CN115935784B - Analysis method and system for fuel data in combustion system of building ceramic kiln - Google Patents
Analysis method and system for fuel data in combustion system of building ceramic kiln Download PDFInfo
- Publication number
- CN115935784B CN115935784B CN202211230486.9A CN202211230486A CN115935784B CN 115935784 B CN115935784 B CN 115935784B CN 202211230486 A CN202211230486 A CN 202211230486A CN 115935784 B CN115935784 B CN 115935784B
- Authority
- CN
- China
- Prior art keywords
- opening
- combustion system
- data set
- data
- control valve
- 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
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P10/00—Technologies related to metal processing
- Y02P10/10—Reduction of greenhouse gas [GHG] emissions
- Y02P10/143—Reduction of greenhouse gas [GHG] emissions of methane [CH4]
Landscapes
- Regulation And Control Of Combustion (AREA)
Abstract
The invention relates to the technical field of combustion of building ceramic kiln, in particular to a method and a system for analyzing fuel data in a combustion system of a building ceramic kiln, wherein the method comprises the following steps: s1, acquiring an initial data set of a combustion system, and performing initial processing on the initial data set to obtain a modeling data set suitable for modeling; s2, optimizing the XGBoost model by adopting a GridSearchCV algorithm, and putting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model; s3, acquiring an online data set of the combustion system, and performing online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; s4, according to the opening difference predicted value and the real-time opening difference value of the combustion system, obtaining an opening analog quantity of a control valve of the combustion system in a gradual approach mode; s5, according to the opening analog quantity of the control valve of the combustion system, a flow valve control instruction is issued to the combustion system.
Description
Technical Field
The invention relates to the technical field of combustion of building ceramic kiln, in particular to a method and a system for analyzing fuel data in a combustion system of a building ceramic kiln.
Background
At present, the kiln in the building ceramic industry in China basically adopts natural gas as fuel, and along with the continuous rising of the price of the natural gas and the national double-carbon requirements, energy conservation and emission reduction become the main targets of the building ceramic producer for reducing the cost, but the natural gas combustion system of the traditional kiln has the following problems: not only has low utilization rate of natural gas, but also has low thermal efficiency and low energy utilization rate; and the air input is too high, so that even if the natural gas burns fully, the heat can be taken away by the redundant air, and the energy waste is caused.
Disclosure of Invention
The invention aims to provide a method and a system for analyzing fuel data in a combustion system of a building ceramic kiln, which integrate an automatic control principle of a gas control valve and a neural network learning model so as to comprehensively control the combustion system, thereby achieving the effects of energy conservation and emission reduction, reducing the manufacturing cost and improving the production efficiency.
To achieve the purpose, the invention adopts the following technical scheme:
a method for analyzing fuel data in a combustion system of a building ceramic kiln comprises the following steps:
s1, acquiring an initial data set of a combustion system, and performing initial processing on the initial data set to obtain a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
s2, optimizing the XGBoost model by adopting a GridSearchCV algorithm, and putting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
s3, acquiring an online data set of the combustion system, and performing online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
s4, according to the opening difference predicted value and the real-time opening difference value of the combustion system, obtaining an opening analog quantity of a control valve of the combustion system in a gradual approach mode; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow;
s5, according to the opening analog quantity of the control valve of the combustion system, a flow valve control instruction is issued to the combustion system.
Preferably, the combustion system comprises:
the natural gas phase analyzer is used for measuring the CH4 content;
the natural gas intelligent meter is used for measuring the natural gas flow X1;
the online oxygen content detection monitor is used for measuring the content of O2 and the relative humidity;
an air-gas smart meter for measuring an air gas flow X2;
the waste gas detector is used for measuring the content of CO;
an exhaust-gas smart meter for measuring an exhaust gas flow X3;
the intelligent control valve for the flow of the natural gas pipe is used for controlling the opening M1 of the flow of the natural gas pipe;
the air pipe flow intelligent control valve is used for controlling the opening M2 of the air pipe flow;
the initial data set and the online data set comprise CH4 content, natural gas flow X1, O2 content, relative humidity, air gas flow X2, CO content, waste gas flow X3, opening M1 of natural gas pipe flow and opening M2 of air pipe flow.
Preferably, in S1, the initial data set is initially processed to obtain a modeling data set suitable for modeling; the method specifically comprises the following steps:
s21, respectively exploring and analyzing historical equipment data and laboratory test data, and removing missing values and abnormal values after analysis to obtain correct historical equipment data and laboratory test data;
s22, carrying out association and combination processing on correct historical equipment data and laboratory test data through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
s23, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
s24, distributing the difference parameters and the corresponding combined data set into a group of modeling data, and taking a plurality of groups of modeling data as a modeling data set;
s25, randomly distributing the modeling data set, and dividing the modeling data set into a 70% training set and a 30% testing set.
Preferably, the difference formula is:
difference parameter = data at current time-data at last time.
Preferably, the step S3 is that the GridSearchCV algorithm is adopted to optimize the XGBoost model, modeling data is put into the optimized XGBoost model for training, and an opening difference prediction model is generated; the method specifically comprises the following steps:
s31, setting parameters of the tree model iteration number n_evators of the XGBoost model, the tree depth max_depth of the tree model and the loss function threshold gamma in splitting nodes to obtain multiple groups of parameters, and generating a corresponding prediction model;
s32, training a plurality of initial prediction models by adopting training sets in sequence, and obtaining corresponding prediction data accuracy through evaluation functions in sequence;
s33, selecting a prediction model corresponding to the highest prediction data accuracy, predicting the prediction model by adopting a test set, and obtaining the corresponding prediction data accuracy through an evaluation function;
s34, outputting the prediction model if the numerical value of the accuracy rate of the prediction data is more than 0.75 and less than 1; if the value of the accuracy of the predicted data is less than 0.75 or greater than 1, discarding the prediction model, and repeating S31 to 33 until a prediction model conforming to S34 is obtained.
Preferably, the evaluation function is:
where R represents the accuracy of the predicted data,represents the opening difference predicted value of the ith, y (i) represents the true data value of the ith,/->Representing the average of all real data values y.
Preferably, in S3, the acquiring an online data set of the combustion system, and performing online processing on the online data set to obtain a test data set suitable for testing; the method specifically comprises the following steps:
s31, carrying out association and combination processing on the online data set through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
s32, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
s33, distributing the difference parameter and the corresponding combined data set into a group of test data, and taking a plurality of groups of test data as the test data set.
Preferably, the step S4 is used for obtaining the opening analog quantity of the control valve of the combustion system in a gradually approaching mode according to the opening difference predicted value and the real-time opening difference value of the combustion system; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow; the method specifically comprises the following steps:
s41, judging a difference value between the opening difference predicted value and a real-time opening difference value of the combustion system, and acquiring an analog quantity of a control valve corresponding to the real-time opening difference value of the combustion system when the difference value is equal to 0; when the difference is not equal to 0, S42 is performed;
s42, judging whether the real-time opening difference value of the combustion system is larger than 0, and when the real-time opening difference value is larger than 0, controlling the opening analog quantity stepping of the intelligent control valve of the natural gas pipe to be 0.01% and controlling the opening analog quantity stepping of the intelligent control valve of the air pipe to be-0.01%; when the real-time opening difference value is smaller than 0, controlling the opening analog stepping of the intelligent control valve of the natural gas pipe flow to be-0.01%, and controlling the opening analog stepping of the intelligent control valve of the air pipe flow to be 0.01%; outputting the opening analog quantity of the modified intelligent control valve for the flow of the natural gas pipe and the opening analog quantity of the intelligent control valve for the flow of the air pipe;
s43, updating a real-time opening difference value of the combustion system according to the opening analog quantity of the intelligent control valve for controlling the modified natural gas pipe flow and the opening analog quantity of the intelligent control valve for controlling the air pipe flow, and repeating the step S41 until the opening analog quantity of the control valve of the combustion system is equal to the opening difference predicted value, so as to obtain the opening analog quantity of the control valve of the combustion system.
The analysis system of fuel data in the combustion system of the building ceramic kiln adopts the analysis method of fuel data in the combustion system of the building ceramic kiln, which comprises the following steps:
the data acquisition module is used for acquiring an initial data set of the combustion system, and carrying out initial processing on the initial data set to acquire a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
the model training module is used for optimizing the XGBoost model by adopting a GridSearchCV algorithm, and inputting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
the model prediction module is used for acquiring an online data set of the combustion system, and carrying out online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
the instruction construction module is used for obtaining the opening analog quantity of the control valve of the combustion system in a gradual approach mode according to the opening difference predicted value and the real-time opening difference value of the combustion system; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow;
and the instruction issuing module is used for issuing a flow valve control instruction to the combustion system according to the opening analog quantity of the control valve of the combustion system.
One of the above technical solutions has the following beneficial effects: on the basis of effective data, the opening difference predicted value is obtained through training and predicting an opening difference predicted model, then the opening difference predicted value and the real-time opening difference value of the combustion system are gradually and gradually forced to obtain the opening analog quantity of the control valve of the combustion system, namely, the optimal ratio of natural gas to air is formulated, and the problems that the original system cannot optimize the utilization rate of the natural gas and has low energy utilization rate are solved.
Drawings
FIG. 1 is a flow chart of a method of analyzing fuel data in a combustion system of a ceramic kiln in a building in accordance with the present invention;
FIG. 2 is a schematic diagram of the fuel data analysis system in a combustion system of a ceramic kiln of the present invention;
in the accompanying drawings: the system comprises a data acquisition module 1, a model training module 2, a model prediction module 3, an instruction construction module 4 and an instruction issuing module 5.
Detailed Description
The technical scheme of the invention is further described below by the specific embodiments with reference to the accompanying drawings.
In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "axial", "radial", "circumferential", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element in question must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, unless otherwise indicated, the meaning of "a plurality" is two or more.
In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
As shown in fig. 1, a method for analyzing fuel data in a combustion system of a ceramic kiln comprises the following steps:
s1, acquiring an initial data set of a combustion system, and performing initial processing on the initial data set to obtain a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
s2, optimizing the XGBoost model by adopting a GridSearchCV algorithm, and putting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
s3, acquiring an online data set of the combustion system, and performing online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
s4, according to the opening difference predicted value and the real-time opening difference value of the combustion system, obtaining an opening analog quantity of a control valve of the combustion system in a gradual approach mode; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow;
s5, according to the opening analog quantity of the control valve of the combustion system, a flow valve control instruction is issued to the combustion system.
The invention aims to provide a data analysis method for combustion values and the like in a building ceramic kiln combustion system, which is characterized in that on the basis of adopting effective data, an opening difference prediction value is obtained through training and predicting an opening difference prediction model, then the opening difference prediction value and a real-time opening difference value of the combustion system are subjected to gradual approach to obtain an opening analog quantity of a control valve of the combustion system, namely, the optimal ratio of natural gas to air is formulated, and the problems that the original system cannot optimize the utilization rate of the natural gas and has low energy utilization rate are solved.
Further, the system is controlled by a fully-automatic gas flow regulating control valve, and the natural gas flow in a natural gas pipe and the air flow in an air pipe in the combustion system are timely and automatically regulated according to the change of the natural gas combustion value, so that the problems of heat removal and energy waste of redundant air in the original system are solved.
In conclusion, the system integrates the automatic control principle of the gas control valve and the neural network learning model so as to comprehensively control the combustion system, thereby achieving the effects of energy conservation and emission reduction, reducing the manufacturing cost and improving the production efficiency.
Further illustratively, in this embodiment, kiln data is obtained by automatically collecting the combustion system every 30 seconds, wherein the basic sensing and control equipment and data provided with the combustion system include, but are not limited to, the following data:
the natural gas phase analyzer is used for measuring the CH4 content;
the natural gas intelligent meter is used for measuring the natural gas flow X1;
the online oxygen content detection monitor is used for measuring the content of O2 and the relative humidity;
an air-gas smart meter for measuring an air gas flow X2;
the waste gas detector is used for measuring the content of CO;
an exhaust-gas smart meter for measuring an exhaust gas flow X3;
the intelligent control valve for the flow of the natural gas pipe is used for controlling the opening M1 of the flow of the natural gas pipe;
the air pipe flow intelligent control valve is used for controlling the opening M2 of the air pipe flow;
the initial data set and the online data set comprise CH4 content, natural gas flow X1, O2 content, relative humidity, air gas flow X2, CO content, waste gas flow X3, opening M1 of natural gas pipe flow and opening M2 of air pipe flow.
To illustrate further, in S1, the initial data set is initially processed to obtain a modeling data set suitable for modeling; the method specifically comprises the following steps:
s21, respectively exploring and analyzing historical equipment data and laboratory test data, and removing missing values and abnormal values after analysis to obtain correct historical equipment data and laboratory test data;
the method specifically comprises the following steps:
(1) and obtaining statistical indexes of maximum value, maximum value bit number, minimum value position, 25% quantile, median, 75% quantile, mean value, average absolute deviation, variance, standard deviation, skewness and kurtosis of each variable by using a vector function of pandas, wherein the statistical indexes are used for knowing the total sample number and the statistical condition of the number of each variable.
(2) The unit inconsistency in the laboratory history filling data is changed into a unified unit, for example, the unit of oxygen content is unified to mg/L, and g/L is replaced to mg/L.
(3) The partial time data deletion severely requires the data deletion method as follows: counting the missing unit cells of the row, and deleting the row when the missing unit cells are more than 30%; if the character data is missing, filling the character data by using the mode value of the character value of the column, wherein the method comprises counting the number of each group of characters of the column, and assigning the first arranged value into the null value.
(4) The method for obviously filling errors in partial data information comprises the following steps: calculating an average value (mean) and a standard value (sigma) of each column of data (datan), and eliminating when each abnormal value (dif) is more than 3 times of the standard value, wherein the formula is as follows: diff { data } n Mean > 3 x sigma, which is filled with the mode value of the column data.
S22, carrying out association and combination processing on correct historical equipment data and laboratory test data through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
the basic parameters of the associated merging data are as follows:
s23, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
wherein, the difference formula is:
difference parameter = data at current time-data at last time.
S24, distributing the difference parameters and the corresponding combined data set into a group of modeling data, and taking a plurality of groups of modeling data as a modeling data set;
firstly, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter, and taking the difference parameter as modeling data required by an opening difference prediction model, wherein one group of modeling data is longitudinally displayed as follows:
s25, randomly distributing the modeling data set, and dividing the modeling data set into a 70% training set and a 30% testing set.
In a further explanation, the step S3 is to optimize the XGBoost model by adopting a GridSearchCV algorithm, and put modeling data into the optimized XGBoost model for training to generate an opening difference prediction model; the method specifically comprises the following steps:
s31, setting parameters of the tree model iteration number n_evators of the XGBoost model, the tree depth max_depth of the tree model and the loss function threshold gamma in splitting nodes to obtain multiple groups of parameters, and generating a corresponding prediction model;
s32, training a plurality of initial prediction models by adopting training sets in sequence, and obtaining corresponding prediction data accuracy through evaluation functions in sequence; in order to make the CO content value in the waste gas approach zero, natural gas is fully combusted, waste is reduced, and efficiency is improved, so that the difference value between the opening M1 of the intelligent natural gas pipe flow control valve and the opening M2 of the intelligent air pipe flow control valve is required to be optimal. In this embodiment, the other items in the training set are used as the input value X to obtain the opening difference prediction value of the output value, i.e. the optimal "difference value between opening M2 and M1 (dif_m1_m2)", so as to achieve the best benefit.
S33, selecting a prediction model corresponding to the highest prediction data accuracy, predicting the prediction model by adopting a test set, and obtaining the corresponding prediction data accuracy through an evaluation function;
s34, outputting the prediction model if the numerical value of the accuracy rate of the prediction data is more than 0.75 and less than 1; if the value of the accuracy of the predicted data is less than 0.75 or greater than 1, discarding the prediction model, and repeating S31 to 33 until a prediction model conforming to S34 is obtained.
Further described, the evaluation function is:
where R represents the accuracy of the predicted data,represents the opening difference predicted value of the ith, y (i) represents the true data value of the ith,/->Representing the average of all real data values y.
To further illustrate, in S3, the acquiring an online data set of the combustion system, performing online processing on the online data set to obtain a test data set suitable for testing; the method specifically comprises the following steps:
s31, carrying out association and combination processing on the online data set through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
s32, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
s33, distributing the difference parameter and the corresponding combined data set into a group of test data, and taking a plurality of groups of test data as the test data set.
In a further explanation, S4, according to the opening difference predicted value and the real-time opening difference value of the combustion system, the opening analog quantity of the control valve of the combustion system is obtained in a gradual approach mode; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow; the method specifically comprises the following steps:
s41, judging a difference value between the opening difference predicted value and a real-time opening difference value of the combustion system, and acquiring an analog quantity of a control valve corresponding to the real-time opening difference value of the combustion system when the difference value is equal to 0; when the difference is not equal to 0, S42 is performed;
s42, judging whether the real-time opening difference value of the combustion system is larger than 0, and when the real-time opening difference value is larger than 0, controlling the opening analog quantity stepping of the intelligent control valve of the natural gas pipe to be 0.01% and controlling the opening analog quantity stepping of the intelligent control valve of the air pipe to be-0.01%; when the real-time opening difference value is smaller than 0, controlling the opening analog stepping of the intelligent control valve of the natural gas pipe flow to be-0.01%, and controlling the opening analog stepping of the intelligent control valve of the air pipe flow to be 0.01%; outputting the opening analog quantity of the modified intelligent control valve for the flow of the natural gas pipe and the opening analog quantity of the intelligent control valve for the flow of the air pipe;
s43, updating a real-time opening difference value of the combustion system according to the opening analog quantity of the intelligent control valve for controlling the modified natural gas pipe flow and the opening analog quantity of the intelligent control valve for controlling the air pipe flow, and repeating the step S41 until the opening analog quantity of the control valve of the combustion system is equal to the opening difference predicted value, so as to obtain the opening analog quantity of the control valve of the combustion system.
The analysis system of fuel data in the combustion system of the building ceramic kiln adopts the analysis method of fuel data in the combustion system of the building ceramic kiln, which comprises the following steps:
the data acquisition module 1 is used for acquiring an initial data set of the combustion system, and carrying out initial processing on the initial data set to obtain a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
the model training module 2 is used for optimizing the XGBoost model by adopting a GridSearchCV algorithm, and inputting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
the model prediction module 3 is used for acquiring an online data set of the combustion system, and carrying out online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
the instruction construction module 4 is used for obtaining the opening analog quantity of the control valve of the combustion system in a gradual approach mode according to the opening difference predicted value and the real-time opening difference value of the combustion system; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow;
the instruction issuing module 5 is used for issuing a flow valve control instruction to the combustion system according to the opening analog quantity of the control valve of the combustion system.
According to the system, on the basis of effective data, an opening difference predicted value is obtained through training and predicting an opening difference predicted model, and then the opening difference predicted value and a real-time opening difference value of the combustion system are subjected to gradual approach to obtain an opening analog quantity of a control valve of the combustion system. Furthermore, the system integrates the automatic control principle of the gas control valve and a neural network learning model so as to comprehensively control the combustion system, thereby achieving the effects of energy conservation and emission reduction, reducing the manufacturing cost and improving the production efficiency. The technical principle of the present invention is described above in connection with the specific embodiments. The description is made for the purpose of illustrating the general principles of the invention and should not be taken in any way as limiting the scope of the invention. Other embodiments of the invention will occur to those skilled in the art from consideration of this specification without the exercise of inventive faculty, and such equivalent modifications and alternatives are intended to be included within the scope of the invention as defined in the claims.
Claims (8)
1. The method for analyzing the fuel data in the combustion system of the building ceramic kiln is characterized by comprising the following steps of:
s1, acquiring an initial data set of a combustion system, and performing initial processing on the initial data set to obtain a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
s2, optimizing the XGBoost model by adopting a GridSearchCV algorithm, and putting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
s3, acquiring an online data set of the combustion system, and performing online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
s4, according to the opening difference predicted value and the real-time opening difference value of the combustion system, obtaining an opening analog quantity of a control valve of the combustion system in a gradual approach mode; the opening analog quantity of the control valve of the combustion system comprises the opening analog quantity of the intelligent control valve of the natural gas pipe flow and the opening analog quantity of the intelligent control valve of the air pipe flow; the method specifically comprises the following steps:
s41, judging a difference value between the opening difference predicted value and a real-time opening difference value of the combustion system, and acquiring an analog quantity of a control valve corresponding to the real-time opening difference value of the combustion system when the difference value is equal to 0; when the difference is not equal to 0, S42 is performed;
s42, judging whether the real-time opening difference value of the combustion system is larger than 0, and when the real-time opening difference value is larger than 0, controlling the opening analog quantity stepping of the intelligent control valve of the natural gas pipe to be 0.01% and controlling the opening analog quantity stepping of the intelligent control valve of the air pipe to be-0.01%; when the real-time opening difference value is smaller than 0, controlling the opening analog stepping of the intelligent control valve of the natural gas pipe flow to be-0.01%, and controlling the opening analog stepping of the intelligent control valve of the air pipe flow to be 0.01%; outputting the opening analog quantity of the modified intelligent control valve for the flow of the natural gas pipe and the opening analog quantity of the intelligent control valve for the flow of the air pipe;
s43, updating a real-time opening difference value of the combustion system according to the opening analog quantity of the intelligent control valve for controlling the modified natural gas pipe flow and the opening analog quantity of the intelligent control valve for controlling the air pipe flow, and repeating the step S41 until the opening analog quantity of the control valve of the combustion system is equal to the opening difference predicted value, so as to obtain the opening analog quantity of the control valve of the combustion system;
s5, according to the opening analog quantity of the control valve of the combustion system, a flow valve control instruction is issued to the combustion system.
2. The method of analyzing fuel data in a combustion system of a ceramic kiln of a building according to claim 1, wherein the combustion system comprises:
the natural gas phase analyzer is used for measuring the CH4 content;
the natural gas intelligent meter is used for measuring the natural gas flow X1;
the online oxygen content detection monitor is used for measuring the content of O2 and the relative humidity;
an air-gas smart meter for measuring an air gas flow X2;
the waste gas detector is used for measuring the content of CO;
an exhaust-gas smart meter for measuring an exhaust gas flow X3;
the intelligent control valve for the flow of the natural gas pipe is used for controlling the opening M1 of the flow of the natural gas pipe;
the air pipe flow intelligent control valve is used for controlling the opening M2 of the air pipe flow;
the initial data set and the online data set comprise CH4 content, natural gas flow X1, O2 content, relative humidity, air gas flow X2, CO content, waste gas flow X3, opening M1 of natural gas pipe flow and opening M2 of air pipe flow.
3. The method for analyzing fuel data in a combustion system of a ceramic kiln according to claim 2, wherein in S1, the initial data set is initially processed to obtain a modeling data set suitable for modeling; the method specifically comprises the following steps:
s21, respectively exploring and analyzing historical equipment data and laboratory test data, and removing missing values and abnormal values after analysis to obtain correct historical equipment data and laboratory test data;
s22, carrying out association and combination processing on correct historical equipment data and laboratory test data through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
s23, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
s24, distributing the difference parameters and the corresponding combined data set into a group of modeling data, and taking a plurality of groups of modeling data as a modeling data set;
s25, randomly distributing the modeling data set, and dividing the modeling data set into a 70% training set and a 30% testing set.
4. A method of analyzing fuel data in a combustion system of a ceramic kiln according to claim 3, wherein the difference formula is:
difference parameter = data at current time-data at last time.
5. The method for analyzing fuel data in a combustion system of a building ceramic kiln according to claim 4, wherein the step S2 is characterized in that a GridSearchCV algorithm is adopted to optimize an XGBoost model, modeling data are put into the optimized XGBoost model for training, and an opening difference prediction model is generated; the method specifically comprises the following steps:
s31, setting parameters of the tree model iteration number n_evators of the XGBoost model, the tree depth max_depth of the tree model and the loss function threshold gamma in splitting nodes to obtain multiple groups of parameters, and generating a corresponding prediction model;
s32, training a plurality of initial prediction models by adopting training sets in sequence, and obtaining corresponding prediction data accuracy through evaluation functions in sequence;
s33, selecting a prediction model corresponding to the highest prediction data accuracy, predicting the prediction model by adopting a test set, and obtaining the corresponding prediction data accuracy through an evaluation function;
s34, outputting the prediction model if the numerical value of the accuracy rate of the prediction data is more than 0.75 and less than 1; if the value of the accuracy of the predicted data is less than 0.75 or greater than 1, discarding the prediction model, and repeating S31 to 33 until a prediction model conforming to S34 is obtained.
6. The method of claim 5, wherein the evaluation function is:
where R represents the accuracy of the predicted data,represents the opening difference predicted value of the ith, y (i) represents the true data value of the ith,/->Representing the average of all real data values y.
7. The method for analyzing fuel data in a combustion system of a ceramic kiln according to claim 6, wherein in S3, the online data set of the combustion system is obtained, and the online data set is processed to obtain a test data set suitable for testing; the method specifically comprises the following steps:
s31, carrying out association and combination processing on the online data set through association conditions to obtain a combined data set; wherein the association condition includes time, equipment number and pipeline number;
s32, calculating data characteristics of the combined data set through a difference formula to obtain a difference parameter;
s33, distributing the difference parameter and the corresponding combined data set into a group of test data, and taking a plurality of groups of test data as the test data set.
8. A system for analyzing fuel data in a combustion system of a ceramic kiln in a building, characterized in that a method for analyzing fuel data in a combustion system of a ceramic kiln in a building according to any one of claims 1 to 7 is used, comprising:
the data acquisition module is used for acquiring an initial data set of the combustion system, and carrying out initial processing on the initial data set to acquire a modeling data set suitable for modeling; wherein the initial dataset comprises historical device data and laboratory test data;
the model training module is used for optimizing the XGBoost model by adopting a GridSearchCV algorithm, and inputting modeling data into the optimized XGBoost model for training to obtain an opening difference prediction model;
the model prediction module is used for acquiring an online data set of the combustion system, and carrying out online processing on the online data set to acquire a test data set suitable for testing; putting the test data set into an opening difference prediction model to obtain an opening difference prediction value; wherein the opening difference predicted value represents the difference between the opening M1 of the natural gas pipe flow and the opening M2 of the air pipe flow;
the instruction construction module is used for obtaining the opening analog quantity of the control valve of the combustion system in a gradual approach mode according to the opening difference predicted value and the real-time opening difference value of the combustion system; the analog quantity of the control valve of the combustion system comprises the analog quantity of the opening of the intelligent control valve of the natural gas pipe flow and the analog quantity of the opening of the intelligent control valve of the air pipe flow;
and the instruction issuing module is used for issuing a flow valve control instruction to the combustion system according to the opening analog quantity of the control valve of the combustion system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211230486.9A CN115935784B (en) | 2022-09-30 | 2022-09-30 | Analysis method and system for fuel data in combustion system of building ceramic kiln |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211230486.9A CN115935784B (en) | 2022-09-30 | 2022-09-30 | Analysis method and system for fuel data in combustion system of building ceramic kiln |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115935784A CN115935784A (en) | 2023-04-07 |
| CN115935784B true CN115935784B (en) | 2023-07-18 |
Family
ID=86552899
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211230486.9A Active CN115935784B (en) | 2022-09-30 | 2022-09-30 | Analysis method and system for fuel data in combustion system of building ceramic kiln |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115935784B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116300666A (en) * | 2023-05-24 | 2023-06-23 | 科大智能物联技术股份有限公司 | Power plant boiler operation control method based on XGBoost optimization |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111377595A (en) * | 2020-05-08 | 2020-07-07 | 蚌埠凯盛工程技术有限公司 | Method and system for controlling gas supply of glass kiln in real time |
| CN114036758A (en) * | 2021-11-12 | 2022-02-11 | 华电新疆哈密煤电开发有限公司 | Boiler combustion state dynamic display method based on numerical simulation and machine learning |
| CN115081697A (en) * | 2022-06-09 | 2022-09-20 | 佛山众陶联供应链服务有限公司 | Method and equipment for predicting firing curve based on raw materials and computer storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2441686C (en) * | 2003-09-23 | 2004-12-21 | Westport Research Inc. | Method for controlling combustion in an internal combustion engine and predicting performance and emissions |
-
2022
- 2022-09-30 CN CN202211230486.9A patent/CN115935784B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111377595A (en) * | 2020-05-08 | 2020-07-07 | 蚌埠凯盛工程技术有限公司 | Method and system for controlling gas supply of glass kiln in real time |
| CN114036758A (en) * | 2021-11-12 | 2022-02-11 | 华电新疆哈密煤电开发有限公司 | Boiler combustion state dynamic display method based on numerical simulation and machine learning |
| CN115081697A (en) * | 2022-06-09 | 2022-09-20 | 佛山众陶联供应链服务有限公司 | Method and equipment for predicting firing curve based on raw materials and computer storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Experimental and modeling study of liquid fuel injection and combustion in diesel engines with a common rail injection system;Leilei Xu 等;《Applied Energy》;第230卷;287-304 * |
| 压缩天然气发动机空气管理执行器驱动控制研究;宋君花 等;《车用发动机》(第1期);23-26、31 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115935784A (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111804146B (en) | Intelligent ammonia injection control method and intelligent ammonia injection control device | |
| CN109583585B (en) | Construction method of power station boiler wall temperature prediction neural network model | |
| JP2019527413A (en) | Computer system and method for performing root cause analysis to build a predictive model of rare event occurrences in plant-wide operations | |
| JP2014105989A (en) | Energy consumption prediction method of building power equipment | |
| CN115935784B (en) | Analysis method and system for fuel data in combustion system of building ceramic kiln | |
| CN114548494B (en) | Visual cost data prediction intelligent analysis system | |
| CN105317476A (en) | Turbine flow curve identification and optimization method based on feature flow area | |
| CN116186624A (en) | Boiler assessment method and system based on artificial intelligence | |
| CN115688581A (en) | Oil gas gathering and transportation station equipment parameter early warning method, system, electronic equipment and medium | |
| JP2008112428A (en) | Statistical prediction method and apparatus for influent water quality in water treatment facilities | |
| CN113962140A (en) | Method for optimizing steam turbine valve flow characteristic function based on GA-LSTM | |
| CN112365082A (en) | Public energy consumption prediction method based on machine learning | |
| CN118213997A (en) | A method for urban power grid load forecasting based on AHP-grey fuzzy algorithm | |
| CN114969068B (en) | Urban pressure pipe network real-time flow monitoring data analysis method and system | |
| CN116629406A (en) | A Method for Predicting Carbon Emissions from Electricity Consumption Based on Improved Prophet Model | |
| CN113836794B (en) | Soft and hard combined fly ash carbon content online monitoring method | |
| CN114332515B (en) | On-line judging method for operation safety of coal mill based on intuitionistic fuzzy clustering | |
| Arakelyan et al. | Analysis of the DCS historical data for estimation of input signal significance | |
| CN118536410B (en) | Big data driven modeling-based energy consumption optimization decision analysis method and system | |
| CN111210147A (en) | Method and system for evaluating performance of sintering process based on time series feature extraction | |
| CN118211811B (en) | Visual digital management system and method based on gas management | |
| CN114413249B (en) | Data analysis method for power station boiler efficiency benchmarking optimization | |
| CN118039003B (en) | Silicon content prediction method for digital twin system of blast furnace based on distribution and graph convolution | |
| CN111520740B (en) | Method for coordinately optimizing operation of multiple porous medium combustors | |
| CN117557400A (en) | Tree growth intelligent monitoring system based on cloud computing platform |
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 |