CN117906376B - Method and system for monitoring carbon emission of rotary kiln - Google Patents
Method and system for monitoring carbon emission of rotary kiln Download PDFInfo
- Publication number
- CN117906376B CN117906376B CN202410302064.0A CN202410302064A CN117906376B CN 117906376 B CN117906376 B CN 117906376B CN 202410302064 A CN202410302064 A CN 202410302064A CN 117906376 B CN117906376 B CN 117906376B
- Authority
- CN
- China
- Prior art keywords
- value
- carbon dioxide
- target
- monitoring point
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 185
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 28
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 26
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims abstract description 270
- 229910002092 carbon dioxide Inorganic materials 0.000 claims abstract description 135
- 239000001569 carbon dioxide Substances 0.000 claims abstract description 135
- 230000002159 abnormal effect Effects 0.000 claims abstract description 38
- 238000012937 correction Methods 0.000 claims abstract description 23
- 238000002485 combustion reaction Methods 0.000 claims abstract description 12
- 230000005856 abnormality Effects 0.000 claims description 31
- 238000003062 neural network model Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 abstract description 10
- 230000000875 corresponding effect Effects 0.000 description 20
- 230000008859 change Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000000306 recurrent effect Effects 0.000 description 3
- 239000004568 cement Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Landscapes
- Regulation And Control Of Combustion (AREA)
Abstract
The invention relates to the field of carbon emission monitoring, in particular to a method and a system for monitoring carbon emission of a rotary kiln. The method comprises the following steps: acquiring carbon dioxide concentration values, wind direction included angles and position information of each monitoring point at different moments; calculating a first error value of the monitoring point; weighting the second correlation, and calculating a second error value of the monitoring point; acquiring an abnormal monitoring point and correcting a carbon dioxide concentration value of the abnormal monitoring point to obtain a carbon dioxide concentration correction value; and inputting the carbon dioxide concentration correction value sequences of all monitoring points at the same moment into a preset neural network, and judging the combustion state at the moment. By the technical scheme, the real-time performance and accuracy of the carbon emission monitoring of the rotary kiln can be improved.
Description
Technical Field
The present invention relates generally to the field of carbon emission monitoring. More particularly, the invention relates to a method and a system for monitoring carbon emission of a rotary kiln.
Background
The rotary kiln is a large heating device for cement production. The cement raw material needs to be burned in the rotary kiln, and a large amount of carbon dioxide gas and a small amount of other gases are discharged in the burning process. The carbon dioxide emission of the rotary kiln is monitored, and whether the carbon dioxide emission reaches the standard can be judged. The sensor can be interfered by electromagnetic factors, environment factors and the like, so that the sensor monitors that the carbon dioxide concentration value is abnormal.
The method for judging the abnormality of the monitoring data in the prior art comprises the following steps: and judging which monitoring point has abnormality in the carbon dioxide concentration value acquired by the sensor by comparing the carbon dioxide concentration values acquired by the sensors of the plurality of monitoring points. However, in the process of collecting the carbon dioxide concentration value, the sensors of different monitoring points are easily influenced by factors such as wind direction, the azimuth of the monitoring point relative to the rotary kiln and the like, so that the carbon dioxide concentration value collected by the sensors is greatly different from the real carbon dioxide concentration value, and the abnormal condition of the carbon dioxide concentration value cannot be judged directly by using a conventional method.
Disclosure of Invention
In order to solve one or more of the above technical problems, the present invention provides a method and a system for monitoring carbon emission of a rotary kiln.
In a first aspect, the invention discloses a method for monitoring carbon emission of a rotary kiln, which comprises the following steps: collecting a carbon dioxide concentration value and an air direction included angle of a target monitoring point at a target moment; at the target moment, taking a first difference absolute value between the true value and the predicted value as a first error value according to the true value of the acquired carbon dioxide concentration difference value of the target monitoring point and the calculated predicted value; calculating the influence degree of target monitoring points at the target moment, traversing all the monitoring points, and constructing a first influence degree sequence; constructing a first concentration sequence according to the carbon dioxide concentration values of all monitoring points at the target moment; calculating sequence correlation of the first influence degree sequence and the first concentration sequence as first correlation; deleting the carbon dioxide concentration value of the target monitoring point in the first concentration sequence to construct a second concentration sequence; deleting the influence degree of the target monitoring point in the first influence degree sequence, and constructing a second influence degree sequence; calculating a sequence correlation of the second influence sequence and the second concentration sequence as a second correlation; taking the product of the second difference absolute value of the first correlation and the second correlation and the first correlation as a second error value of the target monitoring point; calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment based on the first error value and the second error value, traversing all the monitoring points to obtain the carbon dioxide abnormality degree corresponding to each monitoring point, and responding to the carbon dioxide abnormality degree being larger than a preset threshold value to obtain the abnormal monitoring point; correcting the carbon dioxide concentration values of the abnormal monitoring points to obtain carbon dioxide concentration correction values, traversing all the abnormal monitoring points, constructing carbon dioxide concentration values corresponding to all the monitoring points at the same moment into a moment concentration sequence, traversing all the moments, and obtaining a moment concentration sequence corresponding to each moment; and inputting the time concentration sequence of the target time into a preset neural network model, and outputting a combustion state result of the target time.
In one embodiment, calculating the first error value comprises the steps of: acquiring an adjacent historical time period of a target time, calculating a difference value between a carbon dioxide concentration value of the target historical time and a carbon dioxide concentration value of a time before the target historical time in the adjacent historical time period as a true value of a carbon dioxide concentration difference value, and calculating a difference value between a wind direction included angle of the target historical time and a wind direction included angle of the time before the target historical time in the adjacent historical time period as a wind direction difference value, wherein the wind direction included angle is a minimum included angle between a wind direction monitored by a target monitoring point and a horizontal line of the rotary kiln; performing linear fitting on the carbon dioxide concentration difference value and the wind direction difference value in the adjacent historical time period based on a least square method to obtain a fitting straight line, wherein the wind direction difference is an independent variable, and the carbon dioxide concentration difference value is a dependent variable; inputting the wind direction difference of the target moment into a fitting straight line to obtain a predicted value of the carbon dioxide concentration difference value of the target moment; the absolute value of the first difference between the real value and the predicted value is taken as the first error value.
In one embodiment, the calculating the influence degree of the target monitoring point at the target moment satisfies the following steps: acquiring position information of a target monitoring point, wherein the position information comprises a linear distance between the target monitoring point and the rotary kiln and an azimuth angle of the target monitoring point relative to the rotary kiln; calculating the influence degree of the target monitoring point based on the position information of the target monitoring point and the wind direction included angle of the target monitoring point, wherein the influence degree of the target monitoring point meets the relation:
wherein, Represents the/>Time 1/>Influence degree of monitoring point, th/>The moment is the target moment, the/>The monitoring point is the target monitoring point,/>Represents the/>Linear distance between monitoring point and rotary kiln,/>Represents the/>Time 1/>Azimuth angle of monitoring point relative to rotary kiln,/>Represents the/>Time 1/>The wind direction included angle of the monitoring point.
In one embodiment, the first correlation satisfies the relationship:
wherein, Represents the/>First correlation of time of day,/>Represents the/>Time first concentration sequence,/>Represent the firstTime first influence sequence,/>Represents the/>Covariance of the first concentration sequence and the first influence sequence at the moment,Represents the/>Square difference of time first concentration sequence,/>Represents the/>The square difference of the time instant first influence degree sequence.
In one embodiment, the calculating the carbon dioxide abnormality degree of the target monitoring point at the target time includes the steps of: respectively normalizing the first error value and the second error value, and calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment based on the normalized first error value and second error value, wherein the carbon dioxide abnormality degree meets the relation:
wherein, Represents the/>Time 1/>Monitoring point carbon dioxide abnormality degree,/>Represents the/>Time 1/>First error value of monitoring point,/>Represents the/>Time 1/>And a second error value of the monitoring point.
In one embodiment, the obtaining the carbon dioxide concentration correction value includes the steps of: acquiring a carbon dioxide concentration value of each historical moment in a preset number of historical moments adjacent to the moment of the abnormal monitoring point, and averaging the carbon dioxide concentration values of each historical moment to be used as an adjacent concentration average; calculating a third difference absolute value of the adjacent concentration mean value and the carbon dioxide concentration value at each historical moment respectively, taking the percentage of the third difference absolute value to the adjacent concentration mean value as the weight of the corresponding historical moment, and calculating a carbon dioxide concentration correction value based on the carbon dioxide concentration value at each historical moment and the corresponding weight, wherein the carbon dioxide concentration correction value satisfies the relation:
wherein, Represents the/>Time carbon dioxide concentration correction value,/>A sequence number indicating the time at which the abnormality monitoring point is located,Represents the/>Carbon dioxide concentration value at time,/>Represents the/>Weights corresponding to historical moments,/>Front/>, representing moment of abnormality monitoring pointTime,/>Indicating the total number of historical moments.
In one embodiment, the preset neural network model includes an input layer, a hidden layer, and an output layer; the input layer is used for receiving the input of the time concentration sequence at the target time, and the hidden layer is used for extracting the characteristics of the input information and inputting the extracted characteristics into the output layer to output a combustion state result label at the target time; the loss function of the preset neural network model uses cross entropy loss, a gradient descent algorithm is used for training the model, and network parameters of the preset neural network model are updated iteratively; and stopping updating when the neural network model reaches the set maximum training times or the network loss value is smaller than the set loss value, so as to obtain the trained preset neural network model.
In one embodiment, the time instant concentration sequence of the target time instant includes: in response to the absence of an abnormal monitoring point in the target moment, taking the element in the moment concentration sequence as a carbon dioxide concentration value corresponding to each normal monitoring point at the target moment; and responding to the existence of the abnormal monitoring points in the target time, wherein the elements in the time concentration sequence are the carbon dioxide concentration value corresponding to each normal monitoring point and the carbon dioxide concentration correction value corresponding to each abnormal monitoring point at the target time.
In a second aspect, the invention discloses a rotary kiln carbon emission monitoring system, comprising: a processor; and a memory storing computer instructions for a rotary kiln carbon emission monitoring method, which when executed by the processor, cause the apparatus to perform the rotary kiln carbon emission monitoring method.
The invention has the following technical effects:
According to the method, the influence of the wind direction and the azimuth of the monitoring point relative to the rotary kiln on the carbon dioxide concentration data collected by the sensor is analyzed, and the first error value and the second error value are calculated according to the analysis result. The accuracy of judging the carbon dioxide concentration abnormal data can be improved according to the first error value and the second error value, the carbon dioxide concentration value corresponding to the abnormal monitoring point is corrected, the accuracy of the whole data is improved, and the instantaneity and the accuracy of the carbon emission monitoring of the rotary kiln are further improved.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the invention are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
Fig. 1 is a flowchart of a method for monitoring carbon emission of a rotary kiln according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a system for monitoring carbon emission of a rotary kiln according to an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that when the terms "first," "second," and the like are used in the claims, the specification and the drawings of the present invention, they are used merely for distinguishing between different objects and not for describing a particular sequential order. The terms "comprises" and "comprising" when used in the specification and claims of the present invention are taken to specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The invention provides a method for monitoring carbon emission of a rotary kiln. As shown in fig. 1, a method for monitoring carbon emission of a rotary kiln includes steps S1 to S5, which are described in detail below.
S1, acquiring carbon dioxide concentration values, wind direction included angles and position information of each monitoring point at different moments.
It should be noted that, it is difficult to accurately monitor the carbon emission in the rotary kiln by using a single sensor, so that a plurality of sensors at different positions are selected for data acquisition.
In one embodiment, each monitoring point is provided with a carbon dioxide sensor and a magnetic wind direction sensor, an infrared carbon dioxide sensor is used for collecting carbon dioxide concentration values, a magnetic wind direction sensor is used for collecting wind direction data, and an included angle between a wind taking direction and a horizontal direction is set to be smaller. The position information of the monitoring points is kept unchanged, namely the monitoring points are fixed, and the position information comprises the linear distance between the monitoring points and the rotary kiln and the azimuth angle of the monitoring points relative to the rotary kiln.
S2, calculating a first error value of the monitoring point.
Under the non-ideal state in the same raw material combustion ratio and within a specified time: the wind direction changes continuously, and the wind direction data in different time points can influence the concentration of carbon dioxide. In order to eliminate the influence factors of the wind direction difference of different time points on the carbon dioxide concentration change so as to accurately judge whether the carbon dioxide concentration of each monitoring point at each moment is abnormal or not, the invention analyzes the change of the carbon dioxide concentration caused by the wind direction difference of each monitoring point at each time point and the adjacent time point to obtain a first error of the carbon dioxide concentration value.
In one embodiment, an adjacent historical time period of the target time is obtained, a difference value between a carbon dioxide concentration value of the target historical time and a carbon dioxide concentration value of a time before the target historical time in the adjacent historical time period is calculated to be a true value of a carbon dioxide concentration difference value, and the adjacent historical time period of the target time can be set by a person skilled in the art, and the invention is set to 10, namely, a carbon dioxide concentration value and a wind direction included angle of each time in the first 10 times adjacent to the target time are selected, wherein the wind direction included angle is a minimum included angle between a wind direction monitored by a target monitoring point and a horizontal line of the rotary kiln. Calculating a difference value between a wind direction included angle at a target historical moment and a wind direction included angle at a moment before the target historical moment in an adjacent historical time period as a wind direction difference value; performing linear fitting on the carbon dioxide concentration difference value and the wind direction difference value in the adjacent historical time period based on a least square method to obtain a fitting straight line, wherein the wind direction difference is an independent variable, and the carbon dioxide concentration difference value is a dependent variable; inputting the wind direction difference of the target moment into a fitting straight line to obtain a predicted value of the carbon dioxide concentration difference value of the target moment; the absolute value of the first difference between the real value and the predicted value is taken as the first error value.
S3, weighting the second correlation, and calculating a second error value of the monitoring point.
In one embodiment, the influence degree of the target monitoring point at the target moment is calculated based on the position information of the target monitoring point and the wind direction included angle of the target monitoring point, and the influence degree satisfies the relation:
wherein, Represents the/>Time 1/>Influence degree of monitoring point, th/>The moment is the target moment, the/>The monitoring point is the target monitoring point,/>Represents the/>Linear distance between monitoring point and rotary kiln,/>Represents the/>Time 1/>Azimuth angle of monitoring point relative to rotary kiln,/>Represents the/>Time 1/>The wind direction included angle of the monitoring point.
And traversing all monitoring points at the target moment to construct a first influence sequence.
According to the carbon dioxide concentration values of all monitoring points at the target moment, a first concentration sequence is constructed, the sequence correlation of the first influence degree sequence and the first concentration sequence is calculated to be used as a first correlation, and the first correlation satisfies the relation:
wherein, Represents the/>First correlation of time of day,/>Represents the/>Time first concentration sequence,/>Represent the firstTime first influence sequence,/>Represents the/>Covariance of the first concentration sequence and the first influence sequence at the moment,Represents the/>Square difference of time first concentration sequence,/>Represents the/>The square difference of the time instant first influence degree sequence.
When (when)When the first concentration sequence and the first influence degree sequence are positively correlated, the closer the first correlation is to 1, the greater the degree of influence of the first concentration sequence by the first influence degree sequence, namely the greater the possibility of abnormality in the first concentration sequence, whereas the closer the first correlation is to 0, the lesser the degree of influence of the first concentration sequence by the first influence degree sequence, namely the lesser the possibility of abnormality in the first concentration sequence.
Deleting the carbon dioxide concentration value of the target monitoring point in the first concentration sequence, constructing a second concentration sequence, deleting the influence degree of the target monitoring point in the first influence degree sequence, constructing a second influence degree sequence, calculating the sequence correlation of the second influence degree sequence and the second concentration sequence as a second correlation, and taking the product of the absolute value of the second difference value of the first correlation and the second correlation and the first correlation as a second error value of the target monitoring point.
And traversing all the monitoring points to obtain a second error value corresponding to each monitoring point.
S4, acquiring an abnormal monitoring point and correcting the carbon dioxide concentration value of the abnormal monitoring point to obtain a carbon dioxide concentration correction value.
In one embodiment, the first error value reflects the degree of abnormality in the timing of the carbon dioxide concentration of the target monitoring point and the second error value reflects the degree of abnormality in the carbon dioxide concentration of the target monitoring point at the target time as compared to the carbon dioxide concentrations of the other monitoring points at the target time.
Based on the normalized first error value and the normalized second error value, calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment, wherein the carbon dioxide abnormality degree meets the relation:
wherein, Represents the/>Time 1/>Monitoring point carbon dioxide abnormality degree,/>Represents the/>Time 1/>First error value of monitoring point,/>Represents the/>Time 1/>And a second error value of the monitoring point.
Traversing all monitoring points to obtain the carbon dioxide abnormality degree corresponding to each monitoring point, and responding to the carbon dioxide abnormality degree being larger than a preset threshold value to obtain abnormal monitoring points, wherein the carbon dioxide concentration value monitored by the monitoring points is abnormal data when the carbon dioxide abnormality degree is larger than 0.5, and the monitoring points are abnormal monitoring points, and the preset threshold value can be set by a person skilled in the art.
The correction of the carbon dioxide concentration value of each abnormal monitoring point comprises the following steps:
S41, acquiring a carbon dioxide concentration value of each history time in a preset number of history times adjacent to the time of the abnormal monitoring point, averaging the carbon dioxide concentration values of each history time as an adjacent concentration average, and exemplarily, acquiring 3 history times adjacent to the time of the abnormal monitoring point, namely the first Time of day, th/>Time and/>At time, a carbon dioxide concentration value at each historical time is obtained, i.e./>Carbon dioxide concentration value at time/>First/>Carbon dioxide concentration value at time/>And/>Carbon dioxide concentration value at time/>And the moment/>, where the abnormal monitoring point is locatedNearest neighbor of the first and secondThe carbon dioxide concentration value at the moment is most closely related to the carbon dioxide concentration value at the moment of the abnormal monitoring point, and the like, and weights are given to the carbon dioxide concentration values at each historical moment according to different tight layers.
S42, calculating third difference absolute values of the adjacent concentration mean value and the carbon dioxide concentration value at each historical moment respectively, taking the percentage of the third difference absolute value to the adjacent concentration mean value as the weight of the corresponding historical moment, and calculating a carbon dioxide concentration correction value based on the carbon dioxide concentration value at each historical moment and the corresponding weight, wherein the carbon dioxide concentration correction value satisfies the relation:
wherein, Represents the/>Time carbon dioxide concentration correction value,/>A sequence number indicating the time at which the abnormality monitoring point is located,Represents the/>Carbon dioxide concentration value at time,/>Represents the/>Weights corresponding to historical moments,/>Front/>, representing moment of abnormality monitoring pointTime,/>Indicating the total number of historical moments.
And traversing all the abnormal monitoring points to obtain the carbon dioxide concentration correction value of each abnormal monitoring point.
S5, inputting the carbon dioxide concentration correction value sequences of all monitoring points at the same moment into a preset neural network, and judging the combustion state at the moment.
In one embodiment, the time concentration sequence of the target time is input into a preset neural network model, and the combustion state result of the target time is output.
After correction of the carbon dioxide concentration of all abnormal monitoring points is completed, all carbon dioxide concentration values at the same moment are built into a moment concentration sequence, and illustratively, in response to the fact that no abnormal monitoring point exists in the target moment, elements in the moment concentration sequence are carbon dioxide concentration values corresponding to each normal monitoring point in the target moment; and responding to the existence of the abnormal monitoring points in the target time, wherein the elements in the time concentration sequence are the carbon dioxide concentration value corresponding to each normal monitoring point and the carbon dioxide concentration correction value corresponding to each abnormal monitoring point at the target time.
Inputting a time concentration sequence of a target time into a preset neural network model, wherein the preset neural network model comprises an input layer, a hidden layer and an output layer; the input layer is used for receiving the input of the time concentration sequence of the target time, and the hidden layer is used for extracting the characteristics of the input information and then inputting the extracted characteristics into the output layer so as to output the combustion state result label of the target time.
Illustratively, the preset neural network may be RNN (Recurrent Neural Network loop neural network), LSTM (Long-Short Term Memory Long-short-term memory recurrent neural network), TCN (Temporal Convolutional Network time domain convolutional network), GRU (Gate Recurrent Unit gate loop unit), and the like, and the preset neural network of the present invention adopts RNN.
The training set of the preset neural network is a time concentration sequence corresponding to a plurality of historical time, and the labels are 0 and 1, wherein 0 represents that the combustion state is sufficient, 1 represents that the combustion state is insufficient, and particularly, carbon dioxide concentration data of each historical time monitoring point can be marked by a person skilled in the art according to actual conditions.
The loss function of the preset neural network uses cross entropy loss, a gradient descent algorithm training model is used, and network parameters of the preset neural network model are updated iteratively; and stopping updating when the neural network model reaches the set maximum training frequency or the network loss value is smaller than the set loss value, so as to obtain the trained preset neural network model, wherein the network loss value is smaller than 0.0001 or the training frequency reaches 200 times.
The real-time concentration sequence is input into a trained preset neural network, so that the combustion state result at the time can be directly obtained, and the monitoring of the carbon emission of the rotary kiln is completed.
The embodiment of the invention also discloses a system for monitoring the carbon emission of the rotary kiln, referring to fig. 2, comprising a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions realize the method for monitoring the carbon emission of the rotary kiln according to the invention when the computer program instructions are executed by the processor.
The above system further comprises other components well known to those skilled in the art, such as a communication bus and a communication interface, the arrangement and function of which are known in the art and therefore are not described in detail herein.
In the context of this patent, the foregoing memory may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium may be any suitable magnetic or magneto-optical storage medium, such as, for example, resistance change Memory RRAM (Resistive Random Access Memory), dynamic Random Access Memory DRAM (Dynamic Random Access Memory), static Random Access Memory SRAM (Static Random-Access Memory), enhanced dynamic Random Access Memory EDRAM (ENHANCED DYNAMIC Random Access Memory), high-Bandwidth Memory HBM (High-Bandwidth Memory), hybrid storage cube HMC (Hybrid Memory Cube), or any other medium that may be used to store the desired information and that may be accessed by an application, a module, or both. Any such computer storage media may be part of, or accessible by, or connectable to, the device.
While various embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Many modifications, changes, and substitutions will now occur to those skilled in the art without departing from the spirit and scope of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention.
The above embodiments are not intended to limit the scope of the present invention, so: all equivalent changes in structure, shape and principle of the invention should be covered in the scope of protection of the invention.
Claims (5)
1. The method for monitoring the carbon emission of the rotary kiln is characterized by comprising the following steps of:
Collecting a carbon dioxide concentration value and an air direction included angle of a target monitoring point at a target moment;
At the target moment, taking a first difference absolute value between the true value and the predicted value as a first error value according to the true value of the acquired carbon dioxide concentration difference value of the target monitoring point and the calculated predicted value;
Calculating the influence degree of target monitoring points at the target moment, traversing all the monitoring points, and constructing a first influence degree sequence;
constructing a first concentration sequence according to the carbon dioxide concentration values of all monitoring points at the target moment;
Calculating sequence correlation of the first influence degree sequence and the first concentration sequence as first correlation;
deleting the carbon dioxide concentration value of the target monitoring point in the first concentration sequence to construct a second concentration sequence;
Deleting the influence degree of the target monitoring point in the first influence degree sequence, and constructing a second influence degree sequence;
Calculating a sequence correlation of the second influence sequence and the second concentration sequence as a second correlation;
Taking the product of the second difference absolute value of the first correlation and the second correlation and the first correlation as a second error value of the target monitoring point;
Calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment based on the first error value and the second error value, traversing all the monitoring points to obtain the carbon dioxide abnormality degree corresponding to each monitoring point, and responding to the carbon dioxide abnormality degree being larger than a preset threshold value to obtain the abnormal monitoring point;
Correcting the carbon dioxide concentration values of the abnormal monitoring points to obtain carbon dioxide concentration correction values, traversing all the abnormal monitoring points, constructing carbon dioxide concentration values corresponding to all the monitoring points at the same moment into a moment concentration sequence, traversing all the moments, and obtaining a moment concentration sequence corresponding to each moment;
inputting a time concentration sequence of the target time into a preset neural network model, and outputting a combustion state result of the target time;
Calculating the first error value comprises the steps of:
Acquiring an adjacent historical time period of a target time, calculating a difference value between a carbon dioxide concentration value of the target historical time and a carbon dioxide concentration value of a time before the target historical time in the adjacent historical time period as a true value of a carbon dioxide concentration difference value, and calculating a difference value between a wind direction included angle of the target historical time and a wind direction included angle of the time before the target historical time in the adjacent historical time period as a wind direction difference value, wherein the wind direction included angle is a minimum included angle between a wind direction monitored by a target monitoring point and a horizontal line of the rotary kiln;
performing linear fitting on the carbon dioxide concentration difference value and the wind direction difference value in the adjacent historical time period based on a least square method to obtain a fitting straight line, wherein the wind direction difference is an independent variable, and the carbon dioxide concentration difference value is a dependent variable;
Inputting the wind direction difference of the target moment into a fitting straight line to obtain a predicted value of the carbon dioxide concentration difference value of the target moment;
Taking the absolute value of the first difference between the true value and the predicted value as a first error value;
the influence degree of the target monitoring point at the target moment is calculated to satisfy the following steps:
acquiring position information of a target monitoring point, wherein the position information comprises a linear distance between the target monitoring point and the rotary kiln and an azimuth angle of the target monitoring point relative to the rotary kiln;
calculating the influence degree of the target monitoring point based on the position information of the target monitoring point and the wind direction included angle of the target monitoring point, wherein the influence degree of the target monitoring point meets the relation:
wherein, Represents the/>Time 1/>Influence degree of monitoring point, th/>The moment is the target moment, the/>The monitoring point is the target monitoring point,/>Represents the/>Linear distance between monitoring point and rotary kiln,/>Represents the/>Time 1/>Azimuth angle of monitoring point relative to rotary kiln,/>Represents the/>Time 1/>The wind direction included angle of the monitoring point;
The step of calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment comprises the following steps:
Respectively normalizing the first error value and the second error value, and calculating the carbon dioxide abnormality degree of the target monitoring point at the target moment based on the normalized first error value and second error value, wherein the carbon dioxide abnormality degree meets the relation:
wherein, Represents the/>Time 1/>Monitoring point carbon dioxide abnormality degree,/>Represents the/>Time 1/>First error value of monitoring point,/>Represents the/>Time 1/>A second error value of the monitoring point;
the step of obtaining the carbon dioxide concentration correction value comprises the following steps:
Acquiring a carbon dioxide concentration value of each historical moment in a preset number of historical moments adjacent to the moment of the abnormal monitoring point, and averaging the carbon dioxide concentration values of each historical moment to be used as an adjacent concentration average;
calculating a third difference absolute value of the adjacent concentration mean value and the carbon dioxide concentration value at each historical moment respectively, taking the percentage of the third difference absolute value to the adjacent concentration mean value as the weight of the corresponding historical moment, and calculating a carbon dioxide concentration correction value based on the carbon dioxide concentration value at each historical moment and the corresponding weight, wherein the carbon dioxide concentration correction value satisfies the relation:
wherein, Represents the/>Time carbon dioxide concentration correction value,/>Sequence number indicating time of abnormal monitoring point,/>Represents the/>Carbon dioxide concentration value at time,/>Represents the/>Weights corresponding to historical moments,/>Front/>, representing moment of abnormality monitoring pointTime,/>Indicating the total number of historical moments.
2. The method for monitoring carbon emissions in a rotary kiln according to claim 1, wherein the first correlation satisfies a relationship:
wherein, Represents the/>First correlation of time of day,/>Represents the/>Time first concentration sequence,/>Represents the/>Time first influence sequence,/>Represents the/>Covariance of time instant first concentration sequence and first influence sequence,/>Represents the/>Square difference of time first concentration sequence,/>Represents the/>The square difference of the time instant first influence degree sequence.
3. The method for monitoring carbon emission of a rotary kiln according to claim 1, wherein the preset neural network model comprises an input layer, a hidden layer and an output layer;
the input layer is used for receiving the input of the time concentration sequence at the target time, and the hidden layer is used for extracting the characteristics of the input information and inputting the extracted characteristics into the output layer to output a combustion state result label at the target time;
The loss function of the preset neural network model uses cross entropy loss, a gradient descent algorithm is used for training the model, and network parameters of the preset neural network model are updated iteratively;
And stopping updating when the neural network model reaches the set maximum training times or the network loss value is smaller than the set loss value, so as to obtain the trained preset neural network model.
4. The method for monitoring carbon emission of a rotary kiln according to claim 1, wherein the time-of-day concentration sequence of the target time-of-day comprises:
In response to the absence of an abnormal monitoring point in the target moment, taking the element in the moment concentration sequence as a carbon dioxide concentration value corresponding to each normal monitoring point at the target moment;
And responding to the existence of the abnormal monitoring points in the target time, wherein the elements in the time concentration sequence are the carbon dioxide concentration value corresponding to each normal monitoring point and the carbon dioxide concentration correction value corresponding to each abnormal monitoring point at the target time.
5. A rotary kiln carbon emissions monitoring system, comprising:
A processor; and a memory storing computer instructions for a rotary kiln carbon emission monitoring method, which when executed by the processor, cause the apparatus to perform a rotary kiln carbon emission monitoring method according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410302064.0A CN117906376B (en) | 2024-03-18 | 2024-03-18 | Method and system for monitoring carbon emission of rotary kiln |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410302064.0A CN117906376B (en) | 2024-03-18 | 2024-03-18 | Method and system for monitoring carbon emission of rotary kiln |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117906376A CN117906376A (en) | 2024-04-19 |
| CN117906376B true CN117906376B (en) | 2024-05-24 |
Family
ID=90690876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410302064.0A Active CN117906376B (en) | 2024-03-18 | 2024-03-18 | Method and system for monitoring carbon emission of rotary kiln |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117906376B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2063035A1 (en) * | 1969-09-26 | 1971-07-02 | Gen Electric | |
| US4022569A (en) * | 1975-12-05 | 1977-05-10 | Alcan Research And Development Limited | Calcination of coke |
| JP2000130732A (en) * | 1998-10-30 | 2000-05-12 | Nagano Kogyo Kk | Failure diagnosys method for incinerator |
| KR20060119521A (en) * | 2005-05-20 | 2006-11-24 | 엘지전자 주식회사 | A control method from wrong using of gasoven range |
| CN102629104A (en) * | 2011-12-01 | 2012-08-08 | 燕山大学 | Calcination predictive control system and method for rotary cement kiln |
| CN103307873A (en) * | 2013-06-26 | 2013-09-18 | 中冶长天国际工程有限责任公司 | Method for detecting flue-gas temperature in rotary kiln |
| CN105605934A (en) * | 2015-12-10 | 2016-05-25 | 泰安中联水泥有限公司 | Natural gas kiln drying system |
| CN111868268A (en) * | 2018-03-19 | 2020-10-30 | 杰富意钢铁株式会社 | Molten metal component estimation device, molten metal component estimation method, and molten metal production method |
| CN114812162A (en) * | 2022-05-17 | 2022-07-29 | 天津水泥工业设计研究院有限公司 | Method and system for judging kiln conditions of rotary cement kiln and information data processing terminal |
| CN115681945A (en) * | 2022-11-10 | 2023-02-03 | 哈尔滨锅炉厂有限责任公司 | Online monitoring method for temperature of fire-facing side wall of supercritical carbon dioxide boiler hearth |
| CN115759450A (en) * | 2022-11-28 | 2023-03-07 | 广东省科学院江门产业技术研究院有限公司 | Carbon emission accounting early warning analysis system based on big data |
| CN116519555A (en) * | 2023-04-25 | 2023-08-01 | 山东圣文环保科技有限公司 | Dilution method flue gas monitoring system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114545866A (en) * | 2020-11-11 | 2022-05-27 | 台泥资讯股份有限公司 | Method for controlling coal consumption system |
-
2024
- 2024-03-18 CN CN202410302064.0A patent/CN117906376B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2063035A1 (en) * | 1969-09-26 | 1971-07-02 | Gen Electric | |
| US4022569A (en) * | 1975-12-05 | 1977-05-10 | Alcan Research And Development Limited | Calcination of coke |
| JP2000130732A (en) * | 1998-10-30 | 2000-05-12 | Nagano Kogyo Kk | Failure diagnosys method for incinerator |
| KR20060119521A (en) * | 2005-05-20 | 2006-11-24 | 엘지전자 주식회사 | A control method from wrong using of gasoven range |
| CN102629104A (en) * | 2011-12-01 | 2012-08-08 | 燕山大学 | Calcination predictive control system and method for rotary cement kiln |
| CN103307873A (en) * | 2013-06-26 | 2013-09-18 | 中冶长天国际工程有限责任公司 | Method for detecting flue-gas temperature in rotary kiln |
| CN105605934A (en) * | 2015-12-10 | 2016-05-25 | 泰安中联水泥有限公司 | Natural gas kiln drying system |
| CN111868268A (en) * | 2018-03-19 | 2020-10-30 | 杰富意钢铁株式会社 | Molten metal component estimation device, molten metal component estimation method, and molten metal production method |
| CN114812162A (en) * | 2022-05-17 | 2022-07-29 | 天津水泥工业设计研究院有限公司 | Method and system for judging kiln conditions of rotary cement kiln and information data processing terminal |
| CN115681945A (en) * | 2022-11-10 | 2023-02-03 | 哈尔滨锅炉厂有限责任公司 | Online monitoring method for temperature of fire-facing side wall of supercritical carbon dioxide boiler hearth |
| CN115759450A (en) * | 2022-11-28 | 2023-03-07 | 广东省科学院江门产业技术研究院有限公司 | Carbon emission accounting early warning analysis system based on big data |
| CN116519555A (en) * | 2023-04-25 | 2023-08-01 | 山东圣文环保科技有限公司 | Dilution method flue gas monitoring system |
Non-Patent Citations (3)
| Title |
|---|
| 加权数据融合算法在液化气监测中的应用;刘鸿飞;杨建栋;方丽英;王普;;南京理工大学学报;20160430;第40卷(第02期);全文 * |
| 回转窑煤粉高效低NO_x燃烧的最佳二次风温研究;徐顺生;黄日升;李罗军;邓波;;硅酸盐通报;20130831;第32卷(第08期);全文 * |
| 回转窑运行状态监测系统人机界面的设计;罗会铭;罗小燕;;江西教育学院学报;20071231;第28卷(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117906376A (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110441065B (en) | Gas turbine on-line detection method and device based on LSTM | |
| Lei et al. | Mutual information based anomaly detection of monitoring data with attention mechanism and residual learning | |
| CN115018021B (en) | Machine room abnormity detection method and device based on graph structure and abnormity attention mechanism | |
| CN111814956B (en) | Multi-task learning air quality prediction method based on multi-dimensional secondary feature extraction | |
| CN111931806A (en) | Equipment fault diagnosis method and device for multi-sensor data fusion | |
| CN116032557B (en) | Method and device for updating deep learning model in network security anomaly detection | |
| CN110618610A (en) | Gas turbine NOx standard exceeding early warning method and device based on LSTM deep learning | |
| CN105607631B (en) | The weak fault model control limit method for building up of batch process and weak fault monitoring method | |
| Zhao et al. | The application of BP neural networks to analysis the national vulnerability. | |
| WO2023063887A2 (en) | Method and apparatus for predicting state of wind turbine blade, and device and storage medium therefor | |
| CN115936262B (en) | Yield prediction method, system and medium based on big data environment interference | |
| CN117992895B (en) | Oil and gas pipeline area risk monitoring method and system based on big data | |
| CN114429308A (en) | Enterprise security risk assessment method and system based on big data | |
| CN107222169B (en) | A kind of photovoltaic array real time monitoring and fault detection method based on outlier detection | |
| CN117906376B (en) | Method and system for monitoring carbon emission of rotary kiln | |
| CN116485049B (en) | Electric energy metering error prediction and optimization system based on artificial intelligence | |
| CN115099493B (en) | Forest fire spreading rate prediction method in any direction based on CNN | |
| CN116205885A (en) | Abnormal state detection method, system, electronic device and readable storage medium | |
| CN114046816B (en) | Sensor signal fault diagnosis method based on lightweight gradient lifting decision tree | |
| CN112184037B (en) | Multi-modal process fault detection method based on weighted SVDD | |
| CN114266925B (en) | DLSTM-RF-based user electricity stealing detection method and system | |
| CN117686682B (en) | Indoor gas fault monitoring method and system | |
| CN117290792B (en) | Air pressure forecasting system and method based on machine learning | |
| Oluyomi et al. | Detection of False Data Injection in Smart Water Metering Infrastructure | |
| CN117134350A (en) | Ultra-short-term wind power segmentation prediction method and device under extreme weather condition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |