CN-122015491-A - Self-adaptive partition combustion-supporting control method for smelting furnace
Abstract
The invention discloses a self-adaptive zonal combustion-supporting control method for a smelting furnace, and relates to the technical field of smelting furnace temperature control. The method comprises the steps of acquiring a thermal image containing pixel coordinates and temperature through a collecting device, presetting a partition template and corresponding to a combustion fan one by one, calculating comprehensive proper matching degree based on temperature distribution focal power and uniformity balance, temperature gradient directivity and smoothness and combining dynamic weights, dynamically adjusting the partition according to expert experience, targeting and global optimal repartitioning strategies when the matching degree is lower than a threshold value, and controlling the deflection angle of the combustion fan to enable flame to accurately cover the partition after adjustment. The method realizes partition self-adaptive optimization and accurate combustion supporting, improves heating efficiency and temperature uniformity, and reduces energy consumption.
Inventors
- BI JIACHENG
- WANG PENG
Assignees
- 四川华洁铝业有限公司
Dates
- Publication Date
- 20260512
- Application Date
- 20260209
Claims (9)
- 1. The self-adaptive zonal combustion supporting control method for the smelting furnace is characterized by comprising the following steps of: s1, acquiring a thermal image of a smelting furnace through an acquisition device, wherein each pixel point in the thermal image is provided with a unique coordinate and a corresponding temperature value, each pixel point is expressed as ((x, y) T), wherein (x, y) is a two-dimensional plane coordinate of the pixel point, and T is a real-time temperature value under the coordinate; S2, partitioning the thermal image by adopting a preset initial partitioning template, numbering each partition, wherein the partition numbers are in one-to-one correspondence with the combustion-supporting fans, and the combustion-supporting fans corresponding to the numbers are responsible for combustion supporting of the corresponding partitions; S3, along with temperature change in the melting process of the regenerated aluminum, obtaining the coincidence degree of the cold material heating efficiency and the partition contour of each partition, the temperature abnormal gradient strength and the overheat area according to the temperature data in the thermal image, substituting the dynamic weight adjusted along with the process stage into a preset comprehensive adaptation degree function, and calculating to obtain the comprehensive adaptation degree of each partition; S4, if the comprehensive adaptation degree is lower than a preset adaptation degree threshold, adjusting the size and/or shape of the corresponding partition to realize real-time change of the partition; s5, controlling the corresponding combustion-supporting fans to rotate according to the adjusted subareas to change the air outlet angle, and enabling the corresponding combustors of the combustion-supporting fans to rotate by the same angle as the air outlet angle, so that flames generated by the combustion-supporting fans cover the adjusted subareas.
- 2. The self-adaptive zonal combustion supporting control method of a smelting furnace according to claim 1, wherein the calculation of the coincidence degree of the cold charge heating efficiency and the zonal profile comprises the following steps: S30, identifying a melting region, a cold material region and a overheating region on the surface of a molten pool in the thermal image through an image processing technology, wherein each region is defined by a pixel point set, pixel groups corresponding to different temperature states on the surface of the molten pool respectively, and realizing accurate distinction through pixel point temperature characteristic differences, namely a melting region pixel set P melt = { ((x, y), T) |T epsilon [ T1, T2] }, and a cold material region pixel set = { ((X, y), T) |t < T1}, the superheat region pixel set P over = { ((x, y), T) |t > T2}, wherein T1 is the melting critical melting temperature of the secondary aluminum, and T2 is the preset upper temperature limit; S31, calculating cold material heating efficiency, wherein the calculation formula of the cold material heating efficiency is as follows: wherein For the purpose of the heating efficiency of the cold material, As a set of pixels in the cold-charge region, For the current temperature of the cold-material region pixels, For the initial temperature of the cold-charge area pixels, The total energy consumption of the combustion-supporting fans in the corresponding subareas; S32, calculating the intersection ratio of the partition contour and the natural contour of the cold material region based on the temperature abrupt change characteristics of the pixel points by adopting an image similarity algorithm, and evaluating the rationality of the partition shape by the intersection ratio.
- 3. The self-adaptive zonal combustion supporting control method of smelting furnace according to claim 1, wherein the calculation of the temperature anomaly gradient strength comprises the following steps: B1. Performing pixel level operation on the thermal image to obtain a temperature gradient of each pixel point, wherein the temperature gradient comprises a direction parameter and a intensity parameter of temperature change; the temperature gradient vector of each pixel point is calculated through a basic gradient vector formula of a Sobel operator, and a gradient module value is further calculated based on the temperature gradient vector, wherein the gradient module value is used for quantifying the intensity of temperature change; B2. Comparing the temperature gradient field of the current subarea with a preset ideal gradient mode pixel by pixel, wherein the ideal gradient mode is a gradient distribution mode which points to the center of an unmelted solid cold material area from a burner nozzle and has a clear and smooth temperature gradient transition zone; B3. If the gradient modulus value exceeds a preset gradient threshold value and the angle difference between the direction and the ideal gradient mode is larger than the preset angle threshold value, judging the pixel point as an abnormal gradient pixel; B4. And calculating the sum of gradient modulus values of all abnormal gradient pixels in the partition as the temperature abnormal gradient intensity, and quantifying the disturbance degree of the temperature gradient in the partition.
- 4. A smelting furnace adaptive zonal combustion control method according to claim 2 or 3, wherein the comprehensive fitness function is: ; Wherein, the In order to integrate the degree of adaptation, 、 、 、 Is a dynamic weight and the sum is 1, For the heating efficiency of the cold charge, For the cross-over ratio to characterize the zonal contour fitness, For the intensity of the temperature anomaly gradient, The dynamic weight is adjusted according to the process stage of melting the regenerated aluminum for the overheating area of the high-temperature area converted based on the pixel set P over of the overheating area.
- 5. The self-adaptive zoning combustion supporting control method of the smelting furnace according to claim 1, wherein the regulation of the zoning in the step S4 adopts one or more of the following repartitioning strategies, namely, targeted repartitioning based on cold and hot area identification of temperature data of pixels of a thermal image, global optimal repartitioning based on an optimization algorithm aiming at maximizing comprehensive adaptation degree and expert experience repartitioning based on a rule base combined with a smelting process rule; The strategy priority of partition adjustment is that expert experience repartitioning based on a rule base is executed preferentially, local partition details are optimized through targeting repartitioning based on cold and hot area identification, and overall fitness is checked through global optimal repartitioning based on an optimization algorithm, so that each partition is ensured to be free of overlapping and coverage dead areas.
- 6. The self-adaptive zonal combustion supporting control method of smelting furnace according to claim 5, wherein the targeted repartition based on cold and hot area identification comprises the following steps: C1. Processing the thermal image by adopting an image segmentation algorithm, and accurately identifying a cold material area, a melting area and a superheat area based on the temperature value of each pixel ((x, y) and T); C2. the partition is adjusted according to a preset repartitioning principle, wherein the repartitioning principle comprises that one partition covers a P cold pixel set corresponding to an independent cold material area as completely as possible, the partition boundary is aligned with a natural contour formed by a cold/hot area based on pixel point temperature distribution, and the occurrence of the partition with too few pixel points or too long and narrow shape is avoided; C3. Through merging, splitting or deformation repartition, adjacent and internal pixel points ((x, y), T) are merged in regions with uniform temperature and small temperature gradient, splitting is carried out on the regions containing P cold pixel sets corresponding to a plurality of independent large-area cold material regions, and the boundary of the regions is dynamically adjusted according to the natural contour formed by the cold material regions based on the abrupt change of the temperature of the pixel points, so that deformation is realized.
- 7. The self-adaptive zonal combustion supporting control method of smelting furnace according to claim 5, wherein the global optimal repartition based on the optimization algorithm comprises the following steps: D1. Taking the comprehensive fitness function as an objective function, wherein the objective is to maximize the total fitness of the whole hearth; D2. Setting the optimization variable as the boundary position of the subarea, wherein the constraint conditions comprise that the subarea covers the whole surface of the molten pool and has no overlap, the area of the subarea is not smaller than the minimum area covered by the effective flame of the burner, and the shape of the subarea is not excessively singular; D3. And adopting a heuristic optimization algorithm to iterate and generate an optimal partition scheme, wherein the heuristic optimization algorithm comprises a genetic algorithm or a particle swarm algorithm.
- 8. The self-adaptive zonal combustion supporting control method of smelting furnace according to claim 5, wherein the expert experience repartition based on the rule base comprises the following steps: Z1. if the new cold material is detected near the furnace door, merging at least two subareas near the furnace door, and adjusting the angle of the corresponding combustion-supporting fan to enable the flame of the burner to be adapted to cover the merged subareas; Z2, if the temperature standard deviation of a certain partition is continuously higher than a preset temperature threshold value, splitting the partition into at least two partitions; And Z3, if the system is in the heat preservation stage, automatically switching to a preset large-area low-power heat preservation partition template.
- 9. The self-adaptive zonal combustion supporting control method for a smelting furnace according to claim 1, wherein the controlling of the left-right up-down deflection angle of the combustion supporting fan in S5 comprises the following steps: S50, calculating the geometric center of the adjusted subarea, taking the geometric center as a target point of aiming of the flame of the burner, and matching an air outlet angle by the combustion-supporting fan by taking the target point as a reference; s51, calculating a left-right deflection angle and an up-down pitching angle required by the combustion-supporting fan according to the shape, the size and the inherent parameters of the burner of the adjusted subarea, and ensuring that the air outlet direction is matched with the flame injection direction of the burner so as to cover the whole subarea; S52, controlling the combustion-supporting fan to smoothly transit from the current angle to the calculated target angle, wherein the angle change rate does not exceed a preset angle change threshold value, so that the burner flame stably covers the subareas under the assistance of the air outlet of the combustion-supporting fan.
Description
Self-adaptive partition combustion-supporting control method for smelting furnace Technical Field The invention relates to the technical field of smelting furnace temperature control, in particular to a self-adaptive zonal combustion-supporting control method for a smelting furnace. Background In the technical field of temperature control of a secondary aluminum smelting furnace, the traditional temperature control mostly adopts a fixed partition template or an integral unified temperature control mode, and has a plurality of technical limitations and application pain points that the distribution of a cold material area, a melting area and a superheating area on the surface of a molten pool continuously changes due to random addition positions of the cold material in the process of smelting secondary aluminum and dynamic change of melting progress, and the fixed partition template cannot adjust the size and the shape of a partition according to a real-time temperature field, so that the situation that the partition is disjointed from an actual temperature area frequently occurs, the cold material area cannot be accurately covered and heated to cause low melting efficiency, and the melting area or the superheating area is continuously heated to cause energy waste and even cause local burning loss of a furnace body; meanwhile, the traditional control method lacks accurate control on the directionality and smoothness of the temperature gradient, a scientific and quantitative adaptation degree assessment system is not established, the heating effect is judged only by experience, abnormal conditions such as temperature mutation, gradient disturbance and the like cannot be effectively identified, further abnormal conditions such as uneven temperature distribution of a molten pool are caused, regenerated aluminum melting consistency is poor, material defects are prone to occur, in addition, the angle of a combustion-supporting fan is mostly fixed or is regulated later, flame coverage is difficult to adapt to a dynamically-changed partition, the problems of inaccurate heating and large energy consumption are further aggravated, the prior art fails to combine a core requirement dynamic optimization control strategy of each process stage of the melting of the regenerated aluminum, multi-objective balance of heating efficiency, melting quality, energy consumption control and equipment protection cannot be achieved, and practical requirements of high efficiency, energy conservation and accurate temperature control in industrial production are difficult to meet. Disclosure of Invention The invention aims to provide a self-adaptive zonal combustion-supporting control method for a smelting furnace, which solves the problems in the background technology. The invention is realized by the following technical scheme: a self-adaptive zonal combustion-supporting control method for a smelting furnace comprises the following steps: s1, acquiring a thermal image of a smelting furnace through an acquisition device, wherein each pixel point in the thermal image is provided with a unique coordinate and a corresponding temperature value, each pixel point is expressed as ((x, y) T), wherein (x, y) is a two-dimensional plane coordinate of the pixel point, and T is a real-time temperature value under the coordinate; S2, partitioning the thermal image by adopting a preset initial partitioning template, numbering each partition, wherein the partition numbers are in one-to-one correspondence with the combustion-supporting fans, and the combustion-supporting fans corresponding to the numbers are responsible for combustion supporting of the corresponding partitions; S3, along with temperature change in the melting process of the regenerated aluminum, obtaining the coincidence degree of the cold material heating efficiency and the partition contour of each partition, the temperature abnormal gradient strength and the overheat area according to the temperature data in the thermal image, substituting the dynamic weight adjusted along with the process stage into a preset comprehensive adaptation degree function, and calculating to obtain the comprehensive adaptation degree of each partition; S4, if the adaptation degree is lower than a preset adaptation degree threshold, adjusting the size and/or shape of the corresponding partition to realize real-time change of the partition; And S5, controlling the corresponding combustion-supporting fans to rotate according to the adjusted subareas to change the air outlet angle, and enabling the corresponding combustors of the combustion-supporting fans to rotate by the same angle as the air outlet angle, so that flames generated by the combustion-supporting fans cover the adjusted subareas. Further, the method comprises the steps of, The cold material heating efficiency and the partitioned contour matching degree calculation comprises the following steps: S30, identifying a melting region, a cold material r