Evaluation method for collaborative emission reduction effect of atmospheric pollutants and greenhouse gases

Abstract

The invention discloses an atmospheric pollutant and greenhouse gas collaborative emission reduction effect evaluation method, and relates to the technical field of atmospheric pollutant emission reduction, and the method comprises the following steps: at the initial stage of the whole process, setting an initial reference flow rate for flue gas according to a historical average empirical value, and introducing the flue gas into a flue; when flue gas enters a flue at an initial reference flow rate, multiple types of sensors are arranged at key positions of the flue, and real-time monitoring and high-frequency sampling are carried out on the flue gas flow state; and carrying out comprehensive analysis on the collected flue gas flow state data by utilizing a pre-trained deep learning model, and identifying a potential turbulent flow risk in the flue gas. The turbulent flow risk is intelligently identified and regulated through the deep learning model, the flue gas flow rate is dynamically optimized, the SCR reaction efficiency and the flow field uniformity are effectively improved, the problems of NOx exceeding and ammonia escape are solved, and cooperative emission reduction of pollutants and greenhouse gas and multi-target optimization of system operation are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of air pollution emission reduction, and in particular to a method for evaluating the synergistic emission reduction effect of air pollutants and greenhouse gases. Background Art

[0002] The evaluation of the synergistic reduction effect of atmospheric pollutants and greenhouse gases refers to the process of systematically analyzing and quantifying the combined reduction effect of measures on traditional atmospheric pollutants (such as sulfur dioxide, nitrogen oxides, particulate matter, etc.) and greenhouse gases (such as carbon dioxide, methane, etc.) when formulating and implementing environmental protection or emission reduction measures. By evaluating the comprehensive impact of these synergistic emission reduction measures in terms of economy, environment and society, the best strategy for reducing emissions of conventional pollutants and greenhouse gases can be found, helping decision makers achieve a comprehensive balance between improving environmental quality and addressing climate change.

[0003] In industrial parks, some enterprises have installed a selective catalytic reduction (SCR) system to reduce nitrogen oxide (NOx) emissions. The system reduces pollutants by injecting ammonia into the flue gas at a suitable temperature of 250°C to 400°C, and reducing NOx to harmless nitrogen and water under the action of a catalyst.

[0004] The prior art has the following deficiencies: The existing technology usually uses a constant flow rate to pass the flue gas into the flue. However, when there is local turbulence in the flue gas flow field, continuing to maintain a constant flow rate will lead to uneven distribution of flue gas flow velocity in the SCR reactor, resulting in significant differences in NOx concentration in different areas. At this time, the ammonia injection system is still added according to the "average NOx concentration", resulting in insufficient ammonia injection in high-concentration areas, NOx cannot fully react, and direct emissions exceed the standard. At the same time, turbulent disturbances will also affect the uniform mixing of ammonia and NOx, and local reaction conditions will be limited, so that part of the ammonia is discharged without participating in the reaction, causing ammonia escape and secondary pollution.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a method for evaluating the synergistic emission reduction effect of atmospheric pollutants and greenhouse gases. Through the identification and quantitative analysis of potential turbulence risks by deep learning models, the turbulence formation trend can be accurately predicted in advance, and the flue gas inlet flow rate can be dynamically and intelligently adjusted to adapt to complex and changeable working conditions. This data-driven active optimization strategy not only reduces the excessive local NOx emissions and ammonia escape caused by turbulence, effectively improves the flow field uniformity and reaction efficiency in the SCR reactor, but also reduces the secondary pollution and catalyst loss caused by ammonia escape, and realizes multi-objective synergistic optimization in terms of economy, environmental benefits and equipment stability, and fully reflects the comprehensive effect of synergistic emission reduction of atmospheric pollutants and greenhouse gases to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for evaluating the synergistic emission reduction effect of atmospheric pollutants and greenhouse gases, comprising the following steps: At the beginning of the whole process, a starting reference flow rate is set for the flue gas according to the historical average experience value and it is introduced into the flue; When the flue gas enters the flue at the initial reference flow rate, multiple types of sensors are deployed at key locations in the flue to conduct real-time monitoring and high-frequency sampling of the flue gas flow state; Use pre-trained deep learning models to comprehensively analyze the collected smoke flow state data to identify potential turbulence risks in the smoke; When the turbulence risk is identified, the severity of the turbulence is further quantified and analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. By dynamically adjusting the flue gas flow rate, the uniformity of the flow field in the flue is improved and the risk of flue gas turbulence is eliminated.

[0008] Preferably, the historical average empirical value refers to the average value of the flue gas flow rate obtained based on actual operating condition data during the long-term operation of the SCR system.

[0009] Preferably, a pre-trained deep learning model is used to comprehensively analyze the collected smoke flow state data to identify potential turbulence risks in the smoke. The specific steps are: The real-time acquired smoke flow state data is preprocessed, and features reflecting potential turbulence risks in the smoke flow process are extracted from the preprocessed data, wherein the extracted features include the irregularity of the smoke flow trajectory and the ratio of the flue gas flow velocity fluctuation amplitude to the average flow velocity per unit time. Within the monitoring period, after in-depth analysis of the extracted features, a smoke chaos degree reference value and a flow velocity pulsation intensity reference value are generated respectively. The smoke chaos degree reference value and the flow velocity pulsation intensity reference value are used to quantify the irregular disorder of the smoke flow trajectory in the flue and the fluctuation strength of the smoke flow velocity.

[0010] Preferably, the reference value of the flue gas chaos degree and the reference value of the flow velocity pulsation intensity are input into the trained deep learning model as feature vectors, and the hidden danger degree coefficient is output by the deep learning model. Based on the hidden danger degree coefficient, the potential turbulence risk in the flue gas flow process is intelligently predicted.

[0011] Preferably, the hidden danger degree coefficient generated when predicting the potential turbulence risk in the flue gas using the deep learning model is compared and analyzed with a preset hidden danger degree reference threshold to identify the turbulence risk. The specific steps are as follows: If the hidden danger degree coefficient is greater than the hidden danger degree reference threshold, a risk signal is generated, indicating that there is a potential turbulence risk in the smoke; if the hidden danger degree coefficient is less than or equal to the hidden danger degree reference threshold, a normal signal is generated, indicating that the smoke is flowing normally.

[0012] Preferably, after the turbulence risk is identified, the severity of the turbulence is quantitatively analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. The specific steps are as follows: After identifying the turbulence risk in the flue gas, the currently calculated hidden danger degree coefficient is compared with the preset hidden danger degree reference threshold, and a flow rate control factor is constructed according to the degree of deviation to quantify the turbulence risk level. The calculation expression of the flow rate control factor is: ,in: is the flow rate control factor, is the hidden danger degree coefficient output by the current deep learning model, is the reference threshold of hidden danger level, is a very small positive number, is the response sensitivity coefficient, which is used to adjust the sensitivity to abnormalities. is the hyperbolic tangent function; In order to obtain the flow rate control factor After that, based on the current starting reference flow rate, the target flow rate after intelligent control is calculated to achieve precise intervention in the flue flow field. The control expression of the target flow rate is: ,in: is the initial reference flow rate, The maximum control range coefficient is used to limit the flow rate variation range to prevent excessive control. It is the target flow rate after intelligent correction.

[0013] Preferably, during the monitoring period, the specific steps of generating a reference value of smoke chaos after in-depth analysis of the irregularity of the smoke flow trajectory are as follows: From the preprocessed flue gas velocity vector sequence, the velocity direction, amplitude and gradient change information with representative spatial distribution are extracted to construct a multidimensional state vector group. ,in: Indicated in i The instantaneous flue gas velocity amplitude at each spatial sampling point; Indicated in i The local velocity gradient of each spatial sampling point is used to characterize the intensity of local flow velocity changes; Indicated in i The angle between the velocity vector direction of each spatial sampling point and the reference axis reflects the degree of flow deviation. In the state lattice, the curvature change trend of the continuous trajectory is analyzed, and the smoke chaos reference value is constructed by the sum of the curvature changes of the continuous trajectory, which reflects the geometric distortion degree of the trajectory in the high-dimensional phase space, that is, the nonlinearity, complexity and irregularity of the smoke flow. The construction expression of the smoke chaos reference value is: ,in: is the reference value of smoke chaos degree, N represents the number of trajectory points of the spatial sampling points, Indicates i The trajectory curvature at the spatial sampling point is calculated as: ,in: and are the first-order and second-order guided quantities of the state vector group, respectively. Indicates i The arc length position of the curve of the spatial sampling point, It represents the rate of change of curvature along the trajectory, reflecting the degree of local trajectory distortion.

[0014] Preferably, within the monitoring period, the specific steps of generating a flow velocity pulsation intensity reference value after in-depth analysis of the ratio of the flue gas flow velocity fluctuation amplitude to the average flow velocity per unit time are as follows: During the monitoring period, a continuous sequence of flow rate data points is collected , the velocity variation frequency response is calculated based on the velocity data point sequence, and the calculation expression is: ,in: is the velocity variation frequency response, Indicates k The instantaneous flow velocity value at the sampling moment, n is the number of sampling points in the monitoring period, the numerator is the discrete second-order derivative approximation of the flow velocity, which represents the rate of change of the acceleration trend of the flow velocity. The denominator is As a normalization term, the sum of adjacent velocities is introduced to control the influence of numerical scale; After obtaining the velocity variation frequency response, we further combined the fluctuation energy density of the velocity data to construct a velocity pulsation intensity reference value that enhances perception. The constructed expression is: ,in: Indicates the reference value of flow velocity pulsation intensity. Indicates the instantaneous jump of flow velocity within a unit sampling interval, indicating the first-order gradient of flow velocity, Emphasizes sensitivity to sharp jumps in fluctuating energy.

[0015] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention can accurately predict the turbulence formation trend in advance and dynamically and intelligently adjust the flue gas flow rate to adapt to complex and changeable working conditions by identifying and quantitatively analyzing potential turbulence risks through deep learning models. This data-driven active optimization strategy not only reduces the NOx local emission exceeding the standard and ammonia escape caused by turbulence, effectively improves the flow field uniformity and reaction efficiency in the SCR reactor, but also reduces the secondary pollution and catalyst loss caused by ammonia escape, and realizes multi-objective collaborative optimization in terms of economy, environmental benefits and equipment stability, and fully reflects the comprehensive effect of coordinated emission reduction of atmospheric pollutants and greenhouse gases. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 The present invention is a method flow chart of a method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases. DETAILED DESCRIPTION

[0018] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0019] The present invention provides Figure 1 A method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases is shown, comprising the following steps: At the beginning of the whole process, a starting reference flow rate is set for the flue gas according to the historical average experience value and it is introduced into the flue; The historical average empirical value refers to the average value of the flue gas flow rate obtained based on actual operating data during the long-term operation of the SCR system. This value is stable and adaptable under most operating conditions, and can effectively guarantee the operation effect of the system. When setting the initial baseline flow rate of flue gas, the historical operating data of the equipment under different load, temperature, pressure and other conditions are usually referred to, and the flow rate range with the best performance or the most commonly used is selected as the empirical value. For example, a boiler system has been operating most stably at a flue gas flow rate of 12m / s at 75% load in the past year, with high NOx conversion rate and low ammonia escape rate. Then this flow rate of 12m / s can be used as the "historical average empirical value" for future startup or default operation. This value has both a practical basis and is convenient for use as a reference standard in the absence of real-time control.

[0020] When the flue gas enters the flue at the initial reference flow rate, multiple types of sensors are deployed at key locations in the flue to conduct real-time monitoring and high-frequency sampling of the flue gas flow state; The purpose of this step is to provide comprehensive and dynamic information on flue gas operating conditions: on the one hand, it can understand the distribution of flue gas flow rate, pressure and temperature at different times and spaces; on the other hand, it can understand the mixing and conversion of NOx and ammonia in the SCR reaction area.

[0021] In the SCR exhaust gas treatment system, "key positions" usually refer to areas in the flue that best represent the overall flow field characteristics or are prone to flow field anomalies. For example, the initial flow velocity and temperature distribution of the flue gas just entering the flue can be captured at the inlet section, so as to timely understand the emission stability of the boiler or other equipment; in areas with obvious geometric changes such as bends or expansion and contraction sections, turbulence, eddies or local pressure changes are prone to occur, so they need to be monitored; sensors should also be deployed in front and behind the catalyst bed to evaluate the gas composition and temperature differences in the actual reaction zone; in addition, the outlet section is an important observation point for comprehensive evaluation of the overall reaction efficiency, NOx removal rate and ammonia escape. By setting up monitoring points in these areas, it is easy to discover possible flow field anomalies, catalyst failure or local overheating problems at the first time, thereby providing accurate data support for subsequent intelligent control.

[0022] In order to accurately grasp the state of flue gas flow, it is necessary to deploy multiple types of sensors for "multi-dimensional" monitoring. Flow rate sensors or pitot tubes are used to capture flue gas velocity distribution and its fluctuations; pressure or differential pressure sensors help determine local turbulence, blockage or leakage in the flue; temperature sensors can assess whether the thermal conditions of combustion and catalytic reactions are in the ideal range; NOx concentration sensors are key to assessing denitrification efficiency and NOx distribution; ammonia sensors or gas analyzers can monitor ammonia escape and assist in optimizing ammonia injection; if necessary, sulfur dioxide, carbon monoxide, oxygen and other gas analyzers can be deployed for comprehensive diagnosis of flue gas composition. These sensors work together to sample at high frequency and transmit data to the control system in real time.

[0023] Use pre-trained deep learning models (such as those based on CNN, RNN, Transformer, etc.) to comprehensively analyze the collected flue gas flow state data, identify potential turbulence risks in the flue gas, and evaluate the potential impact of turbulence on ammonia escape; Use the pre-trained deep learning model to comprehensively analyze the collected smoke flow state data to identify potential turbulence risks in the smoke. The specific steps are: Preprocess the smoke flow state data acquired in real time, and extract features reflecting potential turbulence hazards in the smoke flow process from the preprocessed data, wherein the extracted features include the irregularity of the smoke flow trajectory and the ratio of the flue gas flow velocity fluctuation amplitude to the average flow velocity per unit time. Within the monitoring period, after in-depth analysis of the extracted features, generate a smoke chaos degree reference value and a flow velocity pulsation intensity reference value, respectively. The smoke chaos degree reference value and the flow velocity pulsation intensity reference value are used to quantify the irregularity of the smoke flow trajectory in the flue and the flue gas flow velocity fluctuation intensity. The reference value of flue gas chaos degree and the reference value of flow velocity pulsation intensity are input into the trained deep learning model as feature vectors. The hidden danger degree coefficient is output by the deep learning model, and the potential turbulence risk in the flue gas flow process is intelligently predicted based on the hidden danger degree coefficient.

[0024] The preprocessing of the real-time acquired flue gas flow state data usually includes steps such as data denoising, outlier detection and removal, missing value filling, data smoothing, and data format conversion. Among them, data denoising is used to eliminate interference signals in the sensor acquisition process to ensure the signal quality during subsequent analysis; outlier detection and removal can eliminate extreme abnormal points caused by instrument errors or short-term failures to prevent model misjudgment; missing value filling can repair the missing parts of the sampling through interpolation or statistical methods to ensure data integrity; data smoothing helps to eliminate local sharp fluctuations or jitters and improve the stability and reliability of the data; data format conversion ensures that the preprocessed data can be seamlessly connected to the subsequent deep learning model training and analysis process in a suitable format. This series of preprocessing steps provides high-quality and highly reliable data support for subsequent turbulent intelligent analysis, risk assessment, and flow rate control.

[0025] When the flue gas is passed into the flue, if its flow trajectory shows obvious irregularity, that is, it is characterized by drastic changes, frequent disturbances, variable directions or nonlinear deviations, it usually indicates that there is a high potential turbulence risk in the flue gas. Because in a stable laminar state, the flue gas flow trajectory should be relatively smooth, consistent in direction, and gradually changing in speed; once the trajectory is twisted, vortexed, falling off, backflow and other phenomena occur, it means that the velocity gradient, pressure distribution or thermal disturbance in the local area has reached a certain threshold, and the possibility of inducing turbulence is greatly increased. The irregularity of the trajectory reflects the unevenness of energy transfer in the flow field and the complexity of fluid disturbances. It is an important precursor feature before the formation of turbulence. Therefore, monitoring and analyzing the irregularity of the flue gas flow trajectory is one of the key means to identify turbulence risks, early warning and implement intelligent regulation.

[0026] During the monitoring period, the specific steps for generating a reference value of smoke chaos after in-depth analysis of the irregularity of the smoke flow trajectory are as follows: From the preprocessed flue gas velocity vector sequence, the velocity direction, amplitude and gradient change information with representative spatial distribution are extracted to construct a multidimensional state vector group. ,in: Indicated in i The instantaneous flue gas velocity amplitude at each spatial sampling point; Indicated in i The local velocity gradient of each spatial sampling point is used to characterize the intensity of local flow velocity changes; Indicated in i The angle between the velocity vector direction of each spatial sampling point and the reference axis reflects the degree of flow deviation. All state vector groups It is mapped into a high-dimensional phase space to form a dot-lattice trajectory group, and the distribution form of the trajectory group will provide a geometric basis for the next step of chaos calculation. The role of this step is to establish the geometric mapping relationship of the flow trajectory in the state space and provide the original structure for analyzing its geometric irregularity.

[0027] In the state lattice, the curvature change trend of the continuous trajectory is analyzed, and the smoke chaos reference value is constructed by the sum of the curvature changes of the continuous trajectory, which reflects the geometric distortion degree of the trajectory in the high-dimensional phase space, that is, the nonlinearity, complexity and irregularity of the smoke flow. The construction expression of the smoke chaos reference value is: ,in: is the reference value of smoke chaos degree, N represents the number of trajectory points of the spatial sampling points, Indicates i The trajectory curvature at the spatial sampling point is calculated as: ,in: and are the first-order and second-order guided quantities of the state vector group, respectively. Indicates i The arc length position of the curve of the spatial sampling point, It represents the rate of change of curvature along the trajectory, reflecting the degree of distortion of the local trajectory.

[0028] It can be seen from the smoke chaos reference value that during the monitoring period, the greater the performance value of the smoke chaos reference value generated after in-depth analysis of the irregularity of the smoke flow trajectory, the greater the risk of potential turbulence in the smoke. Specifically, the smoke chaos reference value is an indicator generated after in-depth analysis of the irregularity and nonlinear disorder of the smoke flow trajectory. It essentially reflects the distortion and complexity of the smoke trajectory in the high-dimensional state space. When the smoke chaos reference value performance value is large, it means that the movement trajectory of the smoke is highly irregular, the state is changeable, and there is a strong disorder. The trajectory changes are difficult to predict and control, showing obvious turbulent characteristics, indicating that there is a high potential turbulence risk in the smoke; on the contrary, if the performance value of the smoke chaos reference value is small, it means that the smoke movement trajectory is relatively regular, continuous and stable, the operation state of the smoke in the flow field is relatively stable, the risk of potential turbulence inside the flue is small, and the flow state is within the normal or stable range. Therefore, the potential turbulence risk of flue gas under actual operating conditions can be intuitively and accurately quantified and evaluated by the reference value of flue gas chaos.

[0029] When the flue gas is introduced into the flue, if the ratio of the fluctuation amplitude of the flue gas velocity per unit time to its average velocity increases significantly, this usually indicates that the flue gas flow state tends to be unstable and there is a potential risk of turbulence. This is because in the laminar or steady flow state, the flue gas velocity changes relatively smoothly and the fluctuation amplitude is small; when the velocity fluctuates violently, it means that the local gas is affected by factors such as disturbance, shear force or temperature gradient, resulting in the appearance of velocity shear layer, vortex structure or recirculation zone in the flow field, which are the precursors of turbulence.

[0030] During the monitoring period, the specific steps for generating a reference value of the velocity pulsation intensity after in-depth analysis of the ratio of the flue gas velocity fluctuation amplitude to the average velocity per unit time are as follows: During the monitoring period, a continuous sequence of flow rate data points is collected , the velocity variation frequency response is calculated based on the velocity data point sequence, and the calculation expression is: ,in: is the velocity variation frequency response, Indicates k The instantaneous flow velocity value at the sampling moment, n is the number of sampling points in the monitoring period, the numerator is the discrete second-order derivative approximation of the flow velocity, which represents the rate of change of the acceleration trend of the flow velocity. The denominator is As a normalization term, the sum of adjacent velocities is introduced to control the influence of numerical scale; This step captures the trend of local nonlinear disturbances in the velocity sequence. The more drastic the second-order changes (such as pulses and mutations), the faster and stronger the flue gas velocity fluctuation response is, reflecting the potential turbulence inducement.

[0031] After obtaining the velocity variation frequency response, we further combined the fluctuation energy density of the velocity data to construct a velocity pulsation intensity reference value that enhances perception. The constructed expression is: ,in: Indicates the reference value of flow velocity pulsation intensity. Indicates the instantaneous jump of flow velocity within a unit sampling interval, indicating the first-order gradient of flow velocity, Emphasizes sensitivity to sharp jumps in fluctuating energy (more emphasis on abnormal disturbances than square), It is used to maintain the stability of the calculation results, prevent the energy density from being too large or over-responding to small jumps. The logarithmic transformation is used to compress the numerical scale and enhance the distinction in the high disturbance area, while having fault tolerance for low disturbance; This step integrates the two dimensions of frequent jumps and strong disturbances to construct a flow velocity pulsation intensity reference value with "pulsation sensitivity". The higher the value, the more unstable the local airflow is, which is likely to develop into turbulence. It is the core decision-making indicator for quantifying the flue gas pulsation intensity.

[0032] It can be seen from the flow velocity pulsation intensity reference value that, during the monitoring period, the greater the performance value of the flow velocity pulsation intensity reference value generated after in-depth analysis of the ratio of the flue gas velocity fluctuation amplitude to the average flow velocity per unit time, the greater the risk of potential turbulence in the flue gas. The flow velocity pulsation intensity reference value essentially reflects the intensity and severity of the fluctuation of the flue gas velocity during the monitoring period. When the value of the flow velocity pulsation intensity reference value is large, it means that there are significant and frequent fluctuations in the flue gas velocity during the monitoring period, that is, the velocity fluctuates violently in a short period of time and even shows a large jump or irregular movement trend. This phenomenon indicates that the flue gas flow state is unstable, and the airflow structure may have characteristics such as vortex, backflow or shear disturbance, thereby increasing the potential risk of turbulence. On the contrary, when the performance value of the flow velocity pulsation intensity reference value is small, it means that the flue gas velocity changes relatively smoothly and regularly, and the flue gas presents a relatively stable flow state in the flue, and the risk of potential turbulence is significantly reduced. Therefore, the reference value of flow velocity pulsation intensity can more accurately reflect the stability of the flue gas flow field and the strength of turbulence risk, providing an important reference for further flue gas regulation and turbulence risk management.

[0033] A trained deep learning model refers to an artificial intelligence model that has been trained with a large amount of sample data and has fixed parameters, and can give stable prediction results under given input conditions. Specifically, the model is usually based on a neural network structure, such as a multi-layer perceptron (MLP), a convolutional neural network (CNN) or an attention mechanism (Transformer) framework. Through the previous massive working condition data learning, it has mastered the characteristic laws and complex mapping relationships closely related to the occurrence of turbulence in the flue gas flow process. In actual operation, the historical or simulated flue gas flow data will first be labeled or scored. For example, the significant turbulence phenomenon that has occurred is defined as a high-risk case, and the relatively stable flow process is defined as a low-risk case. After that, through the iterative training process, the model will continuously adjust the internal weights so that when it sees new inputs (such as feature vectors such as "smoke chaos reference value" and "flow velocity pulsation intensity reference value"), it can also output a quantifiable "hazard degree coefficient" more accurately. Such a formed deep learning model often has a strong generalization ability. Even in the face of previously unseen working condition changes, it can also capture the correlation between features and give relatively reliable risk prediction results. It is worth emphasizing that "trained" does not mean that it stops evolving. If new measured data or extreme working conditions are continuously added, online learning or incremental training can be used to allow the model to continuously improve itself during operation, thereby maintaining high accuracy and robustness when facing dynamic situations such as multi-source flue gas merging, non-constant load or seasonal climate influences.

[0034] In specific use, when the "smoke chaos reference value" and "flow velocity pulsation intensity reference value" are used as input features, the deep learning model will comprehensively consider the different aspects of information contained in these two indicators: the former mainly measures the degree of chaos of smoke flow trajectory and energy distribution, reflecting the possibility of irregular structures such as turbulence, vortex or shear layer inside the smoke; the latter focuses more on the violent fluctuations and acceleration changes of flow velocity in a limited time or space interval, and quantifies the intensity of fluctuations and whether pulse jumps are frequent. After these input features are transformed by multiple layers of nonlinear neural networks, they interact with a large amount of implicit knowledge and weights stored in the model, and finally output a continuous "hidden danger degree coefficient". The higher the value of the hidden danger degree coefficient, the greater the probability of turbulence under the current working conditions, and the need to trigger a higher level of early warning or linkage control strategy; if the hidden danger degree coefficient is low, it means that the overall smoke flow remains stable, and there is no need to perform large-scale flow rate regulation or emergency working condition switching for the time being. Such a prediction mechanism not only allows the production management system to perceive turbulence risks in advance, but can also be combined with the automated scheduling algorithm to make timely adjustments to flue gas flow, ammonia injection amount, and temperature control, thereby reducing the occurrence of problems such as NOx exceeding the standard and ammonia escape, enabling the SCR system to achieve high-efficiency, low-emission and stable operation within a wider range of operating conditions.

[0035] The deep learning model is not specifically limited here, and can achieve the reference value of smoke chaos degree and flow velocity pulsation intensity reference value Conduct comprehensive analysis to generate hidden danger degree coefficient In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; the hidden danger degree coefficient The generated expression is: , where are the reference values ​​of smoke chaos degree and flow velocity pulsation intensity reference value The preset scaling factor of The preset proportional coefficient refers to the value of each indicator (such as chaos index) when comprehensively calculating the hidden danger degree coefficient. and flow velocity pulsation intensity reference value ) is the weight factor assigned to and . Their numerical values ​​are usually determined by experience, test data or model tuning results, and are used to balance the proportion of different indicators in the final turbulence assessment. If a certain indicator has a relatively greater impact on the turbulence risk, it can be assigned a higher proportional coefficient; otherwise, it can be assigned a lower proportional coefficient. The purpose of this is to allow the final "hidden danger degree coefficient" to more accurately reflect the contribution of each indicator to the overall turbulence risk, so as to achieve better prediction and evaluation results when making a comprehensive judgment.

[0036] It can be seen from the hidden danger degree coefficient that during the monitoring period, the greater the performance value of the smoke chaos degree reference value generated after the in-depth analysis of the irregularity of the smoke flow trajectory, the greater the performance value of the flow velocity pulsation intensity reference value generated after the in-depth analysis of the ratio of the flue gas flow velocity fluctuation amplitude to the average flow velocity per unit time. That is, the greater the performance value of the hidden danger degree coefficient generated when the potential turbulence risk in the smoke is predicted using the pre-trained deep learning model, the greater the risk of potential turbulence in the smoke, and vice versa.

[0037] The hidden danger degree coefficient generated when predicting the potential turbulence risk in the flue gas using the deep learning model is compared and analyzed with the pre-set hidden danger degree reference threshold to identify the turbulence risk. The specific steps are as follows: If the hidden danger degree coefficient is greater than the hidden danger degree reference threshold, a risk signal is generated, indicating that there is a potential turbulence risk in the smoke; if the hidden danger degree coefficient is less than or equal to the hidden danger degree reference threshold, a normal signal is generated, indicating that the smoke is flowing normally.

[0038] When the turbulence risk is identified, the severity of the turbulence is further quantified and analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. By dynamically adjusting the flue gas flow rate, the uniformity of the flow field in the flue is improved, and the risk of flue gas turbulence is eliminated; Specifically, the rate at which smoke enters the flue and the flow field distribution can be dynamically changed by adjusting the speed of the induced draft fan inverter, opening and closing the pipeline valve or the angle of the guide plate, etc.

[0039] When the turbulence risk is identified, the severity of the turbulence is quantitatively analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. The specific steps are as follows: After identifying the turbulence risk in the flue gas, the currently calculated hidden danger degree coefficient is compared with the preset hidden danger degree reference threshold, and a flow rate control factor is constructed according to the degree of deviation to quantify the turbulence risk level. The calculation expression of the flow rate control factor is: ,in: is the flow rate control factor, is the hidden danger degree coefficient output by the current deep learning model, is the reference threshold of hidden danger level, It is a very small positive number to prevent the denominator from being zero and to enhance numerical stability. It is usually set to 0.0001. is the response sensitivity coefficient, which is used to adjust the sensitivity to abnormalities. It is a hyperbolic tangent function, which is used to smoothly map the result to (-1, 1) to prevent extreme jumps; This step constructs a continuous, nonlinear, and abnormally sensitive flow rate control factor. When the hidden danger degree coefficient is slightly higher than the hidden danger degree reference threshold, the output is a mild response; when the hidden danger degree coefficient is significantly higher than the hidden danger degree reference threshold, the output is rapidly amplified and the drive system enters a strong control state, ensuring that the control behavior has the progressive characteristics of early warning, inhibition, and intervention.

[0040] The flow rate control factor After that, based on the current starting reference flow rate, the target flow rate after intelligent control is calculated to achieve precise intervention in the flue flow field. The control expression of the target flow rate is: ,in: is the initial reference flow rate, which is derived from the optimal value under historical stable conditions. is the maximum control amplitude coefficient , used to limit the flow rate range and prevent excessive regulation, It is the target flow rate after intelligent correction; This step realizes the dynamic nonlinear regulation of flue gas flow rate. When the target flow rate The flow rate will be lower than the initial reference flow rate, thereby extending the reactor residence time and alleviating the turbulence trend; when the system tends to stabilize When the flue gas flow rate returns to the baseline state, the SCR efficiency and energy balance are maintained. ,This structure also allows forward speed regulation and has good scalability and adaptability.

[0041] Through the above scheme, real-time intelligent monitoring and precise control of the flue gas flow state can be achieved, effectively overcoming the limitations and negative effects of the traditional constant flow rate and average ammonia injection strategy, and significantly improving the refined control level and response timeliness of NOx removal in the process of SCR system treating flue gas. Specifically, through the identification and quantitative analysis of potential turbulence risks by the deep learning model, the turbulence formation trend can be accurately predicted in advance, and the flue gas inlet flow rate can be dynamically and intelligently adjusted to adapt to complex and changeable working conditions. This data-driven active optimization strategy not only reduces the excessive local NOx emissions and ammonia escape caused by turbulence, effectively improves the flow field uniformity and reaction efficiency in the SCR reactor, but also reduces the secondary pollution and catalyst loss caused by ammonia escape, thereby achieving multi-objective collaborative optimization in terms of economy, environmental benefits and equipment stability, and fully reflects the comprehensive effect of coordinated emission reduction of atmospheric pollutants and greenhouse gases.

[0042] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0043] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0044] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0045] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0046] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0047] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0048] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0049] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0050] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases, characterized in that: The following steps are involved: At the beginning of the whole process, a starting reference flow rate is set for the flue gas according to the historical average experience value and it is introduced into the flue; When the flue gas enters the flue at the initial reference flow rate, multiple types of sensors are deployed at key locations in the flue to conduct real-time monitoring and high-frequency sampling of the flue gas flow state; Use pre-trained deep learning models to comprehensively analyze the collected smoke flow state data to identify potential turbulence risks in the smoke; When the turbulence risk is identified, the severity of the turbulence is further quantified and analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. By dynamically adjusting the flue gas flow rate, the uniformity of the flow field in the flue is improved and the risk of flue gas turbulence is eliminated.

2. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 1, characterized in that: The historical average empirical value refers to the average value of the flue gas flow rate obtained based on actual operating data during the long-term operation of the SCR system.

3. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 1, characterized in that: Use the pre-trained deep learning model to comprehensively analyze the collected smoke flow state data to identify potential turbulence risks in the smoke. The specific steps are: The real-time acquired smoke flow state data is preprocessed, and features reflecting potential turbulence risks in the smoke flow process are extracted from the preprocessed data, wherein the extracted features include the irregularity of the smoke flow trajectory and the ratio of the flue gas flow velocity fluctuation amplitude to the average flow velocity per unit time. Within the monitoring period, after in-depth analysis of the extracted features, a smoke chaos degree reference value and a flow velocity pulsation intensity reference value are generated respectively. The smoke chaos degree reference value and the flow velocity pulsation intensity reference value are used to quantify the irregular disorder of the smoke flow trajectory in the flue and the fluctuation strength of the smoke flow velocity.

4. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 3 is characterized in that: The reference value of flue gas chaos degree and the reference value of flow velocity pulsation intensity are input into the trained deep learning model as feature vectors. The hidden danger degree coefficient is output by the deep learning model, and the potential turbulence risk in the flue gas flow process is intelligently predicted based on the hidden danger degree coefficient.

5. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 4, characterized in that: The hidden danger degree coefficient generated when predicting the potential turbulence risk in the flue gas using the deep learning model is compared and analyzed with the pre-set hidden danger degree reference threshold to identify the turbulence risk. The specific steps are as follows: If the hidden danger degree coefficient is greater than the hidden danger degree reference threshold, a risk signal is generated, indicating that there is a potential turbulence risk in the smoke; if the hidden danger degree coefficient is less than or equal to the hidden danger degree reference threshold, a normal signal is generated, indicating that the smoke is flowing normally.

6. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 5, characterized in that: When the turbulence risk is identified, the severity of the turbulence is quantitatively analyzed, and the preset flow rate of the flue gas is intelligently controlled according to the turbulence risk level. The specific steps are as follows: After identifying the turbulence risk in the flue gas, the currently calculated hidden danger degree coefficient is compared with the preset hidden danger degree reference threshold, and a flow rate control factor is constructed according to the degree of deviation to quantify the turbulence risk level. The calculation expression of the flow rate control factor is: ,in: is the flow rate control factor, is the hidden danger degree coefficient output by the current deep learning model, is the reference threshold of hidden danger level, is a very small positive number, is the response sensitivity coefficient, which is used to adjust the sensitivity to abnormalities. is the hyperbolic tangent function; In order to obtain the flow rate control factor After that, based on the current starting reference flow rate, the target flow rate after intelligent control is calculated to achieve precise intervention in the flue flow field. The control expression of the target flow rate is: ,in: is the initial reference flow rate, The maximum control range coefficient is used to limit the flow rate variation range to prevent excessive control. It is the target flow rate after intelligent correction.

7. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 3, characterized in that: During the monitoring period, the specific steps for generating a reference value of smoke chaos after in-depth analysis of the irregularity of the smoke flow trajectory are as follows: From the preprocessed flue gas velocity vector sequence, the velocity direction, amplitude and gradient change information with representative spatial distribution are extracted to construct a multidimensional state vector group. ,in: Indicated in i The instantaneous flue gas velocity amplitude at each spatial sampling point; Indicated in i The local velocity gradient of each spatial sampling point is used to characterize the intensity of local flow velocity changes; Indicated in i The angle between the velocity vector direction of each spatial sampling point and the reference axis reflects the degree of flow deviation. In the state lattice, the curvature change trend of the continuous trajectory is analyzed, and the smoke chaos reference value is constructed by the sum of the curvature changes of the continuous trajectory, which reflects the geometric distortion degree of the trajectory in the high-dimensional phase space, that is, the nonlinearity, complexity and irregularity of the smoke flow. The construction expression of the smoke chaos reference value is: ,in: is the reference value of smoke chaos degree, N represents the number of trajectory points of the spatial sampling points, Indicates i The trajectory curvature at the spatial sampling point is calculated as: ,in: and are the first-order and second-order guided quantities of the state vector group, respectively. Indicates i The arc length position of the curve of the spatial sampling point, It represents the rate of change of curvature along the trajectory, reflecting the degree of distortion of the local trajectory.

8. The method for evaluating the synergistic reduction effect of atmospheric pollutants and greenhouse gases according to claim 3 is characterized in that: During the monitoring period, the specific steps for generating a reference value of the velocity pulsation intensity after in-depth analysis of the ratio of the flue gas velocity fluctuation amplitude to the average velocity per unit time are as follows: During the monitoring period, a continuous sequence of flow rate data points is collected , the velocity variation frequency response is calculated based on the velocity data point sequence, and the calculation expression is: ,in: is the velocity variation frequency response, Indicates k The instantaneous flow velocity value at the sampling moment, n is the number of sampling points in the monitoring period, the numerator is the discrete second-order derivative approximation of the flow velocity, which represents the rate of change of the acceleration trend of the flow velocity. The denominator is As a normalization term, the sum of adjacent velocities is introduced to control the influence of numerical scale; After obtaining the velocity variation frequency response, we further combined the fluctuation energy density of the velocity data to construct a velocity pulsation intensity reference value that enhances perception. The constructed expression is: ,in: Indicates the reference value of flow velocity pulsation intensity. Indicates the instantaneous jump of flow velocity within a unit sampling interval, indicating the first-order gradient of flow velocity, Emphasizes sensitivity to sharp jumps in fluctuating energy.

Images