Biomass boiler denitrification control system based on low temperature condition monitoring

Abstract

The present invention discloses a biomass boiler denitrification control system based on low temperature working condition monitoring, which relates to the field of denitrification control technology, including: obtaining flue gas monitoring data of biomass outlet flue gas; performing dust removal treatment on biomass outlet flue gas according to a multi-level dust removal decision model and a bag filter to obtain dust-removed flue gas; making control decisions on a desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust removal flue gas, and establishing a desulfurization and denitrification control decision space; performing multi-level optimization on the desulfurization and denitrification control decision space according to a desulfurization and denitrification multi-feature prediction model and a desulfurization and denitrification multi-feature optimization factor, and determining a desulfurization and denitrification control strategy; performing desulfurization and denitrification treatment on the dust removal flue gas according to the desulfurization and denitrification control strategy and a desulfurization and denitrification integrated reactor. The present invention solves the technical problems of low denitrification efficiency, high energy consumption, and high risk of secondary pollutant generation in the prior art, and achieves the technical effect of ensuring that flue gas meets emission standards.

Description

Technical Field

[0001] The present invention relates to the technical field of denitration control, and in particular to a biomass boiler denitration control system based on low-temperature operating condition monitoring. Background Art

[0002] In the application of biomass boilers, with the increasingly stringent environmental protection requirements and the widespread application of biomass energy, sulfur dioxide and nitrogen oxides contained in boiler flue gas are still the main pollutants. If these pollutants are not effectively removed, they will have a serious impact on the environment and human health. Most of the existing denitrification technologies rely on traditional selective catalytic reduction technology. However, under low temperature conditions, the denitrification efficiency is low, and additional energy is required to maintain the reaction conditions, resulting in excessive energy consumption. In addition, the secondary pollutants that may be generated during the denitrification process aggravate environmental pollution, and the dust in the flue gas often affects the activity of the catalyst and reduces the denitrification effect. Summary of the invention

[0003] The present application provides a biomass boiler denitrification control system based on low-temperature operating condition monitoring, which is used to solve the technical problems of low denitrification efficiency, high energy consumption and high risk of secondary pollutant generation in the prior art.

[0004] In view of the above problems, the present application provides a biomass boiler denitrification control system based on low temperature condition monitoring.

[0005] The present application provides a biomass boiler denitrification control system based on low temperature condition monitoring, the system comprising:

[0006] A low-temperature operating condition monitoring module, which is used to monitor the biomass boiler in real time and obtain the flue gas monitoring data of the biomass outlet flue gas; a dust removal processing module, which is used to perform dust removal processing on the biomass outlet flue gas based on the flue gas monitoring data, according to a multi-level dust removal decision model and a bag filter, to obtain dust-removed flue gas; a desulfurization and denitrification control decision module, which is used to make control decisions on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust-removed flue gas, and establish a desulfurization and denitrification control decision space; a desulfurization and denitrification optimization module, which is used to perform multi-level optimization on the desulfurization and denitrification control decision space according to a desulfurization and denitrification multi-feature prediction model and a desulfurization and denitrification multi-feature optimization factor, and determine a desulfurization and denitrification control strategy; a desulfurization and denitrification processing module, which is used to perform desulfurization and denitrification processing on the dust-removed flue gas according to the desulfurization and denitrification control strategy and the desulfurization and denitrification integrated reactor.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The present application monitors the biomass boiler in real time to obtain the flue gas monitoring data of the biomass outlet flue gas; based on the flue gas monitoring data, the biomass outlet flue gas is subjected to dust removal treatment according to a multi-level dust removal decision model and a bag filter to obtain dust-removed flue gas; control decisions are made on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust-removed flue gas, and a desulfurization and denitrification control decision space is established; multi-level optimization is performed on the desulfurization and denitrification control decision space according to a desulfurization and denitrification multi-feature prediction model and a desulfurization and denitrification multi-feature optimization factor to determine a desulfurization and denitrification control strategy; the dust-removed flue gas is subjected to desulfurization and denitrification treatment according to the desulfurization and denitrification control strategy and the desulfurization and denitrification integrated reactor. The present invention solves the technical problems of low denitrification efficiency, high energy consumption and high risk of secondary pollutant generation in the prior art. The flue gas data is monitored in real time by a low-temperature operating condition monitoring module, and a multi-level dust removal decision model and a bag dust collector are combined to perform flue gas dust removal treatment to obtain dust-removed flue gas. The control strategy is optimized according to the dust-removed flue gas data, and the decision space is optimized through a multi-feature prediction model to determine the optimal desulfurization and denitrification strategy. Finally, the flue gas is efficiently desulfurized and denitrified according to the optimization strategy, thereby achieving the technical effect of ensuring that the flue gas meets the emission standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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 creative work.

[0010] Figure 1 A schematic diagram of the structure of a biomass boiler denitrification control system based on low temperature condition monitoring provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a desulfurization and denitrification control decision space flow chart for establishing a desulfurization and denitrification control system based on low-temperature condition monitoring in a biomass boiler denitrification control system provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: low temperature condition monitoring module 11 , dust removal processing module 12 , desulfurization and denitrification control decision module 13 , desulfurization and denitrification optimization module 14 , desulfurization and denitrification processing module 15 . DETAILED DESCRIPTION

[0013] The present application provides a biomass boiler denitrification control system based on low-temperature operating condition monitoring, aiming to solve the technical problems of low denitrification efficiency, high energy consumption and high risk of secondary pollutant generation in the prior art. The flue gas data is monitored in real time through the low-temperature operating condition monitoring module, and the flue gas dust removal treatment is carried out in combination with a multi-level dust removal decision model and a bag dust collector to obtain dust-removed flue gas. The control strategy is optimized according to the dust-removed flue gas data, and the decision space is optimized through a multi-feature prediction model to determine the optimal desulfurization and denitrification strategy. Finally, the flue gas is efficiently desulfurized and denitrified according to the optimized strategy, thereby achieving the technical effect of ensuring that the flue gas meets the emission standards.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides a biomass boiler denitrification control system based on low temperature condition monitoring, the system comprising:

[0017] The low temperature operating condition monitoring module 11 is used to monitor the biomass boiler in real time and obtain the flue gas monitoring data of the biomass outlet flue gas.

[0018] In the embodiment of the present application, the low temperature condition monitoring module collects data on the temperature, flow rate, pressure and flue gas components (such as NOx, SOx, particulate matter, etc.) of the flue gas in real time through various sensors installed at the outlet of the biomass boiler. The flue gas monitoring data of the flue gas at the biomass outlet is obtained through the collection of sensors. Among them, the sensors at the outlet of the biomass boiler include temperature sensors, pressure sensors, gas composition sensors, etc.

[0019] The dust removal processing module 12 is used to perform dust removal processing on the biomass outlet flue gas based on the flue gas monitoring data, according to a multi-level dust removal decision model and a bag filter to obtain dust-removed flue gas.

[0020] In the embodiment of the present application, the dust removal processing module processes the flue gas at the biomass boiler outlet based on a multi-level dust removal decision model by receiving input from the flue gas monitoring data. First, the flue gas monitoring data is input into the dust removal decision layer to determine whether the dust removal process needs to be started. Then, the data and decision enter the dust removal prediction layer to generate a prediction result of the dust removal effect, including the predicted dust removal rate and the probability of pollutant generation. Then, the prediction result is verified by the dust removal verification layer. If the verification passes, the dust removal strategy is determined. Finally, the dust removal strategy is applied to the bag filter to perform dust removal on the flue gas to obtain dust-removed flue gas.

[0021] Furthermore, in the system provided in the embodiment of the application, the dust removal processing module 12 is also used for:

[0022] The multi-level dust removal decision model includes a dust removal decision layer, a dust removal prediction layer and a dust removal verification layer; the flue gas monitoring data is input into the dust removal decision layer to obtain a dust removal decision; the flue gas monitoring data and the dust removal decision are input into the dust removal prediction layer to obtain a dust removal prediction result, wherein the dust removal prediction result includes a predicted dust removal rate and a predicted probability of dust removal pollutant generation; the dust removal prediction result is input into the dust removal verification layer to obtain a dust removal verification result; when the dust removal verification result is passed, the dust removal decision is output as a dust removal strategy, and the biomass outlet flue gas is subjected to dust removal treatment according to the dust removal strategy and the bag filter to obtain the dust-removed flue gas.

[0023] In the embodiment of the present application, the multi-level dust removal decision model includes a dust removal decision layer, a dust removal prediction layer and a dust removal verification layer. The dust removal decision layer and the dust removal prediction layer are obtained through training of the neural network model, while the dust removal verification layer is verified based on preset rules.

[0024] When training the dust removal decision layer, first obtain a large amount of flue gas monitoring data and dust removal decisions under corresponding circumstances from the historical database, use these data as training data, and train the model through supervised learning algorithms such as support vector machines to obtain the dust removal decision layer. After training, the dust removal decision layer can determine the dust removal prediction based on the flue gas monitoring data.

[0025] When training the dust removal prediction layer, first obtain a large amount of flue gas monitoring data and dust removal decisions under corresponding circumstances from the historical database, and at the same time obtain the dust removal effect achieved when taking corresponding dust removal decision measures, including dust removal rate and dust removal pollutant generation probability. The dust removal prediction layer is obtained by training the support vector machine using the obtained data. The trained dust removal prediction layer can determine the dust removal prediction result based on the flue gas monitoring data and dust removal decisions.

[0026] The dust removal verification layer is based on preset rules. At this layer, the rules pre-set by technical experts are used to determine whether the prediction results meet the actual needs. For example, it verifies whether the predicted dust removal rate meets the standard and whether the probability of secondary pollutant generation is within an acceptable range. If the verification result passes, the current dust removal strategy will continue to be executed; if the verification result fails, the dust removal decision or prediction result needs to be adjusted until the requirements are met.

[0027] To obtain dust-removed flue gas, first receive the flue gas monitoring data from the low-temperature condition monitoring module, which includes the flue gas temperature, flow rate, particle concentration, flue gas composition, etc. These flue gas characteristic data are input into the dust removal decision layer, which uses the neural network model to learn the relationship between flue gas characteristics and dust removal strategies based on historical data, and outputs dust removal decisions. The output of this step is the dust removal strategy, which determines the bag filter parameters, such as filter bag size, wind speed, etc.

[0028] Next, the flue gas monitoring data and the dust removal decision from the dust removal decision layer are input into the dust removal prediction layer. This layer is also trained through a neural network model, using the flue gas characteristic data and dust removal decisions to predict the dust removal effect. The output of the dust removal prediction layer is the dust removal prediction result, including the predicted dust removal rate (i.e. the percentage of particulate matter that is expected to be removed) and the predicted probability of dust removal pollutant generation (i.e. the predicted probability of secondary pollutants that may be generated during the dust removal process).

[0029] The dust removal prediction results are then input into the dust removal verification layer for verification to check whether the predicted dust removal rate meets the standard and whether the predicted probability of secondary pollutant generation is within an acceptable range. If the verification passes, the current dust removal strategy is considered effective; if the verification fails, the dust removal decision needs to be readjusted or the prediction results need to be recalculated until the verification passes.

[0030] When the dust removal verification result is passed, the final dust removal strategy is output, and the bag dust collector and other equipment are enabled to perform actual dust removal operations according to the strategy. The bag dust collector effectively removes particulate matter in the flue gas through its filter bag technology, and finally obtains dust-free flue gas.

[0031] Furthermore, in the system provided in the embodiment of the application, the dust removal processing module 12 is also used for:

[0032] When the predicted dust removal rate is greater than the expected dust removal rate, and the predicted dust removal pollutant generation probability is less than the dust removal pollutant generation probability threshold, the dust removal verification result is passed; otherwise, the dust removal verification result is failed.

[0033] In the embodiment of the present application, when the predicted dust removal rate is greater than the expected dust removal rate, that is, the actual predicted dust removal effect exceeds the set target, and the predicted dust removal pollutant generation probability is less than the dust removal pollutant generation probability threshold, that is, the predicted secondary pollutant generation probability is lower than the acceptable threshold, the dust removal verification result is passed. This means that the dust removal operation can not only achieve or exceed the expected dust removal effect, but also will not generate too many secondary pollutants. Among them, the expected dust removal rate is pre-set by technical experts.

[0034] Otherwise, if the predicted dust removal rate does not reach the expected dust removal rate or the predicted dust removal pollutant generation probability exceeds the threshold, the dust removal verification result is failed. In this case, the dust removal strategy is considered invalid.

[0035] Furthermore, in the system provided in the embodiment of the application, the dust removal processing module 12 is also used for:

[0036] When the dust removal verification result is failed, the dust removal decision is adjusted to establish a dust removal decision adjustment space; the dust removal decision adjustment space is optimized and screened according to the dust removal prediction layer and the dust removal verification layer to obtain the dust removal strategy.

[0037] In an embodiment of the present application, when the dust removal verification result is failed, it means that the predicted dust removal rate does not reach the expected dust removal rate, or the predicted dust removal pollutant generation probability exceeds the set dust removal pollutant generation probability threshold. At this time, the current dust removal decision is adjusted. The first step of the adjustment is to establish a dust removal decision adjustment space through an optimization algorithm, such as particle swarm optimization or genetic algorithm. This space contains different combinations of multiple decision variables such as dust removal equipment type and equipment parameters. These adjustment parameters include filter bag size, wind speed, air flow, etc.

[0038] Next, these adjusted decision combinations are screened for optimal performance. First, each adjusted decision combination is input into the dust removal prediction layer, which predicts the dust removal effect of each combination based on the flue gas monitoring data and the current dust removal strategy, including the predicted dust removal rate and the predicted probability of dust removal pollutant generation.

[0039] The prediction results are then verified through the dust removal verification layer. The predicted dust removal rate is compared with the expected dust removal rate, and the predicted probability of dust removal pollutant generation is checked to see if it is lower than the set threshold. If the predicted dust removal rate is greater than the expected dust removal rate, and the predicted probability of dust removal pollutant generation is less than the dust removal pollutant generation probability threshold, the verification passes, indicating that the decision combination meets the expected requirements. Otherwise, continue to adjust the decision, and re-predict and verify until a decision combination that meets the conditions is found.

[0040] Finally, the strategy with a passing dust removal verification result is used as the final dust removal strategy. This strategy meets the rules of the dust removal verification layer and ensures that the predicted dust removal effect and secondary pollutant generation meet environmental protection requirements.

[0041] The desulfurization and denitrification control decision module 13 is used to make control decisions on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust removal flue gas, and establish a desulfurization and denitrification control decision space.

[0042] In the embodiment of the present application, the multivariate control elements of the desulfurization and denitrification integrated reactor are first obtained. Then, the dust removal flue gas detection data is used as the control retrieval constraint, and the control parameters of the multivariate control elements are used as the control retrieval target to perform control record retrieval and obtain a set of feasible desulfurization and denitrification control schemes. Then, based on these scheme sets, the desulfurization and denitrification trigger domain is established through trigger interval feature identification to ensure optimized operation within this range. Finally, control decisions are made based on the trigger domain to generate a desulfurization and denitrification control decision space.

[0043] Further, such as Figure 2 As shown, in the system provided by the application embodiment, the desulfurization and denitrification control decision module 13 is also used for:

[0044] The multivariate control elements of the desulfurization and denitrification integrated reactor are obtained, wherein the multivariate control elements include desulfurizer composition, desulfurizer dosage, denitrification catalyst composition, denitrification catalyst dosage, reaction time and reaction temperature; the dust removal flue gas detection data is used as a desulfurization and denitrification control retrieval constraint, and the control parameters of the multivariate control elements are used as a desulfurization and denitrification control retrieval target; control records of the desulfurization and denitrification integrated reactor are retrieved according to the desulfurization and denitrification control retrieval constraint and the desulfurization and denitrification control retrieval target to obtain a desulfurization and denitrification control scheme set; trigger interval feature identification is performed on the desulfurization and denitrification control scheme set according to the multivariate control elements to establish a desulfurization and denitrification trigger domain; control decision is made according to the desulfurization and denitrification trigger domain to generate the desulfurization and denitrification control decision space.

[0045] In an embodiment of the present application, the multi-element control elements pre-set in the integrated desulfurization and denitrification reactor are first obtained, wherein the multi-element control elements include desulfurizer composition, desulfurizer dosage, denitrification catalyst composition, denitrification catalyst dosage, reaction time and reaction temperature.

[0046] Next, the dust removal flue gas detection data is used as the desulfurization and denitrification control retrieval constraint, and the control parameters of the multi-element control elements are used as the desulfurization and denitrification control retrieval target. After that, the control record of the desulfurization and denitrification integrated reactor is retrieved according to the desulfurization and denitrification control retrieval constraint and the desulfurization and denitrification control retrieval target, and multiple historical desulfurization and denitrification control schemes corresponding to the dust removal flue gas detection data (or similar historical dust removal flue gas detection data) are obtained, and these multiple historical desulfurization and denitrification control schemes are integrated as a desulfurization and denitrification control scheme set.

[0047] Afterwards, the trigger interval characteristics of the desulfurization and denitrification control scheme set are identified, that is, the maximum and minimum values ​​of the multivariate control elements in the desulfurization and denitrification control scheme set are determined, and the maximum and minimum values ​​of the multivariate control elements including desulfurizer composition, desulfurizer dosage, denitrification catalyst composition, denitrification catalyst dosage, reaction time and reaction temperature are determined, and the range between the maximum and minimum values ​​of these multivariate control elements is used as the desulfurization and denitrification trigger domain.

[0048] Finally, control decisions are made based on the established desulfurization and denitrification trigger domain, that is, the desulfurization and denitrification trigger domain is randomly selected according to the multivariate control elements to obtain multiple desulfurization and denitrification control decisions, each of which includes desulfurization agent composition parameters, desulfurization agent dosage parameters, denitrification catalyst composition parameters, denitrification catalyst dosage parameters, reaction time parameters, and reaction temperature parameters. Finally, the obtained multiple desulfurization and denitrification control decisions are added to the desulfurization and denitrification control decision space. The desulfurization and denitrification control decision space includes multiple desulfurization and denitrification control decisions.

[0049] The desulfurization and denitrification optimization module 14 is used to perform multi-level optimization on the desulfurization and denitrification control decision space according to the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification multi-feature optimization factors to determine the desulfurization and denitrification control strategy.

[0050] In the embodiment of the present application, three optimization factors are first set, namely, the overall sulfur and nitrate removal rate, the desulfurization and denitrification energy efficiency ratio, and the secondary pollution risk, to provide a basis for optimizing the control decision. Then, based on the desulfurization and denitrification multi-feature prediction model, according to the set expected constraints, the control decision space is optimized to select the solution that best meets the environmental protection and energy efficiency requirements. Finally, the desulfurization and denitrification control strategy is determined by establishing a fitness function and maximizing it.

[0051] Furthermore, in the system provided in the application embodiment, the desulfurization and denitrification optimization module 14 is also used for:

[0052] The desulfurization and denitrification multi-feature optimization factors include the overall sulfur and nitrate removal rate, the desulfurization and denitrification energy efficiency ratio and the secondary pollution risk; according to the desulfurization and denitrification multi-feature optimization factors, the desulfurization and denitrification expectation constraints are set; based on the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification expectation constraints, the desulfurization and denitrification control decision space is expected to be optimized, and the desulfurization and denitrification expectation optimization space is established; according to the desulfurization and denitrification multi-feature optimization factors, a desulfurization and denitrification fitness function is established, and the desulfurization and denitrification fitness function is:

[0053] ;

[0054] Among them, SNFIT represents the fitness of desulfurization and denitrification, CW represents the first weight of desulfurization and denitrification, EW represents the second weight of desulfurization and denitrification, RW represents the third weight of desulfurization and denitrification, the value ranges of CW, EW and RW are all from 0 to 1, the sum of CW, EW and RW is 1, g(SNC) represents the normalized predicted sulfur and nitrate overall removal rate, g(SNE) represents the normalized predicted desulfurization and denitrification energy efficiency ratio, and g(SNR) represents the normalized predicted secondary pollution risk coefficient; according to the desulfurization and denitrification fitness function, the desulfurization and denitrification expected optimization space is optimized to maximize the desulfurization and denitrification fitness to generate the desulfurization and denitrification control strategy.

[0055] In the embodiment of the present application, the multi-feature optimization factors for desulfurization and denitrification include the overall sulfur and nitrate removal rate, the desulfurization and denitrification energy efficiency ratio and the risk of secondary pollution.

[0056] Next, according to the multi-feature optimization factors of desulfurization and denitrification, technical experts set the expected constraints of desulfurization and denitrification, which include the overall sulfur and nitrate removal rate constraint, the desulfurization and denitrification energy efficiency ratio constraint and the secondary pollution risk constraint. Among them, the overall sulfur and nitrate removal rate constraint is the lowest removal rate threshold, ensuring that the desulfurization and denitrification process can achieve the required pollutant removal effect. The desulfurization and denitrification energy efficiency ratio constraint is the maximum value of the energy efficiency ratio, ensuring that the process can save energy to the maximum extent while ensuring the denitrification efficiency. The secondary pollution risk constraint is a maximum allowable value to ensure that the amount of by-products generated is lower than the environmental protection requirements and reduce secondary pollution.

[0057] Then, based on the multi-feature prediction model of desulfurization and denitrification and the expected constraints of desulfurization and denitrification, the desulfurization and denitrification control decision space is optimized, and the expected optimization space of desulfurization and denitrification is established.

[0058] Then, according to the multi-feature optimization factors of desulfurization and denitrification, the desulfurization and denitrification fitness function is established. The specific expression of the desulfurization and denitrification fitness function is: ; Among them, SNFIT represents the fitness of desulfurization and denitrification, CW represents the first weight of desulfurization and denitrification, EW represents the second weight of desulfurization and denitrification, RW represents the third weight of desulfurization and denitrification, the value range of CW, EW, and RW is 0 to 1, and the sum of CW, EW, and RW is 1. g(SNC) represents the normalized predicted sulfur and nitrate overall removal rate, g(SNE) represents the normalized predicted desulfurization and denitrification energy efficiency ratio, and g(SNR) represents the normalized predicted secondary pollution risk coefficient. Among them, CW, EW, and RW are pre-set by technical experts.

[0059] Finally, the desulfurization and denitrification fitness function is used to maximize the fitness of the desulfurization and denitrification expected optimization space. Specifically, by traversing the desulfurization and denitrification expected optimization space, the desulfurization and denitrification fitness corresponding to each decision is calculated, and the decision with the largest desulfurization and denitrification fitness is selected as the desulfurization and denitrification control strategy.

[0060] Furthermore, in the system provided in the application embodiment, the desulfurization and denitrification optimization module 14 is also used for:

[0061] According to the desulfurization and denitrification control decision space, extract the first desulfurization and denitrification control decision; based on the dust removal flue gas detection data and the first desulfurization and denitrification control decision, according to the desulfurization and denitrification multi-feature prediction model, obtain the first desulfurization and denitrification prediction result; judge whether the first desulfurization and denitrification prediction result meets the desulfurization and denitrification expected constraint; if the first desulfurization and denitrification prediction result meets the desulfurization and denitrification expected constraint, add the first desulfurization and denitrification control decision to the desulfurization and denitrification expected optimization space; if the first desulfurization and denitrification prediction result does not meet the desulfurization and denitrification expected constraint, eliminate the first desulfurization and denitrification control decision; according to the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification expected constraint, continue to perform expected optimization on the desulfurization and denitrification control decision space to generate the desulfurization and denitrification expected optimization space.

[0062] In the embodiment of the present application, the desulfurization and denitrification control decision space is firstly randomly selected to extract the first desulfurization and denitrification control decision.

[0063] Next, based on the dust removal flue gas detection data and the first desulfurization and denitrification control decision, the first desulfurization and denitrification prediction result is obtained according to the desulfurization and denitrification multi-feature prediction model. The desulfurization and denitrification multi-feature prediction model is pre-trained, including a sulfur and nitrate overall removal rate prediction model, a desulfurization and denitrification energy efficiency ratio prediction model, and a secondary pollution risk prediction model. The dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the desulfurization and denitrification multi-feature prediction model for prediction to obtain the first desulfurization and denitrification prediction result.

[0064] Then, the first desulfurization and denitrification prediction result is compared with the desulfurization and denitrification desired constraint. If the first desulfurization and denitrification prediction result meets the desulfurization and denitrification desired constraint, it indicates that the control decision is feasible, and the first desulfurization and denitrification control decision is added to the desulfurization and denitrification desired optimization space. If the first desulfurization and denitrification prediction result does not meet the desulfurization and denitrification desired constraint, the control decision is eliminated.

[0065] In each round of screening, by extracting new control decisions from the desulfurization and denitrification control decision space, combined with the desulfurization and denitrification multi-feature prediction model and expected constraints, the remaining control schemes are continued to be optimized. Each newly extracted control decision will go through the same steps, namely prediction, judgment whether it meets the expected constraints, and adding it to the expected optimization space or eliminating it. Through this iterative process, the desulfurization and denitrification control decisions that do not meet the expected constraints of desulfurization and denitrification are eventually removed, and the remaining desulfurization and denitrification control decisions constitute the desulfurization and denitrification expected optimization space.

[0066] Furthermore, in the system provided in the application embodiment, the desulfurization and denitrification optimization module 14 is also used for:

[0067] The desulfurization and denitrification multi-feature prediction model includes a sulfur and nitrate overall removal rate prediction model, a desulfurization and denitrification energy efficiency ratio prediction model and a secondary pollution risk prediction model; the dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the sulfur and nitrate overall removal rate prediction model to obtain a first predicted sulfur and nitrate overall removal rate; the dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the desulfurization and denitrification energy efficiency ratio prediction model to obtain a first predicted desulfurization and denitrification energy efficiency ratio; the dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the secondary pollution risk prediction model to obtain a first predicted secondary pollution risk coefficient; the first predicted sulfur and nitrate overall removal rate, the first predicted desulfurization and denitrification energy efficiency ratio and the first predicted secondary pollution risk coefficient are output as the first desulfurization and denitrification prediction result.

[0068] In an embodiment of the present application, the multi-feature prediction model for desulfurization and denitrification includes a sulfur-nitrate overall removal rate prediction model, a desulfurization and denitrification energy efficiency ratio prediction model, and a secondary pollution risk prediction model.

[0069] Among them, the overall sulfur and nitrate removal rate prediction model, the desulfurization and denitrification energy efficiency ratio prediction model and the secondary pollution risk prediction model are all pre-trained. When training the overall sulfur and nitrate removal rate prediction model, first obtain the dust removal flue gas detection data set, the desulfurization and denitrification control decision and the corresponding sulfur and nitrate overall removal rate set from the historical database. Each dust removal flue gas detection data, the corresponding desulfurization and denitrification control decision and the corresponding sulfur and nitrate overall removal rate constitute a training data set, and these data obtained from the historical database form a training data set. Use these training data sets to train the multi-layer perceptron model to obtain the overall sulfur and nitrate removal rate prediction model, which can predict the predicted sulfur and nitrate overall removal rate based on the input dust removal flue gas detection data and desulfurization and denitrification control decision. Similarly, the data used in the training of the desulfurization and denitrification energy efficiency ratio prediction model include the dust removal flue gas detection data set, the desulfurization and denitrification control decision and the corresponding desulfurization and denitrification energy efficiency ratio set. By using these data to train the convolutional neural network, a desulfurization and denitrification energy efficiency ratio prediction model is obtained. The secondary pollution risk prediction model is obtained by training the long short-term memory network using the dust removal flue gas detection data set, desulfurization and denitrification control decisions and the corresponding secondary pollution risk coefficient set in historical data.

[0070] The dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the sulfur-nitrate overall removal rate prediction model to obtain the first predicted sulfur-nitrate overall removal rate; the dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the desulfurization and denitrification energy efficiency ratio prediction model to obtain the first predicted desulfurization and denitrification energy efficiency ratio; the dust removal flue gas detection data and the first desulfurization and denitrification control decision are input into the secondary pollution risk prediction model to obtain the first predicted secondary pollution risk coefficient.

[0071] Finally, the first predicted sulfur-nitrate overall removal rate, the first predicted desulfurization and denitrification energy efficiency ratio and the first predicted secondary pollution risk coefficient are integrated and output as the first desulfurization and denitrification prediction result.

[0072] The desulfurization and denitrification treatment module 15 is used to perform desulfurization and denitrification treatment on the dust removal flue gas according to the desulfurization and denitrification control strategy and the desulfurization and denitrification integrated reactor.

[0073] In an embodiment of the present application, the desulfurization and denitrification treatment module performs desulfurization and denitrification treatment on the dust-removed flue gas according to the desulfurization and denitrification control strategy and the desulfurization and denitrification integrated reactor. Specifically, the dust-removed flue gas is sent to the desulfurization and denitrification integrated reactor for desulfurization and denitrification treatment according to the desulfurization and denitrification control strategy. During desulfurization, a desulfurizer (such as a calcium-based desulfurizer, ammonia water, etc.) is used to react chemically with sulfur dioxide in the flue gas to generate harmless by-products (such as calcium sulfate). During denitrification, a denitrification catalyst (such as vanadium pentoxide, titanium dioxide, etc.) is used to promote the reaction of nitrogen oxides with a reducing agent (such as ammonia) to generate harmless nitrogen and water vapor.

[0074] After treatment in the integrated desulfurization and denitrification reactor, the sulfur dioxide and nitrogen oxides in the dust removal flue gas are effectively removed, and the pollutant content in the flue gas is significantly reduced.

[0075] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0076] The present application monitors the biomass boiler in real time to obtain the flue gas monitoring data of the biomass outlet flue gas; based on the flue gas monitoring data, the biomass outlet flue gas is subjected to dust removal treatment according to a multi-level dust removal decision model and a bag filter to obtain dust-removed flue gas; control decisions are made on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust-removed flue gas, and a desulfurization and denitrification control decision space is established; multi-level optimization is performed on the desulfurization and denitrification control decision space according to a desulfurization and denitrification multi-feature prediction model and a desulfurization and denitrification multi-feature optimization factor to determine a desulfurization and denitrification control strategy; the dust-removed flue gas is subjected to desulfurization and denitrification treatment according to the desulfurization and denitrification control strategy and the desulfurization and denitrification integrated reactor. The present invention solves the technical problems of low denitrification efficiency, high energy consumption and high risk of secondary pollutant generation in the prior art. The flue gas data is monitored in real time by a low-temperature operating condition monitoring module, and a multi-level dust removal decision model and a bag dust collector are combined to perform flue gas dust removal treatment to obtain dust-removed flue gas. The control strategy is optimized according to the dust-removed flue gas data, and the decision space is optimized through a multi-feature prediction model to determine the optimal desulfurization and denitrification strategy. Finally, the flue gas is efficiently desulfurized and denitrified according to the optimization strategy, thereby achieving the technical effect of ensuring that the flue gas meets the emission standards.

[0077] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0078] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0079] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. Biomass boiler denitrification control system based on low temperature condition monitoring, characterized in that: The system comprises: A low temperature condition monitoring module, which is used to monitor the biomass boiler in real time and obtain flue gas monitoring data of the biomass outlet flue gas; A dust removal processing module, the dust removal processing module is used to perform dust removal processing on the biomass outlet flue gas based on the flue gas monitoring data, according to a multi-level dust removal decision model and a bag filter, to obtain dust-removed flue gas; A desulfurization and denitrification control decision module, which is used to make control decisions on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust removal flue gas, and establish a desulfurization and denitrification control decision space; A desulfurization and denitration optimization module, which is used to perform multi-level optimization on the desulfurization and denitration control decision space according to the desulfurization and denitration multi-feature prediction model and the desulfurization and denitration multi-feature optimization factors to determine the desulfurization and denitration control strategy; A desulfurization and denitration treatment module, which is used to perform desulfurization and denitration treatment on the dust removal flue gas according to the desulfurization and denitration control strategy and the desulfurization and denitration integrated reactor; The desulfurization and denitrification optimization module is used to perform multi-level optimization on the desulfurization and denitrification control decision space according to the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification multi-feature optimization factor to determine the desulfurization and denitrification control strategy, including: The desulfurization and denitrification multi-characteristic optimization factors include the overall sulfur and nitrate removal rate, the desulfurization and denitrification energy efficiency ratio and the secondary pollution risk; According to the desulfurization and denitrification multi-feature optimization factor, setting the desulfurization and denitrification expectation constraints; Based on the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification expected constraints, the desulfurization and denitrification control decision space is expected to be optimized to establish a desulfurization and denitrification expected optimization space; According to the desulfurization and denitrification multi-feature optimization factor, a desulfurization and denitrification fitness function is established, and the desulfurization and denitrification fitness function is: SNFIT=CW*g(SNC)+EW*g(SNE)-RW*g(SNR); Among them, SNFIT represents the fitness of desulfurization and denitrification, CW represents the first weight of desulfurization and denitrification, EW represents the second weight of desulfurization and denitrification, RW represents the third weight of desulfurization and denitrification, the value range of CW, EW, and RW is 0 to 1, and the sum of CW, EW, and RW is 1. g(SNC) represents the normalized predicted overall sulfur and nitrate removal rate, g(SNE) represents the normalized predicted desulfurization and denitrification energy efficiency ratio, and g(SNR) represents the normalized predicted secondary pollution risk coefficient; According to the desulfurization and denitrification fitness function, the desulfurization and denitrification desired optimization space is optimized for maximizing the desulfurization and denitrification fitness to generate the desulfurization and denitrification control strategy; Based on the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification expectation constraints, the desulfurization and denitrification control decision space is expected to be optimized, and the desulfurization and denitrification expectation optimization space is established, including: Extracting a first desulfurization and denitrification control decision according to the desulfurization and denitrification control decision space; Based on the dust removal flue gas detection data and the first desulfurization and denitrification control decision, and according to the desulfurization and denitrification multi-feature prediction model, a first desulfurization and denitrification prediction result is obtained; Determining whether the first desulfurization and denitrification prediction result meets the desulfurization and denitrification expected constraints; If the first desulfurization and denitrification prediction result satisfies the desulfurization and denitrification expectation constraint, adding the first desulfurization and denitrification control decision to the desulfurization and denitrification expectation optimization space; If the first desulfurization and denitrification prediction result does not meet the desulfurization and denitrification expected constraints, eliminating the first desulfurization and denitrification control decision; Continue to perform expected optimization on the desulfurization and denitrification control decision space according to the desulfurization and denitrification multi-feature prediction model and the desulfurization and denitrification expected constraints to generate the desulfurization and denitrification expected optimization space; The desulfurization and denitrification multi-feature prediction model includes a sulfur and nitrate overall removal rate prediction model, a desulfurization and denitrification energy efficiency ratio prediction model and a secondary pollution risk prediction model; The multi-level dust removal decision model includes a dust removal decision layer, a dust removal prediction layer and a dust removal verification layer.

2. The biomass boiler denitrification control system based on low temperature condition monitoring according to claim 1 is characterized in that: The desulfurization and denitrification control decision module is used to make control decisions on the desulfurization and denitrification integrated reactor according to the dust removal flue gas detection data of the dust removal flue gas, and establish a desulfurization and denitrification control decision space, including: Obtaining the multi-element control elements of the desulfurization and denitration integrated reactor, wherein the multi-element control elements include desulfurizer composition, desulfurizer dosage, denitration catalyst composition, denitration catalyst dosage, reaction time and reaction temperature; The dust removal flue gas detection data is used as a desulfurization and denitrification control retrieval constraint, and the control parameters of the multi-element control element are used as a desulfurization and denitrification control retrieval target; According to the desulfurization and denitrification control retrieval constraint and the desulfurization and denitrification control retrieval target, the control record of the desulfurization and denitrification integrated reactor is retrieved to obtain a desulfurization and denitrification control scheme set; Identify the trigger interval characteristics of the desulfurization and denitrification control scheme set according to the multivariate control elements to establish a desulfurization and denitrification trigger domain; A control decision is made according to the desulfurization and denitrification trigger domain to generate the desulfurization and denitrification control decision space.

3. The biomass boiler denitrification control system based on low temperature condition monitoring according to claim 1 is characterized in that: Based on the dust removal flue gas detection data and the first desulfurization and denitrification control decision, and according to the desulfurization and denitrification multi-feature prediction model, a first desulfurization and denitrification prediction result is obtained, including: Inputting the dust removal flue gas detection data and the first desulfurization and denitrification control decision into the sulfur-nitrate overall removal rate prediction model to obtain a first predicted sulfur-nitrate overall removal rate; Inputting the dust removal flue gas detection data and the first desulfurization and denitrification control decision into the desulfurization and denitrification energy efficiency ratio prediction model to obtain a first predicted desulfurization and denitrification energy efficiency ratio; Inputting the dust removal flue gas detection data and the first desulfurization and denitrification control decision into the secondary pollution risk prediction model to obtain a first predicted secondary pollution risk coefficient; The first predicted overall sulfur-nitrate removal rate, the first predicted desulfurization and denitrification energy efficiency ratio and the first predicted secondary pollution risk coefficient are output as the first desulfurization and denitrification prediction result.

4. The biomass boiler denitrification control system based on low temperature condition monitoring according to claim 1, characterized in that: The dust removal processing module is used to perform dust removal processing on the biomass outlet flue gas based on the flue gas monitoring data, according to a multi-level dust removal decision model and a bag filter to obtain dust-removed flue gas, including: Inputting the flue gas monitoring data into the dust removal decision layer to obtain a dust removal decision; Inputting the flue gas monitoring data and the dust removal decision into the dust removal prediction layer to obtain a dust removal prediction result, wherein the dust removal prediction result includes a predicted dust removal rate and a predicted probability of dust removal pollutant generation; Inputting the dust removal prediction result into the dust removal verification layer to obtain a dust removal verification result; When the dust removal verification result is passed, the dust removal decision is output as a dust removal strategy, and the biomass outlet flue gas is subjected to dust removal treatment according to the dust removal strategy and the bag filter to obtain the dust-removed flue gas.

5. The biomass boiler denitrification control system based on low temperature condition monitoring according to claim 4 is characterized in that: The dust removal verification layer includes dust removal verification rules, and the dust removal verification rules include: When the predicted dust removal rate is greater than the expected dust removal rate, and the predicted dust removal pollutant generation probability is less than the dust removal pollutant generation probability threshold, the dust removal verification result is passed; otherwise, the dust removal verification result is failed.

6. The biomass boiler denitrification control system based on low temperature condition monitoring according to claim 4, characterized in that: When the dust removal verification result is not passed, adjusting the dust removal decision and establishing a dust removal decision adjustment space; The dust removal decision adjustment space is optimized and screened according to the dust removal prediction layer and the dust removal verification layer to obtain the dust removal strategy.

Images