Method for predicting FeO content of sintered ore

Abstract

The invention discloses a method for predicting the FeO content of sintered ore, which comprises the following steps of: 1, detecting and analyzing a temperature field of a tail section of a sintering machine: installing a tail thermal imager system, timely and stably acquiring temperature characteristic data of the tail section of the sintering machine, detecting the temperature field of the tail section, and constructing a characteristic image temperature analysis model; step 2, defining and monitoring sintering material layer thermal state process characteristic parameters, according to a sintering ore bed section temperature distribution state, defining thermal state process parameters such as a material layer burn-through index, a red fire layer thickness, a reaction zone thickness and a vertical sintering speed, and carrying out real-time analysis and monitoring; the device has the function of counting the burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed in a continuous time period, and draws and displays a trend curve of the sintering characteristic index. The invention relates to the field of sintered ore content testing, in particular to a method for predicting the FeO content of sintered ore.

Description

Technical Field

[0001] The invention relates to the field of sintered ore content testing, and in particular to a method for predicting the FeO content of sintered ore. Background Art

[0002] The sintering production process is long and has many links. The sintering process is complex, nonlinear, time-varying and uncertain, and is a typical complex nonlinear controlled object. Among them, the FeO content of sintered ore is an important process parameter and quality parameter for evaluating the sintering process and the quality of sintered ore (strength, yield, metallurgical properties). Appropriate and stable FeO content is the basis for ensuring stable and smooth blast furnace production. Ironmaking production practice shows that a 1% change in FeO in sintered ore can affect blast furnace output and coke ratio by 1%-1.5%.

[0003] At present, the fluctuation level of FeO content in sintered ore of various domestic sintering machines is about ±1%, accounting for 80%, which is too large and needs to be strengthened. The application of the prediction method of FeO content in sintered ore will help to improve the stability of the sintering process and sintered ore quality to a new level, further reduce the solid fuel consumption of sintered ore, and be conducive to the long-term stable operation of blast furnaces and the continuous improvement of overall benefits.

[0004] Sintering plants generally adopt the method of "visual inspection + chemical inspection and verification" to determine the FeO content of sintered ore, which brings great difficulties to the control of the production process.

[0005] With the continuous development of technologies such as real-time infrared imaging and computer image recognition and analysis technologies and the lowering of application thresholds, in recent years, some domestic sintering plants have begun or are implementing related technologies for sintering fire prevention. There is an online prediction method for sintering FeO content based on tail section infrared thermal imaging and convolutional neural network. The residual convolutional neural network (Resnet-18) structure is used to establish a sintered ore FeO content prediction model: the cross-sectional image is marked by the sintered ore FeO content collected at the production site, and the convolutional neural network algorithm is used to train the image set. The online prediction system for sintered ore FeO content is developed using C# and Python mixed programming.

[0006] For a long time, visual inspection of sintered ore FeO and other materials at the tail section of the machine has been a basic and important skill for sintering pyrotechnicians. However, due to the influence of personal vision, color recognition ability, experience, and limited visual inspection frequency, the skills of pyrotechnicians are limited by human power and have reached a level that is difficult to improve further. Although chemical inspection of FeO is accurate, the time lag is too long, which also brings difficult-to-eliminate problems such as large fluctuations in the process and low product quality stability caused by long information lag.

[0007] At present, infrared imaging and computer image recognition and analysis technologies are used to analyze and model the characteristic images captured by the tail of the machine based on the general sintering theory in the industry. However, different sintering plants in different regions actually have different raw material conditions, and cannot perform appropriate analysis and modeling based on actual process parameters such as fuel and moisture, which affects the accuracy of the prediction results. Summary of the invention

[0008] It is not possible to give suggestions for rational adjustments based on actual production, and manual adjustments are required based on production experience. The method of the present invention can provide real-time feedback on production conditions, and provide advance guidance for adjusting production parameters such as sintering machine speed, round roller distribution, and fuel ratio, thereby reducing the risk of manual misjudgment.

[0009] To solve the above problems, the technical solution adopted by the present invention is as follows: The present invention is a method for predicting the FeO content of sintered ore, comprising the following steps:

[0010] Step 1: Detection and analysis of the temperature field of the tail section of the sintering machine. By installing a thermal imager system at the tail section, the temperature characteristic data of the tail section of the sintering machine can be collected in a timely and stable manner to realize the temperature field detection of the tail section and build a characteristic image temperature analysis model.

[0011] Step 2: Definition and monitoring of characteristic parameters of the thermal process of the sintering material layer. According to the temperature distribution state of the cross section of the sintering ore bed, the thermal process parameters of the material layer such as the burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed are defined, and real-time analysis and monitoring are performed. The burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed in a continuous time period are statistically analyzed, and the trend curve of the sintering characteristic index is drawn and displayed;

[0012] Step 3: Detection of uniformity characteristic parameters of the sintering material layer in the width direction. According to the temperature distribution state of the sintering ore bed section, define the quantitative expression method of the uniformity of the material layer in the width direction, the edge benefit index and other uniformity characteristic parameters, and monitor in real time to statistically analyze the change trend of the uniformity characteristic parameters in the width direction of the material layer in a continuous time period.

[0013] Step 4: Prediction model and application of FeO content in sintered ore. By collecting and analyzing characteristic data such as the temperature of the tail section of the sintering machine and combining the process parameters that affect the FeO content in sintered ore, a mathematical model for predicting the FeO content in sintered ore is established to display the change curve of the predicted FeO content. In a continuous time period, the FeO content in sintered ore is detected and compared with the predicted results to improve the accuracy of the FeO±0.5 content prediction model. Based on the predicted results of the FeO content, targeted measures are taken to adjust the production operation parameters in time, significantly shorten the adjustment cycle of the FeO content in sintered ore, and improve the stability of the FeO content in sintered ore.

[0014] Furthermore, the construction of the characteristic image temperature analysis model in step 1 needs to realize the temperature result analysis and display of any point, horizontal / vertical line and selected area within the full visual range.

[0015] Furthermore, the characteristic image temperature analysis model constructed in step 1 also needs to realize the analysis and display of the highest temperature distribution state of the sintered ore bed cross section, the longitudinal temperature distribution state with equal spacing in the width direction, and the red-zone distribution state.

[0016] Furthermore, in the step 3, after the variation trend of the uniformity characteristic parameters of the material layer width direction in the continuous time period is statistically analyzed, the trend curves of the sintering uniformity index and the edge effect need to be drawn and displayed.

[0017] Furthermore, in step 4, the process parameters affecting the FeO content of the sintered ore include carbon blending parameters, moisture parameters and permeability.

[0018] The beneficial effects achieved by the present invention using the above structure are as follows:

[0019] 1. This method collects thermal images of the tail section and transmits them to the big data server in real time, performs preprocessing on the images such as "best image screening" and "cross-section image correction", and then performs image preprocessing and feature extraction on the best image. The mathematical relationship between its characteristic value and parameters and indicators such as sintered ore FeO is analyzed to accurately predict and judge FeO and provide production operation guidance. Combined with data tracking and big data storage, the raw material parameters and operating parameters corresponding to the sintered ore of the tail section can also be traced back, and a deep learning model for predicting sintered ore FeO based on the tail section characteristic parameters, raw material parameters and operating parameters is established.

[0020] 2. Further improving the accuracy will help improve the sintering process and the stability of sintered ore quality to a new level, further reduce the solid fuel consumption of sintered ore, and help the long-term stable operation of the blast furnace and the continuous improvement of the overall efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The figure is a schematic diagram of the overall process of a method for predicting the FeO content in sintered ore proposed in the present invention. DETAILED DESCRIPTION

[0022] As the instruction manual Figure 1 As shown, the present invention is a method for predicting the FeO content of sintered ore, comprising the following steps:

[0023] Step 1: Detection and analysis of the temperature field of the tail section of the sintering machine. By installing a thermal imager system at the tail of the machine, timely and stably collect the temperature characteristic data of the tail section of the sintering machine, realize the temperature field detection of the tail section, build a characteristic image temperature analysis model, and realize the temperature result analysis and display of any point, horizontal / vertical line and selected area within the full visual range. It is also necessary to realize the analysis and display of the highest temperature distribution state of the sintering ore bed section, the longitudinal temperature distribution state with equal spacing in the width direction, and the red heat distribution state;

[0024] Step 2: Definition and monitoring of characteristic parameters of the thermal process of the sintering material layer. According to the temperature distribution state of the cross section of the sintering ore bed, the thermal process parameters of the material layer such as the burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed are defined, and real-time analysis and monitoring are performed. The burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed in a continuous time period are statistically analyzed, and the trend curve of the sintering characteristic index is drawn and displayed;

[0025] Step 3: Detection of uniformity characteristic parameters of the sintering material layer in the width direction. According to the temperature distribution state of the sintering ore bed section, define the quantitative expression method of the uniformity of the material layer in the width direction, the uniformity characteristic parameters such as the edge benefit index, and monitor in real time. Statistical trend of the uniformity characteristic parameters of the material layer in the width direction in a continuous time period, and draw and display the trend curve of the sintering uniformity index and the edge effect;

[0026] Step 4: Prediction model and application of FeO content in sintered ore. By collecting and analyzing characteristic data such as the temperature of the tail section of the sintering machine, combined with the process parameters affecting the FeO content in sintered ore, including carbon parameters, moisture parameters and permeability, a mathematical model for predicting the FeO content in sintered ore is established to display the change curve of the predicted FeO content. In a continuous time period, the FeO content in sintered ore is detected and compared with the predicted results to improve the accuracy of the FeO±0.5 content prediction model. According to the prediction results of the FeO content, targeted measures are taken to adjust the production operation parameters in time, significantly shorten the adjustment cycle of the FeO content in sintered ore, and improve the stability of the FeO content in sintered ore.

[0027] It should be understood that although this specification is described according to implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A method for predicting the FeO content of sintered ore, characterized in that: The process includes the following steps: Step 1: Detection and analysis of the temperature field of the tail section of the sintering machine. By installing a thermal imager system at the tail section, the temperature characteristic data of the tail section of the sintering machine can be collected in a timely and stable manner to detect the temperature field of the tail section and build a characteristic image temperature analysis model. Step 2: Definition and monitoring of characteristic parameters of the thermal process of the sintering material layer. According to the temperature distribution state of the cross section of the sintering ore bed, the thermal process parameters of the material layer such as the burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed are defined, and real-time analysis and monitoring are performed. The burn-through index, the red fire layer thickness, the reaction zone thickness and the vertical sintering speed in a continuous time period are statistically analyzed, and the trend curve of the sintering characteristic index is drawn and displayed; Step 3: Detection of uniformity characteristic parameters of the sintering material layer in the width direction. According to the temperature distribution state of the sintering ore bed section, define the quantitative expression method of the uniformity of the material layer in the width direction, the edge benefit index and other uniformity characteristic parameters, and monitor in real time to statistically analyze the change trend of the uniformity characteristic parameters in the width direction of the material layer in a continuous time period. Step 4: Prediction model and application of FeO content in sintered ore. By collecting and analyzing characteristic data such as the temperature of the tail section of the sintering machine and combining the process parameters that affect the FeO content in sintered ore, a mathematical model for predicting the FeO content in sintered ore is established to display the change curve of the predicted FeO content. In a continuous time period, the FeO content in sintered ore is detected and compared with the predicted results to improve the accuracy of the FeO±0.5 content prediction model. Based on the predicted results of the FeO content, targeted measures are taken to adjust the production operation parameters in time, significantly shorten the adjustment cycle of the FeO content in sintered ore, and improve the stability of the FeO content in sintered ore.

2. The method for predicting the FeO content of sintered ore according to claim 1, characterized in that: The construction of the characteristic image temperature analysis model in step 1 needs to realize the temperature result analysis and display of any point, horizontal / vertical line and selected area within the whole visual range.

3. The method for predicting the FeO content of sintered ore according to claim 2, characterized in that: The characteristic image temperature analysis model constructed in step 1 also needs to realize the analysis and display of the highest temperature distribution state of the sintered ore bed cross section, the longitudinal temperature distribution state with equal spacing in the width direction, and the red-zone distribution state.

4. The method for predicting the FeO content of sintered ore according to claim 3, characterized in that: In the step 3, after the variation trend of the uniformity characteristic parameters of the material layer width direction in the continuous time period is statistically analyzed, the trend curves of the sintering uniformity index and the edge effect need to be drawn and displayed.

5. The method for predicting the FeO content of sintered ore according to claim 4, characterized in that: In step 4, the process parameters affecting the FeO content of the sintered ore include carbon blending parameters, moisture parameters and permeability.

Images