A method for measuring the soft cross temperature of a blast furnace based on infrared temperature measurement

Abstract

The present invention relates to the field of blast furnace metallurgy, and discloses a method for soft cross temperature measurement of a blast furnace based on infrared temperature measurement. The method includes: First, install a cross temperature measurement device and an infrared temperature measurement device on the blast furnace, and determine the virtual cross temperature measurement points on the infrared image according to the position mapping relationship between the pixel points of the infrared image and the cross temperature measurement points; and train the temperature relationship model between the virtual cross temperature measurement points and the real cross temperature measurement points according to the temperature values of the virtual cross temperature measurement points and the temperature values of the cross temperature measurement points; then use this model on the blast furnace equipped only with infrared temperature measurement to calculate the predicted temperature values of the cross temperature measurement points. By adopting this method, there is no need to install a traditional cross temperature measurement device, convert the temperature measurement result of the original infrared image into the temperature value of the cross temperature measurement point, and use the existing cross temperature measurement temperature model to perform corresponding blast furnace operations and decisions.

Description

Technical Field

[0001] The present invention relates to the field of blast furnace ironmaking in the metallurgical industry, and particularly to a soft cross temperature measurement method for blast furnaces based on infrared temperature measurement. Background Art

[0002] Cross temperature measurement devices have been widely used in bell-less top blast furnaces. They can continuously and accurately measure the temperature distribution of the coal gas flow in the radial direction of the furnace throat. Since the coal gas flow is vigorous in the areas with high temperature, the cross temperature measurement device can effectively monitor the distribution of the coal gas flow at the top of the blast furnace. The cross temperature measurement device is located at the furnace throat position of the blast furnace, and four temperature measurement arms are installed in four directions on the furnace throat circumferential surface. Each temperature measurement wall is equipped with unequal temperature sensors (such as thermocouples), with a total of 17 - 21 temperature measurement sensors. These temperature measurement points can provide real-time temperature data and can comprehensively reflect the distribution of the coal gas flow in the circumferential direction of the furnace throat. The cross temperature measurement device is installed at the top of the closed blast furnace cylinder, and the operating environment is extremely harsh. Therefore, the sensors of the cross temperature measurement device are prone to damage. After damage, maintenance cannot be carried out in a timely manner, and the sensors must be replaced until the large-period overhaul of the blast furnace.

[0003] The non-contact infrared thermal imaging temperature measurement method was developed in the 1990s and has now been widely used in the temperature measurement of the blast furnace top, and can observe the temperature field distribution in the blast furnace internal space. Since the infrared focal plane array of the thermal imaging sensing element does not need to contact the target in the furnace internal space, it will not affect the temperature field distribution and the sensor life. It uses infrared thermal radiation for temperature measurement and has the advantages of fast response, long sensor life, non-consumable, high measurement temperature, and can achieve real-time continuous measurement.

[0004] According to the infrared radiation temperature measurement principle, it can be known that the temperature measurement result has a great relationship with the emissivity of the target object. Through actual measurement data, it is found that when the emissivity of the infrared detector is set to 0.97, the temperature measurement result of the infrared detector is much lower than the cross temperature measurement result. Therefore, the operator cannot directly use the temperature measurement result of the original infrared image to replace the cross temperature measurement and perform corresponding blast furnace operations and decisions using the existing temperature model of the cross temperature measurement. Summary of the Invention

[0005] In view of the above-mentioned defects of the prior art, the purpose of the present invention is to combine the furnace top infrared temperature measurement technology with the cross temperature measurement technology, without installing the traditional cross temperature measurement device, convert the temperature measurement result of the original infrared image into the temperature value of the set cross temperature measurement points, and perform corresponding blast furnace operations and decisions using the existing cross temperature measurement temperature model.

[0006] To achieve the above purpose, the present invention provides a soft cross temperature measurement method for blast furnaces based on infrared temperature measurement, including the following steps:

[0007] Step S10: Simultaneously set up an infrared temperature measurement device and a cross temperature measurement device on the blast furnace; determine the virtual cross temperature measurement points on the infrared image according to the position mapping relationship between the pixel points of the infrared image and the cross temperature measurement points; and train the temperature relationship model between the virtual cross temperature measurement points and the real cross temperature measurement points based on the temperature values of the virtual cross temperature measurement points and the cross temperature measurement points.

[0008] Step S20: Only set up an infrared temperature measurement device on the blast furnace; set the positions of the cross temperature measurement points, determine the virtual cross temperature measurement points on the infrared image according to the position mapping relationship between the pixel points of the infrared image and the cross temperature measurement points; extract the temperature values of the virtual cross temperature measurement points from the infrared image, and then output the predicted temperature values of the cross temperature measurement points through the temperature relationship model.

[0009] Furthermore, the step S10 includes:

[0010] Step S101: Simultaneously install a cross temperature measurement device and an infrared temperature measurement device on the blast furnace.

[0011] Step S102: Establish a plane coordinate system of the blast furnace according to the installation plane of the cross temperature measurement device in the furnace.

[0012] Step S103: Establish a perspective transformation matrix according to the installation position of the infrared temperature measurement device on the furnace top, and correct the infrared image I1 into an aerial view temperature image I2 located in the plane coordinate system of the blast furnace.

[0013] Step S104: Find the corresponding pixel point coordinates in the aerial view temperature image I2 according to the installation position of the cross temperature measurement device in the furnace, and find the corresponding pixel point coordinates in the infrared image I1 through these pixel point coordinates to establish virtual cross temperature measurement points.

[0014] Step S105: Read the temperature values of the virtual cross temperature measurement points to form a first temperature sequence.

[0015] Step S106: Read the temperature values of the cross temperature measurement points through the cross temperature measurement device to form a second temperature sequence.

[0016] Step S107: Use the first temperature sequence as the input and the second temperature sequence as the output to train the temperature relationship model between the virtual cross temperature measurement points and the real cross temperature measurement points.

[0017] Furthermore, in step S10, the plane coordinate system of the blast furnace is set with the origin of the blast furnace top coordinate located at the center of the blast furnace on the horizontal plane where the temperature measurement point of the cross temperature measurement device is located.

[0018] Furthermore, the step S103 includes:

[0019] Select the coordinates of four points on the circumference of the furnace shell in the blast furnace as A(-R, 0), B(R, 0), C(0, R), D(0, -R), where R is the inner diameter of the blast furnace. The coordinates of these four points on the infrared image are obtained by manually measuring or calculating the upper, lower, left, and right boundary points through an image algorithm to get A'(XA, YA), B'(XB, YB), C'(XC, YC), D'(XD, YD);

[0020] According to the principle of perspective transformation, calculate the perspective transformation matrix P1 from the blast furnace coordinate system to the infrared image I1 coordinate system, and obtain the bird's-eye view temperature image I2 of the original infrared image I1 transformed to the horizontal coordinate system of the blast furnace based on the perspective transformation matrix.

[0021] Furthermore, in the step S107, the temperature relationship model is based on the MLP neural network model.

[0022] Furthermore, the step S20 includes:

[0023] Step S201: Install an infrared temperature measurement device at the top of the blast furnace to obtain the infrared image of the furnace top;

[0024] Step S202: Establish a perspective transformation matrix according to the installation position of the infrared temperature measurement device at the furnace top, and correct the infrared image I3 to the bird's-eye view temperature image I4 located in the blast furnace plane coordinate system;

[0025] Step S203: Set the position of the cross temperature measurement points;

[0026] Step S204: Find the pixel point coordinates corresponding to the cross temperature measurement points from the bird's-eye view temperature image I4, and then establish virtual cross temperature measurement points in the infrared image I3 through the perspective transformation matrix;

[0027] Step S205: Extract the temperature values of the virtual cross temperature measurement points to form a third temperature sequence;

[0028] Step S206: Input the third temperature sequence into the temperature relationship model of the virtual cross temperature measurement points and the real cross temperature measurement points that have been trained, and output a fourth temperature sequence composed of the predicted temperature values of each cross temperature measurement point.

[0029] The present invention achieves the following technical effects:

[0030] By adopting this method, there is no need to install a traditional cross temperature measurement device. The original infrared image is obtained through an infrared temperature measurement device, the temperature measurement result of the original infrared image is converted into the temperature value of the cross temperature measurement points, and the existing cross temperature measurement temperature model is used to perform corresponding blast furnace operations and decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1It is the flowchart of the soft cross temperature measurement method for blast furnaces based on infrared temperature measurement of the present invention;

[0032] Figure 2 It is the position mapping diagram of the infrared image and the bird's-eye temperature image of the present invention;

[0033] Figure 3 It is the neural network model involved in the present invention. Specific embodiments

[0034] To further illustrate the embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used to explain the operating principle of the embodiments in conjunction with the relevant descriptions in the specification. With reference to these contents, those of ordinary skill in the art should be able to understand other possible embodiments and the advantages of the present invention.

[0035] Now, the present invention will be further described in conjunction with the drawings and specific embodiments.

[0036] As Figure 1 shown, the present application proposes a soft cross temperature measurement method for blast furnaces based on infrared temperature measurement, including the following steps:

[0037] Step S10, install an infrared temperature measurement device and a cross temperature measurement device on the blast furnace at the same time; determine the virtual cross temperature measurement points on the infrared image according to the position mapping relationship between the pixel points of the infrared image and the cross temperature measurement points; and train the temperature relationship model between the virtual cross temperature measurement points and the real cross temperature measurement points according to the temperature values of the virtual cross temperature measurement points and the cross temperature measurement points.

[0038] Step S20, only install an infrared temperature measurement device on the blast furnace; set the positions of the cross temperature measurement points, determine the positions of the virtual cross temperature measurement points on the infrared image according to the position mapping relationship between the pixel points of the infrared image and the cross temperature measurement points; extract the temperature values of the virtual cross temperature measurement points from the infrared image, and then output the predicted temperature values of the cross temperature measurement points through the temperature relationship model.

[0039] Specifically, step S10 includes:

[0040] Install a cross temperature measurement device and an infrared temperature measurement device on the blast furnace at the same time; establish a plane coordinate system of the blast furnace according to the installation plane of the cross temperature measurement device in the furnace; establish a perspective transformation matrix according to the installation position of the infrared temperature measurement device on the furnace top, and correct the infrared image I1 into a bird's-eye temperature image I2 located in the plane coordinate system of the blast furnace; find the corresponding pixel point coordinates in the bird's-eye temperature image I2 according to the installation position of the cross temperature measurement device in the furnace, and establish virtual cross temperature measurement points according to these pixel point coordinates. Read the temperature values of the virtual cross temperature measurement points to form a first temperature sequence; read the temperature values of the cross temperature measurement points through the cross temperature measurement device to form a second temperature sequence.

[0041] Specifically, in this embodiment, a specific implementation solution for the above steps is given: Since the infrared temperature measurement camera cannot be installed on the center line of the blast furnace top and take pictures completely parallel to the horizontal plane, but is installed on the side wall of the blast furnace at a certain angle to the horizontal plane, the concentric circle pattern of the cross temperature measurement is not a circle in the infrared image, but a pattern after perspective transformation of the concentric circles, as Figure 2 shown. Therefore, the first step is to solve the perspective transformation matrix P from the actual coordinates of the blast furnace to the infrared image coordinates.

[0042] Establish a blast furnace coordinate system. Let the origin of the blast furnace top coordinate be located at the center of the blast furnace on the installation plane of the cross temperature measurement device (i.e., the horizontal plane where the temperature measurement points of the cross temperature measurement device are located), as Figure 2 shown. The coordinates of four points on the circumference of the inner furnace shell in the blast furnace are A(-R, 0), B(R, 0), C(0, R), D(0, -R), where R is the inner diameter of the blast furnace. The coordinates of these four points in the infrared image can be obtained by manually measuring or calculating the upper, lower, left, and right boundary points through an image algorithm as A’(XA, YA), B’(XB, YB), C’(XC, YC), D’(XD, YD), with the point in the upper left corner of the image as the origin. According to the principle of perspective transformation, the perspective transformation matrix P1 from the blast furnace coordinate system to the infrared image I1 coordinate system can be calculated, and the bird's-eye view temperature image I2 of the original infrared image I1 converted to the horizontal coordinate system of the blast furnace can be obtained according to the perspective transformation matrix.

[0043] Thus, a virtual cross temperature measurement scheme can be designed. For example, a temperature measurement point is set every 0.8 meters in the 4 directions of 0°, 90°, 180°, and 270°, and 5 points are set in each direction. The coordinates of the temperature measurement points are:

[0044]

[0045] Then its corresponding points in the infrared image I3 are

[0046]

[0047] which are the position coordinates of the virtual cross temperature measurement points.

[0048] This virtual cross temperature measurement scheme can adopt the original cross temperature measurement scheme, thus facilitating the use of the original temperature model for blast furnace operation and decision-making. A new cross temperature measurement scheme can also be set, and a new temperature model can be trained to support blast furnace operation and decision-making.

[0049] Then, calculate the conversion relationship between the two temperatures. Considering that the conversion relationship between the two may be non - linear, a neural network model with multiple hidden layers is designed to represent this relationship. The input layer of the network is the result of infrared temperature measurement, and the output layer is the measured temperature result of the corresponding cross - temperature measurement point obtained by fitting. The network contains 2 hidden layers, and each hidden layer contains 10 nodes, as Figure 3 shown.

[0050] In this embodiment, the following training scheme for the temperature relationship model between the virtual cross - temperature measurement point and the real cross - temperature measurement point is given: Sample the temperatures of 20 cross - temperature measurement points and their corresponding infrared temperature measurement results at intervals of 2 hours. They form a sample data. In this implementation scheme, 12,000 groups of sample data are sampled for 50 days to perform cross - validation on the temperature correction network. We randomly use 80% of the data as the training set to train the temperature correction network, 10% of the data as the validation set to control the number of training cycles, and 10% of the data as the test set to test the generalization effect of the model. The loss function uses the mean squared error (MSE), and its definition is:

[0051]

[0052] where

[0053] M is the number of samples in a certain sample set, and the sample set includes the training set, validation set, and test set;

[0054] T(k), T S (k) are the true cross - temperature measurement values and the measured temperature values output by the network for the k - th measurement, respectively.

[0055] After training for 100 epochs using the Adam optimization algorithm and the error change of the validation set tending to 0, the training is terminated, and the weight and bias parameters are saved to the model file Model.

[0056] In the specific implementation process, to train the temperature relationship model between the virtual cross - temperature measurement point and the real cross - temperature measurement point, neural network models such as MLP (Multi - Layer Perceptron) can be used for training. MLP multi - layer perceptron is a forward - structured artificial neural network ANN that maps a set of input vectors to a set of output vectors. MLP can be regarded as a directed graph composed of multiple node layers, where each layer is fully connected to the next layer. Except for the input nodes, each node is a neuron with a non - linear activation function.

[0057] Specifically, step 20 includes:

[0058] Step S201: Install an infrared temperature measurement device on the blast furnace top;

[0059] Step S202: According to the installation position of the infrared temperature measurement device on the furnace top, establish a perspective transformation matrix to correct the infrared image I3 into an aerial temperature image I4 located in the blast furnace plane coordinate system;

[0060] Step S203: Set the positions of the cross temperature measurement points;

[0061] Step S204: Find the pixel point coordinates corresponding to the cross temperature measurement points in the aerial temperature image I4, and then establish virtual cross temperature measurement points in the infrared image I3 through the perspective transformation matrix;

[0062] Step S205: Extract the temperature values of the virtual cross temperature measurement points to form a third temperature sequence;

[0063] Step S206: Input the third temperature sequence into the trained temperature relationship model between the virtual cross temperature measurement points and the real cross temperature measurement points, and output a fourth temperature sequence composed of the predicted temperature values of each cross temperature measurement point.

[0064] The fourth temperature sequence is the predicted temperature values of the cross temperature measurement points. Based on these predicted temperature values, the corresponding blast furnace operations and decisions can be executed using the existing temperature model for cross temperature measurement.

[0065] Although the present invention is specifically shown and described in combination with the preferred embodiments, those skilled in the art should understand that various changes can be made to the present invention in form and detail without departing from the spirit and scope of the present invention defined by the appended claims, and all of them fall within the protection scope of the present invention.

Claims

1. A blast furnace soft cross temperature measurement method based on infrared temperature measurement, characterized in that: The following steps are involved: Step S10: installing an infrared temperature measuring device and a cross temperature measuring device on the blast furnace at the same time; determining a virtual cross temperature measuring point on the infrared image based on a mapping relationship between the positions of the pixels of the infrared image and the positions of the cross temperature measuring points; And according to the temperature value of the virtual cross temperature measuring point and the temperature value of the cross temperature measuring point, the temperature relationship model of the virtual cross temperature measuring point and the real cross temperature measuring point is trained; Step S20, only an infrared temperature measuring device is installed on the blast furnace; The position of the cross temperature measurement point is set, and the virtual cross temperature measurement point on the infrared image is determined according to the mapping relationship between the pixel points of the infrared image and the position of the cross temperature measurement point; Extract the temperature value of the virtual cross temperature measurement point from the infrared image, and then output the predicted temperature value of the cross temperature measurement point through the temperature relationship model; The step S10 includes: Step S101: installing a cross temperature measuring device and an infrared temperature measuring device on the blast furnace at the same time; Step S102: establishing a plane coordinate system of the blast furnace according to the installation plane of the cross temperature measuring device in the furnace; Step S103: establishing a perspective transformation matrix based on the installation position of the infrared temperature measuring device on the furnace roof, and correcting the infrared image I1 into a bird's-eye view temperature image I2 located in the blast furnace plane coordinate system; Step S104: According to the installation position of the cross temperature measuring device in the furnace, the pixel coordinates corresponding to the cross temperature measuring point are found in the bird's-eye view temperature image I2. Based on these pixel coordinates, the corresponding pixel coordinates are found in the infrared image I1 to establish a virtual cross temperature measuring point; Step S105: reading the temperature values of the virtual cross temperature measurement points to form a first temperature sequence; Step S106: reading the temperature values of the cross temperature measuring points by a cross temperature measuring device to form a second temperature sequence; Step S107: using the first temperature sequence as input and the second temperature sequence as output, training a temperature relationship model between the virtual cross temperature measurement point and the real cross temperature measurement point.

2. The blast furnace soft cross temperature measurement method based on infrared temperature measurement according to claim 1, characterized in that: In step S10, the blast furnace plane coordinate system is set so that the blast furnace top coordinate origin is located at the center of the blast furnace on the plane at the cross temperature measurement installation height.

3. The blast furnace soft cross temperature measurement method based on infrared temperature measurement according to claim 2, characterized in that: The step S103 includes: The coordinates of four points on the circumference of the blast furnace shell are selected as A(-R, 0), B(R.0), C(0, R), and D(0, -R), where R is the inner diameter of the blast furnace. The coordinates of these four points on the infrared image are obtained by manually measuring or calculating the upper, lower, left, and right boundary points using image algorithms to obtain A'(XA, YA), B'(XB, YB), C'(XC, YC), and D'(XD, YD); According to the perspective transformation principle, the perspective transformation matrix P1 from the blast furnace coordinate system to the infrared image I1 coordinate system is calculated, and the bird's-eye view temperature image I2 of the original infrared image I1 converted to the horizontal coordinate system of the blast furnace is obtained according to the perspective transformation matrix.

4. The blast furnace soft cross temperature measurement method based on infrared temperature measurement according to claim 1, characterized in that: In step S107, the temperature relationship model is based on an MLP neural network model.

5. The blast furnace soft cross temperature measurement method based on infrared temperature measurement according to claim 1, characterized in that: The step S20 includes: Step S201: installing an infrared temperature measuring device on the top of a blast furnace; Step S202: establishing a perspective transformation matrix based on the installation position of the infrared temperature measuring device on the furnace roof, and correcting the infrared image I3 into a bird's-eye view temperature image I4 located in the blast furnace plane coordinate system; Step S203: setting the position of the cross temperature measurement point; Step S204: Find the pixel coordinates corresponding to the cross temperature measurement point in the bird's-eye view temperature image I4, and then create a virtual cross temperature measurement point in the infrared image I3 using the perspective transformation matrix; Step S205: extracting the temperature values of the virtual cross temperature measurement points to form a third temperature sequence; Step S206: input the third temperature sequence into the trained temperature relationship model of the virtual cross temperature measurement points and the real cross temperature measurement points, and output a fourth temperature sequence consisting of the predicted temperature values of each cross temperature measurement point.

Images