Multi-sensor fused intelligent control system and method for titanium reduction process of fine iron powder

Abstract

The invention belongs to the technical field of metallurgical process control, and discloses a multi-sensor fused intelligent control system and method for the titanium reduction process of fine iron powder. The method aims at solving the problem that local carbon deposition is difficult to monitor and inhibit in real time in a traditional process. The method comprises the following steps of: constructing an in-furnace reaction thermodynamic constraint diagram by acquiring in-furnace temperature distribution, gas component distribution, a furnace wall vibration signal and a local pressure signal, and dividing an in-furnace region into a plurality of reaction front subregions according to the in-furnace reaction thermodynamic constraint diagram; in each partition, the system constructs a carbon deposition triggering network based on vibration-pressure coupling, the carbon deposition triggering probability is accurately calculated, and then a targeted self-adaptive gas injection strategy is generated. By globally optimizing gas injection parameters, real-time inhibition of local carbon deposition and effective separation of titanium in the titanium reduction process of the fine iron powder are achieved. According to the invention, the agglomeration phenomenon of the fine iron powder is reduced, the smoothness of a gas channel is kept, and the titanium recovery rate and the product quality are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of metallurgical process control, more particularly, the present application relates to a multi-sensor fusion intelligent control system and method for titanium reduction in iron concentrate. BACKGROUND

[0002] Titanium, as a metal material, has excellent characteristics such as low density, high strength, and corrosion resistance, and is widely used in the fields of aerospace, chemical industry, medical treatment, etc. The titanium reduction in iron concentrate is an important metallurgical process for extracting titanium from titanium-containing iron ore. The carbon thermal reduction method is usually used to reduce titanium in the +4 state to a low valence state under high temperature conditions, so as to realize effective separation from iron. In the carbon thermal reduction process, controlling the appropriate CO / CO2 ratio is crucial to maintaining the smooth progress of the reduction reaction. Early processes mainly relied on manual experience control, and the state in the furnace was roughly judged by observing the flame color at the furnace mouth and measuring the outlet gas temperature. With the application of online gas analyzers, real-time monitoring of the CO / CO2 ratio became possible, and the control accuracy was improved. Subsequent advanced control strategies based on model predictive control (MPC) have been applied in some large production lines, but due to limitations in model accuracy and monitoring methods, the problem of local carbon deposition cannot be solved.

[0003] In the titanium reduction process of iron concentrate, local carbon deposition (carbon precipitation) is a key step that restricts production efficiency and product quality. Existing technologies use single temperature monitoring or simple gas analysis methods, which cannot capture the precursor characteristics and dynamic evolution process of carbon deposition in the micro region of the furnace. In actual production, when the reduction reaction reaches the middle and late stages, an imbalance in the CO / CO2 ratio often leads to the formation of a large amount of carbon deposits in the region. These carbon deposits quickly adhere to the surface of the iron concentrate powder, causing strong agglomeration between the powder particles. After the formation of the agglomeration region, the gas channel is gradually blocked, causing uneven gas distribution in the furnace, forming "gas flow dead zones" and "gas flow dominant channels", further exacerbating the carbon deposition and agglomeration problem. This vicious cycle forms a local high-temperature zone in the furnace, with a steep temperature gradient, which not only affects the chemical reduction process of titanium, but also leads to the formation of titanium-carbon-iron compounds that are difficult to separate. The existing CO / CO2 ratio control system usually relies on analysis of the gas composition at the outlet of the furnace body, which has serious time lag and spatial averaging effects, making the control signal unable to accurately reflect the local state in the furnace. When the system detects an anomaly and adjusts the gas injection parameters, the problem area has often already developed to an extent that is difficult to recover. In addition, the conventional control strategy uses unified parameter adjustment for the entire furnace, which lacks the ability to intervene specifically in the local carbon deposition area, causing a small controllable problem to evolve into a systemic obstacle that affects the entire reduction process, ultimately leading to a series of technical and economic problems such as low titanium recovery rate, high energy consumption, and poor batch stability in the product.

[0004] In view of this, the present application proposes a multi-sensor fusion intelligent control system and method for iron concentrate powder titanium reduction process to solve the above problems. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, in order to achieve the above-mentioned purpose, the present application provides the following technical scheme: a multi-sensor fusion intelligent control method for iron concentrate powder titanium reduction process, comprising: Step 1: Obtain the furnace temperature distribution data, furnace gas composition distribution data, furnace wall vibration signal and local pressure signal in the reaction furnace at each moment in the iron concentrate powder titanium reduction process; Step 2: According to the spatial distribution characteristics of the furnace temperature distribution data and the furnace gas composition distribution data, a furnace reaction thermodynamic constraint graph is constructed; Step 3: According to the furnace reaction thermodynamic constraint graph, the furnace area is divided into multiple reaction front subareas; Step 4: In each reaction front subarea, a carbon deposition trigger network based on vibration-pressure coupling is constructed to obtain the carbon deposition trigger probability of each reaction front subarea; Step 5: According to the carbon deposition trigger probability of each reaction front subarea, an adaptive gas injection strategy for each reaction front subarea is generated; Step 6: According to the adaptive gas injection strategy of all reaction front subareas, the furnace gas injection parameters are optimized to realize real-time inhibition of local carbon deposition and effective separation of titanium in the iron concentrate powder titanium reduction process.

[0006] Preferably, the construction of the furnace reaction thermodynamic constraint graph comprises: According to the furnace temperature distribution data, the boundary line between the high-temperature core area and the low-temperature edge area in the furnace is identified, and a furnace temperature gradient boundary graph is constructed; According to the furnace gas composition distribution data, the local extreme point of the CO / CO2 ratio in the furnace is identified, and a furnace gas reaction activity graph is constructed; The furnace temperature gradient boundary graph and the furnace gas reaction activity graph are spatially superimposed to generate an initial thermodynamic constraint graph; In the initial thermodynamic constraint graph, the space conflict area between the boundary line in the temperature gradient boundary graph and the extreme point in the gas reaction activity graph is identified, and the space conflict area is corrected by thermodynamic equilibrium to obtain the furnace reaction thermodynamic constraint graph.

[0007] Preferably, the furnace area is divided into multiple reaction front subareas, comprising: Randomly select multiple initial tracking points in the furnace reaction thermodynamic constraint graph; simulate a propagation process of a virtual reaction front, wherein a propagation direction of the virtual reaction front is determined by a weighted vector of a temperature gradient and a CO / CO2 ratio gradient of a corresponding position point in the in-furnace reaction thermodynamic constraint map, and a propagation speed of the virtual reaction front is determined by a thermodynamic entropy increase rate of the corresponding position point in the in-furnace reaction thermodynamic constraint map; when the thermodynamic entropy increase rate is lower than a preset entropy increase rate threshold during the propagation process of the virtual reaction front, stop the propagation, and mark a region covered by the virtual reaction front as an initial reaction front subzone; perform boundary optimization on all the initial reaction front subzones, wherein the boundary optimization is performed by identifying a continuity of a temperature gradient and a CO / CO2 ratio gradient between adjacent initial reaction front subzones, and smoothing a boundary with a continuity lower than a preset continuity threshold to obtain the reaction front subzone.

[0008] Preferably, the constructing, in each of the reaction front subzones, a vibration-pressure coupling-based carbon deposition trigger network to obtain a carbon deposition trigger probability of each of the reaction front subzones comprises: identifying, in each of the reaction front subzones, local mutation points of a vibration waveform according to a time-domain waveform of the furnace wall vibration signal, and constructing a vibration mutation point sequence; identifying, in each of the reaction front subzones, local pulse points of a pressure waveform according to a time-domain waveform of the in-furnace local pressure signal, and constructing a pressure pulse point sequence; constructing a vibration-pressure coupling network according to a time synchronicity of the vibration mutation point sequence and the pressure pulse point sequence, wherein nodes of the vibration-pressure coupling network are the vibration mutation points and the pressure pulse points, and edges are time delays between the vibration mutation points and the pressure pulse points; calculating the carbon deposition trigger probability of the reaction front subzone according to a connectivity of the vibration-pressure coupling network, wherein the connectivity is quantified by an average path length and a clustering coefficient of the vibration-pressure coupling network, the carbon deposition trigger probability is negatively correlated with the average path length and positively correlated with the clustering coefficient.

[0009] Preferably, the generating an adaptive gas injection strategy for each of the reaction front subzones comprises: identifying, in each of the reaction front subzones, a local flow field bottleneck region in the reaction front subzone according to the in-furnace reaction thermodynamic constraint map, wherein the local flow field bottleneck region is a region with the most serious spatial conflict between a temperature gradient and a CO / CO2 ratio gradient in the in-furnace reaction thermodynamic constraint map; In the local flow field bottleneck region, a disturbance flow induction path is designed, wherein the starting point of the disturbance flow induction path is the center of the local flow field bottleneck region, the ending point is the boundary of the reaction front subregion, and the path direction is determined by the thermodynamic entropy production rate gradient of the corresponding position point in the in-furnace reaction thermodynamic constraint map; According to the disturbance flow induction path, an initial gas injection strategy is generated, wherein the initial gas injection strategy includes injection angle, injection flow rate, and injection frequency, the injection angle is determined by the direction of the disturbance flow induction path, and the injection flow rate and the injection frequency are determined by the size of the carbon deposition trigger probability; According to the prediction suppression effect of the initial gas injection strategy on the carbon deposition trigger probability of the reaction front subregion, the initial gas injection strategy is dynamically adjusted to obtain the adaptive gas injection strategy, wherein the prediction suppression effect is quantified by simulating the influence of the initial gas injection strategy on the thermodynamic entropy production rate of the in-furnace reaction thermodynamic constraint map.

[0010] Preferably, the optimized in-furnace gas injection parameters include: According to the adaptive gas injection strategy of all the reaction front subregions, an initial global flow field distribution map is constructed, wherein the initial global flow field distribution map is generated by superimposing the local flow field vector corresponding to the adaptive gas injection strategy of each reaction front subregion; In the initial global flow field distribution map, a global flow field conflict region is identified, wherein the global flow field conflict region is the region with the most serious direction conflict of flow field vectors in the initial global flow field distribution map; The global flow field conflict region is reconstructed, wherein the flow field reconstruction minimizes the direction conflict of flow field vectors in the global flow field conflict region by iteratively adjusting the adaptive gas injection strategy of adjacent reaction front subregions in the global flow field conflict region; According to the global flow field distribution map after the flow field reconstruction, the in-furnace gas injection parameters are optimized, wherein the optimization is achieved by minimizing the flow field turbulence intensity of the global flow field distribution map and maximizing the uniformity of the thermodynamic entropy production rate of the in-furnace reaction thermodynamic constraint map.

[0011] Preferably, the thermodynamic equilibrium correction is achieved by the following way: In the spatial conflict region, a thermodynamic entropy production optimization model is constructed, wherein the objective function of the thermodynamic entropy production optimization model is the total thermodynamic entropy production of the spatial conflict region, and the constraint condition is the physical feasibility of the temperature value and the CO / CO2 ratio value in the spatial conflict region; solving the optimal solution of the thermodynamic entropy increase optimization model by using an iterative algorithm based on gradient descent, wherein the step length of the iterative algorithm is determined by the modulus of the temperature gradient and the CO / CO2 ratio gradient in the spatial conflict region; According to the optimal solution, the temperature value and the CO / CO2 ratio value in the spatial conflict region are adjusted to obtain a corrected in-furnace reaction thermodynamic constraint graph.

[0012] Preferably, the connectivity of the vibration-pressure coupling network is quantified by: The shortest path length between all pairs of nodes in the vibration-pressure coupling network is calculated to obtain the average path length; For each node, the edge connection density between the node and its neighbor nodes is calculated to obtain the local clustering coefficient of the node; the local clustering coefficients of all nodes are averaged to obtain the clustering coefficient; The connectivity of the vibration-pressure coupling network is quantified according to the weighted sum of the average path length and the clustering coefficient, wherein the weight of the weighted sum is determined by the signal-to-noise ratio of the furnace wall vibration signal and the in-furnace local pressure signal.

[0013] Preferably, the flow field turbulence intensity and the thermodynamic entropy increase rate uniformity are quantified by: According to the flow field vector of each position point in the global flow field distribution graph, the curl and divergence of the flow field vector are calculated to obtain the turbulence intensity of the position point; the turbulence intensities of all position points are averaged to obtain the flow field turbulence intensity; According to the thermodynamic entropy increase rate of each position point in the in-furnace reaction thermodynamic constraint graph, the spatial distribution entropy of the thermodynamic entropy increase rate is calculated to obtain the thermodynamic entropy increase rate uniformity.

[0014] The multi-sensor fusion intelligent control system for iron concentrate titanium reduction process is used to realize the multi-sensor fusion intelligent control method for iron concentrate titanium reduction process, comprising: A data acquisition module is used to acquire the in-furnace temperature distribution data, in-furnace gas composition distribution data, furnace wall vibration signal and in-furnace local pressure signal at each moment in the iron concentrate titanium reduction process; A thermodynamic constraint construction module is used to construct an in-furnace reaction thermodynamic constraint graph according to the spatial distribution characteristics of the in-furnace temperature distribution data and the in-furnace gas composition distribution data; A reaction partition module is used to divide the in-furnace region into multiple reaction front partition regions according to the in-furnace reaction thermodynamic constraint graph; a carbon deposition analysis module, configured to construct a vibration-pressure coupling-based carbon deposition trigger network in each of the reaction front sub-zones, and obtain a carbon deposition trigger probability of each of the reaction front sub-zones; a strategy generation module, configured to generate an adaptive gas injection strategy for each of the reaction front sub-zones according to the carbon deposition trigger probability of each of the reaction front sub-zones; a parameter optimization module, configured to optimize a gas injection parameter in the furnace according to the adaptive gas injection strategy of all the reaction front sub-zones, so as to realize real-time inhibition of local carbon deposition and effective separation of titanium in the titanium reduction process of the iron concentrate powder.

[0015] The technical effects and advantages of the multi-sensor fusion intelligent control system and method for the titanium reduction process of the iron concentrate powder are as follows: The present application improves the prediction and inhibition of local carbon deposition in the high-temperature reduction process, and solves a series of cascade problems caused by the lag of carbon deposition monitoring in the traditional process. By establishing a comprehensive and multi-dimensional real-time perception system of the furnace state, the potential risk area can be accurately identified at the initial formation stage of carbon deposition, and the active control concept of "preventing trouble from the beginning" is realized. The iron concentrate powder agglomeration phenomenon is reduced, the gas passage in the reduction furnace is kept unobstructed, and the reaction gas can uniformly contact the material surface, thereby greatly improving the reduction and separation efficiency of titanium. At the same time, the thermodynamic constraint and vibration-pressure coupling analysis method make the system have sharp insight into the micro changes in the furnace, eliminate the inherent response lag problem of the traditional CO / CO2 ratio control system, and realize the precise adjustment of the reduction atmosphere. In actual production application, this intelligent control method not only significantly improves the recovery rate of titanium in the product, but also reduces energy consumption and equipment wear, reduces unplanned downtime caused by carbon deposition, and improves production continuity and batch stability. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The figure is a schematic diagram of the multi-sensor fusion intelligent control method for the titanium reduction process of the iron concentrate powder of the present application. Figure 2 The figure is a schematic diagram of the multi-sensor fusion intelligent control system for the titanium reduction process of the iron concentrate powder of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] The application example provides a multi-sensor fusion intelligent control system and method for a titanium reduction process of iron concentrate.

[0019] Referring to Figure 1 The application provides a multi-sensor fusion intelligent control method for a titanium reduction process of iron concentrate, including the following steps: Step 1: acquiring the in-furnace temperature distribution data, in-furnace gas composition distribution data, furnace wall vibration signal and local pressure signal in the reaction furnace at each moment in the titanium reduction process of iron concentrate; Step 2: constructing an in-furnace reaction thermodynamic constraint graph according to the spatial distribution characteristics of the in-furnace temperature distribution data and the in-furnace gas composition distribution data; Step 3: dividing the in-furnace region into multiple reaction front sub-regions according to the in-furnace reaction thermodynamic constraint graph; Step 4: constructing a carbon deposition trigger network based on vibration-pressure coupling in each reaction front sub-region to acquire the carbon deposition trigger probability of each reaction front sub-region; Step 5: generating an adaptive gas injection strategy for each reaction front sub-region according to the carbon deposition trigger probability of each reaction front sub-region; Step 6: optimizing the in-furnace gas injection parameters according to the adaptive gas injection strategies of all reaction front sub-regions to realize real-time inhibition of local carbon deposition and effective separation of titanium in the titanium reduction process of iron concentrate.

[0020] The application ensures accurate monitoring and control of the titanium reduction process of iron concentrate through a multi-sensor data fusion technology, provides a theoretical basis for reaction front sub-region division by constructing an in-furnace reaction thermodynamic constraint graph, realizes fine management of the reaction region by reaction front sub-region division based on the in-furnace reaction thermodynamic constraint graph, helps to accurately predict the carbon deposition risk region by constructing a carbon deposition trigger network based on vibration-pressure coupling, realizes accurate intervention on the reaction process according to the carbon deposition trigger probability to generate an adaptive gas injection strategy, optimizes the in-furnace gas injection parameters to ensure that the local carbon deposition is effectively inhibited and the separation efficiency of titanium is improved, and the whole method chain constructs a complete closed-loop system from data acquisition, analysis and processing to control execution, thereby providing an intelligent solution for the titanium reduction process of iron concentrate.

[0021] In the embodiment of the application, the detailed implementation steps of step 1 include: A distributed temperature sensor array is installed in the reaction furnace to collect real-time furnace temperature field data, obtain a high-precision temperature distribution matrix, and perform spatial interpolation and denoising on the temperature distribution matrix to obtain furnace temperature distribution data. A multi-point gas composition sampling probe is arranged in the reaction furnace to perform real-time gas composition analysis, obtain gas composition raw data, and perform calibration and component analysis on the gas composition raw data to obtain furnace gas composition distribution data. A high-sensitivity vibration sensor is installed at a key position on the wall of the reaction furnace to collect raw vibration waveforms of the furnace wall, and the raw vibration waveforms are subjected to frequency spectrum analysis and feature extraction to obtain furnace wall vibration signals. A high-temperature pressure sensor is installed at a key position on the wall of the reaction furnace to collect local pressure raw data, and the local pressure raw data is subjected to dynamic response analysis and trend identification to obtain local pressure signals in the furnace.

[0022] In this embodiment, first design the layout of the distributed temperature sensor array, considering the temperature gradient distribution characteristics and hot spot area distribution law in the reactor, determine the type of temperature sensor (such as thermocouple, platinum resistance, optical fiber temperature sensor, etc.) and technical parameters, prefer to choose high temperature, fast response, strong anti-interference ability of sensor type, determine the spatial layout and installation position of the sensor, ensure that the temperature monitoring density of the key reaction area is high enough, at the same time cover the whole reactor space, install the distributed temperature sensor array, collect real-time temperature field data in the furnace, including the temperature value and its timestamp of each measuring point, the sampling frequency is dynamically adjusted according to the reaction rate, usually 1-10Hz, form the original temperature data matrix, preprocess the original temperature data, including outlier detection and correction, signal denoising (such as median filter, wavelet transform denoising, etc.), data standardization, etc., interpolate the discrete temperature measuring point data through spatial interpolation algorithm (such as Kriging interpolation, radial basis function interpolation, etc.), generate continuous temperature field distribution, get high precision temperature distribution matrix, which represents the temperature value of any position in the reactor, apply filtering algorithm (such as Kalman filter, particle filter, etc.) to denoise the temperature distribution matrix, improve the signal-to-noise ratio and accuracy of temperature data, get the temperature distribution data in the furnace, provide basic data for subsequent thermodynamic constraint graph construction, design multi-point gas composition sampling system, determine the spatial distribution of sampling points, focus on covering the reaction front area, reaction hot spot area and boundary area, select suitable gas analysis technology, such as gas chromatography (GC), mass spectrometry (MS), infrared spectroscopy (IR) or laser absorption spectroscopy (LAS) etc., considering the gas component characteristics, measurement accuracy requirements and response time, etc., set up multi-point gas composition sampling probe in the reactor, the sampling probe material selects special alloy or ceramic material with high temperature resistance and corrosion resistance, the probe design considers the gas flow dynamics characteristics, avoids the interference of sampling process to the reaction flow field, carries out real-time gas composition analysis, monitors the concentration distribution of CO, CO2, O2 and other key gas components, the sampling frequency is set according to the reaction kinetics characteristics, usually 0.5-5Hz, obtain the gas composition raw data, calibrate the gas composition raw data, use standard gas for periodic calibration, establish calibration curve, eliminate instrument drift and system error, improve measurement accuracy by applying multivariate correction algorithm, analyze the components of the calibrated data, calculate the CO / CO2 ratio and other key reaction indicators, evaluate the reaction progress using the chemical equilibrium model, obtain the gas composition distribution data in the furnace, provide gas phase reaction information for the thermodynamic constraint diagram, determine the key positions of the furnace wall vibration monitoring, including the furnace wall near the region with intense reaction, structural support points, and regions prone to carbon deposition in history, select appropriate vibration sensor types such as piezoelectric acceleration sensor, MEMS acceleration sensor, etc., consider factors such as measurement range, frequency response, sensitivity and environmental performance, install high-sensitivity vibration sensors at key positions on the reaction furnace wall, use special high-temperature installation methods and heat insulation measures to protect the sensors, the sensor signals are transmitted to the data acquisition system through anti-interference cables, the original waveform of the furnace wall vibration is collected, the sampling frequency is usually 500-2000Hz to capture high-frequency vibration characteristics, the original vibration waveform is analyzed by frequency spectrum, the frequency domain characteristics of the vibration signal are analyzed by applying fast Fourier transform (FFT), wavelet transform or Hilbert-Huang transform, etc., the characteristic frequency components are identified, the vibration characteristic parameters such as main frequency, frequency band energy distribution, harmonic ratio, etc. are extracted from the frequency spectrum, and the vibration characteristic vector is constructed to obtain the furnace wall vibration signal as an important indicator for early warning of carbon deposition, determine the pressure monitoring points in the key regions of the reaction furnace, focus on the reaction front region, flow field bottleneck region and regions with significant pressure fluctuations in history, select appropriate high-temperature pressure sensors such as ceramic pressure sensors, silicon sapphire pressure sensors, etc., consider factors such as measurement range, accuracy, response time and high-temperature performance, install high-temperature pressure sensors in the key regions of the reaction furnace, use water cooling or air cooling technology to protect the sensors, ensure stable operation in high-temperature environment, the sensor signals are transmitted to the data acquisition system after isolation amplification and filtering, the local pressure raw data is collected, the sampling frequency is usually 50-200Hz to capture medium-frequency pressure fluctuations, the dynamic response analysis is performed on the local pressure raw data, the time series analysis method (such as autoregressive moving average model, exponential smoothing, etc.) is applied to identify the pressure change pattern, the pressure pulsation frequency, amplitude and phase characteristics are calculated, the pressure fluctuation trend and pattern are identified, the mutation points and abnormal fluctuations are detected, and the local pressure signal in the furnace is obtained to provide important input for the carbon deposition trigger network construction.

[0023] In the embodiment of the present application, the detailed implementation steps of step 2 include: According to the temperature distribution data in the furnace, the dividing line between the high-temperature core region and the low-temperature edge region in the furnace is identified, and a temperature gradient boundary map of the furnace is constructed. According to the furnace gas composition distribution data, the local extreme points of the CO / CO2 ratio in the furnace are identified, and a furnace gas reaction activity map is constructed; The furnace temperature gradient boundary map and the furnace gas reaction activity map are spatially superimposed to generate an initial thermodynamic constraint map; In the initial thermodynamic constraint map, the spatial conflict area between the boundary line in the temperature gradient boundary map and the extreme point in the gas reaction activity map is identified, and the spatial conflict area is corrected by thermodynamic equilibrium to obtain a furnace reaction thermodynamic constraint map.

[0024] In this embodiment, first, the numerical analysis method is used to process the furnace temperature distribution data, the temperature gradient calculation and edge detection algorithm are applied to identify the area with significant temperature change, and the dividing line between the high-temperature core area (the most active reaction area) and the low-temperature edge area (the relatively inactive reaction area) is located. The dividing line usually appears as a continuous curved surface with a large temperature gradient. The three-dimensional isosurface technology is used to visualize these dividing lines to form a furnace temperature gradient boundary map, which directly shows the spatial structure and gradient distribution of the furnace temperature field. Then, the furnace gas composition distribution data is analyzed, and the spatial distribution of the CO / CO2 ratio is particularly focused on. The CO / CO2 ratio is a key indicator for measuring the degree of reduction reaction and the risk of carbon deposition. The numerical optimization method is used to identify the local extreme points of the CO / CO2 ratio, including the local maximum point (the active reduction reaction area) and the local minimum point (the lagging reduction reaction area). The interpolation algorithm is applied to generate a continuous CO / CO2 ratio distribution field to form a furnace gas reaction activity map. This map reflects the chemical reaction activity state of different areas. The furnace temperature gradient boundary map and the furnace gas reaction activity map are imported into the same three-dimensional coordinate system for spatial superposition and registration. The spatial analysis technology similar to the geographic information system (GIS) is used to process these two layers to generate a composite layer containing both temperature gradient and gas reaction activity information, forming an initial thermodynamic constraint map. This map comprehensively expresses the constraint conditions of thermodynamics and kinetics on the reaction. In the initial thermodynamic constraint map, the spatial relationship between the temperature gradient boundary line and the CO / CO2 ratio extreme point is analyzed, and the area where there is a contradiction or inconsistency between the two, i.e., the spatial conflict area, is identified. These areas usually exhibit inconsistent reaction directions predicted by the temperature gradient and the gas composition. According to the basic principles of thermodynamics, the spatial conflict area is corrected by thermodynamic equilibrium. A thermodynamic model based on the minimization of Gibbs free energy is established, considering the comprehensive influence of temperature, gas composition, pressure, etc. The equilibrium state of the system at each spatial point is calculated, and the inconsistent part in the initial thermodynamic constraint map is corrected. Finally, a furnace reaction thermodynamic constraint map is formed, which accurately describes the thermodynamic constraint conditions in the titanium reduction process of iron concentrate, providing a theoretical basis for subsequent reaction region division and control strategy formulation.

[0025] In the embodiment of the present application, the detailed implementation steps of step 3 include: a plurality of initial tracking points are randomly selected in the in-furnace reaction thermodynamic constraint map; for each initial tracking point, a propagation process of a virtual reaction front is simulated, wherein a propagation direction of the virtual reaction front is determined by a weighted vector of a temperature gradient and a CO / CO2 ratio gradient at a corresponding position point in the in-furnace reaction thermodynamic constraint map, and a propagation speed is determined by a thermodynamic entropy increase rate at the corresponding position point in the in-furnace reaction thermodynamic constraint map; when the thermodynamic entropy increase rate is lower than a preset entropy increase rate threshold in the propagation process of the virtual reaction front, the propagation is stopped, and a region covered by the virtual reaction front is marked as an initial reaction front subzone; boundary optimization is performed on all the initial reaction front subzones, wherein the boundary optimization is performed by identifying a continuity of a temperature gradient and a CO / CO2 ratio gradient between adjacent initial reaction front subzones, and smoothing a boundary with a continuity lower than a preset continuity threshold to obtain a reaction front subzone.

[0026] In this embodiment, first in the three-dimensional space of the furnace reaction thermodynamic constraint graph, a plurality of initial tracking points are selected by using a hierarchical random sampling method, these points are distributed at different positions in the furnace, cover different temperature intervals and different gas composition regions, to ensure the representativeness and diversity of the tracking points, the number of initial points is adaptively determined according to the size and complexity of the furnace body, and is usually 50-200, for each initial tracking point, the propagation process of the virtual reaction front is simulated by using a method of coupling computational fluid dynamics (CFD) and chemical dynamics, a propagation direction vector field is constructed, the vector field is determined by a weighted combination of a temperature gradient vector and a CO / CO2 ratio gradient vector, and a weight coefficient is determined according to reaction sensitivity analysis, for example, the temperature gradient weight is 0.6, and the CO / CO2 ratio gradient weight is 0.4, a propagation speed field is constructed, the propagation speed is determined by a local thermodynamic entropy increase rate, the higher the entropy increase rate, the more active the reaction, and the faster the propagation speed, the entropy increase rate is calculated by using local temperature, pressure and gas composition, and an iterative method is used to simulate the step-by-step propagation of the virtual reaction front, in each iteration step, the propagation direction and speed of each point on the front are calculated, and the front position is updated, in the propagation process, the thermodynamic entropy increase rate of each front point is monitored in real time, when the entropy increase rate is lower than a preset threshold (for example, 15% of the maximum entropy increase rate), it is determined that the reaction activity of the point is insufficient to continue to propagate, and the propagation process of the point is stopped, when all the points of the entire virtual reaction front stop propagating or reach the furnace wall boundary, the entire region covered between the initial point and the final front line of the virtual reaction front is marked as an initial reaction front partition, and the above process is repeated for all the initial tracking points to obtain a plurality of initial reaction front partitions, these partitions may have overlapping or irregular boundary conditions, the boundary characteristics between adjacent initial reaction front partitions are analyzed, the continuity indexes of the temperature gradient and the CO / CO2 ratio gradient on the boundary are calculated, for example, the angle between the gradient directions, the change rate of the gradient size, and the like, the boundary segments with a continuity lower than a preset threshold are identified, these boundary segments usually represent unreasonable partition division regions, a boundary smoothing algorithm is applied to the boundary segments with low continuity, for example, B-spline curve fitting or morphological smoothing processing, so that the boundary is more in line with the physical meaning, highly similar adjacent partitions are merged, and single partitions with significantly different physical characteristics are segmented, and finally the optimized reaction front partitions are formed, these partitions have the characteristics of clear physical meaning, smooth boundary and relatively uniform internal characteristics, and provide a spatial basis for subsequent carbon deposition risk analysis and control strategy formulation.

[0027] In the embodiment of the present application, the detailed implementation steps of step 4 include: In each reaction front partition, local mutation points of the vibration waveform are identified according to the time domain waveform of the furnace wall vibration signal, and a vibration mutation point sequence is constructed; In each reaction front partition, local pulse points of the pressure waveform are identified according to the time domain waveform of the local pressure signal in the furnace, and a pressure pulse point sequence is constructed; According to the time synchronization of the vibration mutation point sequence and the pressure pulse point sequence, a vibration-pressure coupling network is constructed, wherein the nodes of the vibration-pressure coupling network are the vibration mutation points and the pressure pulse points, and the edges are the time delays between the vibration mutation points and the pressure pulse points. According to the connectivity of the vibration-pressure coupling network, the carbon deposition trigger probability of the reaction front subregion is calculated, wherein the connectivity is quantified by the average path length and the clustering coefficient of the vibration-pressure coupling network, the carbon deposition trigger probability is negatively correlated with the average path length and positively correlated with the clustering coefficient.

[0028] In this embodiment, first, for each reaction front partition, the corresponding furnace wall vibration signal of the region is processed, wavelet analysis or Hilbert-Huang transform and other time-frequency analysis methods are applied to identify the local mutation characteristics in the vibration waveform, including amplitude mutation, frequency mutation and phase mutation, etc. These mutations are usually related to the sudden movement or local reaction state change of the material in the furnace. The timestamp, amplitude and duration of each mutation point are recorded, and a vibration mutation point sequence is formed, which reflects the change process of the dynamic state in the furnace. Similarly, for each reaction front partition, the corresponding local pressure signal in the furnace is processed, and a peak detection algorithm and an abnormal point identification method are used to identify the local pulse points in the pressure waveform, including pressure peaks, pressure valleys and pressure gradient mutation points, etc. These pulse points are usually related to changes in gas flow patterns, chemical reaction rates or local gas flow impacts. The timestamp, amplitude and duration of each pulse point are recorded, and a pressure pulse point sequence is formed, which reflects the change process of the fluid dynamics state in the furnace. The time correlation between the vibration mutation point sequence and the pressure pulse point sequence is analyzed, and the time delay between any two points (one from the vibration sequence and one from the pressure sequence) is calculated. If the time delay is within a preset time window (such as 0-5 seconds), it is considered that there is a potential causal relationship between the two points. A network structure is constructed, in which the network nodes include all the vibration mutation points and pressure pulse points. If the time delay between two nodes meets the correlation condition, an edge is established between the two nodes, and the weight of the edge is determined by the size of the time delay, forming a vibration-pressure coupling network. This network describes the spatio-temporal correlation pattern between vibration events and pressure events. Complex network analysis methods are applied to calculate the key topological properties of the vibration-pressure coupling network, including the average path length (the average shortest path between any two nodes in the network) and the clustering coefficient (describing the aggregation degree of nodes in the network). The shorter the average path length, the faster the propagation speed of vibration events and pressure events, and the tighter the coupling. The higher the clustering coefficient, the more local clusters formed in the network, and the stronger the cooperativity of the system. Based on network theory and experimental verification, a carbon deposition trigger probability calculation model is established, which expresses the quantitative relationship between network topological properties and carbon deposition risk. For example, the carbon deposition trigger probability is negatively correlated with the average path length (the shorter the path, the higher the risk), and positively correlated with the clustering coefficient (the stronger the clustering, the higher the risk). The carbon deposition trigger probability of each reaction front partition is calculated through this model, and the probability value is between 0 and 1, representing the risk degree of carbon deposition in the region. This probability value provides a quantitative basis for subsequent control strategy formulation.

[0029] In the embodiment of the present application, the detailed implementation steps of step 5 include: In each reaction front sub-zone, a local flow field bottleneck region is identified according to the in-furnace reaction thermodynamic constraint map, wherein the local flow field bottleneck region is the region where the spatial conflict between the temperature gradient and the CO / CO2 ratio gradient is the most serious in the in-furnace reaction thermodynamic constraint map; In the local flow field bottleneck region, a turbulence-induced path is designed, wherein the starting point of the turbulence-induced path is the center of the local flow field bottleneck region, the ending point of the turbulence-induced path is the boundary of the reaction front sub-zone, and the path direction is determined by the thermodynamic entropy production rate gradient of the corresponding position point in the in-furnace reaction thermodynamic constraint map; According to the turbulence-induced path, an initial gas injection strategy is generated, wherein the initial gas injection strategy includes an injection angle, an injection flow rate, and an injection frequency, the injection angle is determined by the direction of the turbulence-induced path, and the injection flow rate and the injection frequency are determined by the size of the carbon deposition trigger probability; According to the prediction suppression effect of the initial gas injection strategy on the carbon deposition trigger probability of the reaction front sub-zone, the initial gas injection strategy is dynamically adjusted to obtain an adaptive gas injection strategy, wherein the prediction suppression effect is quantified by simulating the influence of the initial gas injection strategy on the thermodynamic entropy production rate of the in-furnace reaction thermodynamic constraint map.

[0030] In this embodiment, firstly, for each reaction front zone, the in-furnace reaction thermodynamic constraint diagram of that region is analyzed, with particular attention paid to the spatial distribution relationship of the temperature gradient and the CO / CO2 ratio gradient. The spatial conflict index of the two gradient fields is calculated, such as the negative value of the cosine of the angle between the gradient directions. The higher the conflict index, the more inconsistent the directions of the two gradient fields are, and the greater the resistance to fluid flow. Regions with conflict indices exceeding a threshold (e.g., 0.7) are identified. These regions are usually bottleneck areas where gas flow is obstructed and the exchange of reactants and products is hindered. These are marked as local flow field bottleneck regions. Then, within each identified local flow field bottleneck region, gas turbulence inducement is designed. The guiding path starts at the geometric center of the bottleneck region, i.e., the point with the highest conflict index, and ends at the boundary of the reaction front zone, enabling effective gas discharge from the bottleneck region. The path direction is designed based on the thermodynamic entropy increase rate gradient field, arranging the path along the negative direction of the entropy increase rate gradient. This allows the gas flow to move from the high entropy increase region (active reaction zone) to the low entropy increase region (slow reaction zone), promoting the discharge of reaction products and the introduction of fresh reactants. The path shape uses a three-dimensional spline curve to ensure a smooth and continuous path, avoiding flow losses caused by sharp turns. Based on the designed turbulence-induced path, key gas injection parameters are determined, including injection... The injection angle, injection flow rate, and injection frequency are all considered. The injection angle is aligned with the tangent of the turbulence-induced path at the injection point to ensure the initial airflow direction follows the designed path. The injection flow rate is positively correlated with the carbon deposition triggering probability of the reaction front zone; the higher the risk, the stronger the required gas disturbance. A piecewise linear or exponential function is used to establish the mapping relationship between flow rate and probability. The injection frequency is also related to the carbon deposition triggering probability, but upper and lower limits are set to consider the system response characteristics. The combination of these parameters forms the initial gas injection strategy, and a predictive model for the gas injection effect is constructed. This model is based on computational fluid dynamics principles to simulate the effect of gas injection on the flow field and thermodynamic state within the furnace. The effects of the state, especially the change in the thermodynamic entropy increase rate distribution, are considered. The average entropy increase rate change in the bottleneck region before and after injection is calculated as a quantitative indicator of the predicted suppression effect. If the predicted suppression effect is not ideal (e.g., the entropy increase rate is reduced by less than 20%), the parameters of the initial injection strategy are adjusted. Optimization methods such as gradient descent or genetic algorithms are usually used to find the optimal parameter combination. The prediction and adjustment process is repeated until a satisfactory suppression effect is achieved or the maximum number of iterations is reached, ultimately forming an adaptive gas injection strategy. This strategy can adaptively adjust the gas injection parameters according to the characteristics of the bottleneck region and the degree of carbon deposition risk, thereby achieving effective suppression of carbon deposition.

[0031] In this embodiment of the invention, the detailed implementation steps of step 6 include: An initial global flow field distribution map is constructed based on the adaptive gas injection strategies of all reaction front partitions. The initial global flow field distribution map is generated by superimposing the local flow field vectors corresponding to the adaptive gas injection strategies of each reaction front partition. identifying a global flow field conflict region in the initial global flow field distribution map, wherein the global flow field conflict region is a region in the initial global flow field distribution map where the direction of flow field vectors conflict most seriously; performing flow field reconstruction on the global flow field conflict region, wherein the flow field reconstruction minimizes the direction conflict of flow field vectors in the global flow field conflict region by iteratively adjusting adaptive gas injection strategies of adjacent reaction front sub-regions within the global flow field conflict region; optimizing the gas injection parameters in the furnace according to the global flow field distribution map after the flow field reconstruction, wherein the optimization is achieved by minimizing the flow field turbulence intensity of the global flow field distribution map and maximizing the thermodynamic entropy production rate uniformity of the reaction thermodynamic constraint map in the furnace.

[0032] In this embodiment, first, collect the adaptive gas injection strategies of all reaction front sub-zones, use the computational fluid dynamics model to simulate the local flow field generated by each injection strategy in the corresponding sub-zone, calculate the flow velocity vector field, pressure distribution and turbulent characteristics and other flow field parameters, interpolate and superimpose the local flow field vectors of all reaction front sub-zones in the entire furnace space, consider the superposition effect of the flow field and the boundary condition constraint, generate an initial global flow field distribution map covering the entire furnace space, which describes the overall flow pattern of the gas in the furnace when all injection strategies act at the same time, in the initial global flow field distribution map, calculate the direction consistency index of the flow field vector, such as the cosine value of the angle between adjacent grid points, identify the areas where the direction consistency index is below a threshold value (such as 0.3), these areas represent a significant conflict in the direction of the flow field generated by the injection strategies of different reaction front sub-zones, mark as global flow field conflict area, reconstruct the flow field in the identified global flow field conflict area, first analyze the adjacent reaction front sub-zones involved in the conflict area and their injection strategies, construct a conflict resolution priority, and adjust the sub-zone with a lower carbon deposition trigger probability first to ensure the control effect of the high-risk area, for each sub-zone that needs to be adjusted, modify the parameters of its injection strategy, such as injection angle, flow rate and frequency, use iterative optimization methods such as alternating direction multiplier method (ADMM) or collaborative optimization algorithm, adjust the strategy of one sub-zone in each iteration, then recalculate the global flow field, evaluate the change in conflict degree, and the iteration process continues until the flow field direction consistency of the conflict area reaches a satisfactory level or reaches the maximum number of iterations, finally obtain the global flow field distribution map after flow field reconstruction, according to the result of flow field reconstruction, further optimize the specific parameters of each gas injector, including hardware parameters such as nozzle type, caliber size, installation position and injection pressure, and control parameters such as injection timing, duration and alternating mode, the optimization goal is dual: on the one hand, minimize the turbulent intensity of the global flow field to avoid excessive turbulent disturbance that destroys the reaction stability, and on the other hand, maximize the spatial uniformity of the thermodynamic entropy increase rate to promote uniform reaction, the entropy increase rate uniformity is quantified by the inverse of the spatial standard deviation or coefficient of variation of the entropy increase rate, a multi-objective optimization method such as Pareto frontier search or weighted objective function method is used to balance these two possibly conflicting objectives, to obtain the optimal combination of gas injection parameters, and finally form a complete furnace gas injection parameter optimization scheme that considers local control requirements and global synergy effects, which can realize real-time suppression of local carbon deposition and effective separation of titanium in the titanium reduction process of iron concentrate, improve production efficiency and product quality.

[0033] In the embodiment of the present application, the thermodynamic equilibrium correction is realized by the following method: In the space conflict region, a thermodynamic entropy increase optimization model is constructed, wherein a target function of the thermodynamic entropy increase optimization model is a total sum of thermodynamic entropy increases of the space conflict region, and a constraint condition is physical feasibility of temperature values and CO / CO2 ratio values in the space conflict region. An iterative algorithm based on gradient descent is adopted to solve an optimal solution of the thermodynamic entropy increase optimization model, wherein a step length of the iterative algorithm is determined by a modulus length of temperature gradients and CO / CO2 ratio gradients of the space conflict region. According to the optimal solution, the temperature values and the CO / CO2 ratio values in the space conflict region are adjusted to obtain a corrected in-furnace reaction thermodynamic constraint graph.

[0034] In this embodiment, first, the identified temperature gradient and the CO / CO2 ratio gradient are finely divided in the region where there is a spatial conflict, a discretized grid model is constructed, each grid point contains temperature, CO / CO2 ratio and other state variables, based on the principle of the second law of thermodynamics, a thermodynamic entropy increase optimization model is constructed, the objective function of the model is defined as the sum of the thermodynamic entropy increase rates of all grid points in the spatial conflict area, the entropy increase rate is calculated through the Gibbs free energy change rate, and the temperature, pressure, gas composition and other influencing factors are considered, the constraint conditions are set to ensure that the solution result meets the physical feasibility, including temperature constraint (such as the temperature value is not lower than the lowest reaction temperature, and is not higher than the upper limit of equipment tolerance), CO / CO2 ratio constraint (such as the ratio is within the range allowed by thermodynamic equilibrium), continuity constraint of state variables (to prevent the occurrence of physically unreasonable jump), energy conservation and mass conservation constraints, etc., the gradient descent method is used to solve the optimization model, first, the original state value is randomly initialized or used as the starting point, the objective function value and the satisfaction of the constraint condition under the current state are calculated, the gradient of the objective function to each state variable is calculated, the search direction is determined, the step size of the gradient descent is adaptively adjusted, which is proportional to the modulus of the local temperature gradient and the CO / CO2 ratio gradient, a smaller step size is used in the region with larger gradient to improve the precision, and a larger step size is used in the region with smaller gradient to accelerate the convergence, the state variables are updated along the search direction, and it is checked whether the updated state meets the constraint condition, if not, the projection correction is performed to make the state back to the feasible region, the above steps are repeated until the convergence condition (such as the gradient norm is less than a threshold or the iteration number reaches an upper limit) is met, the optimal solution of the optimization model is obtained, according to the optimal solution, the temperature value and the CO / CO2 ratio value in the spatial conflict area are adjusted, so that the adjusted state meets the thermodynamic equilibrium principle, the interpolation method is used in the adjustment process to ensure the smoothness of the spatial distribution and avoid the introduction of new discontinuous points, the adjusted temperature field and gas composition field are remapped to the original furnace reaction thermodynamic constraint graph, the data of the conflict area is updated, and the original data of the non-conflict area remains unchanged, and finally the corrected furnace reaction thermodynamic constraint graph is obtained, which accurately describes the constraint conditions of the furnace reaction under the premise of meeting the thermodynamic principle, and provides a reliable theoretical basis for subsequent reaction front partition.

[0035] In the embodiment of the present application, the connectivity of the vibration-pressure coupled network is quantified in the following way: Calculate the shortest path length between all node pairs in the vibration-pressure coupled network to obtain the average path length; For each node, calculate the edge connection density between the node and its neighbor nodes to obtain the local clustering coefficient of the node; and average the local clustering coefficients of all nodes to obtain the clustering coefficient; According to a weighted sum of the average path length and the clustering coefficient, the connectivity of the vibration-pressure coupled network is quantified, wherein the weights of the weighted sum are determined by the signal-to-noise ratios of the furnace wall vibration signals and the in-furnace local pressure signals.

[0036] In the embodiment, first, the constructed vibration-pressure coupled network is represented in the form of an adjacency matrix or an adjacency list, wherein the matrix elements or the list entries represent whether there is a connection between nodes and the connection strength, for each pair of nodes (i, j) in the network, the shortest path length from node i to node j, i.e. the minimum number of edges passed between the two nodes, is calculated using a breadth-first search or a Dijkstra algorithm, and if there is no path between the two nodes, the distance is defined as infinity or the maximum diameter of the network plus 1, the average value is calculated for all reachable node pairs to obtain the average path length of the entire network, which reflects the efficiency of information propagation in the network, in the carbon deposition triggering network, a shorter average path length means that vibration and pressure events can quickly affect each other, and the triggering process is more agile, for each node i in the network, identify its set of all directly connected neighbor nodes N(i), calculate the actual number of connections E(i) between the nodes in N(i), and calculate the local clustering coefficient of node i, which is defined as the ratio of E(i) to the maximum number of connections that can be formed by the nodes in N(i), for a node with k neighbors, its local clustering coefficient is , which describes the density of the node neighborhood, and the arithmetic average of the local clustering coefficients of all nodes in the network is taken to obtain the clustering coefficient of the entire network, which reflects the tendency of node clustering in the network, in the carbon deposition triggering network, a higher clustering coefficient indicates that vibration and pressure events tend to form closely linked groups, enhancing the triggering effect, the quality of the furnace wall vibration signals and the in-furnace local pressure signals is analyzed, and the signal-to-noise ratio (SNR) of each kind of signal is calculated, the effective component and the noise component in the signal can be estimated using power spectral density analysis or wavelet transform, etc., the weights of the average path length and the clustering coefficient in the connectivity calculation are determined according to the signal quality, the signal with a higher signal-to-noise ratio obtains a larger weight, ensuring that the high-quality signal plays a greater role in the evaluation, and the connectivity quantification formula is constructed as follows: , wherein LT is the connectivity, w1 and w2 are weight coefficients, and w1+w2=1; PU is the average path length, and GU is the clustering coefficient; The calculation formula of the weight coefficient is: , , wherein SNR_Y is the signal-to-noise ratio of the pressure signal, and SNR_Z is the signal-to-noise ratio of the vibration signal; in this way, the connectivity index comprehensively considers the propagation efficiency (the reciprocal of the average path length) and the clustering effect (the clustering coefficient) of the network, and reasonably weights according to the signal quality, providing a reliable network topology characteristic quantification basis for the calculation of the carbon deposition triggering probability.

[0037] In the embodiment of the present application, the flow field turbulence intensity and the thermodynamic entropy increase rate uniformity are quantified in the following way: According to the flow field vector of each position point in the global flow field distribution map, the curl and divergence of the flow field vector are calculated to obtain the turbulence intensity of the position point; the average value of the turbulence intensity of all position points is taken to obtain the flow field turbulence intensity; According to the thermodynamic entropy increase rate of each position point in the furnace reaction thermodynamic constraint map, the spatial distribution entropy of the thermodynamic entropy increase rate is calculated to obtain the thermodynamic entropy increase rate uniformity.

[0038] In the embodiment, first, based on the reconstructed global flow field distribution map, the flow field vector V(x, y, z) of each position point on the spatial discrete grid in the furnace is obtained, including the velocity components in three directions (x, y, z), the curl of the flow field vector is calculated using the numerical differentiation method, represented as ×V(x, y, z), which reflects the local rotation characteristics of the fluid, and the curl is a vector field, which maps the vector field to a vector, and the size of the curl is proportional to the intensity of the fluid rotation; the divergence of the flow field vector is calculated, represented as ·V(x, y, z), which reflects the local expansion or contraction characteristics of the fluid, and the divergence is a scalar field, which maps the vector field to a scalar, and the absolute value of the divergence is proportional to the fluid volume change rate, combining the information of the curl and the divergence, the turbulence intensity index of the position point (x, y, z) is defined: , where α and β are weight coefficients determined according to the relative contribution of rotation and expansion to the formation of turbulence, and usually α>β, because the rotation effect plays a dominant role in the formation of turbulence, the average value of the turbulence intensity index of all grid points in the furnace is taken to obtain the turbulence intensity of the overall flow field: , where N is the total number of grid points, and this index reflects the turbulence degree of the entire flow field, and moderate turbulence is conducive to the mixing of matter and heat, but excessive turbulence will interfere with the reaction stability, the thermodynamic entropy increase rate s(x, y, z) of each position point is extracted from the furnace reaction thermodynamic constraint map, which reflects the local reaction activity and energy conversion efficiency, the spatial distribution uniformity of the thermodynamic entropy increase rate is quantified using the concept of information entropy, first, the entropy increase rate value domain is divided into m equal intervals, the number of grid points in each interval is calculated to obtain the frequency distribution p_i, p_i represents the proportion of grid points whose entropy increase rate falls within the interval [s_i, s_i+1), the distribution entropy is calculated by applying the Shannon entropy formula, when the entropy increase rate is completely uniformly distributed, all p_i are equal, and the distribution entropy reaches the maximum value log m, m is the total number of equal intervals; when the entropy increase rate is completely concentrated in one interval, only one p_i is 1 and the rest are 0, and the distribution entropy reaches the minimum value 0, the calculated distribution entropy H is standardized to the thermodynamic entropy increase rate uniformity index: , the value of the index ranges from [0, 1], the value closer to 1 indicates that the entropy increase rate distribution is more uniform, the value closer to 0 indicates that the distribution is more uneven, and the ideal in-furnace reaction should have a moderate uniform entropy increase rate distribution, avoiding the occurrence of reaction 'hot spots' leading to local overheating, and avoiding the occurrence of reaction 'cold zones' leading to incomplete reaction, in this way, the turbulence intensity TI and the entropy increase rate uniformity HU as two key indicators guide the optimization of gas injection parameters, in the actual optimization process, the upper limit constraint of TI and the lower limit constraint of HU are usually set, or a weighted objective function is constructed is minimized, where γ1 and γ2 are weight coefficients adjusted according to process requirements to balance the requirements of flow field stability and reaction uniformity.

[0039] The above describes the multi-sensor fusion intelligent control method for the iron concentrate titanium reduction process in the embodiments of the application, and the following describes the multi-sensor fusion intelligent control system for the iron concentrate titanium reduction process in the embodiments of the application, please refer to Figure 2 An embodiment of the multi-sensor fusion intelligent control system for the iron concentrate titanium reduction process in the embodiments of the application includes: A data acquisition module for acquiring in-furnace temperature distribution data, in-furnace gas composition distribution data, furnace wall vibration signals and local pressure signals in the reaction furnace at each moment in the iron concentrate titanium reduction process; A thermodynamic constraint construction module for constructing an in-furnace reaction thermodynamic constraint graph according to the spatial distribution characteristics of the in-furnace temperature distribution data and the in-furnace gas composition distribution data; A reaction partition module for dividing the in-furnace area into multiple reaction front partitions according to the in-furnace reaction thermodynamic constraint graph; A carbon deposition analysis module for constructing a vibration-pressure coupled carbon deposition trigger network in each reaction front partition to obtain a carbon deposition trigger probability of each reaction front partition; A strategy generation module for generating an adaptive gas injection strategy for each reaction front partition according to the carbon deposition trigger probability of each reaction front partition; A parameter optimization module for optimizing the in-furnace gas injection parameters according to the adaptive gas injection strategies of all reaction front partitions to realize real-time suppression of local carbon deposition and effective separation of titanium in the iron concentrate titanium reduction process; The various modules are connected through wired and / or wireless means to realize data transmission between the modules.

[0040] The present application obtains comprehensive information of the in-furnace state through multi-sensor data fusion, constructs a reaction thermodynamic constraint graph and divides the reaction front partition, and realizes accurate modeling of the complex reaction process in the furnace; through the vibration-pressure coupled carbon deposition trigger network, the carbon deposition risk is accurately predicted; based on the prediction result, an adaptive gas injection strategy is generated and globally optimized, accurate intelligent control of the iron concentrate titanium reduction process is realized, the local carbon deposition problem is effectively inhibited, the titanium separation efficiency is improved, and the product quality is improved.

[0041] The above are only preferred embodiments of the present application and are not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or replace some technical features with equivalent ones. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0042] It should be noted that the formula in the present specification is a dimensionless value calculated, the formula is a formula obtained by software simulation of a large amount of data to reflect the most real situation, and the preset parameters and threshold values in the formula are set by those skilled in the art according to the actual situation.

[0043] Although the embodiments of the present application have been shown and described, those skilled in the art can understand that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the claims and their equivalents.

Claims

1. A method for intelligent control of a multi-sensor fusion-based titanium reduction process for iron fines, characterized by, The method comprises the following steps: Step 1: obtaining the temperature distribution data, gas composition distribution data, furnace wall vibration signal and local pressure signal of the reaction furnace at each moment during the titanium reduction process of the iron concentrate; Step 2: constructing a furnace reaction thermodynamic constraint graph according to the spatial distribution characteristics of the temperature distribution data and the gas composition distribution data in the furnace; Step 3: dividing the furnace area into multiple reaction front subareas according to the furnace reaction thermodynamic constraint graph; Step 4: constructing a carbon deposition trigger network based on vibration-pressure coupling in each reaction front subarea to obtain the carbon deposition trigger probability of each reaction front subarea; Step 5: generating an adaptive gas injection strategy for each reaction front subarea according to the carbon deposition trigger probability of each reaction front subarea; Step 6: optimizing the gas injection parameters in the furnace according to the adaptive gas injection strategies of all reaction front subareas to realize real-time inhibition of local carbon deposition and effective separation of titanium during the titanium reduction process of the iron concentrate.

2. The process intelligent control method of multi-sensor fusion of iron ore fines and titanium reduction as claimed in claim 1, wherein, The construction of the furnace reaction thermodynamic constraint graph comprises: According to the temperature distribution data in the furnace, the boundary line between the high-temperature core area and the low-temperature edge area in the furnace is identified, and a temperature gradient boundary graph in the furnace is constructed; According to the gas composition distribution data in the furnace, the local extreme points of the CO / CO2 ratio in the furnace are identified, and a gas reaction activity graph in the furnace is constructed; The temperature gradient boundary graph and the gas reaction activity graph in the furnace are spatially superimposed to generate an initial thermodynamic constraint graph; In the initial thermodynamic constraint graph, the spatial conflict area between the boundary line in the temperature gradient boundary graph and the extreme point in the gas reaction activity graph is identified, and the spatial conflict area is corrected by thermodynamic equilibrium to obtain the furnace reaction thermodynamic constraint graph.

3. The process intelligent control method of multi-sensor fusion of iron ore fines and reduction of titanium content as claimed in claim 1, wherein, The division of the furnace area into multiple reaction front subareas comprises: In the furnace reaction thermodynamic constraint graph, a plurality of initial tracking points are randomly selected; For each initial tracking point, the propagation process of a virtual reaction front is simulated, wherein the propagation direction of the virtual reaction front is determined by the weighted vector of the temperature gradient and the CO / CO2 ratio gradient of the corresponding position point in the furnace reaction thermodynamic constraint graph, and the propagation speed of the virtual reaction front is determined by the thermodynamic entropy increase rate of the corresponding position point in the furnace reaction thermodynamic constraint graph; When the thermodynamic entropy increase rate is lower than a preset entropy increase rate threshold during the propagation process of the virtual reaction front, the propagation is stopped, and the area covered by the virtual reaction front is marked as an initial reaction front subarea; The boundary optimization is performed on all the initial reaction front subareas, wherein the boundary optimization is performed by identifying the continuity of the temperature gradient and the CO / CO2 ratio gradient between adjacent initial reaction front subareas, and the boundary with a continuity lower than a preset continuity threshold is smoothed to obtain the reaction front subarea.

4. The process intelligent control method of multi-sensor fusion of iron ore fines and reduction of titanium content as claimed in claim 1, wherein, The construction of the carbon deposition trigger network based on vibration-pressure coupling in each reaction front subarea to obtain the carbon deposition trigger probability of each reaction front subarea comprises: In each of the reaction front sub-zones, according to the time-domain waveform of the furnace wall vibration signal, a local mutation point of the vibration waveform is identified, and a vibration mutation point sequence is constructed; In each of the reaction front sub-zones, according to the time-domain waveform of the local pressure signal in the furnace, a local pulse point of the pressure waveform is identified, and a pressure pulse point sequence is constructed; According to the time synchronization of the vibration mutation point sequence and the pressure pulse point sequence, a vibration-pressure coupling network is constructed, wherein the nodes of the vibration-pressure coupling network are the vibration mutation points and the pressure pulse points, and the edges are the time delays between the vibration mutation points and the pressure pulse points; According to the connectivity of the vibration-pressure coupling network, the carbon deposition trigger probability of the reaction front sub-zone is calculated, wherein the connectivity is quantified by the average path length and the clustering coefficient of the vibration-pressure coupling network.

5. The process intelligent control method of multi-sensor fusion of iron ore fines and reduction of titanium content as claimed in claim 1, wherein, The generation of the adaptive gas injection strategy for each of the reaction front sub-zones includes: In each of the reaction front sub-zones, according to the reaction thermodynamic constraint graph in the furnace, a local flow field bottleneck region in the reaction front sub-zone is identified; In the local flow field bottleneck region, a disturbance-induced path is designed, wherein the starting point of the disturbance-induced path is the center of the local flow field bottleneck region, the ending point is the boundary of the reaction front sub-zone, and the path direction is determined by the thermodynamic entropy increase rate gradient of the corresponding position point in the reaction thermodynamic constraint graph in the furnace; According to the disturbance-induced path, an initial gas injection strategy is generated, wherein the initial gas injection strategy includes injection angle, injection flow rate and injection frequency; According to the prediction suppression effect of the initial gas injection strategy on the carbon deposition trigger probability of the reaction front sub-zone, the initial gas injection strategy is dynamically adjusted to obtain the adaptive gas injection strategy, wherein the prediction suppression effect is quantified by simulating the influence of the initial gas injection strategy on the thermodynamic entropy increase rate of the reaction thermodynamic constraint graph in the furnace.

6. The process intelligent control method of multi-sensor fusion of iron ore fines and reduction of titanium dioxide as claimed in claim 1, wherein, The optimization of the furnace gas injection parameters includes: According to the adaptive gas injection strategy of all the reaction front sub-zones, an initial global flow field distribution graph is constructed; In the initial global flow field distribution graph, a global flow field conflict region is identified; The global flow field conflict region is subjected to flow field reconstruction, wherein the flow field reconstruction minimizes the direction conflict of the flow field vector in the global flow field conflict region by iteratively adjusting the adaptive gas injection strategy of the adjacent reaction front sub-zones in the global flow field conflict region; According to the global flow field distribution graph after the flow field reconstruction, the furnace gas injection parameters are optimized, wherein the optimization is realized by minimizing the flow field turbulence intensity of the global flow field distribution graph and maximizing the uniformity of the thermodynamic entropy increase rate of the reaction thermodynamic constraint graph in the furnace.

7. The intelligent control method of process of iron ore fines de-titanium of multi-sensor fusion as claimed in claim 2, wherein, The thermodynamic equilibrium correction is realized by the following way: In the spatial conflict region, a thermodynamic entropy increase optimization model is constructed, wherein the objective function of the thermodynamic entropy increase optimization model is the total thermodynamic entropy increase of the spatial conflict region, and the constraint condition is the physical feasibility of the temperature value and the CO / CO2 ratio value in the spatial conflict region; An iterative algorithm based on gradient descent is used to solve the optimal solution of the thermodynamic entropy increase optimization model, wherein the step length of the iterative algorithm is determined by the modulus of the temperature gradient and the CO / CO2 ratio gradient in the spatial conflict area; According to the optimal solution, the temperature value and the CO / CO2 ratio value in the spatial conflict area are adjusted to obtain a corrected furnace reaction thermodynamic constraint graph.

8. The process intelligent control method of multi-sensor fusion of iron ore fines and reduction of titanium content as claimed in claim 4 wherein, The connectivity of the vibration-pressure coupling network is quantified by: Calculate the shortest path length between all node pairs in the vibration-pressure coupling network to obtain the average path length; For each node, calculate the edge connection density between the node and its neighbor nodes to obtain the local clustering coefficient of the node; and average the local clustering coefficients of all nodes to obtain the clustering coefficient; The connectivity of the vibration-pressure coupling network is quantified by the weighted sum of the average path length and the clustering coefficient.

9. The intelligent control method of process of iron ore fines de-titanium of multi-sensor fusion as claimed in claim 6, wherein, The flow field turbulence intensity and the thermodynamic entropy increase rate uniformity are quantified by: According to the flow field vector of each position point in the global flow field distribution map, the curl and divergence of the flow field vector are calculated to obtain the turbulence intensity of the position point; and the turbulence intensities of all position points are averaged to obtain the flow field turbulence intensity; According to the thermodynamic entropy increase rate of each position point in the furnace reaction thermodynamic constraint graph, the spatial distribution entropy of the thermodynamic entropy increase rate is calculated to obtain the thermodynamic entropy increase rate uniformity.

10. The multi-sensor fusion intelligent control system for iron concentrate titanium reduction process, which is used to realize the multi-sensor fusion intelligent control method for iron concentrate titanium reduction process according to any one of claims 1 to 9, characterized in that, It comprises: A data acquisition module for acquiring furnace temperature distribution data, furnace gas composition distribution data, furnace wall vibration signals, and local pressure signals in the reaction furnace at each moment during the titanium reduction process of iron concentrate; A thermodynamic constraint construction module for constructing a furnace reaction thermodynamic constraint graph based on the spatial distribution characteristics of the furnace temperature distribution data and the furnace gas composition distribution data; A reaction partitioning module for dividing the furnace area into multiple reaction front partitions according to the furnace reaction thermodynamic constraint graph; A carbon deposition analysis module for constructing a vibration-pressure coupling-based carbon deposition trigger network in each reaction front partition to obtain the carbon deposition trigger probability of each reaction front partition; A strategy generation module for generating an adaptive gas injection strategy for each reaction front partition based on the carbon deposition trigger probability of each reaction front partition; A parameter optimization module for optimizing the furnace gas injection parameters based on the adaptive gas injection strategies of all reaction front partitions to achieve real-time suppression of local carbon deposition and effective separation of titanium during the titanium reduction process of iron concentrate.

Images