Thermal power plant denitration system inlet NOx concentration prediction method and device

Abstract

The embodiment of the invention provides a thermal power plant denitration system inlet NOx concentration prediction method and device considering measurement delay, and the method comprises the steps: determining an influence factor of the inlet NOx concentration of a denitration reactor, and collecting the inlet NOx concentration of the denitration reactor and inlet NOx concentration influence factor data; determining a measurement delay time of the inlet NOx concentration, and correcting the inlet NOx concentration; calculating a correlation coefficient between the NOx concentration of the inlet and each influence factor by adopting a Fourier transform correlation coefficient method, and determining an output variable and an input characteristic parameter variable used for prediction; according to the determined output variable and the input characteristic parameter variable, training and finishing the prediction model by adopting a machine learning algorithm based on physical constraint; and inlet NOx concentration influence factor data at the current moment are collected and input into the prediction model, and the inlet NOx concentration value at the current moment is obtained.

Description

Technical Field

[0001] This document relates to the technical field of flue gas denitration in coal-fired power plants, and particularly relates to a method and device for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay. Background Art

[0002] With the promulgation of a series of national policies and regulations, the emissions of air pollutants from coal-fired power plants have been strictly regulated. A large amount of air pollutants are generated during the coal combustion process, causing serious environmental pollution problems. NOx is one of the main pollutants emitted by coal-fired power plants. Its emission into the atmosphere will undergo a series of physical and chemical reactions to generate various harmful substances, causing great harm to the environment and human body. Accurately measuring the inlet NOx concentration value is of great significance for controlling the outlet NOx concentration emission. The accurate prediction of the real-time value of the inlet NOx has become the key to improving the control effect of the denitration system.

[0003] Currently, at home and abroad, the NOx components in flue gas are mainly measured in real time through a Continuous Emission Monitoring System (CEMS). However, this measurement method has disadvantages such as long data analysis time and serious lag in measurement value feedback, resulting in inaccurate measurement results of the inlet NOx. For this reason, many scholars have carried out research on predicting the denitration inlet NOx concentration. They often use mechanism models, that is, constructing physical models for calculation, and some scholars also use data-driven inlet NOx concentration prediction. However, there are certain problems. For example, when using mechanism models, the model establishment process is relatively complex, the process is long, and parameter definition is difficult. The data-driven model is a model completely based on data, and the model is completely limited by the collected operation data. On the one hand, when the data has noise, the model generalization ability is poor. On the other hand, the data is based on the data measured by CEMS, and this data has a certain delay, which will lead to the mismatch of the variable data time series selected at the same time. Predicting the inlet NOx concentration based on the delayed data is bound to be inaccurate.

[0004] Therefore, in the process of predicting the inlet NOx concentration, both mechanism models and data should be considered. At the same time, the delay in the data should also be removed. For this reason, considering the above factors comprehensively, carrying out the prediction of the denitration inlet NOx concentration is an urgent problem to be solved in the denitration control process of thermal power plants. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and device for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay, aiming to solve the above problems in the prior art.

[0006] The present invention provides a method for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay, including:

[0007] Step 101: Determine the influencing factors of the NOx concentration at the inlet of the denitration reactor, and collect the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet;

[0008] Step 102: Determine the measurement delay time of the inlet NOx concentration, and correct the inlet NOx concentration according to the measurement delay time to obtain the inlet NOx concentration without delay;

[0009] Step 103: Calculate the correlation coefficient between the inlet NOx concentration and each influencing factor by using the Fourier transform correlation coefficient method, obtain the delay time between each influencing factor and the inlet NOx concentration according to the correlation coefficient, correct each influencing factor according to the delay time to obtain the data of the influencing factors of the inlet NOx concentration without delay, and determine the output variable and input characteristic parameter variable used for prediction;

[0010] Step 104: Based on the inlet NOx concentration without delay and the data of the influencing factors of the inlet NOx concentration without delay, and according to the determined output variable and input characteristic parameter variable, complete the training of the prediction model by using a machine learning algorithm based on physical constraints;

[0011] Step 105: Collect the data of the influencing factors of the inlet NOx concentration at the current moment, correct it according to the calculated delay time, and input it into the prediction model to obtain the value of the inlet NOx concentration at the current moment.

[0012] The present invention provides a prediction device for the NOx concentration at the inlet of a thermal power plant denitration system considering measurement delay, including:

[0013] A collection module, configured to determine the influencing factors of the NOx concentration at the inlet of the denitration reactor, and collect the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet;

[0014] A correction module, configured to determine the measurement delay time of the inlet NOx concentration, and correct the inlet NOx concentration according to the measurement delay time to obtain the inlet NOx concentration without delay;

[0015] A calculation module, configured to calculate the correlation coefficient between the inlet NOx concentration and each influencing factor by using the Fourier transform correlation coefficient method, obtain the delay time between each influencing factor and the inlet NOx concentration according to the correlation coefficient, correct each influencing factor according to the delay time to obtain the data of the influencing factors of the inlet NOx concentration without delay, and determine the output variable and input characteristic parameter variable used for prediction;

[0016] A prediction model module is used to complete the training of the prediction model by using a machine learning algorithm based on physical constraints according to the determined output variables and input feature parameter variables based on the no-delay inlet NOx concentration and the no-delay inlet NOx concentration influencing factor data; collect the inlet NOx concentration influencing factor data at the current moment, correct it according to the calculated delay time, and input it into the prediction model to obtain the inlet NOx concentration value at the current moment.

[0017] An embodiment of the present invention further provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps of the above-mentioned method for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay are implemented.

[0018] An embodiment of the present invention further provides a computer-readable storage medium, on which an implementation program for information transmission is stored. When the program is executed by a processor, the steps of the above-mentioned method for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay are implemented.

[0019] By adopting the embodiment of the present invention, both the mechanism model and the operating data of the inlet NOx concentration influencing factors are considered in the prediction process. At the same time, the delay time of the inlet NOx concentration measurement itself and the delay time between the inlet NOx concentration and the influencing factors are also analyzed. The established model improves the accuracy of the inlet NOx modeling and has guiding significance for reducing pollutant emissions and costs of coal-fired units. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 It is a flowchart of the method for predicting the inlet NOx concentration of a denitration system in a thermal power plant considering measurement delay according to an embodiment of the present invention;

[0022] Figure 2 It is a schematic diagram of the process of constructing a machine learning algorithm based on physical constraints according to an embodiment of the present invention;

[0023] Figure 3 It is a schematic diagram of the prediction process of the inlet NOx concentration under all working conditions according to an embodiment of the present invention;

[0024] Figure 4It is a schematic diagram of a device for predicting the NOx concentration at the inlet of a denitration system in a thermal power plant considering measurement delay according to an embodiment of the present invention;

[0025] Figure 5 It is a schematic diagram of an electronic device according to an embodiment of the present invention. Specific embodiments

[0026] In order to enable those skilled in the art of the present technology to better understand the technical solutions in one or more embodiments of this specification, the following will clearly and completely describe the technical solutions in one or more embodiments of this specification in conjunction with the accompanying drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, rather than all embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this document.

[0027] Method embodiments

[0028] According to an embodiment of the present invention, there is provided a method for predicting the NOx concentration at the inlet of a denitration system in a thermal power plant considering measurement delay, Figure 1 It is a flowchart of a method for predicting the NOx concentration at the inlet of a denitration system in a thermal power plant considering measurement delay according to an embodiment of the present invention. As Figure 1 shown, the method for predicting the NOx concentration at the inlet of a denitration system in a thermal power plant considering measurement delay according to an embodiment of the present invention specifically includes:

[0029] Step S101, determine the influencing factors of the NOx concentration at the inlet of the denitration reactor, and collect the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet; specifically including:

[0030] According to the NOx generation mechanism of the flue gas in the thermal power plant, determine that the influencing factors of the inlet NOx concentration include: unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume;

[0031] At a predetermined sampling interval, collect the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet from the DCS system or SIS system of the thermal power plant within a predetermined past time.

[0032] Step S102, determine the measurement delay time of the inlet NOx concentration, and correct the inlet NOx concentration according to the measurement delay time to obtain the inlet NOx concentration without delay; specifically including:

[0033] The measurement delay time t1 is determined by CEMS measurement according to the inlet NOx concentration, and it includes the delay time t_sampling of the gas sampling system, the delay time t_transmission of the transmission pipeline, and the response time t_response of the analyzer. Among them, the delay time t_sampling of the gas sampling system is the delay caused by the sampling pipeline, sampling pump, filter, etc., and is determined by the ratio of the sampling pipeline length L1 to the flow rate V1 of the sampling pump, that is, t_sampling = L1 / V1. The delay time t_transmission of the transmission pipeline is the delay for the measurement signal to be transmitted from the sampling point to the instrument CEMS, and is determined by the signal transmission distance L2 and the transmission rate V2, that is, t_transmission = L2 / V2. The response time t_response of the analyzer is the time required for the CEMS analyzer instrument to change when the input signal changes;

[0034] The measurement delay time t1 is calculated according to Formula 1, and the inlet NOx concentration data collected at the current time t2 is corrected, that is, the inlet NOx concentration value at time t2 is corrected to the inlet NOx concentration value at time (t2 - t1):

[0035] t1 = t_sampling + t_transmission + t_response Formula 1.

[0036] Step S103, calculate the correlation coefficient between the inlet NOx concentration and each influencing factor by the Fourier transform correlation coefficient method, obtain the delay time between each influencing factor and the inlet NOx concentration according to the correlation coefficient, correct each influencing factor according to the delay time to obtain the data of the influencing factors of the inlet NOx concentration without delay, and determine the output variable and input characteristic parameter variable used for prediction; specifically including:

[0037] Let the inlet NOx concentration factor be X(t), and the influencing factors of the inlet NOx concentration be Y(t) = {y1(t), y2(t), y3(t), y4(t), y5(t), y6(t)}, where y1, y2, y3, y4, y5, y6 are the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume respectively, t is the index number, t = 1, 2,..., N, and N is the number of samples of the inlet NOx concentration collected. Then, the Fourier cross-correlation coefficient is determined according to Formula 2:

[0038] C XY (t) = F^-1{X(f)*(Y*(f))} Formula 2;

[0039] Among them, C XY(t) is the Fourier cross-correlation coefficient of X(t) and Y(t); F^{-1} is the inverse Fourier transform; X(f), Y(f) are the Fourier transforms of X(t) and Y(t); Y*(f) is the complex conjugate of Y(t); * is the conjugate of a complex number;

[0040] Respectively obtain the Fourier correlation coefficients of X(t) and Y(t) within the time period from 0 to t3, and obtain the moments t4, t5, t6, t7, t8, t9 corresponding to the maximum values of the Fourier correlation coefficients of y1, y2, y3, y4, y5, y6, which are the delay times of the inlet NOx concentration and its influencing factors;

[0041] According to the calculated delay times t4, t5, t6, t7, t8, t9, correct the data of the influencing factors of the inlet NOx concentration collected at the current moment T, that is, correct the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume at the moment T to the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume at the moments (T - t4), (T - t5), (T - t6), (T - t7), (T - t8), (T - t9);

[0042] Determine that the output variable used for prediction is the corrected inlet NOx concentration value, and the input variables are the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume.

[0043] Step S104, based on the non-delayed inlet NOx concentration and the non-delayed data of the influencing factors of the inlet NOx concentration, according to the determined output variable and input characteristic parameter variables, train the prediction model using a machine learning algorithm based on physical constraints; specifically including:

[0044] Use the Zeldovich mechanism model to construct the prediction model according to factors such as temperature, oxygen concentration, nitrogen concentration, pressure, reaction time, and activation energy;

[0045] Use the BP neural network model as the machine learning algorithm. Among them, the BP neural network model is a three-layer model including an input layer, a hidden layer, and an output layer. The input layer has six input variables, namely the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume. The output layer is the corrected inlet NOx concentration value. The parameters of the BP neural network model are the weights and thresholds between the input layer and the hidden layer, and the weights and thresholds between the hidden layer and the output layer. The correction of the above weights and thresholds is carried out through a global error function, and the global error function is defined as Formulas 3 - 5:

[0046] E(θ) = Data(θ) + λPhysics(θ) Formula 3;

[0047]

[0048]

[0049] Among them, E(θ) is the global error function, Data(θ) is the error function of the BP neural network model; Physics(θ) is the error function of the physical model; λ is the weight coefficient of the physical constraint term, and the λ is obtained by the empirical method or by the genetic optimization algorithm; y i is the actual value of the inlet NOx concentration; is the predicted value of the inlet NOx concentration by the BP neural network model; is the predicted value of the inlet NOx concentration by the physical model;

[0050] According to the minimization of E(θ), the optimal weights and thresholds between the input layer and the hidden layer and the weights and threshold parameters between the hidden layer and the output layer are obtained, and then the prediction model with the optimal physical constraint is determined.

[0051] Step S105: Collect the data of the influencing factors of the inlet NOx concentration at the current moment, correct them according to the calculated delay time, and input them into the prediction model to obtain the inlet NOx concentration value at the current moment. Collect the data of the influencing factors of the inlet NOx concentration at the current moment, correct them according to the calculated delay times t4, t5, t6, t7, t8, t9, and input them into the prediction model with the optimal physical constraint to obtain the inlet NOx concentration value at the current moment.

[0052] According to the specific embodiments provided by the present invention, the technical effects of the embodiments of the present invention are as follows:

[0053] (1) The present invention provides a method for calculating the measurement delay of the inlet NOx concentration, provides a method for eliminating the measurement delay of the NOx concentration from the perspective of the measurement principle, provides a more accurate real-time inlet NOx concentration measurement value, and can better match the real-time status of the inlet NOx concentration.

[0054] (2) The present invention also provides a method for calculating the delay between the influencing factors of the inlet NOx concentration and the inlet NOx concentration based on the Fourier correlation coefficient method. By transforming the time-domain data into frequency-domain data, the data characteristics at different frequencies are obtained, and the characteristics of the data are more prominent. This technology eliminates the delay between the inlet NOx concentration and its influencing factors, making the established model more matching and having higher accuracy.

[0055] (3) The present invention provides a machine learning algorithm based on physical constraints, which comprehensively considers the physical model and the machine learning algorithm model, taking into account both the operating data and the generation mechanism, avoiding the machine learning model from ignoring physical laws due to overfitting the training data, thereby improving the generalization ability and accuracy of the model, especially in the case of insufficient data or high noise. The established model has higher accuracy and the prediction results are more in line with the actual situation.

[0056] The technical solutions of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] As Figure 1 shown, the specific processing includes the following:

[0058] Step S101: According to the generation mechanism of NOx in the flue gas of the thermal power plant, determine the influencing factors of the NOx concentration at the inlet of the denitration reactor, and collect the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the inlet NOx concentration.

[0059] Step S201: According to the measurement principle of the NOx concentration at the inlet of the denitration reactor, determine the measurement delay time of the inlet NOx concentration, and correct the inlet NOx concentration accordingly to obtain the inlet NOx concentration data without delay.

[0060] Step S301: Use the Fourier transform correlation coefficient method to analyze the correlation coefficient between the inlet NOx concentration and each influencing factor. According to the calculated correlation coefficient, obtain the delay time between each influencing factor and the inlet NOx concentration, and correct each influencing factor to obtain the influencing factors of the inlet NOx concentration without delay, thereby determining the output variable and the input feature parameter variable used for prediction.

[0061] Step S401: The prediction model uses a machine learning algorithm based on physical constraints.

[0062] Step S501: Collect the influencing factors of the inlet NOx concentration at the current moment, correct them according to the calculated delay time, and input them into the machine learning algorithm model based on physical constraints to obtain the inlet NOx concentration value at the current moment.

[0063] Specifically, for a domestic 600MW subcritical pressure boiler unit, according to the analysis of the NOx generation mechanism in the flue gas of the thermal power plant, the main NOx influencing factors include unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume.

[0064] Specifically, the data of the inlet NOx concentration and its influencing factors are mainly collected from the operating data of the DCS system or SIS system of the thermal power plant in the past year, with a sampling interval of 1 minute, about 525,600 sample data.

[0065] Specifically, according to the CEMS measurement equipment used in this 600MW unit, the measurement delay time t1 is approximately 62.3s, and the inlet NOx concentration is corrected based on the t1 value.

[0066] Specifically, the delay time calculation of the influencing factors of the inlet NOx concentration adopts the correlation coefficient method of Fourier transform, and the correlation coefficients between the inlet NOx concentration and the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume within 0 - 20s (t3 is taken as 20s) are obtained respectively. Through calculation, the moments corresponding to the maximum correlation coefficients of the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume within 0 - 20s are 15s, 8s, 12s, 5s, 9s, and 18s respectively, and the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume are corrected according to these moments.

[0067] Combined with Figure 2 Specifically, the machine learning algorithm based on physical constraints mainly adds physical model constraints to the loss function of the machine learning algorithm.

[0068] The physical model constraint is mainly constructed by using the Zeldovich mechanism model, and the Zeldovich mechanism model is mainly built according to factors such as temperature, oxygen concentration, nitrogen concentration, pressure, reaction time, and activation energy.

[0069] The machine learning algorithm mainly adopts the BP neural network algorithm. The BP neural network model is a three - layer model including an input layer, a hidden layer, and an output layer. The input layer has six input variables, namely the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume. The output layer is the corrected inlet NOx concentration value. The parameters of the BP neural network are the weights and thresholds between the input layer and the hidden layer, and the weights and thresholds between the hidden layer and the output layer. The correction of the above - mentioned weights and thresholds is mainly carried out through the global error function, and the global error function is defined as:

[0070] E(θ) = Data(θ) + λPhysics(θ)

[0071] (Formula 3)

[0072] In the formula, E(θ) is the global error function, Data(θ) is the error function of the BP neural network model; Physics(θ) is the error function of the physical model; λ is the weight coefficient of the physical constraint term, and the λ can be obtained by the empirical method or through the genetic optimization algorithm.

[0073]

[0074] In the formula, y iis the actual value of the inlet NOx concentration; is the predicted value of the inlet NOx concentration by the BP neural network model.

[0075]

[0076] In the formula, y i is the actual value of the inlet NOx concentration; is the predicted value of the inlet NOx concentration by the physical model.

[0077] According to minimizing E(θ), the optimal weights and thresholds between the input layer and the hidden layer, and the weights and thresholds between the hidden layer and the output layer are obtained, and then the BP neural network model with the optimal physical constraints is determined.

[0078] Specifically, collect the influencing factors of the inlet NOx concentration at the current moment, correct them according to the calculated delay times t4, t5, t6, t7, t8, t9, and input them into the BP neural network model with the optimal physical constraints to obtain the inlet NOx concentration value at the current moment.

[0079] Reference Figure 3 , the present invention can also construct a prediction model for the NOx concentration at the denitration inlet under all working conditions. Considering the load increase, load decrease, and stable load conditions during the actual operation of the unit, in order to construct a prediction model under all working conditions, 3 working conditions need to be comprehensively considered. Specifically, in the collected data, the data under the 3 working conditions of load increase, load decrease, and stable load are collected respectively, and then the BP neural network models under the physical constraints of the 3 working conditions are established respectively. The establishment process is the same as that of Example 1. Example 2 provides a variety of working condition models.

[0080] When predicting the inlet NOx concentration at the current moment, it is necessary to collect the unit load data 5 minutes before and after this time period, judge whether it is in the load increase, load decrease, or stable load stage, and input them into the corresponding models respectively to obtain the predicted value of the inlet NOx concentration at the current moment.

[0081] The embodiment of the present invention provides a method for predicting the NOx concentration at the inlet of a thermal power plant denitration system considering measurement delay. Through the method of the present invention, the delay time of the inlet NOx modeling auxiliary variable relative to the inlet NOx is accurately calculated, the problem of the time series mismatch of the variable data selected by the inlet NOx and its influencing factors at the same time is solved, the pure delay effect is eliminated, the accuracy of the inlet NOx modeling is improved, and it has guiding significance for reducing pollutant emissions and costs of coal-fired units.

[0082] Device Embodiment 1

[0083] According to the embodiment of the present invention, there is provided a device for predicting the NOx concentration at the inlet of a thermal power plant denitration system considering measurement delay, Figure 4It is a schematic diagram of a device for predicting the NOx concentration at the inlet of a thermal power plant denitration system considering measurement delay according to an embodiment of the present invention. As Figure 4 shown, the device for predicting the NOx concentration at the inlet of a thermal power plant denitration system considering measurement delay according to an embodiment of the present invention specifically includes:

[0084] An acquisition module 40, configured to determine the influencing factors of the NOx concentration at the inlet of the denitration reactor, and acquire the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet; specifically for:

[0085] According to the NOx generation mechanism of the thermal power plant flue gas, it is determined that the influencing factors of the inlet NOx concentration include: unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume;

[0086] At a predetermined sampling interval, the NOx concentration at the inlet of the denitration reactor and the data of the influencing factors of the NOx concentration at the inlet within a past predetermined time are acquired from the DCS system or SIS system of the thermal power plant.

[0087] A calibration module 42, configured to determine the measurement delay time of the inlet NOx concentration, and calibrate the inlet NOx concentration according to the measurement delay time to obtain the inlet NOx concentration without delay; specifically for:

[0088] According to the CEMS measurement of the inlet NOx concentration, it is determined that the measurement delay time t1 includes the delay time t_sampling of the gas sampling system, the delay time t_transmission of the transmission pipeline, and the response time t_response of the analyzer. Among them, the delay time t_sampling of the gas sampling system is the delay caused by the sampling pipeline, sampling pump, and filter, etc., and is determined by the ratio of the sampling pipeline length L1 to the flow rate V1 of the sampling pump, that is, t_sampling = L1 / V1. The delay time t_transmission of the transmission pipeline is the delay of the measurement signal from the sampling point to the instrument CEMS, and is determined by the signal transmission distance L2 and the transmission rate V2, that is, t_transmission = L2 / V2. The response time t_response of the analyzer is the time required for the CEMS analyzer instrument to change when the input signal changes;

[0089] According to Formula 1, the measurement delay time t1 is calculated, and the NOx concentration data at the inlet at the current t2 moment is calibrated, that is, the NOx concentration value at the inlet at the t2 moment is calibrated to the NOx concentration value at the inlet at the (t2 - t1) moment:

[0090] t1 = t_sampling + t_transmission + t_response Formula 1;

[0091] A calculation module 44, configured to calculate the correlation coefficients between the inlet NOx concentration and various influencing factors by using the Fourier transform correlation coefficient method, obtain the delay times between various influencing factors and the inlet NOx concentration according to the correlation coefficients, correct various influencing factors according to the delay times to obtain the data of the influencing factors of the inlet NOx concentration without delay, and determine the output variables and input characteristic parameter variables used for prediction; specifically configured to:

[0092] Let the inlet NOx concentration factor be X(t), and the influencing factors of the inlet NOx concentration be Y(t) = {y1(t), y2(t), y3(t), y4(t), y5(t), y6(t)}, where y1, y2, y3, y4, y5, y6 are the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume respectively, t is the index number, t = 1, 2,..., N, and N is the number of samples of the inlet NOx concentration collected. Then, the Fourier cross-correlation coefficient is determined according to Formula 2:

[0093] C XY (t) = F^{-1}{X(f)*(Y*(f))} Formula 2;

[0094] Where, C XY (t) is the Fourier cross-correlation coefficient of X(t) and Y(t); F^{-1} is the inverse Fourier transform; X(f), Y(f) are the Fourier transforms of X(t) and Y(t); Y*(f) is the complex conjugate of Y(t); * is the conjugate of the complex number;

[0095] Respectively obtain the Fourier correlation coefficients of X(t) and Y(t) within the time period from 0 to t3, and obtain the times t4, t5, t6, t7, t8, t9 corresponding to the maximum values of the Fourier correlation coefficients of y1, y2, y3, y4, y5, y6, which are the delay times between the inlet NOx concentration and its influencing factors;

[0096] According to the calculated delay times t4, t5, t6, t7, t8, t9, correct the data of the influencing factors of the inlet NOx concentration collected at the current moment T, that is, correct the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume at the T moment to the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume at the moments of (T - t4), (T - t5), (T - t6), (T - t7), (T - t8), (T - t9);

[0097] Determine that the output variable used for prediction is the corrected value of the inlet NOx concentration, and the input variables are the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume;

[0098] A prediction model module 46, which is used to train the prediction model based on the delay-free inlet NOx concentration and the delay-free inlet NOx concentration influencing factor data, according to the determined output variable and input feature parameter variable, by using a machine learning algorithm based on physical constraints; collect the inlet NOx concentration influencing factor data at the current moment, correct it according to the calculated delay time, and input it into the prediction model to obtain the inlet NOx concentration value at the current moment. Specifically, it is used for:

[0099] Construct the prediction model by using the Zeldovich mechanism model according to factors such as temperature, oxygen concentration, nitrogen concentration, pressure, reaction time, and activation energy;

[0100] Use the BP neural network model as the machine learning algorithm. Among them, the BP neural network model is a three-layer model including an input layer, a hidden layer, and an output layer. The input layer is six input variables of the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and coal grinding volume. The output layer is the corrected inlet NOx concentration value. The parameters of the BP neural network model are the weights and thresholds between the input layer and the hidden layer, and the weights and thresholds between the hidden layer and the output layer. The correction of the above weights and thresholds is carried out through the global error function, and the global error function is defined as Formula 3 - Formula 5:

[0101] E(θ) = Data(θ) + λPhysics(θ) Formula 3;

[0102]

[0103]

[0104] Among them, E(θ) is the global error function, Data(θ) is the BP neural network model error function; Physics(θ) is the physical model error function; λ is the weight coefficient of the physical constraint term, and the λ is obtained by the empirical method or by the genetic optimization algorithm; y i is the actual value of the inlet NOx concentration; is the predicted value of the inlet NOx concentration by the BP neural network model; is the predicted value of the inlet NOx concentration by the physical model;

[0105] According to the minimization of E(θ), obtain the optimal weights and thresholds between the input layer and the hidden layer, and the weights and thresholds between the hidden layer and the output layer parameters, and then determine the prediction model with the optimal physical constraints;

[0106] Collect the data of the influencing factors of the inlet NOx concentration at the current moment, correct it according to the calculated delay times t4, t5, t6, t7, t8, and t9, and input it into the prediction model with the optimal physical constraints to obtain the inlet NOx concentration value at the current moment.

[0107] The embodiment of the present invention is a device embodiment corresponding to the above method embodiment. The specific operations of each module can be understood with reference to the description of the method embodiment and will not be elaborated here.

[0108] Device Embodiment Two

[0109] The embodiment of the present invention provides an electronic device, as Figure 5 shown, including: a memory 50, a processor 52, and a computer program stored on the memory 50 and executable on the processor 52. When the computer program is executed by the processor 52, it implements the steps described in the method embodiment.

[0110] Device Embodiment Three

[0111] The embodiment of the present invention provides a computer-readable storage medium. An implementation program for information transmission is stored on the computer-readable storage medium. When the program is executed by the processor 52, it implements the steps described in the method embodiment.

[0112] The computer-readable storage medium described in this embodiment includes, but is not limited to: ROM, RAM, magnetic disk, optical disk, etc.

[0113] 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting NOx concentration at the inlet of a thermal power plant denitrification system taking into account measurement delay, characterized in that: include: Step 101, determining the influencing factors of the NOx concentration at the inlet of the denitration reactor, and collecting the inlet NOx concentration of the denitration reactor and the influencing factors of the inlet NOx concentration data; Step 102, determining a measurement delay time of an inlet NOx concentration, and correcting the inlet NOx concentration according to the measurement delay time to obtain an inlet NOx concentration without delay; Step 103, using Fourier transform correlation coefficient method to calculate the correlation coefficient between the inlet NOx concentration and each influencing factor, obtaining the delay time between each influencing factor and the inlet NOx concentration according to the correlation coefficient, correcting each influencing factor according to the delay time, obtaining inlet NOx concentration influencing factor data without delay, and determining the output variable and input characteristic parameter variable used for prediction; Step 104, based on the undelayed inlet NOx concentration and the undelayed inlet NOx concentration influencing factor data, according to the determined output variable and the input characteristic parameter variable, a machine learning algorithm based on physical constraints is used to train and complete the prediction model; Step 105 , collecting data of factors affecting the inlet NOx concentration at the current moment, correcting the data according to the calculated delay time, and inputting the data into the prediction model to obtain the inlet NOx concentration value at the current moment.

2. The method according to claim 1, characterized in that Determine the factors affecting the NOx concentration at the inlet of the denitrification reactor, and collect the data of the NOx concentration at the inlet of the denitrification reactor and the factors affecting the NOx concentration at the inlet, including: According to the NOx generation mechanism of flue gas in thermal power plants, the factors affecting the inlet NOx concentration are determined to include: unit load, total air volume, total coal volume, total primary air volume, total secondary air volume and pulverized coal volume; At a predetermined sampling interval, the data of the inlet NOx concentration and the inlet NOx concentration influencing factors of the denitration reactor within a predetermined time in the past are collected from the DCS system or SIS system of the thermal power plant.

3. The method according to claim 1, characterized in that: Determining the measurement delay time of the inlet NOx concentration, and correcting the inlet NOx concentration according to the measurement delay time to obtain the inlet NOx concentration without delay specifically includes: According to the inlet NOx concentration measured by CEMS, the measurement delay time t1 is determined to include the delay time t_sampling of the gas sampling system, the delay time t_transmission of the transmission pipeline and the response time t_response of the analyzer, wherein the delay time t_sampling of the gas sampling system is the delay caused by the sampling pipeline, the sampling pump and the filter, and is determined by the ratio of the sampling pipeline length L1 to the flow rate V1 of the sampling pump, that is, t_sampling=L1 / V1, the delay time t_transmission of the transmission pipeline is the delay of the measurement signal from the sampling point to the instrument CEMS, and is determined by the signal transmission distance L2 and the transmission rate V2, that is, t_transmission=L2 / V2, and the response time t_response of the analyzer is the time required for the CEMS analyzer instrument to respond to changes in the input signal; The measurement delay time t1 is calculated according to formula 1, and the inlet NOx concentration data collected at the current time t2 is corrected, that is, the inlet NOx concentration value at time t2 is corrected to the inlet NOx concentration value at time (t2-t1): t1=t_sampling+t_transmission+t_response Formula 1.

4. The method according to claim 1, characterized in that: Step 103 specifically includes: Assume that the inlet NOx concentration factor is X(t), and the inlet NOx concentration influencing factor is Y(t)={y1(t), y2(t), y3(t), y4(t), y5(t), y6(t)}, wherein y1, y2, y3, y4, y5, y6 are unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume, respectively, t is the index number, t=1, 2, ..., N, N is the number of inlet NOx concentration samples collected, and the Fourier correlation coefficient is determined according to Formula 2: C XY (t) = F^{-1}{X(f)*(Y*(f))} Formula 2; Among them, C XY (t) is the Fourier correlation coefficient of X(t) and Y(t); F^{-1} is the inverse Fourier transform; X(f), Y(f) are the Fourier transforms of X(t) and Y(t); Y*(f) is the complex conjugate of Y(t); * is the conjugate of a complex number; The Fourier correlation coefficients of X(t) and Y(t) in the time range of 0-t3 are obtained respectively, and the time t4, t5, t6, t7, t8, t9 corresponding to the maximum value of the Fourier correlation coefficients of y1, y2, y3, y4, y5, y6 are obtained, which are the delay times between the inlet NOx concentration and its influencing factors; According to the calculated delay times t4, t5, t6, t7, t8, and t9, the data of factors affecting the inlet NOx concentration collected at the current time T are corrected, that is, the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume at the time T are corrected to the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume at the time (T-t4), (T-t5), (T-t6), (T-t7), (T-t8), and (T-t9), respectively; The output variable used for the prediction is determined to be the corrected inlet NOx concentration value, and the input variables are the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume and pulverized coal volume.

5. The method according to claim 1, characterized in that Step 104 specifically includes: The prediction model is constructed using the Zeldovich mechanism model based on factors such as temperature, oxygen concentration, nitrogen concentration, pressure, reaction time, and activation energy; A BP neural network model is used as a machine learning algorithm, wherein the BP neural network model is a three-layer model including an input layer, a hidden layer, and an output layer. The input layer is six input variables including corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume. The output layer is the corrected inlet NOx concentration value. The parameters of the BP neural network model are the weight and threshold between the input layer and the hidden layer, and the weight and threshold between the hidden layer and the output layer. The correction of the above weights and thresholds is corrected by a global error function, and the global error function is defined as Formula 3-Formula 5: E(θ)=Data(θ)+λPhysics(θ) Formula 3; Wherein, E(θ) is the global error function, Data(θ) is the BP neural network model error function; Physics(θ) is the physical model error function; λ is the weight coefficient of the physical constraint term, and the λ is obtained by empirical method or by genetic optimization algorithm; y i is the actual value of inlet NOx concentration; The predicted value of the inlet NOx concentration by BP neural network model; is the predicted value of the physical model for the inlet NOx concentration; By minimizing E(θ), the optimal weight and threshold between the input layer and the hidden layer, and the weight and threshold parameters between the hidden layer and the output layer are obtained, thereby determining the prediction model of the optimal physical constraint.

6. The method according to claim 5, characterized in that The step 105 specifically includes: The data of influencing factors of the inlet NOx concentration at the current moment are collected, and correction is performed according to the calculated delay times t4, t5, t6, t7, t8, and t9, and input into the prediction model of the optimal physical constraint to obtain the inlet NOx concentration value at the current moment.

7. A device for predicting NOx concentration at the inlet of a thermal power plant denitrification system taking into account measurement delay, characterized in that: include: A collection module is used to determine the influencing factors of the NOx concentration at the inlet of the denitrification reactor, and collect the inlet NOx concentration of the denitrification reactor and the influencing factors of the inlet NOx concentration data; A correction module, used to determine a measurement delay time of an inlet NOx concentration, and correct the inlet NOx concentration according to the measurement delay time to obtain an inlet NOx concentration without delay; a calculation module, used for calculating the correlation coefficient between the inlet NOx concentration and each influencing factor by using Fourier transform correlation coefficient method, obtaining the delay time between each influencing factor and the inlet NOx concentration according to the correlation coefficient, correcting each influencing factor according to the delay time, obtaining inlet NOx concentration influencing factor data without delay, and determining the output variable and input characteristic parameter variable used for prediction; A prediction model module is used to complete the prediction model based on the inlet NOx concentration without delay and the inlet NOx concentration influencing factor data without delay, according to the determined output variable and the input characteristic parameter variable, using a machine learning algorithm based on physical constraints to train; collect the inlet NOx concentration influencing factor data at the current moment, and correct it according to the calculated delay time, and input it into the prediction model to obtain the inlet NOx concentration value at the current moment.

8. The device according to claim 7, characterized in that The acquisition module is specifically used for: According to the NOx generation mechanism of flue gas in thermal power plants, the factors affecting the inlet NOx concentration are determined to include: unit load, total air volume, total coal volume, total primary air volume, total secondary air volume and pulverized coal volume; At a predetermined sampling interval, data on the inlet NOx concentration and inlet NOx concentration influencing factors of the denitrification reactor within a predetermined period of time in the past are collected from the DCS system or SIS system of the thermal power plant; The correction module is specifically used for: According to the inlet NOx concentration measured by CEMS, the measurement delay time t1 is determined to include the delay time t_sampling of the gas sampling system, the delay time t_transmission of the transmission pipeline and the response time t_response of the analyzer, wherein the delay time t_sampling of the gas sampling system is the delay caused by the sampling pipeline, the sampling pump and the filter, and is determined by the ratio of the sampling pipeline length L1 to the flow rate V1 of the sampling pump, that is, t_sampling=L1 / V1, the delay time t_transmission of the transmission pipeline is the delay of the measurement signal from the sampling point to the instrument CEMS, and is determined by the signal transmission distance L2 and the transmission rate V2, that is, t_transmission=L2 / V2, and the response time t_response of the analyzer is the time required for the CEMS analyzer instrument to respond to changes in the input signal; The measurement delay time t1 is calculated according to formula 1, and the inlet NOx concentration data collected at the current time t2 is corrected, that is, the inlet NOx concentration value at time t2 is corrected to the inlet NOx concentration value at time (t2-t1): t1=t_sampling+t_transmission+t_response Formula 1; The calculation module is specifically used for: Assume that the inlet NOx concentration factor is X(t), and the inlet NOx concentration influencing factor is Y(t)={y1(t), y2(t), y3(t), y4(t), y5(t), y6(t)}, wherein y1, y2, y3, y4, y5, y6 are unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume, respectively, t is the index number, t=1, 2, ..., N, N is the number of inlet NOx concentration samples collected, and the Fourier correlation coefficient is determined according to Formula 2: C XY (t) = F^{-1}{X(f)*(Y*(f))} Formula 2; Among them, C XY (t) is the Fourier correlation coefficient of X(t) and Y(t); F^{-1} is the inverse Fourier transform; X(f), Y(f) are the Fourier transforms of X(t) and Y(t); Y*(f) is the complex conjugate of Y(t); * is the conjugate of a complex number; The Fourier correlation coefficients of X(t) and Y(t) in the time range of 0-t3 are obtained respectively, and the time t4, t5, t6, t7, t8, t9 corresponding to the maximum value of the Fourier correlation coefficients of y1, y2, y3, y4, y5, y6 are obtained, which are the delay times between the inlet NOx concentration and its influencing factors; According to the calculated delay times t4, t5, t6, t7, t8, and t9, the data of factors affecting the inlet NOx concentration collected at the current time T are corrected, that is, the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume at the time T are corrected to the unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume at the time (T-t4), (T-t5), (T-t6), (T-t7), (T-t8), and (T-t9), respectively; Determine that the output variable used in the prediction is the corrected inlet NOx concentration value, and the input variables are the corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume and pulverized coal volume; The prediction model module is specifically used for: The prediction model is constructed using the Zeldovich mechanism model based on factors such as temperature, oxygen concentration, nitrogen concentration, pressure, reaction time, and activation energy; A BP neural network model is used as a machine learning algorithm, wherein the BP neural network model is a three-layer model including an input layer, a hidden layer, and an output layer. The input layer is six input variables including corrected unit load, total air volume, total coal volume, total primary air volume, total secondary air volume, and pulverized coal volume. The output layer is the corrected inlet NOx concentration value. The parameters of the BP neural network model are the weight and threshold between the input layer and the hidden layer, and the weight and threshold between the hidden layer and the output layer. The correction of the above weights and thresholds is corrected by a global error function, and the global error function is defined as Formula 3-Formula 5: E(θ)=Data(θ)+λPhysics(θ) Formula 3; Wherein, E(θ) is the global error function, Data(θ) is the BP neural network model error function; Physics(θ) is the physical model error function; λ is the weight coefficient of the physical constraint term, and the λ is obtained by empirical method or by genetic optimization algorithm; y i is the actual value of inlet NOx concentration; The predicted value of the inlet NOx concentration by BP neural network model; is the predicted value of the physical model for the inlet NOx concentration; According to minimizing E(θ), the optimal weight and threshold between the input layer and the hidden layer, and the weight and threshold parameters between the hidden layer and the output layer are obtained, thereby determining the prediction model of the optimal physical constraint; The data of influencing factors of the inlet NOx concentration at the current moment are collected, and correction is performed according to the calculated delay times t4, t5, t6, t7, t8, and t9, and input into the prediction model of the optimal physical constraint to obtain the inlet NOx concentration value at the current moment.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of a method for predicting the NOx concentration at the inlet of a thermal power plant denitrification system taking into account measurement delays as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by a processor, the steps of the method for predicting the NOx concentration at the inlet of a thermal power plant denitrification system taking into account measurement delay are implemented as described in any one of claims 1 to 6.

Images