Method for predicting nitrogen oxide emission in waste incineration industry and related device

Abstract

The invention provides a method for predicting nitrogen oxide emission in the waste incineration industry and a related device, and belongs to the technical field of environmental protection. According to the method, feature importance analysis is carried out on all preprocessed data, and key features closely related to nitrogen oxide emission in the waste incineration industry are obtained; carrying out empirical mode decomposition on key characteristics closely related to nitrogen oxide emission in the waste incineration industry and continuous nitrogen oxide emission data in the waste incineration industry to obtain an intrinsic mode function component and a residual component; constructing a training set and a verification set by using the obtained intrinsic mode function component and the residual component; training the LSTM model according to the training set to obtain a trained LSTM model; and predicting the data in the verification set by using the trained LSTM model, and comparing the data with continuous nitrogen oxide emission data in the garbage industry to obtain a nitrogen oxide emission prediction result in the garbage incineration industry. The problem that the accuracy of nitrogen oxide emission prediction in the waste incineration industry is not high is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental protection, and in particular relates to a method and related device for predicting nitrogen oxide emissions in a waste incineration industry. Background Art

[0002] With increasingly stringent environmental regulations and increasing public environmental awareness, the waste incineration industry is facing increasing pressure and regulatory requirements to reduce emissions. In order to reduce nitrogen oxide emissions, the waste incineration industry needs to take a series of effective emission reduction measures, such as optimizing combustion processes and installing advanced denitrification equipment. However, the implementation of these measures often requires a lot of money and resources. If there is a lack of accurate nitrogen oxide emission forecasts as a guide, the waste industry may over-invest or have poor emission reduction effects in the emission reduction process, resulting in increased operating costs and reduced economic benefits.

[0003] At present, machine learning technology has been widely used in many industries. However, in the field of environmental protection, especially in the waste incineration industry, the application of machine learning is still insufficient. Traditional prediction methods that rely on simple statistical models or fixed parameter models are difficult to accurately capture the dynamic changes in nitrogen oxide emissions, and the prediction errors are large. This limitation is particularly prominent in the field of environmental protection, because the data in the environmental protection industry is usually highly complex and dynamic, which will bring huge challenges to the accuracy and stability of the model.

[0004] For example, in the process of municipal solid waste incineration, the sources of waste are wide and the composition is complex. Since waste classification has not yet been refined, there is significant uncertainty in the composition of waste entering the incinerator. Different waste components will undergo different physical and chemical reactions during the incineration process, which directly affects the types and quantities of pollutant emissions, further increasing the difficulty of emission prediction. Summary of the invention

[0005] The purpose of the present invention is to provide a method and related device for predicting nitrogen oxide emissions in the waste incineration industry, so as to solve the problem of low accuracy of nitrogen oxide emission prediction in the waste incineration industry in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for predicting nitrogen oxide emissions in a waste incineration industry, comprising the following steps: Obtain continuous nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data; Preprocess all acquired data and remove abnormal data; The feature importance analysis was performed on all preprocessed data to obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry; The key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data of the waste industry are subjected to empirical mode decomposition to obtain intrinsic mode function components and residual components; The obtained intrinsic mode function components and residual components are used to construct training sets and validation sets; Train the LSTM model according to the training set to obtain a trained LSTM model; The trained LSTM model is used to predict the data in the validation set and compared with the continuous nitrogen oxide emission data of the garbage industry to obtain the prediction results of nitrogen oxide emissions in the garbage incineration industry.

[0007] A further improvement of the present invention is that, in the step of preprocessing the acquired data and removing abnormal data, all acquired data are preprocessed, the abnormal data are first removed to obtain data after the abnormalities are removed, and then the data after the abnormalities are removed are processed for data stability using an emission dynamic balance algorithm to obtain stable data.

[0008] A further improvement of the present invention is that the specific steps of the emission dynamic balance algorithm include: Analyze the time series characteristics of all acquired data in real time to obtain several dynamic features; Normalize the obtained dynamic features; Combine several normalized dynamic features to obtain the overall fluctuation of all acquired data; Perform data stationarity processing on the overall fluctuation size of all acquired data to obtain the original stationarity data; The original stability data is monitored in real time and adjusted dynamically until the overall fluctuation of all acquired data reaches the expected fluctuation range, thus obtaining the final stability data.

[0009] A further improvement of the present invention is that in the step of combining the normalized dynamic features to obtain the overall fluctuation size of all acquired data, the calculation formula for the overall fluctuation size of all acquired data is:

[0010] in, The overall fluctuation size of all data obtained is the weight factor corresponding to each dynamic feature, The sum of the five weight factors is equal to 1, for t The trend change value after normalization at each moment, for t The periodic fluctuation value after normalization at each moment, fort The noise level after normalization at each moment, for t The real-time data change rate after time normalization processing, For working conditions, The value of is 0 or 1. , , , and To dynamically adjust parameters, t Dynamically adjust parameters at all times , , , (t) and Set according to the volatility index; In the step of performing data stationarity processing on the overall fluctuation size of all acquired data, the calculation formula for performing data stationarity processing on the overall fluctuation size of all acquired data is:

[0011] in, for t The data after the moment data stationarity processing, i is a natural number, is the size of the smoothing window, which is used to control the number of data points involved in the smoothing calculation. is the smoothing coefficient, used to adjust the overall fluctuation The degree of influence on the smoothing process, for t The data before the moment data stationarity processing, j An index, indicating moving forward from the current time j time period, The overall fluctuation size of all the data obtained, for t The overall fluctuation of all data obtained at all times.

[0012] A further improvement of the present invention is that, in the step of performing feature importance analysis on all preprocessed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry, an emission pattern mining algorithm is specifically used to perform feature importance analysis on all preprocessed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry.

[0013] A further improvement of the present invention is that the emission pattern mining algorithm specifically comprises the following steps: Constructing a tree model for emission pattern mining; Using all preprocessed data to train the tree model for emission pattern mining, a trained tree model for emission pattern mining is obtained; Calculate the SHAP value based on the trained tree model for emission pattern mining; According to the calculated SHAP value, the contribution of all preprocessed data to the prediction results of nitrogen oxide emissions in the waste incineration industry is determined, and the contribution sizes are sorted to obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry.

[0014] A further improvement of the present invention is that, in the step of constructing a training set and a validation set using the obtained intrinsic mode function components and residual components, 80% of the obtained intrinsic mode function components and residual components are used as a training set, and 20% are used as a validation set.

[0015] In a second aspect, the present invention provides a system for predicting nitrogen oxide emissions in a waste incineration industry, including a data acquisition module, a data preprocessing module, a feature importance analysis module, an empirical mode decomposition module, a data set construction module, a model training module, and a waste incineration industry nitrogen oxide emission prediction module; The data acquisition module is used to acquire continuous high-frequency nitrogen oxide emission data and related flue gas parameter data and production monitoring point data; The data preprocessing module is used to preprocess all acquired data and remove abnormal data; The feature importance analysis module is used to perform feature importance analysis on all pre-processed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry; The empirical mode decomposition module is used to perform empirical mode decomposition on key features closely related to nitrogen oxide emissions in the waste incineration industry and continuous nitrogen oxide emission data in the waste industry to obtain intrinsic mode function components and residual components; The data set construction module is used to construct a training set and a validation set using the obtained intrinsic mode function components and residual components; The module training module is used to train the LSTM model according to the training set to obtain a trained LSTM model; The waste incineration industry nitrogen oxide emission prediction is used to use the trained LSTM model to predict the data in the validation set, and compare it with the continuous waste industry nitrogen oxide emission data to obtain the waste incineration industry nitrogen oxide emission prediction result.

[0016] In a third aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method for predicting nitrogen oxide emissions in the waste incineration industry when executing the computer program.

[0017] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for predicting nitrogen oxide emissions in the waste incineration industry.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention is an improved invention. Compared with the existing method for predicting nitrogen oxide emissions in the waste incineration industry, on the one hand, the present invention first performs feature importance analysis on all preprocessed data when predicting nitrogen oxide emissions in the waste incineration industry, and obtains key features closely related to nitrogen oxide emissions in the waste incineration industry. Compared with the traditional feature selection method based on information gain and mutual information, it can more accurately screen out features that have a key impact on nitrogen oxide emission prediction, thereby improving the accuracy of nitrogen oxide emission prediction, thereby effectively solving the problem of low accuracy of nitrogen oxide emission prediction in the waste incineration industry in the prior art.

[0019] Furthermore, the present invention also discloses the steps of preprocessing all acquired data and clearing abnormal data, firstly, preprocessing all acquired data, clearing abnormal data, obtaining data after clearing abnormalities, and then using emission dynamic balance algorithm to perform data stationarity processing on the data after clearing abnormalities, to obtain stationary data. By performing data stationarity processing on the data after clearing abnormalities, data fluctuation can be reduced, data stationarity can be improved, and thus the accuracy of nitrogen oxide emission prediction in the waste incineration industry can be improved.

[0020] Furthermore, the present invention also discloses real-time monitoring and dynamic adjustment of the original stability data until the overall fluctuation size of all acquired data reaches the expected fluctuation range to obtain the final stability data. Real-time monitoring and dynamic adjustment of the original stability data can ensure that the data remains stable in a complex and changeable waste incineration environment, thereby providing reliable data for subsequent nitrogen oxide emission predictions. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of the method for predicting nitrogen oxide emissions in the waste incineration industry of the present invention; Figure 2 A schematic diagram of a nitrogen oxide emission prediction system for a waste incineration industry according to the present invention; Figure 3 It is a schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0022] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0023] The method for predicting nitrogen oxide emissions in the waste incineration industry proposed in the present invention uses a trained LSTM model to predict the data in the validation set and compares it with the continuous nitrogen oxide emission data of the waste industry to obtain the prediction result of nitrogen oxide emissions in the waste incineration industry. Compared with the prior art, the present invention effectively solves the problem of low accuracy of nitrogen oxide emission prediction in the waste incineration industry in the prior art.

[0024] Embodiment 1: The flowchart of the method for predicting nitrogen oxide emissions in the waste incineration industry of the present invention is as follows Figure 1 As shown, the method for predicting nitrogen oxide emissions in the waste incineration industry of the present invention comprises the following steps: S1. Obtain continuous high-frequency nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data.

[0025] S2. Preprocess all acquired data and remove abnormal data.

[0026] S3. Perform feature importance analysis on all preprocessed data to obtain key features that are closely related to nitrogen oxide emissions in the waste incineration industry.

[0027] S4. Perform empirical mode decomposition on the key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data of the waste industry to obtain the intrinsic mode function components and residual components.

[0028] S5. Use the obtained intrinsic mode function components and residual components to construct training sets and validation sets.

[0029] S6. Train the LSTM model according to the training set to obtain a trained LSTM model.

[0030] S7. Use the trained LSTM model to predict the data in the validation set and compare it with the continuous nitrogen oxide emission data of the garbage industry to obtain the prediction results of nitrogen oxide emissions in the garbage incineration industry.

[0031] Embodiment 2: The schematic diagram of the nitrogen oxide emission prediction system for the waste incineration industry of the present invention is as follows Figure 2 As shown, the waste incineration industry nitrogen oxide emission prediction system of the present invention includes a data acquisition module, a data preprocessing module, a feature importance analysis module, an empirical mode decomposition module, a data set construction module, a model training module and a waste incineration industry nitrogen oxide emission prediction module.

[0032] The data acquisition module is used to obtain continuous high-frequency nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data.

[0033] The data preprocessing module is used to preprocess all acquired data and remove abnormal data.

[0034] The feature importance analysis module is used to perform feature importance analysis on all preprocessed data to obtain key features that are closely related to nitrogen oxide emissions in the waste incineration industry.

[0035] The empirical mode decomposition module is used to perform empirical mode decomposition on key features closely related to nitrogen oxide emissions in the waste incineration industry and continuous nitrogen oxide emission data in the waste industry to obtain intrinsic mode function components and residual components.

[0036] The data set construction module is used to construct a training set and a validation set using the obtained intrinsic mode function components and residual components.

[0037] The module training module is used to train the LSTM model according to the training set to obtain a trained LSTM model.

[0038] The prediction of nitrogen oxide emissions in the waste incineration industry is used to use the trained LSTM model to predict the data in the validation set and compare it with the continuous nitrogen oxide emission data of the waste industry to obtain the prediction results of nitrogen oxide emissions in the waste incineration industry.

[0039] Embodiment 3: The method for predicting nitrogen oxide emissions in the waste incineration industry of the present invention comprises the following steps: S1. Obtain continuous high-frequency nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data.

[0040] First, continuous high-frequency nitrogen oxide emission data and related flue gas parameter data and production monitoring point data are obtained (continuous means that data is obtained every minute, and the data obtained is continuous). Among them, the relevant flue gas parameter data include flue gas temperature, flue gas pressure, flue gas flow rate and flue gas oxygen content. Production monitoring point data include equipment operation status, equipment liquid level data and equipment power data.

[0041] This embodiment specifically adopts CEMS (Continuous Emission Monitoring Systems) to obtain continuous high-frequency nitrogen oxide emission data and related flue gas parameter data, and specifically adopts DCS (Distributed Control System) to obtain production monitoring point data.

[0042] S2. Preprocess all acquired data and remove abnormal data.

[0043] Then, all acquired data are preprocessed to remove abnormal data, including data of shutdown, equipment failure, equipment maintenance, equipment calibration, high-frequency nitrogen oxide emission data, related flue gas parameter data, and maximum and minimum values ​​in production monitoring point data.

[0044] In this step, when preprocessing all the acquired data, first remove the abnormal data to obtain the data after the abnormalities are removed, and then use the emission dynamic balance algorithm to perform data stability processing on the data after the abnormalities are removed to obtain stable data. The specific steps of the emission dynamic balance algorithm include: A. Analyze the time series characteristics of all acquired data in real time to obtain several dynamic features; B. Normalize the obtained dynamic features; C. Combine several normalized dynamic features to obtain the overall fluctuation size of all acquired data; D. Perform data stability processing on the overall fluctuation size of all acquired data to obtain the original stable data; E. Monitor and dynamically adjust the original stability data in real time until the overall fluctuation of all acquired data reaches the expected fluctuation range to obtain the final stability data.

[0045] In step A, several dynamic features include trend change values ​​(using ), periodic fluctuation value (using ), noise level (in ) and real-time data change rate (using The relevant descriptions of trend change value, periodic fluctuation value, noise level and real-time data change rate are as follows: Trend change value: By analyzing long time series data, it is determined whether nitrogen oxide emissions are showing an upward, downward or stable trend, providing a basis for predicting the future trend of nitrogen oxide emissions in the waste incineration industry.

[0046] Periodic fluctuation value: Determining whether there are regular periodic changes in the data, such as daily cycles, weekly cycles, or monthly cycles, helps to discover the periodic activity patterns of emission sources.

[0047] Noise level: Evaluate the noise content in the data, remove or reduce the interference of noise on data analysis, and improve the reliability and accuracy of the data.

[0048] Real-time data change rate: Calculate the rate of change of data at each time point to promptly detect rapid changes in emission data so that appropriate measures can be taken.

[0049] The calculation formula for normalizing the obtained dynamic features in step B is:

[0050] in, for t Several dynamic features after time normalization, for t Several dynamic features before time normalization, is the mean value of each dynamic feature, is the standard deviation.

[0051] Step B normalizes the obtained dynamic features (dynamic features are also called factors) to make the value of each dynamic feature fall within a similar scale range, thereby eliminating the differences in dimensions and value ranges between different factors.

[0052] The calculation formula for the overall fluctuation of all data obtained in step C is:

[0053] in, The overall fluctuation size of all data obtained is the weight factor corresponding to each dynamic feature, for t The trend change value after normalization at each moment, for t The periodic fluctuation value after normalization at each moment, for t The noise level after normalization at each moment, for t The real-time data change rate after normalization, OF is the operating condition, which comprehensively considers the impact of multiple factors such as equipment operation status, production process and environmental conditions on nitrogen oxide emissions in the waste incineration industry, and analyzes the operating condition as an important dynamic feature. , , , and for t Dynamically adjust parameters at all times.

[0054] This embodiment The value range of is [0.10,0.30], The value range of is [0.10,0.25], The value range of is [0.10,0.25], The value range of is [0.10,0.30], The value range of is [0.15,0.40]. , , , and Select one value from the range of values ​​as the weight factor corresponding to each dynamic feature. , , , and The sum of the five weight factors can be equal to 1.

[0055] This embodiment The preferred values ​​are as follows: This embodiment The value is 0.25 The value is 0.20 The value is 0.20 The value of is 0.20, The value of is 0.15, It can be adjusted appropriately according to actual needs.

[0056] This embodiment The value of is 0 or 1. Under normal working conditions, The value of is 1. In other working conditions (such as failure of the incinerator or stopping the feeding of garbage into the incinerator until the garbage in the furnace is completely burned out, etc.), The value of is 0.

[0057] This step uses the volatility index to t Dynamic adjustment coefficient at all times Adjusted, the volatility index ( , VI) is calculated as:

[0058] in, is the standard deviation of all the data obtained, is the absolute value of the change rate of all acquired data, The volatility index.

[0059] This step is t Dynamic adjustment coefficient at all times The specific instructions for adjustment are as follows: according to Adjust the value of t Dynamic adjustment coefficient at all times , , , and ; when When it increases (indicating increased volatility), it increases , reduce This adjustment method can highlight the impact of trend change value, periodic fluctuation value, noise level and real-time data change rate on the overall fluctuation when the fluctuation is large, and relatively reduce the influence weight of the working condition, so that the overall fluctuation size of all acquired data can better reflect the actual fluctuation of the data.

[0060] when When it decreases (indicating a decrease in volatility), , , and ,Increase When the data fluctuation is small, appropriately reduce the influence of trend change value, periodic fluctuation value, noise level and real-time data change rate, and relatively increase the weight of the working condition, so as to ensure the rationality of the calculation of the overall fluctuation size of all acquired data.

[0061] Specifically, when When it is greater than 1.5 (also called high volatility), The value of is 1.2. The value of is 0.7.

[0062] when When it is greater than 0.8 and less than or equal to 1.5 (also called moderate fluctuation), and The value of is 1.

[0063] when When it is less than or equal to 0.8 (also called low volatility), The value of is 0.7. The value of is 1.2.

[0064] The calculation formula for processing the data stability of the overall fluctuation size of all acquired data in step D is:

[0065] in, for t The data after the moment data stationarity processing, i is a natural number, is the size of the smoothing window, which is used to control the number of data points involved in the smoothing calculation. is the smoothing coefficient, used to adjust the overall fluctuation The degree of influence on the smoothing process, for t The data before the moment data stationarity processing, j An index, indicating moving forward from the current time j time period, for t The overall fluctuation of all data obtained at all times.

[0066] The larger the value, the greater the impact of the overall fluctuation of all acquired data on the smoothing effect.

[0067] The specific processing process of step E is described as follows: During the real-time monitoring and dynamic adjustment of the original stationary data, the size of the smoothing window is continuously adjusted. , adjust the smoothing coefficient , dynamically adjusted according to real-time data changes , until the overall fluctuation size of all acquired data reaches the expected fluctuation range, and the final stationary data is obtained.

[0068] S3. Perform feature importance analysis on all preprocessed data to obtain key features that are closely related to nitrogen oxide emissions in the waste incineration industry.

[0069] Specifically, in this step, the emission pattern mining algorithm is used to perform feature importance analysis on all pre-processed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry (in this embodiment, 20 key features closely related to nitrogen oxide emissions in the waste incineration industry are obtained). The specific steps of the emission pattern mining algorithm include: a. Construct a tree model for emission pattern mining; b. Using all preprocessed data to train the tree model for emission pattern mining, to obtain a trained tree model for emission pattern mining; c. Calculate the SHAP (SHapley AdditiveexPlanations) value based on the tree model mined from the trained emission pattern; d. According to the calculated SHAP value, determine the contribution of the preprocessed data to the prediction results of nitrogen oxide emissions in the waste incineration industry, sort the contribution sizes, and obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry.

[0070] The tree model for emission pattern mining in step a is based on the dynamic modeling technology of time series data, which can capture the pattern association between each feature and nitrogen oxide emissions in the waste incineration industry in real time, and then learn the complex nonlinear relationship between each feature and nitrogen oxide emissions in the waste incineration industry. The calculation formula of SHAP value in step c is:

[0071] in, is a feature subset, is the set of all features, and Contains and excludes features respectively The predicted value at i is a natural number, Indicates the first few features, Tree model functions for emission pattern mining.

[0072] The specific process of step d is as follows: According to the calculated SHAP value, the contribution of the preprocessed data to the prediction results of nitrogen oxide emissions from the waste incineration industry is determined, and the contribution sizes are sorted (specifically, this embodiment sorts the contribution sizes from large to small). This embodiment selects the data ranked in the top 20 in terms of contribution size as the key features closely related to nitrogen oxide emissions from the waste incineration industry.

[0073] S4. Perform empirical mode decomposition on the key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data of the waste industry to obtain the intrinsic mode function components and residual components.

[0074] The key features closely related to nitrogen oxide emissions in the waste incineration industry and the real nitrogen oxide emission data of the waste industry are subjected to empirical mode decomposition to obtain the intrinsic mode function (IMF) and residual components.

[0075] Among them, each IMF represents the signal characteristics on different time scales. The high-frequency IMF reflects the short-term fluctuations and noise components in the data, and the low-frequency IMF reflects the long-term trend and main change pattern of the data.

[0076] The specific steps of empirical mode decomposition include: Preprocessing: Preprocess the key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data to ensure the integrity and consistency of the data and provide a basis for subsequent empirical mode decomposition.

[0077] Extreme point detection: For each key feature closely related to nitrogen oxide emissions in the waste incineration industry, detect its local maximum and local minimum points. A local maximum point refers to a point whose value is greater than that of its adjacent points, while the opposite is true for a local minimum point.

[0078] Envelope construction: The cubic spline interpolation method is used to fit the local maximum points and local minimum points respectively to construct the upper envelope and the lower envelope.

[0079] Mean calculation and detail component extraction: Calculate the mean of the upper envelope and the lower envelope, subtract the mean from the continuous garbage industry nitrogen oxide emission data to obtain a detail component.

[0080] Sifting: Check whether the extracted detail component meets the conditions of the intrinsic mode function component, that is, the number of extreme points and zero crossing points is equal or differs by one, and the mean of the upper envelope and the lower envelope is zero. If not, take the detail component as a new signal and repeat the above steps until the conditions of the intrinsic mode function component are met.

[0081] Iterative decomposition: The extracted intrinsic mode function components are stripped from the continuous garbage industry nitrogen oxide emission data, and the above process is repeated for the residual components until the residual components become a monotonic function or a low-frequency signal.

[0082] S5. Use the obtained intrinsic mode function components and residual components to construct training sets and validation sets.

[0083] The obtained intrinsic mode function components and residual components are used to construct a training set and a validation set, and 80% of the obtained intrinsic mode function components and residual components are used as a training set, and 20% are used as a validation set.

[0084] In order to improve the generalization ability and robustness of the LSTM (Long Short-Term Memory) model in the future, this embodiment adopts the time series cross-validation method in the training set, divides the training set into multiple subsets, and uses one of the subsets as the validation set in turn, and the remaining subsets as the training set, and performs multiple LSTM model training and verification.

[0085] S6. Train the LSTM model according to the training set to obtain a trained LSTM model.

[0086] The LSTM model consists of an input gate, a forget gate, an output gate, and a memory unit, which can capture long-term and short-term dependencies in time series data.

[0087] Specifically, after many iterations and experiments, the input dimension of the LSTM model in this step is 390, the number of hidden layer neurons is 128, the number of hidden layers is 2, and the output dimension is 1, which corresponds to the predicted value of nitrogen oxide emissions in the 5th minute.

[0088] In this step, the Adam optimization algorithm is used to train the LSTM model. The Adam optimization algorithm is a variant of stochastic gradient descent (SGD). It combines the concepts of momentum and RMSprop, and the LSTM model can converge faster. In this step, the learning rate of the LSTM model is set to 0.001, and L2 regularization is used to prevent the LSTM model from overfitting.

[0089] S7. Use the trained LSTM model to predict the data in the validation set and compare it with the continuous nitrogen oxide emission data of the garbage industry to obtain the prediction results of nitrogen oxide emissions in the garbage incineration industry.

[0090] The indicator used in this step to evaluate the prediction results of nitrogen oxide emissions from the waste incineration industry is the mean absolute percentage error.

[0091] The mean absolute percentage error is calculated as:

[0092] Where MAPE is the mean absolute percentage error, is the sample size, For the i The true value of the samples, For the i Sample prediction values, For the i The absolute prediction error percentage for each sample.

[0093] In order to verify the effectiveness of the method for predicting nitrogen oxide emissions in the waste incineration industry proposed in the present invention, this embodiment uses four pollutants (NOx, CO, Dust and HCl) from the Zengcheng Line 5 data of a South China Institute project for prediction. This embodiment uses the mean absolute percentage error to evaluate the prediction results of nitrogen oxide emissions in the waste incineration industry, and the evaluation results are shown in Table 1.

[0094] Table 1 Evaluation results of mean absolute percentage error

[0095] The data in Table 1 show that the average absolute percentage errors of the four pollutants (NOx, CO, Dust and HCl) are relatively small, and the average absolute percentage errors of the four pollutants (NOx, CO, Dust and HCl) are all less than 100%, which verifies the effectiveness of the prediction method of the present invention.

[0096] Embodiment 4: See also Figure 3 As shown, the present invention also provides an electronic device 100 for a method for predicting nitrogen oxide emissions in a waste incineration industry; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0097] The memory 101 can be used to store the computer program 103, and the processor 102 implements the steps of the method for predicting nitrogen oxide emissions in the waste incineration industry described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0098] The at least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0099] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for predicting nitrogen oxide emissions in the waste incineration industry, and the processor 102 can execute the plurality of instructions to implement: Obtain continuous high-frequency nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data; Preprocess all acquired data and remove abnormal data; The feature importance analysis was performed on all preprocessed data to obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry; The key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data of the waste industry are subjected to empirical mode decomposition to obtain intrinsic mode function components and residual components; The obtained intrinsic mode function components and residual components are used to construct training sets and validation sets; Train the LSTM model according to the training set to obtain a trained LSTM model; The trained LSTM model is used to predict the data in the validation set and compared with the continuous nitrogen oxide emission data of the garbage industry to obtain the prediction results of nitrogen oxide emissions in the garbage incineration industry.

[0100] Embodiment 5: If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0101] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting nitrogen oxide emissions in the waste incineration industry, characterized in that: The following steps are involved: Obtain continuous nitrogen oxide emission data and related flue gas parameter data as well as production monitoring point data; Preprocess all acquired data and remove abnormal data; The feature importance analysis was performed on all preprocessed data to obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry; The key features closely related to nitrogen oxide emissions in the waste incineration industry and the continuous nitrogen oxide emission data of the waste industry are subjected to empirical mode decomposition to obtain intrinsic mode function components and residual components; The obtained intrinsic mode function components and residual components are used to construct training sets and validation sets; Train the LSTM model according to the training set to obtain a trained LSTM model; The trained LSTM model is used to predict the data in the validation set and compared with the continuous nitrogen oxide emission data of the garbage industry to obtain the prediction results of nitrogen oxide emissions in the garbage incineration industry.

2. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 1 is characterized in that: In the step of preprocessing all acquired data and removing abnormal data, all acquired data are preprocessed, abnormal data are first removed to obtain data after the abnormalities are removed, and then the emission dynamic balance algorithm is used to perform data stability processing on the data after the abnormalities are removed to obtain stable data.

3. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 2 is characterized in that: The specific steps of the emission dynamic balance algorithm include: Analyze the time series characteristics of all acquired data in real time to obtain several dynamic features; Normalize the obtained dynamic features; Combine several normalized dynamic features to obtain the overall fluctuation of all acquired data; Perform data stationarity processing on the overall fluctuation size of all acquired data to obtain the original stationarity data; The original stability data is monitored in real time and adjusted dynamically until the overall fluctuation of all acquired data reaches the expected fluctuation range, thus obtaining the final stability data.

4. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 3 is characterized in that: In the step of combining the normalized dynamic features to obtain the overall fluctuation size of all acquired data, the calculation formula for the overall fluctuation size of all acquired data is: in, The overall fluctuation size of all data obtained is the weight factor corresponding to each dynamic feature, The sum of the five weight factors is equal to 1, for t The trend change value after normalization at each moment, for t The periodic fluctuation value after normalization at each moment, for t The noise level after normalization at each moment, for t The real-time data change rate after time normalization processing, For working conditions, The value of is 0 or 1. , , , (t) and for t Dynamically adjust parameters at all times. t Dynamically adjust parameters at all times , , , (t) and Set according to the volatility index; In the step of performing data stationarity processing on the overall fluctuation size of all acquired data, the calculation formula for performing data stationarity processing on the overall fluctuation size of all acquired data is: in, for t The data after the moment data stationarity processing, i is a natural number, is the size of the smoothing window, which is used to control the number of data points involved in the smoothing calculation. is the smoothing coefficient, used to adjust the overall fluctuation The degree of influence on the smoothing process, for t The data before the moment data stationarity processing, j An index, indicating moving forward from the current time j time period, for t The overall fluctuation of all data obtained at all times.

5. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 1, characterized in that: In the step of performing feature importance analysis on all preprocessed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry, an emission pattern mining algorithm is specifically used to perform feature importance analysis on all preprocessed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry.

6. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 5, characterized in that: The specific steps of the emission pattern mining algorithm include: Constructing a tree model for emission pattern mining; Using all preprocessed data to train the tree model for emission pattern mining, a trained tree model for emission pattern mining is obtained; Calculate the SHAP value based on the trained tree model for emission pattern mining; According to the calculated SHAP value, the contribution of all preprocessed data to the prediction results of nitrogen oxide emissions in the waste incineration industry is determined, and the contribution sizes are sorted to obtain the key features closely related to nitrogen oxide emissions in the waste incineration industry.

7. The method for predicting nitrogen oxide emissions in the waste incineration industry according to claim 1, characterized in that: In the step of constructing a training set and a validation set using the obtained intrinsic mode function components and residual components, specifically 80% of the obtained intrinsic mode function components and residual components are used as a training set, and 20% are used as a validation set.

8. A nitrogen oxide emission prediction system for the waste incineration industry, characterized in that: It includes data acquisition module, data preprocessing module, feature importance analysis module, empirical mode decomposition module, data set construction module, model training module and waste incineration industry nitrogen oxide emission prediction module; The data acquisition module is used to acquire continuous high-frequency nitrogen oxide emission data and related flue gas parameter data and production monitoring point data; The data preprocessing module is used to preprocess all acquired data and remove abnormal data; The feature importance analysis module is used to perform feature importance analysis on all pre-processed data to obtain key features closely related to nitrogen oxide emissions in the waste incineration industry; The empirical mode decomposition module is used to perform empirical mode decomposition on key features closely related to nitrogen oxide emissions in the waste incineration industry and continuous nitrogen oxide emission data in the waste industry to obtain intrinsic mode function components and residual components; The data set construction module is used to construct a training set and a validation set using the obtained intrinsic mode function components and residual components; The module training module is used to train the LSTM model according to the training set to obtain a trained LSTM model; The waste incineration industry nitrogen oxide emission prediction is used to use the trained LSTM model to predict the data in the validation set, and compare it with the continuous waste industry nitrogen oxide emission data to obtain the waste incineration industry nitrogen oxide emission prediction result.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for predicting nitrogen oxide emissions in the waste incineration industry according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting nitrogen oxide emissions in the waste incineration industry according to any one of claims 1 to 7 are implemented.

Images