CN111639820A - Energy consumption parameter optimization method and system for cement kiln production - Google Patents
Energy consumption parameter optimization method and system for cement kiln production Download PDFInfo
- Publication number
- CN111639820A CN111639820A CN202010517817.1A CN202010517817A CN111639820A CN 111639820 A CN111639820 A CN 111639820A CN 202010517817 A CN202010517817 A CN 202010517817A CN 111639820 A CN111639820 A CN 111639820A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- cement
- model
- parameter
- sample set
- 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.)
- Pending
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 133
- 239000004568 cement Substances 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 35
- 238000005457 optimization Methods 0.000 title claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 230000002068 genetic effect Effects 0.000 claims abstract description 23
- 238000012216 screening Methods 0.000 claims abstract description 20
- 230000035945 sensitivity Effects 0.000 claims abstract description 13
- 230000008859 change Effects 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 7
- 238000012935 Averaging Methods 0.000 claims 2
- 230000008569 process Effects 0.000 abstract description 12
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 108090000623 proteins and genes Proteins 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 210000000349 chromosome Anatomy 0.000 description 4
- 230000035772 mutation Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 239000002826 coolant Substances 0.000 description 3
- 238000001816 cooling Methods 0.000 description 3
- 238000010206 sensitivity analysis Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 239000003245 coal Substances 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000004134 energy conservation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000002912 waste gas Substances 0.000 description 1
- 239000002351 wastewater Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Genetics & Genomics (AREA)
- Manufacturing & Machinery (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
Abstract
The invention discloses an energy consumption parameter optimization method and system for cement kiln production, which comprises the following processes: acquiring energy consumption key parameters and historical data of clinker energy consumption of corresponding tons to form a sample set; training a cement energy consumption model based on a BP neural network by using a sample set; the cement energy consumption model takes key parameters as input and ton clinker energy consumption as output; screening out parameters which have great influence on energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model; and optimizing the energy consumption model by using a genetic algorithm to obtain a parameter combination with the lowest energy consumption. The method optimizes the key parameters in the cement production process, provides the recommended values of the key parameters for field operation, reduces the consumption of electric energy in the production process, and helps cement enterprises to reduce the production cost.
Description
Technical Field
The invention belongs to the technical field of optimization of energy consumption in cement kiln production, and particularly relates to an energy consumption parameter optimization method for cement kiln production and a system applying the method.
Background
The cement manufacturing industry is one of the high energy consumption industries in China all the time, the dependence on energy is high, and the energy consumption accounts for 40-60% of the production cost. In recent years, energy conservation of cement enterprises is greatly developed, but compared with the world advanced level, the energy consumption per ton of cement is still different. In the face of huge market pressure, cement enterprises pay more and more attention to energy conservation and consumption reduction. With the rapid spread of energy informatization in cement enterprises, the energy construction of the enterprises enters a new period, and the exploration of energy consumption statistical analysis has very important significance.
The cement burning system is a main energy consumption part in the cement production process, complex physical and chemical reactions are carried out in the cement burning system, numerous links and equipment are involved, the collected data have the characteristics of nonlinearity, strong coupling, numerous variables and the like, and the characteristics of multivariable, nonlinearity and large hysteresis are realized, so that the research on the aspect of cement energy consumption analysis is less. Wherein: the invention patent 201810910130.7 provides a convolutional neural network-based multi-energy-consumption index prediction method for a cement burning process, wherein an established CNN model takes energy consumption related variables as network input to jointly predict unit power consumption and ton coal consumption of a cement burning system, so that a scheduling basis is provided for the cement burning process in time, the problem that only single energy consumption is predicted incompletely is avoided, and the prediction result is closer to the actual situation of comprehensive energy consumption. The invention patent 201710990534.7 provides a cement production power consumption prediction method with a hidden time series deep belief network, which comprises the steps of establishing an HTS-DBN model, carrying out unsupervised forward training on the model, and determining an initial weight and bias; and adopting a BP reverse error correction algorithm to perform supervised reverse fine adjustment on the whole neural network. The method solves the time-varying delay problem, can accurately predict the power consumption of cement production, and provides a basis for scientific production scheduling and reasonable energy planning of cement production. In recent years, with the development of artificial intelligence and the popularization of industrial data acquisition, a Distributed Control System (DCS) has been widely used in various industries. The artificial intelligence analysis optimization method becomes the mainstream of industrial data analysis optimization.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides a method and a system for optimizing energy consumption parameters in cement kiln production, and solves the technical problems of insufficient prediction precision and incomplete prediction parameters of a cement production energy consumption model in the prior art.
In order to solve the technical problems, the invention provides a method for optimizing energy consumption parameters in cement kiln production, which is characterized by comprising the following steps of:
determining key parameters influencing energy consumption;
acquiring key parameters and historical data of clinker energy consumption of corresponding tons to form a sample set;
training a cement energy consumption model based on a BP neural network by using a sample set; the cement energy consumption model takes key parameters as input, and ton clinker energy consumption as output;
screening out parameters which have great influence on energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model;
and optimizing the energy consumption model by using a genetic algorithm to obtain a parameter combination with the lowest energy consumption.
Further, the initial value and the threshold value of the cement energy consumption model are optimized through a genetic algorithm.
Further, the screening out parameters having a large influence on the energy consumption sensitivity by an average value method includes:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: a data column P in a training sample set PjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndthe two new training sample sets are predicted by using the trained BP neural network model, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the change of the variable quantity on the output resultj;
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablej;
Step 5: for MIVjIf the accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
the corresponding k input variables are selected to represent all the input variables to reconstruct the cement energy consumption model.
Correspondingly, the invention also provides an energy consumption parameter optimization system for cement kiln production, which is characterized by comprising a parameter determination module, a sample acquisition module, a model training module, a parameter screening module and a parameter optimization module, wherein:
the parameter determination module is used for determining key parameters influencing energy consumption;
the sample acquisition module is used for acquiring key parameters and historical data corresponding to the ton of clinker energy consumption to form a sample set;
the model training module is used for training a cement energy consumption model based on a BP neural network by utilizing a sample set; the cement energy consumption model takes key parameters as input, and ton clinker energy consumption as output;
the parameter screening module is used for screening out parameters which have large influence on the energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model;
and the parameter optimization module is used for optimizing the energy consumption model by utilizing a genetic algorithm to obtain a parameter combination with the lowest energy consumption.
Further, the initial value and the threshold value of the cement energy consumption model are optimized through a genetic algorithm.
Further, in the parameter screening module, the screening of the parameter having a large influence on the sensitivity to energy consumption by an average value method includes:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: a data column P in a training sample set PjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndthe two new training sample sets are predicted by using the trained BP neural network model, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the change of the variable quantity on the output resultj;
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablej;
Step 5: for MIVjAre arranged in descending order of magnitude ifThe accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
the corresponding k input variables are selected to represent all the input variables to reconstruct the cement energy consumption model.
Compared with the prior art, the invention has the following beneficial effects: the invention provides an energy consumption parameter optimization method for cement kiln production, which can solve the problem that the traditional cement energy consumption modeling is complex and inaccurate, optimize key parameters in the cement production process, provide key parameter recommendation values for field operation, reduce the consumption of electric energy in the production process and help cement enterprises to reduce the production cost.
Drawings
FIG. 1 is a flowchart of a genetic algorithm optimizing BP neural network algorithm;
fig. 2 is a flow chart of the hybrid algorithm.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The invention has the following inventive concept: firstly analyzing influence variables according to a process flow as input, optimizing an initial value and a threshold value of a BP neural network by using a genetic algorithm, then modeling energy consumption and input by using the genetic optimization BP neural network, carrying out sensitivity analysis on energy consumption parameters by using an average value, reestablishing an energy consumption model after screening the parameters, taking the energy consumption predicted by the model as fitness, optimizing the whole energy consumption, and outputting the corresponding energy consumption as a most parameter scheme.
Examples
The invention relates to an energy consumption parameter optimization method for cement kiln production, which comprises the following steps:
step1, carrying out process analysis on energy consumption links of a cement production line, screening out key parameters influencing energy consumption, and using the key parameters as input of a later model;
after analyzing the energy consumption link process generated in the cement production flow, determining key parameters possibly influencing energy consumption comprises the following steps: the method comprises the following steps of feeding coal amount, volume percentage of carbon monoxide, raw material flow rate, pressure of a cooling agent blast pipeline, temperature of air entering a cooling machine, clinker flow rate, outlet pressure of a preheater, waste water temperature of the preheater, clinker amount at an outlet of the cooling machine, clinker temperature at an outlet of the cooling machine, pressure of a cooling agent chimney, waste gas temperature of the cooling agent chimney, surface temperature of the middle section of a decomposing furnace, surface temperature of the middle section of a rotary furnace and ambient temperature.
And 2, acquiring historical data of key parameters and ton clinker energy consumption, carrying out normalization processing on the data to be used as a sample set, and randomly dividing the sample set data into a training set and a testing set, wherein the training set and the testing set are provided.
After analyzing and obtaining the key parameters of the energy consumption, obtaining the key parameters and historical data of the energy consumption (energy consumption for short) of the corresponding ton of clinker through a database collected by a Distributed Control System (DCS) of a cement enterprise.
And respectively carrying out normalization processing on different key parameters and ton clinker energy consumption by adopting a maximum and minimum method to serve as a sample set.
In the formula, yiFor normalized data values, xiFor the original data value, xminIs the minimum value of the data sequence, xmaxIs the maximum value of the data sequence.
And randomly dividing the sample set data into a training set and a testing set.
And 3, firstly, optimizing the initial value and the threshold value of the BP neural network through a genetic algorithm, then taking key parameters as input and ton clinker energy consumption as output, and establishing a cement energy consumption model by utilizing the improved BP neural network.
A flowchart of genetic algorithm optimization BP neural network is shown in fig. 1, and initial values and threshold values are optimized by taking prediction accuracy of the BP neural network as a target, and the specific process comprises the following steps:
and Step1, training a many-to-one energy consumption prediction model by using a BP neural network according to the energy consumption key parameter data and the ton clinker energy consumption as input and output.
Step2, determining the length of the genetic algorithm individuals according to the network structure of the trained BP neural network model, and performing population initialization: and carrying out coding operation on the weight and the threshold value among the input, the output and the hidden layer of the BP neural network in a real number coding mode to obtain a real number code as an individual of the population. And generating an initialization population by adopting a random method according to the set initial population size.
Step3, determining a fitness function, following a principle of high-quality and poor-quality by a genetic algorithm, and representing the adaptability of an individual by using the fitness value of the individual in an iteration process as a unique basis of genetic operation. And taking the reciprocal of the error function of the cement energy consumption predicted value of the individual i (i ═ 1,2,3 … …, n) BP neural network and the expected output value as a fitness function F, wherein the expected output value is the energy consumption data in the test set.
in the formula, n is the total number of the test samples; oiAnd (4) predicting output for the BP neural network of the individual i. y isiThe expected output of the individual i is the energy consumption data value of the test set;
step4, genetic algorithm there are many ways to perform the selection operation during the selection operation. Common selection methods are tournament methods, roulette methods, and the like. The invention discloses a method for selecting roulette, which is a selection strategy based on fitness proportion, wherein each individual i has corresponding selection probability pi:
fi=1/Fi(3)
In the formula, FiIs the fitness value of the individual i, since fitness isThe smaller the value, the better, so the reciprocal of the fitness value is calculated before individual selection; the number of population individuals is N;
step5, performing crossover operation by using a real number crossover method, the kth chromosome αkAnd the l-th chromosome αlThe specific method of interleaving at j bits is as follows:
αkj=αkj(1-b)+αljb (5)
αlj=αlj(1-b)+αkjb (6)
wherein b is a random number between [0, 1 ].
Step6, selecting the jth gene α of the ith individualijAnd carrying out corresponding mutation operation on the strain, wherein the specific mutation operation mode is as follows:
in the formula, αmaxIs gene αijαminIs gene αijThe lower bound of (c). f (g) r2(1-g/Gmax); r2Is a random number; g is the current iteration number; gmaxIs the maximum number of evolutions; r is [0, 1]]Random number of cells in between.
And Step7, obtaining the initial network weight and threshold value assignment of the optimal individual through the processes, taking the initial network weight and threshold value assignment as the optimal initial value and threshold value of the BP neural network, and reestablishing the energy consumption model.
The step3 is to find the optimum for the initial value and the threshold value, because the ordinary BP neural network determines the initial weight value and the threshold value randomly according to experience, and usually within a certain range. The optimal initial value and threshold value are found through the genetic algorithm, and the convergence degree and precision of the optimized bp neural network are improved.
And 4, firstly, carrying out sensitivity analysis on the energy consumption parameters through an average value method, screening out parameters which have large influence on the energy consumption sensitivity, and correcting the energy consumption model to improve the efficiency and the accuracy of the model.
Firstly, sensitivity analysis is respectively carried out on the energy consumption parameters by utilizing an average value method, the energy consumption parameters with qualified sensitivity are screened out, and the energy consumption model established in the third step is corrected, wherein the specific process of the average value method comprises the following steps:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: after the training of the improved BP neural network in the third step is finished, a data column P in a training sample set P isjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndthe two new training sample sets are predicted by using the trained BP neural network model in the third step, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the variable change on the output resultj
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablejThe sign of which indicates the direction in which the input variable is associated with the output variable, and the magnitude of the absolute value of which indicates the input variable to output variableThe degree of influence.
Step 5: for MIVjIf the accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
η therein0And taking 85 percent.
Selecting the corresponding k input variables may represent all the input variables to reconstruct the cement energy consumption model.
And fifthly, optimizing the energy consumption model by using the screened parameter data as input and utilizing a genetic algorithm to find the parameter combination with the lowest energy consumption.
And fourthly, after the data is screened to obtain a final energy consumption model, obtaining a correct input-output relation by adopting the model, further combining a global optimization algorithm, and finding out the optimal operating parameter by taking the output of the energy consumption model as an adaptive value. The flow chart is shown in fig. 2, and the parameters are optimized by taking energy consumption as a target, and the specific steps are as follows:
step 1: carrying out population initialization: and carrying out coding operation on the input key parameters of the improved BP network in a real number coding mode to obtain a real number code as an individual of the population. And generating an initialization population by random generation according to the set initial population size.
Step 2: the genetic algorithm is utilized, the neural network predicted value is taken as search information, the advantage of a specific function is not needed, and the neural network predicted value is directly taken as an adaptive value: firstly, obtaining an initial weight and a threshold value of the neural network at the moment through the individual i, and obtaining a predicted value output by the network after training of the neural network. And taking the prediction result of the improved BP neural network as an individual adaptive value.
Step 3: the selection of the roulette method as a selection operation is still a selection strategy based on fitness scale, in which each individual i has a corresponding selection probability pi:
gi=k/Si(9)
In the formula, SiThe fitness value of the individual i is the network prediction value, the smaller the fitness value is, the better the fitness value is, therefore, before individual selection, the reciprocal of the fitness value is calculated, the number of population individuals is N, and the coefficient is k.
Step4 performing crossover operation by using real number crossover method, chromosome k αkAnd the l-th chromosome αlThe specific method of interleaving at j bits is as follows:
αkj=αkj(1-b)+αljb (11)
αlj=αlj(1-b)+αkjb (12)
wherein b is a random number between [0, 1 ].
Step5 selecting the jth gene α of the ith individualijAnd carrying out corresponding mutation operation on the strain, wherein the specific mutation operation mode is as follows:
in the formula, αmaxIs gene αijαminIs gene αijThe lower bound of (c). f (g) r2(1-g/Gmax); r2Is a random number; g is the current iteration number; gmaxIs the maximum number of evolutions; r is [0, 1]]Random number of cells in between.
Step 6: and finally, searching a global optimal value of the model through the process, and taking a corresponding input value as an optimal solution, namely a cement production recommended parameter recommended by the algorithm.
The invention provides an energy consumption parameter optimization method for cement kiln production, which can solve the problems of complexity and inaccuracy of traditional cement energy consumption modeling, optimize key parameters in a cement production process, provide key parameter recommended values for field operation, reduce the consumption of electric energy in the production process and help cement enterprises to reduce the production cost.
Examples
Correspondingly, the invention also provides an energy consumption parameter optimization system for cement kiln production, which is characterized by comprising a parameter determination module, a sample acquisition module, a model training module, a parameter screening module and a parameter optimization module, wherein:
the parameter determination module is used for determining key parameters influencing energy consumption;
the sample acquisition module is used for acquiring key parameters and historical data corresponding to the ton of clinker energy consumption to form a sample set;
the model training module is used for training a cement energy consumption model based on a BP neural network by utilizing a sample set; the cement energy consumption model takes key parameters as input, and ton clinker energy consumption as output;
the parameter screening module is used for screening out parameters which have large influence on the energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model;
and the parameter optimization module is used for optimizing the energy consumption model by utilizing a genetic algorithm to obtain a parameter combination with the lowest energy consumption.
Further, the initial value and the threshold value of the cement energy consumption model are optimized through a genetic algorithm.
Further, in the parameter screening module, the screening of the parameter having a large influence on the sensitivity to energy consumption by an average value method includes:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: a data column P in a training sample set PjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndthe two new training sample sets are predicted by using the trained BP neural network model, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the change of the variable quantity on the output resultj;
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablej;
Step 5: for MIVjIf the accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
the corresponding k input variables are selected to represent all the input variables to reconstruct the cement energy consumption model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (6)
1. An energy consumption parameter optimization method for cement kiln production is characterized by comprising the following steps:
determining key parameters influencing energy consumption;
acquiring key parameters and historical data of clinker energy consumption of corresponding tons to form a sample set;
training a cement energy consumption model based on a BP neural network by using a sample set; the cement energy consumption model takes key parameters as input and ton clinker energy consumption as output;
screening out parameters which have great influence on energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model;
and optimizing the energy consumption model by using a genetic algorithm to obtain a parameter combination with the lowest energy consumption.
2. The method as claimed in claim 1, wherein the initial values and threshold values of the cement energy consumption model are optimized by genetic algorithm.
3. The method as claimed in claim 1, wherein said selecting parameters with a greater influence on energy consumption sensitivity by averaging comprises:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: a data column P in a training sample set PjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndtwo new training sample sets are formed and are predicted by utilizing the trained BP neural network model, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the variable change on the output resultj;
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablej;
Step 5: for MIVjIf the accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
the corresponding k input variables are selected to represent all the input variables to reconstruct the cement energy consumption model.
4. The energy consumption parameter optimization system for cement kiln production is characterized by comprising a parameter determination module, a sample acquisition module, a model training module, a parameter screening module and a parameter optimization module, wherein:
the parameter determination module is used for determining key parameters influencing energy consumption;
the sample acquisition module is used for acquiring key parameters and historical data corresponding to the ton of clinker energy consumption to form a sample set;
the model training module is used for training a cement energy consumption model based on a BP neural network by utilizing a sample set; the cement energy consumption model takes key parameters as input and ton clinker energy consumption as output;
the parameter screening module is used for screening out parameters which have large influence on the energy consumption sensitivity through an average value method; taking the parameters as the input of a cement energy consumption model;
and the parameter optimization module is used for optimizing the energy consumption model by utilizing a genetic algorithm to obtain a parameter combination with the lowest energy consumption.
5. The system of claim 4, wherein the initial values and threshold values of the cement energy consumption model are optimized by a genetic algorithm.
6. The system of claim 4, wherein the parameter selection module selects the parameter with a greater influence on the sensitivity of energy consumption by an averaging method, and the parameter selection module comprises:
step 1: assume that the original training sample set P has m samples, each sample containing n variables, i.e. sample set P ═ { P ═ P1,P2,....PnThe output is a variable Y ═ Y1,y2,.....ym];
Step 2: a data column P in a training sample set PjEach value of j 1,2,3,... n is added and subtracted by 10% on the original basis, respectively, to form two new data columnsAnd
step 3: will be composed ofAndtwo new training sample sets are formed and are predicted by utilizing the trained BP neural network model, and two groups of prediction results are correspondingly obtainedAndthe difference value obtained by the difference of the two represents the influence change value IV of the variable change on the output resultj;
Step 4: for m output difference values IVjIs summed and averaged to obtain the average influence value MIV of the j-th input variablej;
Step 5: for MIVjIf the accumulated contribution rate of the first k sorted MIV absolute values satisfies the following formula:
the corresponding k input variables are selected to represent all the input variables to reconstruct the cement energy consumption model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010517817.1A CN111639820A (en) | 2020-06-09 | 2020-06-09 | Energy consumption parameter optimization method and system for cement kiln production |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010517817.1A CN111639820A (en) | 2020-06-09 | 2020-06-09 | Energy consumption parameter optimization method and system for cement kiln production |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111639820A true CN111639820A (en) | 2020-09-08 |
Family
ID=72333076
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010517817.1A Pending CN111639820A (en) | 2020-06-09 | 2020-06-09 | Energy consumption parameter optimization method and system for cement kiln production |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111639820A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112275439A (en) * | 2020-10-14 | 2021-01-29 | 济南大学 | Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system |
CN112845610A (en) * | 2020-12-31 | 2021-05-28 | 中冶赛迪重庆信息技术有限公司 | Steel rolling power consumption parameter recommendation method and system |
CN112985096A (en) * | 2021-03-05 | 2021-06-18 | 广州东兆信息科技有限公司 | Kiln ceramic quality monitoring system and method based on Internet of things cloud platform |
CN114545866A (en) * | 2020-11-11 | 2022-05-27 | 台泥资讯股份有限公司 | Method for controlling coal consumption system |
-
2020
- 2020-06-09 CN CN202010517817.1A patent/CN111639820A/en active Pending
Non-Patent Citations (1)
Title |
---|
高倩: "基于大数据的能耗特性分析方法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112275439A (en) * | 2020-10-14 | 2021-01-29 | 济南大学 | Cement raw material vertical mill differential pressure soft measurement modeling method, storage medium and system |
CN114545866A (en) * | 2020-11-11 | 2022-05-27 | 台泥资讯股份有限公司 | Method for controlling coal consumption system |
CN112845610A (en) * | 2020-12-31 | 2021-05-28 | 中冶赛迪重庆信息技术有限公司 | Steel rolling power consumption parameter recommendation method and system |
CN112845610B (en) * | 2020-12-31 | 2023-03-14 | 中冶赛迪信息技术(重庆)有限公司 | Steel rolling power consumption parameter recommendation method and system |
CN112985096A (en) * | 2021-03-05 | 2021-06-18 | 广州东兆信息科技有限公司 | Kiln ceramic quality monitoring system and method based on Internet of things cloud platform |
CN112985096B (en) * | 2021-03-05 | 2021-09-07 | 广州东兆信息科技有限公司 | Kiln ceramic quality monitoring system and method based on Internet of things cloud platform |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111639820A (en) | Energy consumption parameter optimization method and system for cement kiln production | |
Chen et al. | Modelling and optimization of fed-batch fermentation processes using dynamic neural networks and genetic algorithms | |
CN111832101B (en) | Construction method of cement strength prediction model and cement strength prediction method | |
CN106202946A (en) | Clinker free calcium levels Forecasting Methodology based on degree of depth belief network model | |
CN111650834B (en) | Sewage treatment process prediction control method based on extreme learning machine | |
CN105069220B (en) | Microbial fermentation optimization method based on BP neural network immune genetic algorithm | |
CN109919356B (en) | BP neural network-based interval water demand prediction method | |
CN101139661A (en) | Copper flash smelting operation parameter optimization method | |
CN112488315A (en) | Batch scheduling optimization method based on deep reinforcement learning and genetic algorithm | |
CN113589693B (en) | Cement industrial decomposing furnace temperature model predictive control method based on neighborhood optimization | |
CN113094988A (en) | Data-driven slurry circulating pump operation optimization method and system | |
CN113449930A (en) | Optical fiber preform preparation quality prediction method based on BP neural network | |
CN112926896A (en) | Production scheduling method for cigarette cut tobacco production | |
CN115034465B (en) | Coke quality prediction method based on artificial intelligence | |
CN111639821A (en) | Cement kiln production energy consumption prediction method and system | |
CN111833970A (en) | Construction method and application of cement clinker quality characterization parameter prediction model | |
CN110849149A (en) | Energy perception-based sintering batching scheme cascade optimization obtaining method and device | |
CN117236523B (en) | Combined heuristic carbon emission optimization method based on pareto strategy | |
Zhao et al. | Analysis of energy consumption prediction model based on genetic algorithm and wavelet neural network | |
Huang et al. | Machine learning approach for investigating compressive strength of self-compacting concrete containing supplementary cementitious materials and recycled aggregate | |
CN112836902A (en) | Coal combustion calorific capacity prediction method based on improved BP neural network | |
CN109101683B (en) | Model updating method for pyrolysis kettle of coal quality-based utilization and clean pretreatment system | |
CN114004341A (en) | Optical fiber preform preparation process optimization method based on genetic algorithm and BP neural network | |
CN115049123A (en) | Prediction method for silicon content of molten iron in blast furnace based on GA-XGboost model | |
CN114926075A (en) | Mechanical part production scheduling method based on man-hour prediction |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200908 |
|
RJ01 | Rejection of invention patent application after publication |