Resistance adjusting method and system for kiln treatment facility

Abstract

The invention discloses a resistance adjusting method and system for a kiln treatment facility, and relates to the field of fan collaboration.The method comprises the steps that wind pressure characteristics of each fan in a fan group are extracted, a fluid resistance model is constructed, and the fluid resistance model is used for simulating and analyzing the resistance characteristic coefficient of each fan; based on the resistance characteristic coefficient of each fan, dynamically selecting a main control fan and an auxiliary fan from the fan group by using a multi-target optimization technology, and configuring initial frequency parameters for the main control fan and the auxiliary fan according to the kiln pressure signal under the reference working condition; and monitoring the kiln pressure signal in real time, and if it is monitored that the kiln pressure signal is abnormal, adaptively adjusting the initial frequency parameter so as to realize collaborative optimization between the main control fan and the auxiliary fan. By monitoring the kiln pressure signal in real time and dynamically adjusting the initial frequency parameter of the fan, collaborative optimization between the main control fan and the auxiliary fan is effectively achieved, and the response speed and stability of the fan system are improved.

Description

Technical Field

[0001] The present invention relates to the field of fan coordination, and in particular to a resistance adjustment method and system for a kiln treatment facility. Background Art

[0002] The kiln treatment facilities, including the existing denitrification, desulfurization, and dust removal systems, have been operating continuously at full capacity for extended periods. As the operating time increases, the resistance of each facility gradually increases, causing the fan frequency to continuously increase and remain in a high-hertz state for a long time. Operating the fan at a high frequency poses multiple risks, including a significant increase in energy consumption, with the difference in energy consumption becoming more pronounced with higher operating frequency. Furthermore, fan vibration increases, significantly increasing the bearing failure rate. The risk of inverter failure increases significantly at high frequencies, and the overall equipment life is severely impacted.

[0003] Once the fan system stops due to a fault, the kiln smoke will not be able to be discharged normally, and the kiln pressure will rise sharply in a very short time. When the pressure exceeds the safety threshold, the airflow will violently impact the kiln roof and pool wall, especially in the later operation stage of the kiln, which is very likely to cause the kiln roof to collapse or the pool wall to rupture, and then cause glass liquid to leak, which will bring great risks to the operation safety of the kiln and the company's production.

[0004] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0005] In response to the problems in the related art, the present invention proposes a resistance adjustment method and system for kiln treatment facilities to overcome the above-mentioned technical problems existing in the existing related art.

[0006] To this end, the specific technical solutions adopted in the present invention are as follows: According to one aspect of the present invention, a method for adjusting resistance of a kiln treatment facility is provided, the method comprising: S1. Obtain historical operating parameters of the wind turbine group, extract the wind pressure characteristics of each wind turbine in the wind turbine group and construct a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each wind turbine; S2. Based on the resistance characteristic coefficient of each fan, a multi-objective optimization technique is used to dynamically select the main fan and auxiliary fan from the fan group, and initial frequency parameters are configured for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline operating condition; S3. Monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main control fan and the auxiliary fan. If no abnormality is detected in the kiln pressure signal, the main control fan and the auxiliary fan are operated according to the initial frequency parameters.

[0007] Furthermore, the historical operating parameters of the wind turbine group are obtained, the wind pressure characteristics of each wind turbine in the wind turbine group are extracted, and a fluid resistance model is constructed. The resistance characteristic coefficient of each wind turbine is simulated and analyzed using the fluid resistance model, including: S11. Collect historical operating data of a fan group and extract historical wind pressure characteristics from the historical operating data. The fan group includes a denitrification fan, a desulfurization fan, and a dust removal fan. S12. Mapping the historical wind pressure characteristics to a high-dimensional space based on a predefined nonlinear kernel function to obtain high-dimensional features; S13, constructing a ridge regression prediction model, inputting the high-dimensional features into the ridge regression prediction model, and outputting the wind pressure change parameters of each wind turbine in the future time period; S14. Construct a fluid resistance model based on the wind pressure variation parameters in the future time period, simulate and analyze the resistance response coefficient during wind pressure disturbance, and calculate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient.

[0008] Furthermore, a fluid resistance model is constructed based on the wind pressure variation parameters in the future time period, and the resistance response coefficient during wind pressure disturbance is simulated and analyzed. Based on the resistance response coefficient, the resistance characteristic coefficient of each wind turbine is calculated, including: S141. Compare the wind pressure change parameter in the future time period with the historical wind pressure characteristic parameter to obtain the wind pressure change amplitude, and assign a weight to the historical wind pressure characteristic based on the wind pressure change amplitude; S142. Calculate the observed resistance performance coefficient of each wind turbine in the future time period based on the weight coefficient of the historical wind pressure characteristics; S143. Constructing a fluid response model based on a time evolution mechanism using finite element analysis technology, dividing the wind turbine group into grid areas in the fluid response model and performing discretization processing; S144. Inputting the wind pressure variation parameters in the future time period into the fluid response model, and simulating the resistance response coefficient of each grid area of ​​the wind turbine group when the wind pressure is disturbed by the fluid response model; S145. Comprehensively evaluate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient during wind pressure disturbance, the observed resistance performance coefficient, and the weight coefficient of historical wind pressure characteristics.

[0009] Furthermore, based on the resistance characteristic coefficient of each fan, the main fan and auxiliary fan are dynamically selected from the fan group using multi-objective optimization technology. The initial frequency parameters of the main fan and auxiliary fan are configured according to the kiln pressure signal under the baseline working condition, including: S21. Analyze the output air volume of each fan based on the resistance characteristic coefficient of each fan, select a master fan and an auxiliary fan from the fan group based on the output air volume, and respectively establish parameter configuration combinations for the master fan and the auxiliary fan; S22, performing similarity cluster analysis on the parameter configuration combinations, dividing similar features in the parameter configuration combinations into the same group to obtain several cluster groups, scoring each cluster group respectively, and selecting the cluster group within a preset score range as the initial frequency parameter combination; S23. Use the improved firefly algorithm to select the optimal solution from the initial frequency parameter combination, and configure the initial frequency parameters for the main control fan and the auxiliary fan through the PID controller according to the kiln pressure signal under the benchmark working conditions.

[0010] Furthermore, the improved firefly algorithm is used to select the optimal solution from the initial frequency parameter combination, and according to the kiln pressure signal under the benchmark working condition, the initial frequency parameters of the main control fan and auxiliary fan are configured through the PID controller, including: S231, generating an initial position matrix of the firefly population based on the initial frequency parameter combination, and calculating an initial degree fitness value as the brightness of each firefly; S232, iteratively executing the optimization process for each firefly, comparing the fitness value of any firefly with that of the other fireflies, and calculating the moving step length of the current firefly based on the comparison result, and moving the current firefly toward a firefly with a better brightness according to the moving step length; S233. In each iteration, the Pareto front is selected based on the crowding distance to perform population update. The iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination. S234: Use the output initial frequency parameter combination as the initial operation setting of the main fan and the auxiliary fan, and configure the initial frequency parameters for the main fan and the auxiliary fan respectively through the PID controller.

[0011] Furthermore, in each iteration, the Pareto front is selected based on the crowding distance to update the population. The iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination, including: S2331. Perform fast non-dominated sorting based on the fitness of the current firefly and the remaining fireflies, and stratify the fireflies according to their dominance level. S2332. Calculate the firefly crowding index in the same allocation level, and based on the firefly crowding index, select the frontier solution combination from the current firefly population to generate a Pareto solution set, and use the Pareto optimal solution set as the input for the next round of iteration; S2333. Stop after reaching the iteration termination condition, and select the optimal solution from the Pareto solution set that is away from the ideal target point as the initial frequency parameter combination.

[0012] Furthermore, the kiln pressure signal is monitored in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main control fan and the auxiliary fan. If no abnormality is detected in the kiln pressure signal, the main control fan and the auxiliary fan are operated according to the initial frequency parameters, including: S31, collecting kiln signals in real time, and comparing the kiln signals with a preset kiln pressure safety range, and determining whether the kiln pressure signal is abnormal based on the comparison result; S32. When the kiln pressure signal is abnormal, the initial frequency parameters of the main control fan and the auxiliary fan are optimized by pre-building a conflict resolution mechanism to adapt to the abnormal state of the kiln pressure signal; S33. After the adjustment of the main control fan and the auxiliary fan is completed, the kiln pressure stability and the load distribution of the fan group are evaluated. If there is a fluctuation in the fan group load within the prediction period, the load fluctuation is fed back to step S21 to regenerate the initial frequency parameter combination.

[0013] Furthermore, the initial frequency parameters of the main and auxiliary fans are optimized by pre-building a conflict resolution mechanism, including: The initial frequency parameters of the main and auxiliary fans are used as neuron nodes, and a directed graph with directionality and weight association is constructed based on the neuron nodes. Evaluate the neuron node status based on predefined frequency constraints and current limits, and identify abnormal nodes; The directed graph is partitioned using a graph cut algorithm, with the goal of minimizing the graph cut cost, to isolate abnormal nodes from the directed graph, and update the edge weights in the directed graph based on the remaining neuron nodes; When there is an abnormality in the kiln pressure signal, the edge weights of the kiln pressure node and the neuron node are added to the updated directed graph, and the initial frequency parameters of the corresponding main control fan are adjusted first. The kiln pressure signal is re-detected at the preset time point. If the kiln pressure signal has not recovered, the initial frequency parameters of the auxiliary fan are adjusted in sequence.

[0014] Furthermore, the calculation formula of the firefly crowding index is: Where, T ( i j ) indicates the first j The firefly crowding index of the solution, M represents the number of objective functions, d i ( i j +1) indicates that in the objective function d i Better than i j The objective function value of d i ( i j -1) indicates that in the objective function d i Worse than i j The objective function value of Represents the objective function d i The maximum value on Represents the objective function d i The minimum value on e Represents a positive number.

[0015] According to another aspect of the present invention, a resistance adjustment system for a kiln treatment facility is provided, the system comprising: The resistance calculation module is used to obtain the historical operating parameters of the fan group, extract the wind pressure characteristics of each fan in the fan group and build a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each fan; The fan frequency configuration module is used to dynamically select the main fan and auxiliary fan from the fan group based on the resistance characteristic coefficient of each fan using multi-objective optimization technology, and configure the initial frequency parameters for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline working condition; The fan frequency optimization module is used to monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters will be adaptively adjusted to achieve coordinated optimization between the main fan and the auxiliary fan. If there is no abnormality in the kiln pressure signal, the main fan and the auxiliary fan will be operated according to the initial frequency parameters.

[0016] The beneficial effects of the present invention are: 1. The present invention obtains the historical operating parameters of the fan group and constructs a fluid resistance model, which can comprehensively identify the resistance characteristics of each fan, and dynamically configure the initial frequency parameters in combination with a multi-objective optimization method. It can reduce energy consumption and extend equipment life while ensuring system stability. Through real-time monitoring of the kiln pressure signal and adaptive frequency adjustment, efficient coordination between the main control and auxiliary fans is achieved, which can effectively alleviate the safety risks brought about by abnormal fluctuations in kiln pressure, thereby improving the overall safety, stability and economy of kiln operation.

[0017] 2. The present invention effectively realizes the coordinated optimization between the main fan and the auxiliary fan by real-time monitoring of the kiln pressure signal and dynamically adjusting the initial frequency parameters of the fan, thereby improving the response speed and stability of the fan system. By constructing a conflict resolution mechanism, the fan frequency can be adaptively adjusted in time when the kiln pressure signal is abnormal, thereby avoiding problems such as increased energy consumption and equipment failure caused by long-term high-hertz operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 is a flow chart of a resistance adjustment method for a kiln treatment facility according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a specific implementation of a resistance adjustment method for a kiln treatment facility according to an embodiment of the present invention; Figure 3 This is a principle block diagram of a resistance adjustment system for a kiln treatment facility according to an embodiment of the present invention.

[0020] In the picture: 1. Resistance calculation module; 2. Fan frequency configuration module; 3. Fan frequency optimization module. DETAILED DESCRIPTION

[0021] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and the advantages of the present invention.

[0022] According to an embodiment of the present invention, a resistance adjustment method and system for a kiln treatment facility are provided.

[0023] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1-Figure 2 As shown, according to the resistance adjustment method of the kiln treatment facility of an embodiment of the present invention, the method includes: S1. Obtain the historical operating parameters of the fan group, extract the wind pressure characteristics of each fan in the fan group and build a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each fan.

[0024] It should be noted that a fan system that meets the relevant parameters (fan-related parameters: flow rate, wind pressure, and power, among which flow rate is the flue gas flow rate of the kiln; wind pressure is the sum of the resistance of the waste heat boiler, denitrification, desulfurization, and dust removal systems; power is the installed power after meeting the wind pressure and flow rate) is installed at the outlet of each of the three systems of denitrification, desulfurization, and dust removal. This allows the flue gas to pass through its respective treatment and then enter the next system through its respective fans. After operation, the hertz of the three fans is basically maintained between 18 and 22 Hz (set to operate at the same frequency, i.e., the fans of the three facilities have the same frequency). The low hertz operation greatly protects the safe operation of the fans and inverters, especially in the hot summer, which greatly reduces the failure rate of the fans and greatly ensures the safety of kiln production.

[0025] Specifically, a fan group includes denitrification fans, desulfurization fans, and dust removal fans. When collecting historical operating data for the fan group, it is necessary to obtain the operating parameters of the denitrification fans, desulfurization fans, and dust removal fans, including key data indicators such as fan speed, frequency, current, power, wind pressure, temperature, and vibration. At the same time, it is also necessary to record information such as the operating time, load conditions, and operating status changes of each fan, and extract the wind pressure characteristics of each fan under different operating conditions, such as the wind pressure change trend, the relationship between wind pressure and other operating parameters, and the fluctuation characteristics of wind pressure under different load and frequency conditions.

[0026] The historical operating parameters of the fan group are obtained, the wind pressure characteristics of each fan in the fan group are extracted, and a fluid resistance model is constructed. The resistance characteristic coefficient of each fan is simulated and analyzed using the fluid resistance model, including: S11. Collect historical operating data of the wind turbine group and extract historical wind pressure characteristics from the historical operating data; S12. Mapping the historical wind pressure characteristics to a high-dimensional space based on a predefined nonlinear kernel function to obtain high-dimensional features; S13. Build a ridge regression prediction model, input the high-dimensional features into the ridge regression prediction model, and output the wind pressure change parameters of each wind turbine in the future time period.

[0027] It should be noted that the principle of the ridge regression prediction model is to introduce the L2 regularization term (i.e., the sum of the squares of the weight vector multiplied by the regularization coefficient λ) on the basis of standard linear regression. By penalizing excessively large weight coefficients, it alleviates the multicollinearity problem that is prone to occur when high-dimensional feature input is input, thereby improving the model's generalization ability. Specifically, it includes: The high-dimensional features such as wind turbine speed, current, and wind pressure in the historical operating data are standardized to construct a feature matrix and target variable (wind pressure change in the future period). The optimal weight vector is solved by minimizing the loss function. After the model training is completed, the real-time collected wind turbine group feature data is input to output the predicted wind pressure change value of each wind turbine in the future period. The prediction result will serve as the decision basis for subsequent frequency parameter adjustment.

[0028] S14. Construct a fluid resistance model based on the wind pressure variation parameters in the future time period, simulate and analyze the resistance response coefficient during wind pressure disturbance, and calculate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient.

[0029] Among them, the fluid resistance model is constructed based on the wind pressure change parameters in the future time period, the resistance response coefficient during wind pressure disturbance is simulated and analyzed, and the resistance characteristic coefficient of each wind turbine is calculated based on the resistance response coefficient, including: S141. Compare the wind pressure change parameter in the future time period with the historical wind pressure characteristic parameter to obtain the wind pressure change amplitude, and assign a weight to the historical wind pressure characteristic based on the wind pressure change amplitude; S142. Calculate the observed resistance performance coefficient of each wind turbine in the future time period based on the weight coefficient of the historical wind pressure characteristics; S143. Constructing a fluid response model based on a time evolution mechanism using finite element analysis technology, dividing the wind turbine group into grid areas in the fluid response model and performing discretization processing; S144. Inputting the wind pressure variation parameters in the future time period into the fluid response model, and simulating the resistance response coefficient of each grid area of ​​the wind turbine group when the wind pressure is disturbed by the fluid response model; S145. Comprehensively evaluate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient during wind pressure disturbance, the observed resistance performance coefficient, and the weight coefficient of historical wind pressure characteristics.

[0030] It should be noted that the present invention can evaluate the resistance response behavior of wind turbines in different time periods and under different operating conditions, thereby avoiding the errors caused by relying on fixed empirical parameters; by introducing comparative analysis of historical wind pressure characteristics and future change amplitudes, the model's sensitivity to sudden wind pressure disturbances and prediction accuracy can be improved; using finite element technology to discretely model the grid area of ​​the wind turbine group can finely characterize the local fluid behavior, which helps to improve the accuracy of resistance estimation.

[0031] S2. Based on the resistance characteristic coefficient of each fan, the main fan and auxiliary fan are dynamically selected from the fan group using multi-objective optimization technology, and the initial frequency parameters are configured for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline operating conditions.

[0032] Among them, based on the resistance characteristic coefficient of each fan, the multi-objective optimization technology is used to dynamically select the main fan and auxiliary fan from the fan group, and the initial frequency parameters of the main fan and auxiliary fan are configured according to the kiln pressure signal under the benchmark working conditions, including: S21. Analyze the output air volume of each fan based on the resistance characteristic coefficient of each fan, select a master fan and an auxiliary fan from the fan group based on the output air volume, and respectively establish parameter configuration combinations for the master fan and the auxiliary fan; S22, performing similarity cluster analysis on the parameter configuration combinations, dividing similar features in the parameter configuration combinations into the same group to obtain several cluster groups, scoring each cluster group respectively, and selecting the cluster group within a preset score range as the initial frequency parameter combination; S23. Use the improved firefly algorithm to select the optimal solution from the initial frequency parameter combination, and configure the initial frequency parameters for the main control fan and the auxiliary fan through the PID controller according to the kiln pressure signal under the benchmark working conditions.

[0033] Among them, the improved firefly algorithm is used to select the optimal solution from the initial frequency parameter combination, and according to the kiln pressure signal under the benchmark working condition, the initial frequency parameters of the main control fan and auxiliary fan are configured through the PID controller, including: S231, generating an initial position matrix of the firefly population based on the initial frequency parameter combination, and calculating an initial degree fitness value as the brightness of each firefly; S232, iteratively executing the optimization process for each firefly, comparing the fitness value of any firefly with that of the other fireflies, and calculating the moving step length of the current firefly based on the comparison result, and moving the current firefly toward a firefly with a better brightness according to the moving step length; S233. In each iteration, the Pareto front is selected according to the crowding distance to perform population update. The iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination.

[0034] In each iteration, the Pareto front is selected based on the crowding distance to update the population. The iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination, including: S2331. Perform fast non-dominated sorting based on the fitness of the current firefly and the remaining fireflies, and stratify the fireflies according to their dominance level. S2332. Calculate the firefly crowding index in the same allocation level, and based on the firefly crowding index, select the frontier solution combination from the current firefly population to generate a Pareto solution set, and use the Pareto optimal solution set as the input for the next round of iteration.

[0035] It should be noted that the improved firefly algorithm introduces non-dominated sorting and crowding distance evaluation mechanisms in the iterative process, which can more comprehensively explore the solution space and maintain population diversity, significantly improving the optimization efficiency and solution quality; by selecting the frequency parameter combination with the best distance to the target through the Pareto front, it can take into account multiple performance indicators at the same time, and is more suitable for multi-objective control needs under complex working conditions; combined with the kiln pressure reference signal, it is mapped to the initial frequency of the fan through the PID controller to ensure that the adjustment strategy has physical feasibility and response stability.

[0036] The calculation formula of the firefly crowding index is: Where, T ( i j ) indicates the first j The firefly crowding index of the solution, M represents the number of objective functions, d i ( i j +1) indicates that in the objective function d i Better than i j The objective function value of d i ( i j -1) indicates that in the objective function d i Worse than i j The objective function value of Represents the objective function d i The maximum value on Represents the objective function d i The minimum value on e Represents a positive number.

[0037] S2333. Stop after reaching the iteration termination condition, and select the optimal solution from the Pareto solution set that is away from the ideal target point as the initial frequency parameter combination.

[0038] S234: Use the output initial frequency parameter combination as the initial operation setting of the main fan and the auxiliary fan, and configure the initial frequency parameters for the main fan and the auxiliary fan respectively through the PID controller.

[0039] S3. Monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main control fan and the auxiliary fan. If no abnormality is detected in the kiln pressure signal, the main control fan and the auxiliary fan are operated according to the initial frequency parameters.

[0040] Among them, the kiln pressure signal is monitored in real time. If the kiln pressure signal is detected to be abnormal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main control fan and the auxiliary fan. If the kiln pressure signal is normal, the main control fan and the auxiliary fan are operated according to the initial frequency parameters, including: S31, collecting kiln signals in real time, and comparing the kiln signals with a preset kiln pressure safety range, and determining whether the kiln pressure signal is abnormal based on the comparison result; S32. When the kiln pressure signal is abnormal, the initial frequency parameters of the main control fan and the auxiliary fan are optimized by pre-building a conflict resolution mechanism to adapt to the abnormal state of the kiln pressure signal; S33. After the adjustment of the main control fan and the auxiliary fan is completed, the kiln pressure stability and the load distribution of the fan group are evaluated. If there is a fluctuation in the fan group load within the prediction period, the load fluctuation is fed back to step S21 to regenerate the initial frequency parameter combination.

[0041] The initial frequency parameters of the main and auxiliary fans are optimized by pre-building a conflict resolution mechanism, including: The initial frequency parameters of the main and auxiliary fans are used as neuron nodes, and a directed graph with directionality and weight association is constructed based on the neuron nodes. Evaluate the neuron node status based on predefined frequency constraints and current limits, and identify abnormal nodes; The directed graph is partitioned using a graph cut algorithm, with the goal of minimizing the graph cut cost, to isolate abnormal nodes from the directed graph, and update the edge weights in the directed graph based on the remaining neuron nodes; When there is an abnormality in the kiln pressure signal, the edge weights of the kiln pressure node and the neuron node are added to the updated directed graph, and the initial frequency parameters of the corresponding main control fan are adjusted first. The kiln pressure signal is re-detected at the preset time point. If the kiln pressure signal has not recovered, the initial frequency parameters of the auxiliary fan are adjusted in sequence.

[0042] It should be noted that the real-time feedback mechanism based on the kiln pressure signal constructs a directed graph of a neural network with directionality and weight to dynamically identify and isolate frequency abnormal nodes, and then combines the graph cutting algorithm to adaptively optimize the initial frequency parameters of the main control fan and the auxiliary fan, thereby achieving collaborative adjustment with clear division of labor.

[0043] In the conflict resolution mechanism, the master fan assumes primary control responsibility, prioritizing frequency adjustment when kiln pressure deviates to quickly restore it to a safe range. The auxiliary fans, in their follow-up and compensation roles, intervene sequentially if the kiln pressure remains unstable after the master fan's adjustments. These small frequency fluctuations provide auxiliary support and redundancy. When kiln pressure increases, the master fan responds first, and if the pressure is insufficient, the auxiliary fans gradually compensate. When kiln pressure decreases, the master fan prioritizes frequency reduction, followed by the auxiliary fans to prevent over-extraction. If a fan suddenly trips, the master and auxiliary fans synchronize and increase their frequencies to maintain kiln pressure stability and suppress overshoot. If multiple fans trip, the remaining fans automatically assume the master control role and continue system operation. Compared to traditional indiscriminate control methods, this not only improves the system's dynamic response speed and regulation accuracy, but also avoids chaotic frequency adjustment, reducing energy consumption and impact, and further enhancing the safety, reliability, and intelligent level of kiln operation.

[0044] The resistance adjustment method of the kiln treatment facility provided by the present invention will be further described below in conjunction with specific implementation methods.

[0045] For example, each device has its own operating resistance: boiler 1500 Pa, denitrification 600 Pa, desulfurization 200 Pa, dust removal 1200 Pa, kiln and flue resistance 1000 Pa, and the total resistance to be overcome is 4500 Pa.

[0046] If one induced draft fan is used for boiler, denitrification, desulfurization and dust removal, considering their respective resistance and long-term operation reliability, the fan parameters are: wind pressure 10,000 Pa, air volume 300,000 cubic meters, power 1,600 kilowatts, and the hertz during stable long-term operation is 45-48 Hz, with a daily power consumption of about 30,000-35,000 kilowatts.

[0047] The denitrification, desulfurization, and dust removal fans are designed for resistance at each stage based on the following parameters: air pressure of 6,000 Pa, air volume of 300,000 cubic meters, and power of 700 kW. Most of the time, the three fans run at 20-25 Hz. This ensures normal and stable operation even if one fails. If two fail, the remaining fan has a 48 Hz kiln pressure of approximately 10-15 Pa (a much better situation than the aforementioned kiln pressure of 50 Pa if one induced draft fan fails). The three fans have been operating at this low frequency for nearly three years without any problems. They run at 20-25 Hz most of the time, and consume a total of approximately 20,000-25,000 kW of electricity per day.

[0048] The three fans automatically operate according to the kiln pressure signal: When kiln pressure increases, the three fans, starting with the terminal fan, increase their speed by one hertz. For example, the terminal dust collector fan is increased by one hertz first. If the kiln pressure is not normal, the desulfurization fan is increased by one hertz, and finally the denitrification fan is increased. This cycle continues until the kiln pressure returns to normal. When kiln pressure decreases, the hertz is reduced in the opposite direction, starting from the first stage. During normal operation, fine adjustments are made, and most of the time, only 1-3 hertz adjustments are required.

[0049] In addition, the three sets of fans can affect each other. When one of the fans suddenly fails and stops, the remaining fans will automatically increase to the same Hz until the resistance caused by the failure is overcome (the three fan inverters rise and fall to track the kiln pressure signal of the kiln, and the kiln pressure and fan inverter Hz are interlocked. When one of the fans stops, the kiln pressure increases, and the remaining two fans increase their respective Hz according to the kiln pressure signal until the kiln pressure reaches the normal value (reaching the upper limit of 48 Hz, and the operating Hz rises to 48 Hz within 5 minutes). Similarly: if two fans stop, the remaining one will take over (it should be noted that when running three fan systems, the required parameter is that any of the three fans can operate normally at 45 Hz), further ensuring the safe production of the kiln.

[0050] As there are no faults in the fans during on-site operation, the test examples are as follows: when one fan is manually shut down, the kiln pressure is normal when the remaining two fans are automatically adjusted to 35-38 Hz; after two fans are shut down, the remaining fan is automatically increased to 48 Hz within 5 minutes, and the kiln pressure is 10-15 Pa, which does not affect normal production in a short time; the three sets of fans independently overcome the resistance of denitrification, desulfurization and dust removal, and operate at low Hz, ensuring the long-term and stable operation of the treatment facilities, thereby greatly reducing the failure rate of high-load operation of the fans and ensuring the safe production of the kiln.

[0051] According to another embodiment of the present invention, Figure 3 As shown, a resistance adjustment system for a kiln treatment facility is also provided, the system comprising: Resistance calculation module 1 is used to obtain the historical operating parameters of the fan group, extract the wind pressure characteristics of each fan in the fan group and build a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each fan; The fan frequency configuration module 2 is used to dynamically select the main fan and auxiliary fan from the fan group based on the resistance characteristic coefficient of each fan using multi-objective optimization technology, and configure the initial frequency parameters for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline working condition; The fan frequency optimization module 3 is used to monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main fan and the auxiliary fan. If no abnormality occurs in the kiln pressure signal, the main fan and the auxiliary fan are operated according to the initial frequency parameters.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adjusting resistance of a kiln treatment facility, characterized in that: The method includes: S1. Obtain historical operating parameters of the wind turbine group, extract the wind pressure characteristics of each wind turbine in the wind turbine group and construct a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each wind turbine; S2. Based on the resistance characteristic coefficient of each fan, a multi-objective optimization technique is used to dynamically select the main fan and auxiliary fan from the fan group, and initial frequency parameters are configured for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline operating condition; S3. Monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters are adaptively adjusted to achieve coordinated optimization between the main control fan and the auxiliary fan. If no abnormality is detected in the kiln pressure signal, the main control fan and the auxiliary fan are operated according to the initial frequency parameters.

2. The resistance adjustment method of a kiln treatment facility according to claim 1, characterized in that: The process of obtaining historical operating parameters of the fan group, extracting the wind pressure characteristics of each fan in the fan group, and constructing a fluid resistance model, and using the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each fan includes: S11. Collecting historical operation data of a fan group, and extracting historical wind pressure characteristics from the historical operation data, wherein the fan group includes a denitrification fan, a desulfurization fan, and a dust removal fan; S12. Mapping the historical wind pressure characteristics to a high-dimensional space based on a predefined nonlinear kernel function to obtain high-dimensional features; S13, constructing a ridge regression prediction model, inputting the high-dimensional features into the ridge regression prediction model, and outputting the wind pressure change parameters of each wind turbine in the future time period; S14. Construct a fluid resistance model based on the wind pressure variation parameters in the future time period, simulate and analyze the resistance response coefficient during wind pressure disturbance, and calculate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient.

3. The resistance adjustment method of a kiln treatment facility according to claim 2, characterized in that: The fluid resistance model is constructed based on the wind pressure variation parameters in the future time period, the resistance response coefficient during wind pressure disturbance is simulated and analyzed, and the resistance characteristic coefficient of each wind turbine is calculated based on the resistance response coefficient. S141. Compare the wind pressure change parameter in the future time period with the historical wind pressure characteristic parameter to obtain the wind pressure change amplitude, and assign a weight to the historical wind pressure characteristic based on the wind pressure change amplitude; S142. Calculate the observed resistance performance coefficient of each wind turbine in the future time period based on the weight coefficient of the historical wind pressure characteristics; S143. Construct a fluid response model based on a time evolution mechanism using finite element analysis technology, divide the wind turbine group into grid areas in the fluid response model, and perform discretization processing; S144. Inputting the wind pressure variation parameters in the future time period into the fluid response model, and simulating the resistance response coefficient of each grid area of ​​the wind turbine group when the wind pressure is disturbed by the fluid response model; S145. Comprehensively evaluate the resistance characteristic coefficient of each wind turbine based on the resistance response coefficient during wind pressure disturbance, the observed resistance performance coefficient, and the weight coefficient of historical wind pressure characteristics.

4. The resistance adjustment method of a kiln treatment facility according to claim 1, characterized in that: Based on the resistance characteristic coefficient of each fan, the main fan and the auxiliary fan are dynamically selected from the fan group using a multi-objective optimization technology, and the initial frequency parameters of the main fan and the auxiliary fan are configured according to the kiln pressure signal under the baseline working condition. The parameters include: S21. Analyze the output air volume of each fan based on the resistance characteristic coefficient of each fan, select a master fan and an auxiliary fan from the fan group based on the output air volume, and respectively establish parameter configuration combinations for the master fan and the auxiliary fan; S22, performing similarity cluster analysis on the parameter configuration combinations, dividing similar features in the parameter configuration combinations into the same group to obtain several cluster groups, scoring each cluster group respectively, and selecting the cluster group within a preset score range as the initial frequency parameter combination; S23. Use the improved firefly algorithm to select the optimal solution from the initial frequency parameter combination, and configure the initial frequency parameters for the main control fan and the auxiliary fan through the PID controller according to the kiln pressure signal under the benchmark working conditions.

5. The resistance adjustment method of a kiln treatment facility according to claim 4, characterized in that: The improved firefly algorithm is used to select the optimal solution from the initial frequency parameter combination, and the initial frequency parameters of the main control fan and the auxiliary fan are configured by the PID controller according to the kiln pressure signal under the reference working condition. S231, generating an initial position matrix of the firefly population based on the initial frequency parameter combination, and calculating an initial degree fitness value as the brightness of each firefly; S232, iteratively executing the optimization process for each firefly, comparing the fitness value of any firefly with that of the other fireflies, and calculating the moving step length of the current firefly based on the comparison result, and moving the current firefly toward a firefly with a better brightness according to the moving step length; S233. In each iteration, the Pareto front is selected based on the crowding distance to perform population update. The iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination. S234: Use the output initial frequency parameter combination as the initial operation setting of the main fan and the auxiliary fan, and configure the initial frequency parameters for the main fan and the auxiliary fan respectively through the PID controller.

6. The resistance adjustment method of a kiln treatment facility according to claim 5, characterized in that: In each iteration, the Pareto front is selected based on the crowding distance to perform population update, and the iteration stops when the iteration condition is met. The optimal solution of the distance target combination is selected from the Pareto front and used as the initial frequency parameter combination, including: S2331. Perform fast non-dominated sorting based on the fitness of the current firefly and the remaining fireflies, and stratify the fireflies according to their dominance level. S2332. Calculate the firefly crowding index in the same allocation level, and based on the firefly crowding index, select the frontier solution combination from the current firefly population to generate a Pareto solution set, and use the Pareto optimal solution set as the input for the next round of iteration; S2333. Stop after reaching the iteration termination condition, and select the optimal solution from the Pareto solution set that is away from the ideal target point as the initial frequency parameter combination.

7. The resistance adjustment method of a kiln treatment facility according to claim 1, characterized in that: The real-time monitoring of the kiln pressure signal and, if an abnormality is detected in the kiln pressure signal, the adaptive adjustment of the initial frequency parameters to achieve coordinated optimization between the main control fan and the auxiliary fan; if no abnormality is detected in the kiln pressure signal, the operation of the main control fan and the auxiliary fan according to the initial frequency parameters includes: S31, collecting kiln signals in real time, and comparing the kiln signals with a preset kiln pressure safety range, and determining whether the kiln pressure signal is abnormal based on the comparison result; S32. When the kiln pressure signal is abnormal, the initial frequency parameters of the main control fan and the auxiliary fan are optimized by pre-building a conflict resolution mechanism to adapt to the abnormal state of the kiln pressure signal; S33. After the adjustment of the main control fan and the auxiliary fan is completed, the kiln pressure stability and the load distribution of the fan group are evaluated. If there is a fluctuation in the fan group load within the prediction period, the load fluctuation is fed back to step S21 to regenerate the initial frequency parameter combination.

8. The resistance adjustment method of a kiln treatment facility according to claim 7, characterized in that: The optimization of the initial frequency parameters of the main control fan and the auxiliary fan by pre-building a conflict resolution mechanism includes: The initial frequency parameters of the main and auxiliary fans are used as neuron nodes, and a directed graph with directionality and weight association is constructed based on the neuron nodes. Evaluate the neuron node status based on predefined frequency constraints and current limits, and identify abnormal nodes; The directed graph is partitioned using a graph cut algorithm, with the goal of minimizing the graph cut cost, to isolate abnormal nodes from the directed graph, and update the edge weights in the directed graph based on the remaining neuron nodes; When there is an abnormality in the kiln pressure signal, the edge weights of the kiln pressure node and the neuron node are added to the updated directed graph, and the initial frequency parameters of the corresponding main control fan are adjusted first. The kiln pressure signal is re-detected at the preset time point. If the kiln pressure signal has not recovered, the initial frequency parameters of the auxiliary fan are adjusted in sequence.

9. The resistance adjustment method of a kiln treatment facility according to claim 5, characterized in that: The calculation formula of the firefly crowding index is: Where, T ( θ j ) indicates the first j The firefly crowding index of the solution, M represents the number of objective functions, δ i ( θ j +1) indicates that in the objective function δ i Better than θ j The objective function value of δ i ( θ j -1) indicates that in the objective function δ i Worse than θ j The objective function value of Represents the objective function δ i The maximum value on Represents the objective function δ i The minimum value on ε Represents a positive number.

10. A resistance adjustment system for a kiln treatment facility, used to implement the resistance adjustment method for a kiln treatment facility according to any one of claims 1 to 9, characterized in that: The system includes: The resistance calculation module is used to obtain the historical operating parameters of the fan group, extract the wind pressure characteristics of each fan in the fan group and build a fluid resistance model, and use the fluid resistance model to simulate and analyze the resistance characteristic coefficient of each fan; The fan frequency configuration module is used to dynamically select the main fan and auxiliary fan from the fan group based on the resistance characteristic coefficient of each fan using multi-objective optimization technology, and configure the initial frequency parameters for the main fan and auxiliary fan respectively according to the kiln pressure signal under the baseline working condition; The fan frequency optimization module is used to monitor the kiln pressure signal in real time. If an abnormality is detected in the kiln pressure signal, the initial frequency parameters will be adaptively adjusted to achieve coordinated optimization between the main fan and the auxiliary fan. If there is no abnormality in the kiln pressure signal, the main fan and the auxiliary fan will be operated according to the initial frequency parameters.

Images