end point forecasting method based on coding and decoding network
CN115031534A (en) * 2022-04-15 2022-09-09 北京智冶互联科技有限公司 Method, device and equipment for judging zone sintering end point based on thermal imaging
CN117213265B (en) * 2023-11-09 2024-02-20 江苏赛冲智能装备有限公司 Sintering state detection method and system

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101598934A (en) * 2009-07-14 2009-12-09 北京首钢自动化信息技术有限公司 A kind of method for indirectly controlling of sintering end point
CN103031435A (en) * 2012-12-27 2013-04-10 中冶长天国际工程有限责任公司 Control method and system for sintering end point position
CN103472787A (en) * 2013-09-07 2013-12-25 鞍钢股份有限公司 Online judgment method for sintering fuel ratio
CN105039685A (en) * 2015-08-03 2015-11-11 中南大学 Method for soft measurement of iron ore sintering process state
CN106636616A (en) * 2016-12-20 2017-05-10 中国地质大学（武汉） Sintered ore yield prediction method based on bellows waste gas temperature
CN206430762U (en) * 2017-01-17 2017-08-22 马鞍山钢铁股份有限公司 A kind of sintering quality detecting system
CN108469180A (en) * 2018-04-09 2018-08-31 华北理工大学 The method for building up of sintering end point forecasting system based on big data and machine learning
CN110245850A (en) * 2019-05-31 2019-09-17 中国地质大学（武汉） A kind of sintering process operating mode's switch method and system considering timing
CN111128313A (en) * 2019-07-16 2020-05-08 中南大学 Method and system for detecting FeO content of sinter

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542239B (en) * 2010-12-17 2015-04-08 鞍钢股份有限公司 Automatic capturing method for effective image of tail red section of sintering machine
CN109632793A (en) * 2018-12-13 2019-04-16 国网陕西省电力公司 Experiment porch and method for the research of cable tunnel fire temperature field simulated behavior
CN110782067B (en) * 2019-09-23 2022-06-14 中国地质大学（武汉） Sintering end point fluctuation range prediction method based on fuzzy information granulation
CN111401213A (en) * 2020-03-11 2020-07-10 上海中科教育装备集团有限公司 Flame detection experimental device and method for chemical gas detection
CN111505198A (en) * 2020-04-22 2020-08-07 中国特种设备检测研究院 Large-particle coke combustion characteristic testing system

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101598934A (en) * 2009-07-14 2009-12-09 北京首钢自动化信息技术有限公司 A kind of method for indirectly controlling of sintering end point
CN103031435A (en) * 2012-12-27 2013-04-10 中冶长天国际工程有限责任公司 Control method and system for sintering end point position
CN103472787A (en) * 2013-09-07 2013-12-25 鞍钢股份有限公司 Online judgment method for sintering fuel ratio
CN105039685A (en) * 2015-08-03 2015-11-11 中南大学 Method for soft measurement of iron ore sintering process state
CN106636616A (en) * 2016-12-20 2017-05-10 中国地质大学（武汉） Sintered ore yield prediction method based on bellows waste gas temperature
CN206430762U (en) * 2017-01-17 2017-08-22 马鞍山钢铁股份有限公司 A kind of sintering quality detecting system
CN108469180A (en) * 2018-04-09 2018-08-31 华北理工大学 The method for building up of sintering end point forecasting system based on big data and machine learning
CN110245850A (en) * 2019-05-31 2019-09-17 中国地质大学（武汉） A kind of sintering process operating mode's switch method and system considering timing
CN111128313A (en) * 2019-07-16 2020-05-08 中南大学 Method and system for detecting FeO content of sinter

Non-Patent Citations (13)

* Cited by examiner, † Cited by third party
Title
Adaptive prediction system of sintering through point based on self-organize artificial neural network;Q.-M. Feng 等;《Transactions of Nonferrous Metals Society of China 》;20001231;第10卷(第6期);第804-807页 *
Fuzzy-neural-network-based quality prediction system for sintering process;Jun Liao, M. J. 等;《Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251)》;19990602;第5卷;第3221-3244页 *
基于BP神经网络的烧结终点预测模型;谌晓文 等;《烧结球团》;20060831(第4期);第21-26页 *
基于人工神经网络的烧结终点预报系统;汤程等;《微计算机信息》;20070615;第23卷(第17期);第256-257页 *
基于大数据技术的烧结全产线质量智能控制系统;吕庆 等;《钢铁》;20180731;第53卷(第7期);第1-9页 *
基于支持向量机的烧结终点预报与控制研究;吴晓峰;《中国优秀博士学位论文全文数据库工程科技Ⅰ辑》;20100515(第5期);第B023-1页 *
基于机尾断面红外特征构建烧结过程温度三维趋势图的方法研究;吴晓峰;《冶金自动化》;20070325(第02期);第14-18页 *
基于混匀矿烧结基础特性的烧结配矿专家系统研究;刘颂;《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》;20190515(第5期);第B023-50页 *
杨春雨 等.鞍钢烧结生产过程综合智能控制系统.《冶金自动化》.2011,第35卷(第6期), *
烧结生产过程的智能控制技术;杨东晓等;《辽宁科技大学学报》;20081015;第31卷(第05期);第470-474页 *
烧结终点预测模型与控制方法研究;魏玉龙;《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》;20181215(第12期);第B023-5页 *
运用神经网络预报烧结终点;蒋大军;《中国冶金》;20080930(第9期);第10-15页 *
鞍钢烧结生产过程综合智能控制系统;杨春雨 等;《冶金自动化》;20111231;第35卷(第6期);第1-6+20页 *

Also Published As

Publication number Publication date
CN112255364A (en) 2021-01-22

Similar Documents

Publication Publication Date Title
CN112255364B (en) Soft measurement method for real-time quantitative determination of sintering end point state
CN110889091B (en) Machine tool thermal error prediction method and system based on temperature sensitive interval segmentation
JP3660763B2 (en) Inspection pattern inspection method, manufacturing process diagnosis method, and semiconductor substrate manufacturing method
CN110716500B (en) Method and system for determining segmented modeling points of temperature sensitive interval
CN105458363B (en) The bloom chamfering reworking method of cell phone appearance part
CN115015286B (en) Chip detection method and system based on machine vision
CN117195439B (en) Intelligent optimization method for large-shaft-diameter full-section mechanical sealing performance
CN116881530B (en) Device surface defect detection system based on deep learning
CN117115148A (en) Chip surface defect intelligent identification method based on 5G technology
WO2022049549A1 (en) Artificial intelligence based tobacco particle measurement system
US10643326B2 (en) Semiconductor measurement apparatus and computer program
CN107193782A (en) A kind of method of abnormal value removing and correction fitted based on multinomial
CN117557806A (en) Visual monitoring method, device, equipment and storage medium for growth of silicon rod
JP4704897B2 (en) 3D measurement data inspection method using parametric tolerance
CN112085708A (en) Method and equipment for detecting defects of straight line edge in product outer contour
TWI801973B (en) Device and method for estimating error factors
TW201924828A (en) Method for predicting accuracy of electrical discharge machining capable of establishing or modifying a electrical discharge machining accuracy prediction model, and can predict the accuracy of electrical discharge machining
TW201821192A (en) Method for extracting intelligent features for predicting precision of electrical discharge machine and predicting method
US20020051561A1 (en) Image processing apparatus, image processing method, and recording medium recording image processing program
CN117148803B (en) Adjusting control method for automatic centering width adjusting assembly line
CN112001919B (en) Fabric defect detection method based on particle swarm algorithm and two-dimensional digital filtering
CN115205317B (en) Bridge monitoring photoelectric target image light spot center point extraction method
CN115015810A (en) Electric connection defect diagnosis system applied to induction capacitor
WO2022044816A1 (en) Blast furnace condition determination device, operation method of blast furnace, and manufacturing method for molten iron
CN114454179A (en) Automatic correction system and method for joints of intelligent robot hand

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