A method for optimizing kiln flame shape regulation by air-coal ratio test

Abstract

The application discloses a kind of wind coal proportioning test optimization kiln flame form regulation and control methods, it is related to cement rotary kiln combustion regulation and control technical field, the method includes: building flame three-dimensional quantitative evaluation system and calculating comprehensive score, carries out cold state test and establishes sample library, combines hot state calibration condition and fits mapping model and verifies accuracy, is completed parameter optimization by qualified model, issues optimal parameter and iteratively updates model according to measured deviation;The application judges flame state by unified quantitative standard, reduces hot state test burning frequency by cold-hot state mapping, reduces test energy consumption, reduces production disturbance, and the flame regulation and control result is more stable.

Description

Technical Field

[0001] This invention relates to the field of combustion control technology in cement rotary kilns, specifically a method for optimizing the flame morphology control within the kiln through air-coal ratio testing. Background Technology

[0002] Cement clinker calcination is primarily completed within a rotary kiln. The flame pattern formed by pulverized coal combustion at the kiln head directly affects the temperature field distribution within the kiln, ultimately impacting clinker calcination quality, system energy consumption, and the service life of the refractory materials within the kiln. The air-coal ratio is the core method for controlling the flame pattern; its adjustment precision directly affects the operational stability and energy consumption level of the production line. Cold-state flow field testing is a common method for burner performance testing, currently primarily used for burner factory inspection and post-repair structural acceptance.

[0003] In actual production, the optimization of the coal-air mixture ratio mostly relies on hot industrial test combustion. Operators judge the combustion effect by visually observing the flame appearance and combining it with clinker test indicators, repeatedly adjusting parameters iteratively. This method depends on the operator's personal experience, lacks a unified flame evaluation standard, and the operating conditions adjusted by different personnel vary greatly, easily causing fluctuations in clinker quality. Moreover, the effect of hot test combustion can only be verified after the operating conditions stabilize, resulting in a long optimization cycle, high coal consumption during the test process, and frequent parameter adjustments that can disrupt the normal production rhythm. Because the flow field in the room temperature cold state differs significantly from the actual high-temperature combustion conditions, there has been no systematic quantitative method in the industry to correspond them. Cold state tests cannot be directly used to guide the ratio adjustment in hot production, and their application value has not been fully realized. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies by providing a method for optimizing flame morphology control in cement rotary kilns through air-coal ratio testing. This method addresses the problems of relying entirely on manual judgment and high heat consumption in adjusting flames in cement rotary kilns. It first simultaneously collects three types of flame physical field data to establish a unified quantitative scoring standard. Then, it uses multi-section sampling to analyze the combustion performance of different flame segments and accumulates samples through standardized cold-state tests using the isomomentum rule. This invention requires only a small number of hot-state conditions to fit a hot-cold mapping model, and combines coal quality data to correct the optimization range. It also considers two production limitations—emissions and kiln temperature—to screen adjustable parameters. After field operation, the model can be continuously updated based on multiple measured deviations. The entire process eliminates the need for repeated high-temperature firing tests, reducing production disturbance and coal powder consumption. The evaluation results are not affected by operator experience, making it suitable for the daily combustion parameter adjustment work of various pulverized coal rotary kilns.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio test, the method comprising:

[0006] S1. Establish a three-dimensional quantitative evaluation system for flame morphology, and simultaneously collect three types of field data in the kiln: flame temperature field, gas-solid two-phase velocity field, and pulverized coal concentration field. Rely on multi-dimensional field data to fully restore the true combustion state of the flame, avoid the one-sided problem caused by single parameter judgment, extract multiple quantitative indicators, set the weight of each indicator according to the clinker calcination process requirements, and calculate the comprehensive score of flame morphology.

[0007] S2. Before the cold flow field test, the consistency of operating conditions is verified to eliminate the interference of environmental and gas supply fluctuations on the test data and ensure that each group of test data has horizontal comparison value. Within the set parameter range, the characteristic parameters of the cold flow field at the burner outlet corresponding to different air-coal ratios and swirl parameters are tested one by one to form a cold parameter sample library.

[0008] S3. Select multiple calibration conditions covering common loads to conduct hot-state tests and obtain the comprehensive flame score for each condition; use cold-state flow field parameters as input and hot-state flame score as output to fit a nonlinear mapping model, establish the correspondence between room temperature test data and high-temperature actual combustion effect, and predict the flame state without repeatedly conducting high-temperature test combustion. Accuracy verification is completed through reserved independent verification conditions.

[0009] S4. For the target production conditions, with the highest flame comprehensive score as the goal, combined with the preset process boundary constraints, the parameters are optimized by traversing the nonlinear mapping model based on the verified qualified model. At the same time, multiple production restrictions such as combustion effect, environmental emission, and equipment heat resistance are taken into account. Adjustable parameters that are suitable for the site are selected to obtain the optimal air-coal ratio and swirl parameters.

[0010] S5. Send the optimal parameters to the combustion control system and adjust the damper, coal feeding parameters and swirl vane angle accordingly. After the operating conditions stabilize, collect actual flame data to calculate the score. Rely on the deviation between the actual measurement and the prediction to continuously correct the model, maintain the accuracy of parameter prediction in the long term, and trigger model iterative updates according to the deviation between the prediction and the actual score.

[0011] Furthermore, the quantitative indicators in S1 are sampled and extracted from five cross-sections at equal intervals along the flame axis. Segmented sampling can capture the combustion differences in different regions of the front, middle and rear of the flame, and the indicators are consistent with the complete change law of the flame. Among them, the temperature field corresponds to the extraction of flame length coefficient, core high temperature zone ratio and temperature field uniformity, the velocity field corresponds to the extraction of heat and mass exchange efficiency in the recirculation zone, and the concentration field corresponds to the extraction of coal powder burnout gradient. The weights of each indicator are set for the three production targets of high output, low consumption and low nitrogen respectively. When the production focus changes, only the weights need to be adjusted to adapt to different control requirements, without the need to rebuild the evaluation system.

[0012] Furthermore, in S1, when calculating the comprehensive score of flame morphology, the process threshold range corresponding to each index is first normalized to unify the value range of different physical dimension indices, so as to avoid the large magnitude of a single index from interfering with the overall score result. Then, the final score is obtained by weighted summation according to the set weights, and the sum of all weight coefficients is 1.

[0013] Furthermore, the cold flow field test in S2 uses the total momentum of the burner outlet section as the matching benchmark and sets the air supply parameters according to the equal momentum similarity conditions of the hot rated operating condition to ensure that the cold air flow pattern and the high temperature flue gas flow pattern are consistent, so that the relative deviation between the cold outlet momentum and the hot rated operating condition outlet momentum does not exceed the preset allowable range.

[0014] Furthermore, when conducting the cold flow field test in S2, the air supply condition is set according to the same momentum similarity condition of the hot rated operating condition, so that the dynamic characteristics of the cold flow field and the hot combustion flow field are consistent, and the cold test results can truly reflect the flow state at the outlet of the high-temperature burner by relying on the consistency of flow characteristics.

[0015] Furthermore, in the S2 condition consistency verification, the main air supply pipe pressure and the test environment temperature are calibrated and the baseline values ​​are recorded before the test. Before the start of each test, the corresponding operating condition parameters are retested. If the parameters exceed the allowable deviation range of the baseline value, the air supply parameters are adjusted to the baseline range before the test is started to avoid the incomparability of test data caused by environmental and air supply fluctuations.

[0016] Furthermore, the calibration conditions in S3 are evenly distributed according to the load gradient, covering the lowest to the highest commonly used loads of the production line, so that the model has corresponding calibration data in the entire production load range, and can stably predict the flame effect under both high and low load conditions. At the same time, at least one set of conditions is reserved as an independent verification group. The nonlinear mapping model is established by fitting the response surface method. When the relative error between the predicted score and the measured score of the verification condition is no more than 5%, the model is judged to be qualified. If it is not qualified, the calibration conditions are supplemented according to the load gradient and the model is refitted.

[0017] Furthermore, before parameter optimization in S4, the ash content, volatile matter, calorific value and ash composition data of the coal powder entering the kiln are detected by using the preset benchmark coal quality parameters as a reference. The pre-established coal quality-flame characteristic association library is called. When the deviation of the coal quality parameters exceeds the set threshold, the upper and lower limits of the parameter optimization range are adjusted accordingly to offset the flame prediction deviation caused by the fluctuation of coal powder raw materials. The appropriate adjustment parameters can still be output after the coal quality changes.

[0018] Furthermore, the boundary constraint optimization in S4 uses the upper limit of nitrogen oxide emissions at the kiln tail and the upper limit of kiln surface temperature as mandatory constraints. First, parameter combinations that simultaneously satisfy all constraints are screened out, and then the combination with the highest comprehensive flame morphology score is selected as the optimal solution. If there is no combination that satisfies all constraints, non-core constraints are gradually relaxed according to constraint priority and then optimization is carried out again. Priority is given to ensuring the hard requirements of environmental protection and equipment safety, and then pursuing better combustion effect.

[0019] Furthermore, in S5, when the relative deviation between the measured score and the predicted score exceeds 5% after three consecutive adjustments, the model is automatically triggered to iterate and retrain. The retraining retains all the original calibration samples to prevent the loss of historical working condition data after the model is updated, and the model will not fail after the equipment status changes. The newly added measured data is assigned decreasing weights according to the collection time to participate in the fitting, and the original model parameters are directly replaced after the completion.

[0020] Compared with existing technologies, this method for optimizing the flame morphology control in a kiln through air-coal ratio testing has the following advantages:

[0021] I. This invention establishes a quantitative evaluation system for flame morphology that combines temperature field, gas-solid two-phase velocity field, and pulverized coal concentration field, replacing the previous method of judging flame state by human visual inspection. This eliminates the judgment bias caused by differences in personal experience, and the evaluation results are stable and repeatable, which can more accurately reflect the actual combustion state of the flame, providing a unified quantitative judgment basis for the precise adjustment of air-coal ratio.

[0022] Second, this invention establishes a quantitative correspondence between cold-state flow field characteristics and hot-state flame morphology through a small number of hot-state calibration conditions, expanding the application scope of cold-state flow field testing from simple structural acceptance to the proportion optimization stage of daily production. When subsequent changes in operating conditions require parameter adjustments, proportion optimization can be directly completed through cold-state testing, eliminating the need for repeated hot-state combustion tests. This shortens the combustion optimization cycle, reduces coal consumption and interference with normal production during the testing process, and achieves optimization accuracy comparable to the full hot-state combustion test scheme.

[0023] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0025] Figure 1 This is a schematic diagram of the overall process of a method for optimizing the flame morphology control in a kiln through a coal-air ratio test, provided in an embodiment of the present invention.

[0026] Figure 2 is a schematic diagram of the flame morphology comprehensive scoring calculation process of the present invention;

[0027] Figure 3 is a schematic diagram of the closed-loop process of cold-state-hot-state mapping modeling and parameter optimization control in this invention. Detailed Implementation

[0028] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0029] This method is implemented through a cement rotary kiln head combustion system equipped with a swirl-type pulverized coal burner. This system includes angle-adjustable swirl vanes, independent internal and external air supply branches, a coal feeding device, and a combustion control system. This embodiment constructs a multi-physics coupled flame quantification evaluation standard, and establishes parameter mapping relationships by combining cold-state flow field experiments and a small number of hot-state calibration conditions. This achieves low-cost, high-precision optimization of the air-coal ratio and control of the flame morphology. Figure 1 As shown, this method mainly includes five core steps: establishing a flame morphology evaluation system, constructing a cold flow field test and sample library, establishing a cold-to-hot mapping model, parameter optimization, and closed-loop control. The specific implementation process is as follows.

[0030] When constructing a three-dimensional quantitative evaluation system for flame morphology, multiple sets of colorimetric pyrometers were arranged along the flame axis at the kiln head hood to collect temperature field data, and wind speed detection elements were arranged at the burner outlet section to collect gas-solid two-phase velocity field data. Combined with the kiln tail flue gas analysis device and the cross-sectional sampling data, pulverized coal concentration field data was obtained. All three types of data were collected simultaneously, with a unified time reference to ensure data time sequence correspondence. After data collection, multiple cross-sections were taken at equal intervals along the flame axis, and corresponding quantitative indicators were extracted from the three types of physical field data. Three characteristic indicators were extracted from the temperature field data: the flame length coefficient (the ratio of the distance from the highest flame temperature point to the burner nozzle to the kiln's inner diameter), the core high-temperature zone proportion (the proportion of the volume above the set temperature threshold to the total flame volume), and the temperature field uniformity (the ratio of the standard deviation of the cross-sectional temperature at half the flame length to the average temperature of the cross-section). The recirculation zone heat and mass transfer efficiency was extracted from the velocity field data, calculated by combining the axial length, maximum diameter, and average recirculation wind speed of the central recirculation zone, and used to reflect the flame stabilization capability at the flame root. The pulverized coal burnout gradient is extracted from the concentration field data. The slope of the change is then fitted after calculating the pulverized coal burnout rate at each cross section along the axial direction to reflect the smoothness of the combustion process.

[0031] As shown in Figure 2, the process for calculating the individual scores of the above indicators and obtaining the final comprehensive flame morphology score is as follows. When calculating the comprehensive flame morphology score, first, normalization is performed on the corresponding process threshold range for each indicator. The process threshold range for each indicator is determined based on the historical operating data of the corresponding production line and the process requirements of clinker calcination. The reasonable fluctuation range of the indicator under normal production conditions is taken as the upper and lower limits of normalization; values ​​exceeding the upper and lower limits are treated as boundary values. Indicators with higher values ​​and better performance are scored using a positive rule, while indicators with lower values ​​and better performance are scored using a negative rule. Indicators with different dimensions are uniformly converted into standard scores within the same score range. Then, a weighted summation method is used to calculate the final comprehensive flame morphology score. The calculation formula is:

[0032]

[0033] In the formula, The overall score is based on the flame pattern. The index number ranges from 1 to... , This represents the total number of indicators. For the first The weight coefficients corresponding to each indicator, and the sum of all weight coefficients is 1. For the first The scores for each indicator are normalized. Weighting coefficients are pre-set based on the actual process requirements of clinker calcination, with initial values ​​determined jointly by the analytic hierarchy process (AHP) and the experience of on-site process personnel. First, each indicator is ranked according to its impact on calcination quality. Then, a judgment matrix is ​​constructed through pairwise comparisons to obtain the initial weights. Finally, fine-tuning is performed based on the actual on-site operating results. When production targets show a bias, the weighting percentages of each indicator can be adjusted accordingly.

[0034] After establishing the flame evaluation system, preparations were made for cold-state flow field tests. Significant differences exist in the density, velocity, and other physical properties of ambient air and high-temperature combustion flue gas. Cold-state data cannot directly correspond to the hot-state combustion state. Therefore, the test used the total momentum of the burner outlet section as the matching benchmark, setting the air supply parameters according to the equal momentum similarity conditions of the hot-state rated operating condition. The dynamic structure of the flow field is mainly determined by the outlet momentum. When the total outlet momentum is kept consistent between the cold and hot states, the changes in core flow field characteristics such as swirl intensity and recirculation zone distribution exhibit a corresponding relationship. This is the physical basis for cold-state data to map the hot-state flame state. When setting the air supply parameters, the rated outlet momentum was first calculated based on the flue gas density and outlet velocity under the hot-state rated operating condition. Then, the corresponding air supply velocity and pressure under ambient air conditions were calculated, ensuring that the deviation between the cold-state outlet momentum and the hot-state rated operating condition outlet momentum was controlled within a preset allowable range, guaranteeing that the cold-state test data has the basis for mapping the hot-state operating condition.

[0035] Before the formal start of the test, a consistency check of operating conditions was performed. First, standard measuring instruments were used to calibrate the main air supply pipe pressure and the test environment temperature, and the corresponding baseline values ​​were recorded. Subsequently, the corresponding operating parameters were re-measured before each group of tests. If the parameters exceeded the allowable deviation range of the baseline values, the opening of the air supply valve was adjusted first, or the test was started only after the ambient temperature returned to the baseline range. This avoided inconsistencies in the baseline data of each group of tests due to environmental fluctuations, ensuring the comparability of all cold-state test data. Regarding parameter settings, four variables were selected as control parameters: swirl vane angle, the proportion of internal air to total air supply, the proportion of external air to total air supply, and the equivalent wind speed for coal feeding. The upper and lower limits of adjustment for each parameter were set according to the actual operating range on site, and multiple parameter combinations were designed using orthogonal experimental tables. This test design method can reduce the total number of test groups while covering the complete parameter range, improving the efficiency of test execution. Cold-state experiments were conducted group by group, with each parameter combination corresponding to a set of cold-state flow field characteristic parameters. The swirling intensity was calculated using the average ratio of the tangential velocity to the axial velocity of the cross-section. The aspect ratio of the central recirculation zone was taken as the ratio of the maximum diameter of the recirculation zone to its axial length. The coefficient of non-uniformity of the cross-sectional velocity distribution was calculated using the ratio of the standard deviation of the wind speed at each point on the cross-section to the average wind speed. After all tests were completed, all parameter combinations and their corresponding flow field characteristic data were compiled into a cold-state parameter sample library.

[0036] As shown in Figure 3, a complete closed-loop control process is formed, from obtaining sample data from cold-state flow field tests to establishing a mapping model, accuracy verification, parameter optimization, and then issuing execution and iterative updates. When establishing the cold-state and hot-state mapping model, multiple sets of operating conditions covering the commonly used load range of the production line are first selected as calibration conditions. These calibration conditions are evenly distributed according to the load gradient, covering the lowest to highest commonly used load range of the production line, avoiding accuracy degradation due to missing calibration data in local areas. At the same time, at least one set of operating conditions in the intermediate load range is reserved as an independent verification group. The data from the verification group does not participate in model fitting but is only used to verify the generalization accuracy of the model. Hot-state verification tests are conducted group by group. Data is collected only after each set of operating conditions has stabilized. The criteria for determining stable operating conditions are that the fluctuation range of kiln tail temperature, system negative pressure, and clinker quality indicators are all within the allowable range for a continuous period, avoiding distortion of calibration data due to operating condition fluctuations. After the operating conditions stabilize, the corresponding multiphysics field data are collected, and the comprehensive flame morphology score is calculated according to the aforementioned method.

[0037] Using three characteristic parameters of the cold flow field as input variables and the comprehensive score of the hot flame morphology under the corresponding working condition as output variables, a nonlinear mapping model is established using the response surface methodology. The input-output relationship of the combustion flow field exhibits weak nonlinearity; a second-order polynomial model with interaction terms can achieve sufficient fitting accuracy while having low computational cost, making it suitable for field engineering applications. The model's calculation formula is as follows:

[0038]

[0039] In the formula, This is the predicted comprehensive score of the flame shape output by the model. , The input cold flow field characteristic parameters, , The values ​​range from 1 to 3, corresponding to the three flow field characteristic parameters. For the constant term of the model, For the first The coefficient of the first-order term corresponding to the input parameter. For the first The coefficients of the quadratic terms corresponding to the input parameters. For the first Item and the The coefficients of the interaction terms corresponding to the input parameters. All coefficients are obtained by fitting the measured data of the calibration conditions using the least squares method. The fitting process aims to minimize the sum of squared residuals between the predicted and measured values. The values ​​of each coefficient are obtained directly through matrix operations, without the need for manual setting of initial values.

[0040] After model fitting, accuracy verification is performed using a reserved independent validation condition. The cold-state parameters of the validation condition are substituted into the model to obtain a predicted score, which is then compared with the hot-state measured score to calculate the relative error. The formula for calculating the relative error is:

[0041]

[0042] In the formula, The relative error of the prediction result is expressed as a percentage. This is the predicted comprehensive score of the flame shape output by the model. This represents the measured value of the comprehensive flame morphology score obtained from the hot test. When the relative error is no greater than a preset threshold, the model accuracy is considered acceptable and can be used for subsequent parameter optimization. If the accuracy is unacceptable, first analyze the load range with the largest error, then supplement the calibration conditions in the corresponding range according to the load gradient and refit the model until the accuracy of the model across the entire range meets the requirements.

[0043] Before optimizing parameters for the target production conditions, preliminary corrections to coal quality parameters are performed. Historical production data and previous calibration tests are used to compile flame variation patterns corresponding to different coal quality parameters, forming a coal quality-flame characteristic correlation library. The library is established by conducting comparative tests under multiple operating conditions with different coal quality parameters, keeping other operating conditions consistent. The optimal ratio ranges are tested, and the offset of the optimal ratio range for each unit fluctuation in coal quality parameters is statistically analyzed, forming a corresponding relationship. The library stores the impact of fluctuations in parameters such as ash content, volatile matter, and calorific value on the offset of the optimal ratio range. Before optimization, the ash content, volatile matter, calorific value, and ash composition data of the pulverized coal entering the kiln are detected. Using preset benchmark coal quality parameters as a reference, the coal quality-flame characteristic correlation library is called for comparison. When the deviation of coal quality parameters exceeds a set threshold, the upper and lower limits of the parameter optimization range are adjusted accordingly to offset the offset effect of coal quality fluctuations on the optimization results, ensuring that the optimization results are suitable for the current coal quality entering the kiln.

[0044] When formally conducting parameter optimization, the goal is to achieve the highest comprehensive score for flame morphology, while simultaneously using the upper limits of nitrogen oxide emissions at the kiln tail and the upper limit of kiln lining surface temperature as mandatory constraints. These two constraints can be obtained by converting flame parameters with corresponding empirical formulas. Nitrogen oxide emissions are directly related to the proportion and peak temperature of the core high-temperature zone, while the kiln lining surface temperature is directly related to the location and length of the high-temperature zone in the flame. Both can be calculated simultaneously using the conversion relationships attached to the mapping model. The empirical coefficients in the conversion relationships are obtained through regression analysis of measured data from historical operating conditions. Multiple sets of flame parameters and corresponding emission and kiln lining temperature data under stable operating conditions are taken, and linear regression is used to fit the coefficient values. During the optimization process, all feasible parameter combinations are first traversed. The predicted score and constraint index values ​​for each set of parameters are calculated using the mapping model. The parameter combinations that simultaneously satisfy all constraints are first selected, and then the combination with the highest score is chosen as the optimal solution. If no combination satisfies all constraints after traversal, the non-core constraints are gradually relaxed according to constraint priority, and the optimization is restarted to ensure that the final output solution balances combustion performance and process limitations.

[0045] After optimization, the optimal air-coal ratio and swirl parameters are sent to the combustion control system. The system then adjusts the damper opening of the air supply branch, the operating frequency of the coal feeding device, and the angle of the swirl vane actuator. Once the operating conditions stabilize, actual flame data is collected to calculate a comprehensive score, and the deviation between the predicted and measured scores is compared. During daily production, the predicted and measured values ​​for each adjustment are continuously recorded. When the relative deviation between the measured and predicted scores exceeds a preset threshold after a preset number of adjustments, iterative retraining of the model is automatically triggered. Setting continuous triggering criteria eliminates false triggers caused by abnormal fluctuations in single operating conditions, preventing frequent model updates from affecting operational stability. During retraining, all original calibration samples are retained to prevent the model from forgetting historical data. Newly added measured data are assigned decreasing weights according to the acquisition time, with newer data having higher weights, making the model more closely reflect the current equipment status and operating characteristics. The time weight reduction factor is set according to the rate of change of operating conditions. A smaller reduction rate is used when the equipment is stable, and a larger reduction rate is used when the operating conditions fluctuate greatly, ensuring that the model can keep up with the changes in operating conditions without being overly sensitive. After retraining, the original model parameters are directly replaced to complete a model update. When load or coal quality fluctuations occur in subsequent production and the ratio needs to be adjusted, optimization can be completed directly through cold-state tests combined with the updated mapping model, without having to repeat the full-process hot-state calibration.

[0046] This method can reduce the number of hot-state test firings while ensuring optimization accuracy, thereby reducing coal consumption and interference with normal production during the testing process, and improving the objectivity and stability of flame control. In addition to combustion optimization scenarios in new dry-process cement rotary kilns, this method can also be used for flame control in other rotary kiln-type industrial equipment using pulverized coal combustion, and is adaptable to swirl burners of different structures.

[0047] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing the flame morphology control within a kiln through a coal-air mixture ratio experiment, characterized in that, The method includes: S1. Establish a three-dimensional quantitative evaluation system for flame morphology, collect data on flame temperature field, gas-solid two-phase velocity field, and pulverized coal concentration field in the kiln, extract multiple quantitative indicators, set the weight of each indicator according to the clinker calcination process requirements, and calculate the comprehensive score of flame morphology. S2. Before the cold flow field test, the consistency of the operating conditions is verified. Within the set parameter range, the characteristic parameters of the cold flow field at the burner outlet corresponding to different air-coal ratios and swirl parameters are tested one by one to form a cold parameter sample library. S3. Select multiple calibration conditions covering common loads to conduct hot-state tests and obtain the comprehensive flame score for each condition; use cold-state flow field parameters as input and hot-state flame score as output to fit a nonlinear mapping model, and complete the accuracy verification through reserved independent verification conditions; S4. For the target production conditions, with the highest flame comprehensive score as the objective, and combined with the preset process boundary constraints, the parameters are optimized by traversing the nonlinear mapping model based on the verified qualified model to obtain the optimal air-coal ratio and swirl parameters. S5. Send the optimal parameters to the combustion control system and adjust the damper, coal feeding parameters and swirl vane angle accordingly; after stabilization, collect actual flame data to calculate the score, and trigger model iteration update based on the deviation between the predicted and measured scores.

2. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, The quantitative indicators in S1 are sampled and extracted from five cross sections at equal intervals along the flame axis. The temperature field corresponds to the extraction of the flame length coefficient, the proportion of the core high-temperature zone, and the temperature field uniformity. The velocity field corresponds to the extraction of the heat and mass exchange efficiency of the recirculation zone. The concentration field corresponds to the extraction of the coal powder burnout gradient. The weights of each indicator are set according to the three production targets of high output, low consumption, and low nitrogen.

3. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, When calculating the comprehensive score of flame morphology in S1, the process is first normalized based on the process threshold range corresponding to each indicator, and then the final score is obtained by weighted summation according to the set weights, with the sum of all weight coefficients being 1.

4. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, The cold flow field test in S2 uses the total momentum of the burner outlet section as the matching benchmark and sets the air supply parameters according to the equal momentum similarity conditions of the hot rated operating condition, so that the relative deviation between the cold outlet momentum and the hot rated operating condition outlet momentum does not exceed the preset allowable range.

5. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, In the S2 condition consistency verification, the main air supply pipe pressure and test environment temperature are calibrated and the baseline values ​​are recorded before the test. Before the start of each test, the corresponding operating condition parameters are re-measured. If the parameters exceed the allowable deviation range of the baseline values, the air supply parameters are adjusted to the baseline range before the test is started.

6. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, In the S2 condition consistency verification, the main air supply pipe pressure and test environment temperature are calibrated and the baseline values ​​are recorded before the test. Before the start of each test, the corresponding operating condition parameters are re-measured. If the parameters exceed the allowable deviation range of the baseline values, the air supply parameters are adjusted to the baseline range before the test is started.

7. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, The calibration conditions in S3 are evenly distributed according to the load gradient, covering the lowest to the highest commonly used load of the production line, while reserving at least one set of conditions as an independent verification group; the nonlinear mapping model is established by fitting the response surface method. When the relative error between the predicted score and the measured score of the verification condition is no more than 5%, the model is judged to be qualified. If it is not qualified, the calibration conditions are supplemented according to the load gradient and the model is refitted.

8. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, Before parameter optimization in S4, the ash content, volatile matter, calorific value and ash composition data of the coal powder entering the kiln are detected with reference to the preset benchmark coal quality parameters. The pre-established coal quality-flame characteristic association library is called. When the deviation of the coal quality parameters exceeds the set threshold, the upper and lower limits of the parameter optimization range are adjusted accordingly.

9. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, The boundary constraint optimization in S4 uses the upper limit of nitrogen oxide emissions at the kiln tail and the upper limit of kiln skin surface temperature as mandatory constraints. First, parameter combinations that satisfy all constraints are selected, and then the combination with the highest comprehensive flame morphology score is selected as the optimal solution. If there is no combination that satisfies all constraints, the non-core constraints are gradually relaxed according to the constraint priority and then the optimization is repeated.

10. The method for optimizing the flame morphology control in a kiln through a coal-air mixture ratio experiment according to claim 1, characterized in that, In S5, when the relative deviation between the measured score and the predicted score exceeds 5% after three consecutive adjustments, the model is automatically triggered to iterate and retrain. The retraining retains all the original calibration samples, and the newly added measured data is assigned decreasing weights according to the collection time to participate in the fitting. After completion, the original model parameters are directly replaced.

Images