Dioxin closed-loop control method based on real-time detection and dynamic optimization

Abstract

According to the dioxin closed-loop control method based on real-time detection and dynamic optimization, an online monitoring system is deployed, a tunable laser ionization time-of-flight mass spectrometer is arranged at an incineration flue gas discharge outlet to detect the concentration of trichlorobenzene serving as a dioxin indicator in real time, and a sensor is combined to detect working condition parameters; polynomial regression is used for real-time conversion of dioxin toxicity equivalent concentration, and an on-line monitoring-toxicity evaluation integrated system is formed. And a double-layer random forest-eagle mixed optimization model is constructed, the first layer is based on historical working condition data to train a random forest model to predict the dioxin emission trend and the potential standard exceeding risk, and the second layer is used for dynamically optimizing the incineration parameter combination through an improved eagle optimization algorithm and issuing the incineration parameter combination to an incineration system in real time. And meanwhile, a calcium-based retardant directional spraying device is deployed in a flue gas purification section, so that uniform covering and accurate dosage control of the retardant are realized, and the low-temperature resynthesis reaction of dioxin is inhibited. And finally, parameters of the double-layer optimization model are dynamically updated based on the error of an actual detection value and a predicted value, the adaptability of the model to working condition fluctuation is improved, therefore, dioxin emission is effectively reduced, and the environmental protection property and stability of the incineration process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and pollution control, and particularly to a dioxin closed-loop control method based on real-time detection and dynamic optimization. Background Art

[0002] In today's society, with the acceleration of the industrialization process and the continuous expansion of the urbanization scale, the problem of waste treatment has become increasingly prominent. Waste incineration, as a common and effective waste treatment method, is widely used. However, dioxin, a highly toxic substance, will inevitably be produced during the waste incineration process. Dioxin has extremely strong toxicity and poses a serious threat to human health and the ecological environment. For example, it may cause serious consequences such as human endocrine disorders, carcinogenesis, and teratogenesis. Therefore, it is particularly important to effectively monitor and control the dioxin produced during the waste incineration process.

[0003] In terms of dioxin emission control, there are also many deficiencies in the existing control technologies. Some waste incineration plants only rely on experience or simple fixed parameter settings to adjust the incineration process, such as furnace temperature, oxygen content, and activated carbon injection volume, lacking scientific and accurate control strategies. This extensive control method is difficult to adapt to the actual situation of complex and variable waste components and constantly fluctuating incineration conditions, resulting in unstable dioxin emissions and making it difficult to continuously meet the increasingly strict environmental protection standards. At the same time, the means to inhibit the low-temperature resynthesis reaction of dioxin are relatively single and inefficient, and cannot effectively reduce the secondary generation of dioxin during the flue gas purification process.

[0004] In addition, the existing monitoring and control processes are often independent of each other and do not form an organic whole. Monitoring data cannot be timely fed back to guide the adjustment of control strategies, and the implementation effects of control measures cannot be accurately evaluated and optimized through real-time monitoring, making the entire dioxin emission control system lack systematicness and coordination, and it is difficult to achieve the goal of efficient and accurate dioxin emission reduction.

[0005] In summary, the existing technologies have many disadvantages in dioxin monitoring and control, such as poor real-time performance, high cost, inaccurate regulation, and insufficient system coordination. There is an urgent need for a more advanced, efficient, and accurate dioxin emission monitoring and control technology to solve these problems. Summary of the Invention

[0006] To solve the above problems, the present invention proposes the following specific steps for the dioxin closed-loop control method based on real-time detection and dynamic optimization, which is characterized in that:

[0007] Step 1: Deploy an online monitoring system, configure a tunable laser ionization time-of-flight mass spectrometer at the incineration flue gas emission port to detect the concentration of trichlorobenzene, an indicator of dioxins, in real time; detect operating parameters such as the activated carbon injection volume, furnace temperature, and oxygen content through sensors, and combine polynomial regression to convert the dioxin toxicity equivalent concentration in real time, forming an integrated online monitoring-toxicity assessment system;

[0008] Step 2: Construct a double-layer random forest-Aquila hybrid optimization model. In the first layer, train a random forest model based on historical operating data to predict the dioxin emission trend and potential over-standard risk; in the second layer, dynamically optimize the incineration parameter combination through an improved Aquila optimization algorithm, and send the optimization instructions to the incineration system in real time through the industrial Internet of Things;

[0009] Step 3: Deploy a calcium-based blocker directional spraying device in the flue gas purification section, which adopts a worm gear drive multi-directional rotary spraying structure to achieve uniform coverage of the blocker in the flue gas channel and precise control of the dosage; adjust the spraying dosage according to the threshold incineration system redundancy correction control to inhibit the low-temperature re-synthesis reaction of dioxins;

[0010] Step 4: Calculate the error between the actual dioxin emission concentration detection value and the random forest prediction value, dynamically update the parameters of the double-layer optimization model based on the error data, and iterate the incremental data to the random forest to improve the adaptability of the model to operating condition fluctuations.

[0011] The closed-loop control method for dioxins based on real-time detection and dynamic optimization of the present invention has the following beneficial effects:

[0012] 1. By constructing a double-layer random forest-Aquila hybrid optimization model, the first-layer random forest model is trained based on historical operating data, which can accurately predict the dioxin emission trend and potential over-standard risk and issue early warnings. The improved Aquila optimization algorithm in the second layer can dynamically optimize the incineration parameter combination. This combination of accurate prediction and optimization control can timely adjust the incineration parameters according to the actual operating conditions, effectively avoid the over-standard emission of dioxins, improve the stability of the incineration process and energy utilization efficiency at the same time, and reduce energy waste and excessive generation of dioxins caused by unreasonable parameters.

[0013] 2. The present invention deploys a calcium-based blocker directional spraying device with a worm gear drive multi-directional rotary spraying structure in the flue gas purification section, which can achieve uniform coverage of the blocker in the flue gas channel. And accurately control the spraying dosage according to the calculation result of the dioxin emission concentration. When the dioxin emission concentration exceeds a certain threshold, trigger the redundancy correction control to adjust the spraying dosage. This not only improves the contact efficiency between the blocker and the flue gas, but also can accurately inhibit the low-temperature re-synthesis reaction of dioxins, reduce the secondary generation amount of dioxins, and further reduce the dioxin emission. Description of the Drawings

[0014] Figure 1 This is the flowchart of the present invention;

[0015] Figure 2 This is the structural diagram of Step 1 of the present invention. Detailed implementation manners

[0016] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners:

[0017] The present invention proposes a dioxin closed-loop control method based on real-time detection and dynamic optimization. Aiming at the problem of dioxin emissions from waste incineration, a system integrating on-line monitoring and predictive control is constructed. The dioxin concentration is converted by real-time detection of trichlorobenzene and operating conditions parameters, the incineration parameters are optimized by using a double-layer model, a directional spraying device is deployed to inhibit re-synthesis, and the model is updated based on errors, effectively reducing dioxin emissions. The flowchart of the invention is as Figure 1 shown. The steps of the present invention will be introduced in detail below.

[0018] Step 1: Deploy an on-line monitoring system. Configure a tunable laser ionization time-of-flight mass spectrometer at the incineration flue gas emission port to detect the concentration of trichlorobenzene, an indicator of dioxin, in real time; detect the operating conditions parameters such as the activated carbon injection amount, furnace temperature, and oxygen content through sensors, and combine polynomial regression to convert the dioxin toxicity equivalent concentration in real time to form an integrated on-line monitoring-toxicity assessment system. The structural diagram of Step 1 is as Figure 2 shown.

[0019] Step 1.1, Configure a tunable laser ionization time-of-flight mass spectrometer to detect the trichlorobenzene concentration:

[0020] Install a tunable laser ionization time-of-flight mass spectrometer at the incineration flue gas emission port. Use the tunable laser to ionize the molecules in the flue gas, and analyze the ions through time-of-flight mass spectrometry technology to determine the molecular species and concentration.

[0021] When detecting trichlorobenzene, an indicator of dioxin, by adjusting the wavelength of the laser of the mass spectrometer to match the absorption peak of the trichlorobenzene molecule, selective ionization and detection of trichlorobenzene are achieved, and the trichlorobenzene concentration value is output in real time.

[0022] Step 1.2, Integrate the operating conditions parameter sensor network:

[0023] Install a sensor network for collecting the operating conditions parameters of the activated carbon injection amount, furnace temperature, and oxygen content.

[0024] Use a flowmeter to detect the activated carbon injection amount, a temperature sensor to measure the flue gas temperature, and an oxygen content sensor to measure the oxygen content. The sensors transmit the real-time measured data to the incineration system through a data acquisition module to ensure that the monitoring data comprehensively covers the key emission indicators.

[0025] Step 1.3, convert the dioxin toxicity equivalent concentration by combining the polynomial regression model:

[0026] Establish a non-linear function based on the data of historical dioxin toxicity equivalent concentration and the concentration of the indicator trichlorobenzene. By fitting the historical data set, use polynomial regression to establish a non-linear function under the non-linear mapping relationship.

[0027]

[0028] Among them, TEQ is the dioxin toxicity equivalent concentration, and a k is the k-th order coefficient of the polynomial regression model, m is the order, and C TCB is the real-time detected concentration of trichlorobenzene.

[0029] Step 1.4, form an online monitoring - toxicity assessment system:

[0030] The system collects the trichlorobenzene concentration, operating conditions parameters, and dioxin toxicity equivalent concentration information in real time. At the same time, the system alarms for abnormal situations according to the set dioxin toxicity equivalent concentration threshold, forming an integrated online monitoring - toxicity assessment system, providing basic data and assessment basis for subsequent dioxin emission control.

[0031] Step 2: Construct a double-layer random forest - Tianying hybrid optimization model. In the first layer, train a random forest model based on historical operating conditions data to predict the dioxin emission trend and potential over-standard risk; in the second layer, dynamically optimize the incineration parameter combination by improving the Tianying optimization algorithm, and send the optimization instructions to the incineration system in real time through the industrial Internet of Things.

[0032] Step 2.1, the first layer: Train a random forest model based on historical operating conditions data

[0033] Collect historical operating conditions data, including furnace temperature, oxygen content, activated carbon injection amount. At the same time, record the corresponding dioxin emission concentration, and form a data set D with this historical data:

[0034] D = {(T fuenace1 , O 21 , M AC1 , C Dioxin1 ), (T fuenace2 , O 22 , M AC2 , C Dioxin2 ),..., (T fuenacen , O 2n , M ACn , C Dioxinn )}

[0035] Among them, T fuenacen is the furnace temperature of the nth sample, O 2n is the oxygen content of the nth sample, MACn is the activated carbon injection amount for the nth sample, C Dioxinn is the dioxin emission concentration for the nth sample.

[0036] Construct a random forest model: A random forest consists of multiple decision trees. For each decision tree, m samples are randomly drawn with replacement from the dataset D to form a new training subset D i . At the same time, 2 features are randomly selected from all features for splitting at the nodes of the decision tree.

[0037] At the nodes of each decision tree, based on the selected 2 features, the Gini coefficient is used to select the optimal feature for splitting, and the samples are divided into different child nodes until the number of samples in the node is less than the threshold or the depth of the tree reaches the upper limit. Repeat this process to construct multiple decision trees T1, T2,..., T N , where N is the number of decision trees in the random forest.

[0038] For the new operating condition data (T fuenace_new , O 2_new , M AC_new ), input it into each decision tree T i to obtain the predicted dioxin emission concentration C Dioxin_predi , and then integrate the predictions to obtain the final predicted value:

[0039]

[0040] where C Dioxin_predi is the predicted value of dioxin concentration for the ith decision tree, and the final predicted value C Dioxin_pred of the random forest model. According to the set dioxin emission standard threshold C threshold , if the following formula is satisfied:

[0041] C Dioxin_predi ≥C threshold

[0042] then it is judged that there is a potential risk of exceeding the standard, an over-standard warning is triggered, and the second-layer optimization is started.

[0043] Step 2.2, Second layer: Dynamically optimize the incineration parameter combination by improving the Tianying optimization algorithm

[0044] Introduce the Levy flight perturbation mechanism into the basic Tianying optimization algorithm to enhance the global search ability of the algorithm. Let the position of the Tianying individual be X = [x1, x2,..., x d , where d = 3 is the parameter dimension. In each iteration, perform Levy flight perturbation on the position of the Tianying individual:

[0045]

[0046] Among them, X new is the new position vector after Levy flight perturbation, α is a scaling factor, denotes element-wise multiplication, Levy(β) is a random number vector following the Levy distribution, and β is the parameter of the Levy distribution. To balance the global search and local search capabilities of the algorithm, an adaptive inertia weight w is introduced. w changes dynamically with the increase of the iteration number t:

[0047]

[0048] Among them, w max and w min are the maximum and minimum values of the inertia weight respectively, t is the current iteration number, and T max is the maximum iteration number.

[0049] The objective function F(X) is set as:

[0050] F(X) = w1 × C Dioxin (X) + w2 × E(X)

[0051] Among them, C Dioxin (X) is the dioxin emission concentration under the parameter combination X, E(X) is the energy consumption, and w1 and w2 are weight coefficients. By continuously iterating and updating the positions of the eagle individuals through the improved eagle optimization algorithm, the parameter combination that minimizes the objective function F(X) is searched.

[0052] The constraint conditions are set as: the furnace temperature control accuracy is within 5°C, the oxygen content adjustment response time is less than 30 seconds, and the activated carbon injection amount matches the flue gas load.

[0053] Finally, the eagle population is initialized, 50 groups of incineration parameter combinations are randomly generated, F(X) of each group is calculated, the optimal individual is selected to update the position, and iteration is performed until convergence to obtain the combinations of furnace temperature, oxygen content, and activated carbon injection amount for precise control to reduce the dioxin concentration.

[0054] Step 2.3, the optimization instructions are sent to the incineration system in real time through the industrial Internet of Things:

[0055] The optimal incineration parameter combination obtained through the double-layer random forest - eagle hybrid optimization model is sent to the incineration system in real time through the industrial Internet of Things, and the incineration system adjusts the parameters of furnace temperature, oxygen content, and activated carbon injection amount in real time, so as to achieve the purpose of optimizing dioxin emissions and the incineration process.

[0056] Step 3: Deploy a calcium-based blocker directional spraying device in the flue gas purification section, which adopts a worm gear drive multi-directional rotary spraying structure to achieve uniform coverage of the blocker in the flue gas channel and precise dosage control; adjust the spraying dosage according to the threshold incineration system redundancy correction control to inhibit the low-temperature resynthesis reaction of dioxins.

[0057] Step 3.1, Deploy a calcium-based blocker directional spraying device

[0058] Install a specially designed calcium-based blocker directional spraying device in the flue gas purification section. This device adopts a worm gear drive multi-directional rotary spraying structure to ensure uniform coverage of the blocker in the flue gas channel, thus ensuring sufficient contact between the blocker and the flue gas. The worm gear drives a multi-directional nozzle to achieve full coverage of the cross-section of the flue gas channel by adjusting the rotation angle from 0° to 360°. At the same time, dynamically adjust the spraying pressure according to the flue gas flow rate to ensure the particle size of the blocker atomized particles and improve the contact efficiency with the flue gas.

[0059] Step 3.2, Precise dosage control model

[0060] If the dioxin emission concentration calculated in Step 1 exceeds 1.5 times the threshold, the incineration system triggers the redundancy correction control of the blocker spraying dosage, and the blocker spraying dosage calculation formula is corrected as:

[0061]

[0062] Among them, D is the blocker spraying dosage, k is the dosage coefficient, C Dioxin_pred is the predicted value of the dioxin emission concentration of the random forest in Step 2, M ac is the activated carbon injection amount output by the Tianying hybrid optimization model in Step 2, T fuenace is the current furnace temperature, T opt is the optimal furnace temperature setting value output by the Tianying hybrid optimization model in Step 2, Q is the oxygen content output by the Tianying hybrid optimization model in Step 2, and γ is the temperature correction coefficient.

[0063] Step 4: Calculate the error between the actual dioxin emission concentration detection value and the random forest prediction value, dynamically update the parameters of the double-layer optimization model based on the error data, and iterate the incremental data to the random forest to improve the adaptability of the model to working condition fluctuations.

[0064] Calculate the closed-loop control error rate E:

[0065]

[0066] Among them, TED actual is the actually detected dioxin toxicity equivalent concentration, TED predictedIt is the dioxin toxicity equivalent concentration predicted by the model. This formula is used to measure the deviation between the predicted value of the model and the actual detected value. When the calculated E value exceeds 8%, it indicates that the parameters of the random forest model need to be adjusted. The newly added operating condition data and historical data are merged in a 7:3 ratio to generate an incremental data set, and the random forest parameters are adjusted.

[0067] As described above, it is only a preferred embodiment of the present invention, and it is not a limitation of the present invention in any other form. Any modification or equivalent change made according to the technical essence of the present invention still belongs to the scope protected by the present invention.

Claims

1. A closed-loop control method for dioxins based on real-time detection and dynamic optimization, the specific steps are as follows, and it is characterized in that: Step 1: Deploy an online monitoring system, configure a tunable laser ionization time-of-flight mass spectrometer at the incineration flue gas emission port to detect the concentration of trichlorobenzene, an indicator of dioxins, in real time; detect operating condition parameters such as the activated carbon injection amount, furnace temperature, and oxygen content through sensors, and combine polynomial regression to convert the dioxin toxicity equivalent concentration in real time to form an integrated online monitoring-toxicity assessment system; Step 2: Construct a double-layer random forest-Aquila hybrid optimization model. In the first layer, train a random forest model based on historical operating condition data to predict the dioxin emission trend and potential over-standard risk; in the second layer, dynamically optimize the incineration parameter combination through an improved Aquila optimization algorithm, and send the optimization instructions to the incineration system in real time through the industrial Internet of Things; Step 3: Deploy a calcium-based blocker directional spraying device in the flue gas purification section, adopt a worm gear drive multi-directional rotary spraying structure to achieve uniform coverage of the blocker in the flue gas channel and precise dosage control; adjust the spraying dosage according to the threshold incineration system redundancy correction control to inhibit the low-temperature re-synthesis reaction of dioxins; Step 4: Calculate the error between the actual dioxin emission concentration detection value and the random forest prediction value, dynamically update the parameters of the double-layer optimization model based on the error data, and iterate the incremental data to the random forest to improve the adaptability of the model to operating condition fluctuations.

2. The dioxin closed-loop control method based on real-time detection and dynamic optimization according to claim 1, wherein: The deployment of the online monitoring system in Step 1 is expressed as: Step 1.1, configure a tunable laser ionization time-of-flight mass spectrometer to detect the trichlorobenzene concentration: Install a tunable laser ionization time-of-flight mass spectrometer at the incineration flue gas emission port, ionize the molecules in the flue gas with a tunable laser, analyze the ions through time-of-flight mass spectrometry technology to determine the molecular species and concentration; When detecting trichlorobenzene, an indicator of dioxins, by adjusting the wavelength of the mass spectrometer laser, make the mass spectrometer match the absorption peak of trichlorobenzene molecules to achieve selective ionization and detection of trichlorobenzene, and output the trichlorobenzene concentration value in real time; Step 1.2, integrate the operating condition parameter sensor network: Install a sensor network for collecting operating condition parameters such as the activated carbon injection amount, furnace temperature, and oxygen content; Use a flowmeter to detect the activated carbon injection amount, a temperature sensor to measure the flue gas temperature, and an oxygen content sensor to measure the oxygen content; the sensors transmit the real-time measured data to the incineration system through a data acquisition module to ensure that the monitoring data comprehensively covers key emission indicators; Step 1.3, convert the dioxin toxicity equivalent concentration by combining a polynomial regression model: Establish a non-linear function through the data of historical dioxin toxicity equivalent concentration and the indicator trichlorobenzene concentration, fit the historical data set, and use polynomial regression to establish a non-linear function under a non-linear mapping relationship; where TEQ is the concentration of dioxin toxicity equivalent, and a k is the coefficient of the k-th order of the polynomial regression model, m is the order, C TCB is the real-time detected concentration of trichlorobenzene; Step 1.4, form an online monitoring-toxicity assessment system: The system collects the trichlorobenzene concentration, operating condition parameters, and dioxin toxicity equivalent concentration information in real time. At the same time, the system alarms abnormal situations according to the set dioxin toxicity equivalent concentration threshold to form an integrated online monitoring-toxicity assessment system, providing basic data and assessment basis for subsequent dioxin emission control.

3. The dioxin closed-loop control method based on real-time detection and dynamic optimization according to claim 1, characterized in that: The construction of the double-layer random forest-Aquila hybrid optimization model in Step 2 is expressed as follows: Step 2.1, First layer: Train a random forest model based on historical operating conditions data Collect historical operating conditions data, including furnace temperature, oxygen content, and activated carbon injection volume; at the same time, record the corresponding dioxin emission concentration, and form the historical data into a data set D: D = {(T fuenace1 , O 21 , M AC1 , C Dioxin1 ), (T fuenace2 , O 22 , M AC2 , C Dioxin2 ),..., (T fuenacen , O 2n , M ACn , C Dioxinn )} Among them, T fuenacen is the furnace temperature of the nth sample, O 2n is the oxygen content of the nth sample, M ACn is the activated carbon injection amount of the nth sample, C Dioxinn is the dioxin emission concentration of the nth sample; Construct a random forest model: A random forest consists of multiple decision trees. For each decision tree, randomly sample m samples from the dataset D with replacement to form a new training subset D i ; at the same time, randomly select 2 features from all features to be used for splitting at the nodes of the decision tree; At each node of each decision tree, based on the selected two features, the Gini coefficient is used to select the optimal feature for splitting, and the samples are divided into different child nodes until the number of samples in the node is less than the threshold or the depth of the tree reaches the upper limit; repeat this process to construct multiple decision trees T1, T2,..., T N , where N is the number of decision trees in the random forest; For the new operating condition data (T fuenace_new , O 2_new , M AC_new ), input it into each decision tree T i to obtain the predicted dioxin emission concentration C Dioxin_predi , and then integrate the predictions to obtain the final predicted value: Among them, C Dioxin_predi is the predicted value of dioxin concentration of the i-th decision tree, and the final predicted value C Dioxin_pred of the random forest model. According to the set dioxin emission standard threshold C threshold , if the following formula is satisfied: C Dioxin_predi ≥ C threshold Then it is judged that there is a potential risk of exceeding the standard, an over-standard warning is triggered, and the second-layer optimization is started; Step 2.2, Second layer: Dynamically optimize the incineration parameter combination through an improved Tianying optimization algorithm Introduce the Levy flight perturbation mechanism into the basic Tianying optimization algorithm to enhance the global search ability of the algorithm; let the position of the Tianying individual be X = [x1, x2,..., x d , where d = 3 is the parameter dimension; in each iteration, perform Levy flight perturbation on the position of the Tianying individual: Among them, X new is the new position vector after being perturbed by Levy flight. α is a scaling factor, denotes element-wise multiplication. Levy(β) is a random number vector following the Levy distribution, and β is the parameter of the Levy distribution. To balance the global search and local search capabilities of the algorithm, an adaptive inertia weight w is introduced. w changes dynamically with the increase of the iteration number t: where w max and w min are the maximum and minimum values of the inertia weight respectively, t is the current iteration number, and T max is the maximum iteration number; The objective function F(X) is set as: F(X) = w1 × C Dioxin (X) + w2 × E(X) Among them, C Dioxin (X) is the dioxin emission concentration under the parameter combination X, E(X) is the energy consumption, and w1 and w2 are weight coefficients; the positions of the eagle individuals are continuously iteratively updated through the improved eagle optimization algorithm to find the parameter combination that minimizes the objective function F(X); The constraint conditions are set as: the furnace temperature control accuracy is within 5°C, the oxygen content adjustment response time is less than 30 seconds, and the activated carbon injection volume matches the flue gas load; Finally, initialize the Tianying population, randomly generate 50 groups of incineration parameter combinations, calculate F(X) for each group, select the optimal individual to update the position, and iterate until convergence to obtain a combination of furnace temperature, oxygen content, and activated carbon injection volume for precise control to reduce the dioxin concentration; Step 2.3, The optimization instructions are sent to the incineration system in real time through the industrial Internet of Things: The optimal incineration parameter combination obtained through the double-layer random forest-Tianying hybrid optimization model is sent to the incineration system in real time through the industrial Internet of Things, and the incineration system adjusts the furnace temperature, oxygen content, and activated carbon injection volume parameters in real time, so as to achieve the purpose of optimizing dioxin emissions and the incineration process.

4. The dioxin closed-loop control method based on real-time detection and dynamic optimization according to claim 1, characterized in that: The redundant correction control for adjusting the spray dosage in Step 3 is expressed as follows: Step 3.1, Deploy a calcium-based blocker directional spraying device Install a specially designed calcium-based blocker directional spraying device in the flue gas purification section. The device adopts a worm gear drive multi-directional rotary spraying structure to ensure uniform coverage of the blocker in the flue gas channel, so as to ensure full contact between the blocker and the flue gas; the worm gear drives a multi-directional nozzle, and the cross-section of the flue gas channel is fully covered by adjusting the rotation angle from 0° to 360°. At the same time, the spraying pressure is dynamically adjusted according to the flue gas flow rate to ensure the particle size of the blocker atomization particles and improve the contact efficiency with the flue gas; Step 3.2, Precise dosage control model If the dioxin emission concentration calculated in Step 1 exceeds 1.5 times the threshold, the incineration system triggers the redundant correction control of the blocker spray dosage, and the blocker spray dosage calculation formula is corrected as: Among them, D is the blocker spraying dose, k is the dose coefficient, C Dioxin_pred is the predicted value of dioxin emission concentration of the random forest in step 2, M ac is the activated carbon injection amount output by the Tianying hybrid optimization model in step 2, T fuenace is the current furnace temperature, T opt is the optimal furnace temperature setting value output by the Tianying hybrid optimization model in step 2, Q is the oxygen content output by the Tianying hybrid optimization model in step 2, and γ is the temperature correction coefficient.

5. The dioxin closed-loop control method based on real-time detection and dynamic optimization according to claim 1, wherein: The update of the double-layer optimization model parameters in Step 4 is expressed as follows: Calculate the closed-loop control error rate E: Among them, TED actual is the actually detected dioxin toxicity equivalent concentration, and TED predicted is the dioxin toxicity equivalent concentration predicted by the model. This formula is used to measure the deviation degree between the model prediction value and the actually detected value. When the calculated E value exceeds 8%, it indicates that the parameters of the random forest model need to be adjusted; the newly added operating condition data and historical data are merged in a ratio of 7:3 to generate an incremental data set, and the random forest parameters are adjusted.

Images