CN103571994B - Infrared steel slag detection method of converter - Google Patents

Abstract

The invention discloses a converter infrared steel slag detection method. In the first step, when the steel tapping start signal in the production process of the converter is detected, the converter infrared steel slag detection device is put into operation; in the second step, after the device is put into operation, the infrared camera The image of the converter is measured in real time, and the industrial computer processes the image. The processing steps are: firstly, the image is segmented, and the collected image includes: steel slag, molten steel and background, and the image pixels are processed by histogram technology to find a suitable threshold , use this threshold to divide steel slag and molten steel; then calculate the percentage content of steel slag, the percentage content of steel slag is the percentage of the steel slag area in the steel flow area in the steel flow area, corresponding to the digital image, the steel slag area is the sum of steel slag pixels, steel The flow area is the sum of steel flow pixels; the third step is to judge whether the set steel slag alarm limit is exceeded according to the steel slag percentage value, and if the limit value is exceeded, an alarm signal will be output and pouring will be stopped.

Description

Infrared steel slag detection method for converter

technical field

The present invention relates to a detection method for converter steel slag.

Background technique

In steelmaking production, excessive slag flowing out from the ladle has a bad influence on the quality of steel. Excessive slag will reduce the purity of steel, accelerate the erosion of refractory materials, and reduce the service life of equipment. In order to improve the quality of molten steel, during the process of pouring molten steel into the converter, the operator needs to observe the change of molten steel all the time. When the brightness of poured molten steel increases, it proves that steel slag has appeared. Different steelmaking processes are used, and the surrounding environmental conditions are different. In addition to molten steel, a large amount of suspended dust and smoke will also be produced. In this case, it is difficult for the naked eye of the operator to accurately judge the appearance of steel slag according to the change of the brightness of molten steel. Even in a clean environment, only a few experienced operators can accurately distinguish the brightness of molten steel and steel slag. differentiate. Therefore, the process is largely determined by the skill and experience of the operator.

After retrieval and analysis, the relevant reports on the detection of molten steel slag in converter at home and abroad mainly describe the experience and effect of related equipment on site, and there are no relevant reports on the detection method and device of converter steel slag.

Contents of the invention

The object of the present invention is to provide a converter infrared steel slag detection method, which can realize the measurement of the steel slag during the tapping process of the converter.

In order to achieve the above-mentioned purpose of the invention, the present invention adopts the following technical solutions:

A converter infrared steel slag detection method,

The first step is to monitor the tapping start signal during the production process of the converter. If the signal is not detected, it means that the converter has not started pouring, and the infrared steel slag detection device of the converter remains in a waiting state; when the signal is detected, it means that the converter has started to pour. Steel, converter infrared steel slag detection device put into operation;

The converter infrared steel slag detection device includes an infrared CCD camera, an optical transmitter, an optical cable, an optical receiver, and a central control cabinet; the infrared CCD camera is a thermal imager used to detect converter infrared radiation, and the infrared CCD camera outputs electrical signals through The optical transmitter converts the electrical signal into an optical signal, and then transmits it to the optical receiver through an optical cable, and then converts it into an electrical signal and then inputs it to the central control cabinet; the central control cabinet includes an industrial computer, an image acquisition card, and a data acquisition card And the control card, the image acquisition card collects infrared images, the data acquisition card collects the temperature, the industrial computer processes the data input by the image acquisition card and the data acquisition card, and then recognizes and judges the steel slag, and controls the external execution parts through the output of the control card;

In the second step, after the converter infrared steel slag detection device is put into operation, the infrared CCD camera measures the converter image in real time, and the industrial computer processes the image. The processing steps are:

2.1 Image Segmentation Processing

The collected image contains three kinds of objects: steel slag, molten steel and background. The pixels of the image are processed by histogram technology to find a suitable threshold, and this threshold is used to divide steel slag and molten steel. The histogram of the image with obvious contrast with the background The two peaks correspond to a large number of points inside and outside the object, and the valley between the two peaks corresponds to a relatively small number of points near the edge of the object. The gray threshold T ₁ can be used to calculate the area of the object:

Among them: T is the threshold value, that is, the valley; D is the gray level;

D in the above formula can be integrated along T to ∞ to obtain two thresholds, the threshold T ₁ of background and steel flow; the threshold T ₂ of steel slag and molten steel;

Define collections:

is a subset of steel slag pixels;

is a subset of molten steel pixels;

is the collected steel flow pixel set;

is the background pixel set;

is the pixel set of a frame image;

First, divide the background and steel flow into two subsets in the pixel set of the collected image:

       

       

Secondly, from the steel flow set, two subsets of steel slag and molten steel are divided:

       

       

2.2 Calculation of percentage content of steel slag

The steel slag percentage content is the percentage of the steel slag area in the steel flow area in the steel flow area. Corresponding to the digital image, the steel slag area is the sum of the steel slag pixels, and the steel flow area is the sum of the steel flow pixels;

in steel stream collection The total number of pixels in is q ; in the slag set The total number of pixels in m is m , and in the molten steel set The total number of pixels in is n , defined by the stream set: get:

;

Define the percentage content of steel slag as :

After the steel slag subset and the molten steel subset are divided, the total number of pixels m in the steel slag subset and the total number of pixels n in the molten steel subset are counted respectively, and the percentage content of steel slag is obtained by substituting into the above formula for calculation;

The third step is to judge whether the set steel slag alarm limit is exceeded according to the percentage content of steel slag. When the steel slag content exceeds the alarm limit, the industrial computer outputs an alarm signal and controls the execution part to stop pouring through the output of the control card, and enters the first step to continue Wait for the converter to start pouring signal; when the steel slag content does not exceed the alarm limit, continue to work according to the second step.

The invention collects the infrared image of the converter in real time, separates molten steel and steel slag from the image background by means of image segmentation, and then realizes the calculation of the percentage content of steel slag by means of statistical recognition, and finally realizes the detection of steel slag. The method of the invention has been verified by on-site use, and has high detection accuracy, the steel slag identification accuracy rate reaches 99%, and the operation is reliable.

Description of drawings

Fig. 1 is the structure schematic diagram of converter infrared steel slag detection device of the present invention;

Fig. 2 is the flow chart of converter infrared steel slag detection method of the present invention;

Figure 3 is the original image of the converter infrared image;

Fig. 4 is the binary image of the molten steel separated from the infrared image of the converter;

Fig. 5 is the binary image of the steel slag separated from the infrared image of the converter;

In the figure: 1 infrared CCD camera, 2 optical transmitter, 3 optical cable, 4 optical receiver, 5 central control cabinet; 5-1 industrial computer, 5-2 image acquisition card, 5-3 data acquisition card, 5- 4 control cards.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

Referring to Fig. 1, a kind of converter infrared steel slag detection device comprises infrared CCD camera 1, optical end transmitter 2, optical cable 3, optical end receiver 4, central control cabinet 5; Described infrared CCD camera 1 is for detecting converter infrared The main function of the radiation thermal imager is to convert the infrared radiation emitted by the converter into electrical signals through the detection unit.

The infrared CCD camera 1 outputs an electrical signal through the optical transmitter 2 to convert the electrical signal into an optical signal, then transmits it to the optical receiver 4 through the optical cable 3, and then converts it into an electrical signal and then inputs it to the central control cabinet 5. The optical transmitter 2, the optical cable 3 and the optical receiver 4 are a photoelectric signal conversion and transmission device. Since there are many interference sources in the converter production site, converting the converter image signal into an optical signal for transmission can effectively eliminate the interference on site and improve the stability of the signal.

Described central control cabinet 5 comprises industrial computer 5-1, image acquisition card 5-2, data acquisition card 5-3 and control card 5-4, and image acquisition card 5-2 collects infrared images, and data acquisition card 5-3 collects Temperature, the industrial computer 5-1 processes the data input by the image acquisition card 5-2 and the data acquisition card 5-3, and then recognizes and judges the steel slag, outputs an alarm signal, and controls the external execution components through the output of the control card 5-4.

A converter infrared steel slag detection method is carried out on the basis of a converter infrared steel slag detection device, the steps are: see Fig. 2,

The first step is to monitor the tapping start signal during the production process of the converter. If the signal is not detected, it means that the converter has not started pouring, and the infrared steel slag detection device of the converter remains in a waiting state; when the signal is detected, it means that the converter has started to pour. Steel, converter infrared steel slag detection device put into operation;

In the second step, after the converter infrared steel slag detection device is put into operation, the infrared CCD camera 1 measures the converter image in real time, and the industrial computer 5-1 processes the image by means of statistical pattern recognition. The processing steps are:

2.1 Image Segmentation Processing

In this step individual objects are detected and their images are separated from the rest of the scene. The collected image contains three kinds of objects: steel slag, molten steel and background. Steel slag and molten steel have a strong contrast due to the difference in emissivity, but the edge of the image of steel slag and molten steel is not steep. Therefore, the selection of the gray threshold is very important Extracting the boundary of an object has a big impact on positioning the overall size. The pixels of the image are processed by the histogram technique, and a suitable threshold can be found, which can be used to divide steel slag and molten steel. The two peaks of the histogram of the image with obvious contrast with the background correspond to a large number of points inside and outside the object, and the valley between the two peaks corresponds to a relatively small number of points near the edge of the object. The gray threshold T can be used to determine the area of the object Calculation:

Where: T is the threshold, that is, the valley;

D is the gray level;

From the above calculation results, the threshold corresponding to the bottom of the histogram can be found. This threshold increases from T to When , the calculated area is only slightly reduced. After finding the appropriate threshold T , the threshold T can be used to divide steel slag and molten steel. D in the above formula can be integrated along T to ∞ to obtain two thresholds, the threshold T ₁ of background and steel flow; the threshold T ₂ of steel slag and molten steel.

Define collections:

is a subset of steel slag pixels;

is a subset of molten steel pixels;

is the collected steel flow pixel set;

is the background pixel set;

is the pixel set of a frame image;

First, in the pixel set I of the collected image, two subsets are divided, the background pixel set B and the steel flow pixel set M:

       

       

Secondly, from the steel flow set, two subsets of steel slag and molten steel are divided:

       

       

2.2 Calculation of percentage content of steel slag

The steel slag percentage content is the percentage of the steel slag area in the steel flow area in the steel flow area. Corresponding to the digital image, the steel slag area is the sum of the steel slag pixels, and the steel flow area is the sum of the steel flow pixels;

in steel stream collection The total number of pixels in is q ; in the slag set The total number of pixels in m is m , and in the molten steel set The total number of pixels in is n , defined by the stream set: The total number of pixels in is get:

;

Define the percentage content of steel slag as :

After the steel slag subset and the molten steel subset are separated, the total number of pixels m in the steel slag subset and the total number of pixels in the molten steel subset n are counted respectively, and then substituted into the above formula to obtain the percentage content of steel slag.

The third step is to judge whether the set steel slag alarm limit is exceeded according to the steel slag percentage content value. When the steel slag content exceeds the alarm limit value, the industrial computer 5-1 outputs an alarm signal and controls the execution part to stop pouring through the control card 5-4 output , enter the first step and continue to wait for the converter pouring signal; when the steel slag content does not exceed the alarm limit, continue to work according to the second step.

When the converter infrared steel slag detection device is put into operation, the limit value of steel slag percentage content (that is, the steel slag alarm limit) is a variable, which is determined by the process engineer according to the needs of the process.

Example

A converter infrared steel slag detection method, the composition of the converter infrared steel slag detection device is shown in Figure 1.

In the first step, after the operation of the converter infrared slag detection device starts, the central control cabinet continuously detects the tapping start signal during the converter production process through the data acquisition card. If the signal is not detected, it means that the converter has not yet started pouring. The central control The control cabinet issues instructions to keep the converter infrared steel slag detection device in a waiting state and continue to detect the steel tapping start signal; when the signal is detected, it means that the converter begins to tap steel, and the central control cabinet sends an instruction to enable the converter infrared steel slag detection device to start and enter the first stage. Two steps work.

In the second step, after the infrared steel slag detection device of the converter is put into operation, the infrared image of the molten steel flowing out of the converter is first collected through the image acquisition card, and the original infrared image of the converter is obtained as shown in Figure 3. After obtaining the image data, through image segmentation The method is to separate the background image from the original image and extract the binary image containing molten steel, as shown in Figure 4, the white area is the molten steel area, so that the location of the molten steel in the image can be known; A binary image containing steel slag is extracted from the image, as shown in Figure 5.

After obtaining the areas of molten steel and steel slag, the calculation of the percentage content of steel slag is realized by statistically calculating the respective areas. The area of molten steel and steel slag is realized by calculating the number of respective pixels. The total number of pixels in is q ; in the slag set The total number of pixels in m is m , and in the molten steel set The total number of pixels in is n , defined by the stream set: The total number of pixels in is get:

;

Define the percentage content of steel slag as :

The total number m of pixels in the steel slag subset and the total number n of pixels in the molten steel subset are counted separately, and then substituted into the above formula to obtain the percentage content of steel slag.

In the third step, after obtaining the percentage content of steel slag, judge whether it has exceeded the value according to the steel slag alarm limit set according to the converter production process requirements. When the steel slag content exceeds the alarm limit, the converter infrared steel slag detection device outputs an alarm and stops pouring, enters the first step and continues to wait for the converter to start pouring signal; when the steel slag content does not exceed the alarm limit, continue to work according to the second step.

The above are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in this document. within the scope of protection of the invention.

Claims (1)

1. A converter infrared steel slag detection method is characterized in that: The first step is to monitor the tapping start signal during the production process of the converter. If the signal is not detected, it means that the converter has not started pouring, and the infrared steel slag detection device of the converter remains in a waiting state; when the signal is detected, it means that the converter has started to pour. Steel, converter infrared steel slag detection device put into operation; The converter infrared steel slag detection device comprises an infrared CCD camera (1), an optical transmitter (2), an optical cable (3), an optical receiver (4), and a central control cabinet (5); the infrared CCD camera (1) is A thermal imager used to detect the infrared radiation of the converter, the infrared CCD camera (1) outputs an electrical signal through the optical transmitter (2), converts the electrical signal into an optical signal, and then transmits it to the optical receiver (4) through the optical cable (3) ), then converted into an electrical signal and then input to the central control cabinet (5); the central control cabinet (5) includes an industrial computer (5-1), an image acquisition card (5-2), a data acquisition card (5-3 ) and control card (5-4), the image acquisition card (5-2) collects infrared images, the data acquisition card (5-3) collects temperature, and the industrial computer (5-1) controls the image acquisition card (5-2) and After the data input by the data acquisition card (5-3) is processed, the steel slag is identified and judged, and the external execution components are controlled through the output of the control card (5-4); In the second step, after the converter infrared steel slag detection device is put into operation, the infrared CCD camera (1) measures the converter image in real time, and the industrial computer (5-1) processes the image, and the processing steps are: 2.1 Image Segmentation Processing The collected image contains three kinds of objects: steel slag, molten steel and background. The pixels of the image are processed by histogram technology to find a suitable threshold, and this threshold is used to divide steel slag and molten steel. The histogram of the image with obvious contrast with the background The two peaks correspond to a large number of points inside and outside the object, and the valley between the two peaks corresponds to a relatively small number of points near the edge of the object, calculated using the following formula: $\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{\mathrm{<span\; class="notranslate">\; T</span>}}^{<span\; class="notranslate">\; \&infin;</span>}\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; D</span>}$ Among them: T is the threshold value, that is, the valley; D is the gray level; D in the above formula can be integrated along T to ∞ to obtain two thresholds, the threshold T ₁ of background and steel flow; the threshold T ₂ of steel slag and molten steel; Define collections: S(s ₁ ,s ₂ ,…,s _m ) is a subset of steel slag pixels; L(l ₁ ,l ₂ ,…,l _n ) is a subset of molten steel pixels; M(m ₁ ,m ₂ ...,m _q ) is the collected steel flow pixel set; B(b ₁ ,b ₂ ,…,b _l ) is the background pixel set; I(i ₁ ,i ₂ ,…i _x ) is the pixel set of a frame image; First, divide the background and steel flow into two subsets in the pixel set of the collected image: $<span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Subset;</span>\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span>$ $<span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ></span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Subset;</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; )</span>$ Secondly, from the steel flow set, two subsets of steel slag and molten steel are divided: $<span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; ></span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Subset;</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; )</span>$ ${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; <</span><span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Subset;</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$ 2.2 Calculation of steel slag percentage content The steel slag percentage content is the percentage of the steel slag area in the steel flow area in the steel flow area. Corresponding to the digital image, the steel slag area is the sum of the steel slag pixels, and the steel flow area is the sum of the steel flow pixels; The total number of pixels in the steel flow set M(m ₁ ,m ₂ ...,m _q ) is q; the total number of pixels in the steel slag set S(s ₁ ,s ₂ ,...,s _m ) is m, and the molten steel set L(l ₁ ,l ₂ ,…,l _n ) the total number of pixels is n, which is defined by the steel flow set: M(m ₁ ,m ₂ …,m _q ) to get: q=n+m; Define the percentage content of steel slag as ξ: $\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 100</span>}<span\; class="notranslate">\; \%</span>$ After the steel slag subset and the molten steel subset are divided, the total number of pixels m in the steel slag subset and the total number of pixels n in the molten steel subset are counted respectively, and the percentage content of steel slag is obtained by substituting into the above formula for calculation; The third step is to judge whether the set steel slag alarm limit is exceeded according to the steel slag percentage content value. When the steel slag content exceeds the alarm limit value, the industrial computer (5-1) outputs an alarm signal and controls it through the control card (5-4). The execution part stops pouring, enters the first step and continues to wait for the converter to start pouring signal; when the steel slag content does not exceed the alarm limit, continue to work according to the second step.