Anode furnace oxidation-reduction end point judgment system based on image spectrum color peak

Abstract

The invention discloses an anode furnace redox end point judgment system based on an image spectrum color peak. The system comprises an anode furnace flame generation module, an image information acquisition module and an image analysis processing module, the anode furnace flame generation module is used for generating flames during operation of an anode furnace; the image information acquisition module is used for acquiring a high-definition flame image during operation of the anode furnace; and the image sorting module is used for determining an oxidation-reduction end point of the anode furnace according to the high-definition flame image. According to the end point judgment system of the image spectrum color peak, non-contact and interference-free detection of the reaction end point can be realized through an image acquisition and processing technology.

Description

Technical Field

[0001] The invention belongs to the field of anode furnace redox endpoint judgment, and in particular relates to an anode furnace redox endpoint judgment system based on image spectrum color peaks. Background Art

[0002] In the copper smelting process, redox reaction is a key step for converting copper ore into pure copper. First, the copper ore is oxidized to separate copper ions from other elements (such as sulfur, iron, etc.). Then, the oxidized copper ions are reduced to pure copper through a reduction reaction. To ensure the purity of copper, the correct redox balance is important. If oxidation is excessive, copper oxides that are difficult to reduce will be formed, thereby reducing the purity of copper. If reduction is excessive, it may lead to the reduction of other metal impurities, thus affecting the quality of the product. Therefore, judging the end point of redox is crucial in the copper smelting process.

[0003] At present, the redox endpoint judgment of the anode furnace still mainly relies on manual sampling and manual experience to judge whether the anode furnace has reached the redox endpoint. This will cause workers to misjudge, sampling and test results to be delayed, and operators spend a lot of time and energy to monitor and judge the redox endpoint, which is inefficient and prone to negligence, resulting in increased production costs and increased production cycles, and will also have a certain impact on the life of the furnace. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes an anode furnace redox endpoint judgment system based on image spectrum color peaks, which can realize intelligent judgment of the anode furnace redox endpoint.

[0005] To achieve the above object, the present invention provides an anode furnace redox endpoint judgment system based on image spectrum color peaks, comprising: an anode furnace flame generation module, an image information acquisition module and an image analysis module;

[0006] The anode furnace flame generation module is used to generate flames when the anode furnace is running;

[0007] The image information acquisition module is used to acquire high-definition flame images when the anode furnace is running;

[0008] The image analysis module is used to determine the anode furnace redox endpoint based on the high-definition flame image.

[0009] Optionally, the anode furnace flame generation module includes: an anode furnace and a flame image acquisition observation hole, through which the flame in the anode furnace is observed.

[0010] Optionally, the image information acquisition module includes: a high-definition image acquisition device, an image processing submodule, electrical wires and data cables;

[0011] The high-definition image acquisition device is used to acquire high-definition flame images;

[0012] The image processing submodule is used to process the high-definition flame image;

[0013] The electric wire is used to supply the power required by the high-definition image acquisition device;

[0014] The data line is used to transmit the high-definition flame image.

[0015] Optionally, the image processing submodule includes: an image filtering unit and an image enhancement unit;

[0016] The image filtering unit is used to filter the flame image;

[0017] The image enhancement unit is used to enhance the filtered flame image by using a logarithmic grayscale transformation method to obtain the high-definition flame image.

[0018] Optionally, the logarithmic grayscale transformation method is:

[0019] s=clog(1+r)

[0020] Among them, s is the image after logarithmic transformation, c is a constant, and r is the original image.

[0021] Optionally, the image analysis module includes: a spectrum peak value acquisition unit, a fitting curve slope unit, and a judgment unit;

[0022] The spectrum peak value acquisition unit is used to acquire the frequency curve of the high-definition flame image by using fast Fourier transform, and extract the spectrum peak value of the frequency curve;

[0023] The fitting curve slope unit is used to fit the curve formed by the peak-to-peak values ​​of the spectrum and obtain the slope of the fitting curve;

[0024] The judgment unit is used to judge the slope of the fitting curve using a BP neural network model to determine the redox endpoint of the anode furnace, wherein the BP neural network model is obtained by training a training set, and the training set includes the slope of the historical fitting curve.

[0025] Optionally, the frequency curve is three color frequency curves of RGB.

[0026] Optionally, the Boltzmann method is used to fit the curve formed by the peak-to-peak values ​​of the spectrum.

[0027] Compared with the prior art, the present invention has the following advantages and technical effects:

[0028] 1. The endpoint judgment system of the image spectrum color peak in the present invention can realize non-contact and non-interference detection of the reaction endpoint through image acquisition and processing technology.

[0029] 2. High sensitivity: The endpoint judgment system of the image spectrum color peak can detect slight changes with high sensitivity, and it is easier to find changes in the endpoint signal than traditional methods. This can improve the accuracy of experimental results, and help reduce the consumption of measurement consumables and reduce production costs.

[0030] 3. Real-time monitoring and data processing: The system can monitor the image information of the reaction process in real time and extract the color peak information and features through image processing algorithms. This can realize real-time monitoring and data processing of the reaction kinetics process, provide reliable feedback for the smelting process, and improve the controllability and repeatability of the smelting process.

[0031] 4. High precision and accuracy: The endpoint judgment system of the image spectrum color peak can achieve high-precision and accurate judgment of the reaction endpoint by accurately analyzing and modeling the changes in the color peak.

[0032] 5. Automation and intelligence: The system can be linked with computers or automation equipment to achieve automatic control and feedback. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0034] Figure 1 Schematic diagram of a system for determining the redox endpoint of an anode furnace based on image spectrum color peaks according to an embodiment of the present invention;

[0035] Figure 2 This is a structural diagram of the anode furnace redox endpoint judgment system according to an embodiment of the present invention;

[0036] Figure 3 The R peak image and fitting curve diagram of the oxidation end point reduction start flame image of the embodiment of the present invention;

[0037] Figure 4 The R peak image and the fitting curve diagram of the flame image in the restoration process of the embodiment of the present invention;

[0038] Figure 5 It is the R peak image and fitting curve diagram of the flame image close to the restoration end point according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0041] The present invention proposes an anode furnace redox endpoint judgment system based on image spectrum color peaks, such as Figure 1 As shown, it specifically includes: an anode furnace flame generation module, an image information acquisition module and an image analysis module;

[0042] The anode furnace flame generation module is used to generate flames when the anode furnace is running;

[0043] The image information acquisition module is used to acquire high-definition flame images when the anode furnace is running;

[0044] The image analysis module is used to determine the anode furnace redox endpoint based on the high-definition flame image.

[0045] Specifically, the front-end anode furnace flame generating module, the flame image acquisition module and the PC terminal data analysis and processing module, the front-end anode furnace flame generating module is used to generate flame when the anode furnace is running; the flame image acquisition module is used to acquire flame images when the anode furnace is running; the PC terminal data analysis and processing module obtains the flame image color of the end point of the anode furnace oxidation-reduction by analyzing the different colors of the flame images at each stage of the anode furnace operation, thereby judging the end point of the anode furnace oxidation-reduction.

[0046] like Figure 2 As shown, the hardware part of the present invention includes: a main control room connection line 1, an air cooling sleeve bracket 2, an anode furnace body 3, a rotating gear 4, a smoke outlet 5, a hydraulic cylinder 6, a feed port cover 7, an oxidant 8, a reducing agent delivery pipeline, a copper outlet 9, a high-speed camera and an air cooling device 10, and a main control room host 11. The air cooling sleeve bracket 2 is tilted to adjust the nut to align with the observation hole of the anode furnace body 3 so that the flame image can be fully captured. The collected image is stored in the main control room 11 host.

[0047] Furthermore, the anode furnace flame generation module comprises: an anode furnace and a flame image acquisition observation hole, through which the flame in the anode furnace is observed.

[0048] Specifically, the front-end anode furnace flame generation module includes: an anode furnace 1, a flame 2 in the anode furnace, and a flame image acquisition observation hole 7 for observing the flame;

[0049] Further, the image information acquisition module includes: a high-definition image acquisition device, an image processing submodule, electric wires and data cables;

[0050] The high-definition image acquisition device is used to acquire high-definition flame images;

[0051] The image processing submodule is used to process the high-definition flame image;

[0052] The electric wire is used to supply the power required by the high-definition image acquisition device;

[0053] The data line is used to transmit the high-definition flame image.

[0054] Specifically, the flame image acquisition module includes: brackets 3 and 4 for fixing the high-definition image acquisition device; a high-definition image acquisition device 5 for acquiring high-definition flame images; a wire 8 for supplying the power required by the high-definition image acquisition device; and a data line 9 for transmitting high-definition image data to the PC.

[0055] Furthermore, the image processing submodule includes: an image filtering unit and an image enhancement unit;

[0056] The image filtering unit is used to filter the flame image;

[0057] The image enhancement unit is used to enhance the filtered flame image by using a logarithmic grayscale transformation method to obtain the high-definition flame image.

[0058] Furthermore, the logarithmic grayscale transformation method is:

[0059] s=clog(1+r)

[0060] Among them, s is the image after logarithmic transformation, c is a constant, and r is the original image.

[0061] Specifically, this transformation uses a narrowband low-grayscale input image value to map to a broadband output value. The corresponding is a high adjustment value of the input grayscale. This transformation can be used to expand dark pixels in a compressed high-value image. The corresponding is the adjustment value of the anti-logarithmic transformation. All curves of a logarithmic function can complete the diffusion / compression of the image grayscale. The logarithmic function has an important feature, that is, it compresses the dynamic range of the image pixel value to a large extent. A typical example of its application is the Fourier spectrum, whose pixel values ​​have a large dynamic range.

[0062] Furthermore, the image analysis module includes: a spectrum peak value acquisition unit, a fitting curve slope unit, and a judgment unit;

[0063] The spectrum peak value acquisition unit is used to acquire the frequency curve of the high-definition flame image by using fast Fourier transform, and extract the spectrum peak value of the frequency curve;

[0064] The fitting curve slope unit is used to fit the curve formed by the peak-to-peak values ​​of the spectrum and obtain the slope of the fitting curve;

[0065] The judgment unit is used to judge the slope of the fitting curve using a BP neural network model to determine the redox endpoint of the anode furnace, wherein the BP neural network model is obtained by training a training set, and the training set includes the slope of the historical fitting curve.

[0066] Specifically, the PC terminal data analysis and processing module is: the computer 6 is used to analyze and process the flame image collected by the high-definition image acquisition device. Affected by the acquisition environment, it is necessary to process the image collected by the high-definition image acquisition device, and perform image filtering on the collected initial flame image to remove environmental noise, so as to obtain a more obvious flame image, and then analyze and collect multiple data quantities of the processed flame image, including corner points, area and circularity, and HSV three-channel quantity features. Ensure the accuracy of the flame image.

[0067] Furthermore, the frequency curves are three color frequency curves of RGB.

[0068] Furthermore, the Boltzmann method is used to fit the curve formed by the peak-to-peak values ​​of the spectrum.

[0069] Specifically, the processed image will be divided into three color frequency curves of RGB through fast Fourier transform, and the peak value of the spectrum spectrum will be extracted. After collecting data ten times, the Boltzmann method will be used to fit the curve and obtain the slope at the end of the curve.

[0070] Then, the slope of the fitting curve is trained by the BP neural network using the adopted Boltzmann method. During the training process, a corresponding database of the slope and the redox stage is established. After the training is completed, the slope of the fitting curve is judged by the trained neural network system to determine whether the oxidation or reduction end point has been reached. If the end point has been reached, the neural network will send a signal to the worker to inform him that the oxidation or reduction end point has been reached.

[0071] If the endpoint is not reached, the neural network will evaluate when the oxidation or reduction endpoint will be reached and report the estimated time to reach the endpoint to the worker.

[0072] Among them, the anode furnace redox process progresses, and the image of the change trend of the image R peak is as follows Figure 3-5 shown.

[0073] The purpose of this method is to monitor and judge the flame redox phase by analyzing flame images and using neural networks. Such a system can be used in industrial production, for example, to monitor the combustion state during furnace combustion to ensure the stability and efficiency of the combustion process.

[0074] The advantage of this technology is that it can monitor the oxidation-reduction stage of the flame in real time and accurately, improving production efficiency and safety. At the same time, by using image processing and neural network algorithms, this process can be completed automatically, reducing labor costs and the risk of human misjudgment.

[0075] The image processing algorithm analyzes and processes the images collected by the image acquisition device 10 in actual production, obtains a clear image of the sample flame in actual production, and compares it with the endpoint image to determine whether the anode furnace has reached the redox endpoint. If the redox endpoint is reached, the algorithm makes a judgment and issues a reminder to the worker. If the endpoint is not reached, the algorithm predicts the endpoint and issues a reminder to the operator. The image data is stored in the database for subsequent iteration and update of the algorithm.

[0076] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Anode furnace redox endpoint judgment system based on image spectrum color peak, characterized in that: include: Anode furnace flame generation module, image information acquisition module and image analysis module; The anode furnace flame generation module is used to generate flames when the anode furnace is running; The image information acquisition module is used to acquire high-definition flame images when the anode furnace is running; The image analysis module is used to determine the anode furnace redox endpoint based on the high-definition flame image.

2. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 1 is characterized in that: The anode furnace flame generation module comprises: an anode furnace and a flame image acquisition observation hole, through which the flame in the anode furnace is observed.

3. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 2 is characterized in that: The image information acquisition module includes: a high-definition image acquisition device, an image processing submodule, electric wires and data cables; The high-definition image acquisition device is used to acquire high-definition flame images; The image processing submodule is used to process the high-definition flame image; The electric wire is used to supply the power required by the high-definition image acquisition device; The data line is used to transmit the high-definition flame image.

4. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 3 is characterized in that: The image processing submodule includes: an image filtering unit and an image enhancement unit; The image filtering unit is used to filter the flame image; The image enhancement unit is used to enhance the filtered flame image by using a logarithmic grayscale transformation method to obtain the high-definition flame image.

5. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 4 is characterized in that: The logarithmic grayscale transformation method is: s=clog(1+r) Among them, s is the image after logarithmic transformation, c is a constant, and r is the original image.

6. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 1, characterized in that: The image analysis module includes: a spectrum peak value acquisition unit, a fitting curve slope unit, and a judgment unit; The spectrum peak value acquisition unit is used to acquire the frequency curve of the high-definition flame image by using fast Fourier transform, and extract the spectrum peak value of the frequency curve; The fitting curve slope unit is used to fit the curve formed by the peak-to-peak values ​​of the spectrum and obtain the slope of the fitting curve; The judgment unit is used to judge the slope of the fitting curve using a BP neural network model to determine the redox endpoint of the anode furnace, wherein the BP neural network model is obtained by training a training set, and the training set includes the slope of the historical fitting curve.

7. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 6 is characterized in that: The frequency curves are three color frequency curves of RGB.

8. The anode furnace redox endpoint judgment system based on image spectrum color peak according to claim 6, characterized in that: The Boltzmann method is used to fit the curve formed by the peak-to-peak values ​​of the spectrum.

Images