CN116629432A - Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection - Google Patents
Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection Download PDFInfo
- Publication number
- CN116629432A CN116629432A CN202310617519.3A CN202310617519A CN116629432A CN 116629432 A CN116629432 A CN 116629432A CN 202310617519 A CN202310617519 A CN 202310617519A CN 116629432 A CN116629432 A CN 116629432A
- Authority
- CN
- China
- Prior art keywords
- feature
- thermal power
- coal consumption
- generating unit
- power generating
- 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
- 239000003245 coal Substances 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000010276 construction Methods 0.000 title claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 238000012544 monitoring process Methods 0.000 claims abstract description 10
- 230000002068 genetic effect Effects 0.000 claims abstract description 9
- 238000007621 cluster analysis Methods 0.000 claims abstract description 5
- 230000001419 dependent effect Effects 0.000 claims abstract description 5
- 238000000605 extraction Methods 0.000 claims abstract 2
- 210000000349 chromosome Anatomy 0.000 claims description 21
- 108090000623 proteins and genes Proteins 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000035772 mutation Effects 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000002411 adverse Effects 0.000 claims description 2
- 230000010485 coping Effects 0.000 claims 1
- 238000010248 power generation Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 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
- 238000007418 data mining Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Optimization (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Algebra (AREA)
- Development Economics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an intelligent prediction method for coal consumption of a thermal power generating unit based on feature construction and feature selection, which comprises the following steps: collecting historical monitoring data of the thermal power unit through a sensor, and performing cluster analysis by using a K-means algorithm to divide different basic working conditions of the thermal power unit; using feature construction to reflect this delay or advance time phenomenon in hopes of getting more potential premium subsets; removing nonsensical redundant features by using a genetic algorithm to realize extraction and simplification of feature subsets; and establishing a bagging tree regression prediction model by using the screened feature subset as independent variable coal consumption as a dependent variable, and predicting the coal consumption of the thermal power unit in real time according to the regression prediction model for subsequent real-time monitoring data. The intelligent prediction method provided by the application is efficient, accurate, simple and feasible.
Description
Technical Field
The application relates to the field of data mining, in particular to an intelligent prediction method for coal consumption of a thermal power generating unit based on feature construction and feature selection.
Background
Clean energy represented by hydropower, wind power, solar power generation and the like has the characteristics of remarkable intermittence and uncertainty, and the generated energy is easily influenced by factors such as climate, season, period and the like.
In order to fully utilize clean energy as much as possible, improve the duty ratio of the clean energy, avoid the occurrence of large-scale phenomena of wind discarding, water discarding and light discarding, and use thermal power as auxiliary and standby energy to improve the overall stability of the power grid with the help of an auxiliary service mechanism of an electric power market; when the clean energy supply is sufficient, the thermal power generation capacity is reduced, when the clean energy supply is insufficient, the thermal power generation capacity is increased, and the thermal power is converted into dynamic supplementary power for clean energy power generation from the current basic power source role. In other words, in the electric power system, thermal power needs to provide necessary foundation and guarantee for renewable energy access in power grid peak shaving.
This means that the role of thermal power will be changed from the past steady operation to provide a base load in the electrical load to a fast adjustment to cope with peak load in the electrical load, and such a significant change of thermal power generation operation mode brings new challenges to thermal power plant operation optimization and plant scheduling. How to realize high-efficiency operation, energy conservation and emission reduction under the condition of power grid peak shaving is one management problem which a thermal power enterprise must solve. The establishment of an accurate and effective coal consumption prediction model of the thermal power generating unit is one of the most fundamental problems to be solved for realizing the above-mentioned aim.
In recent years, intelligent management deployment including distributed control systems, monitoring information systems, factory information systems, and field bus control systems has been receiving increasing attention from thermal power enterprises. These systems enable thermal power enterprises to record real-time operating conditions and power generation data of thermal power units. Data-driven techniques based on operational data are often used for modeling and analysis of various fields of thermal power plants, such as fault diagnosis of coal-fired power generation units, determination of reference intervals of coal-fired power generation units, and optimization of turbine system independent variable reference values for operators. Compared with the traditional mechanism-based modeling, the data driving method has greater universality and accuracy.
However, few studies have focused on coal consumption prediction for thermal power plants. Through searching, the closest study in the literature is to propose a study on the least squares support vector machine theory in predicting coal consumption (Zhang l., zhou l., & Zhang y. Et al consumption prediction based on least squares support vector machine [ C ]. IOP Conference Series: earth and Environmental Science,2019,227 (3), 032007). In terms of index selection of the coal consumption prediction model, they formulated 20 features directly based on domain knowledge, and this experience-based feature allocation missed the possibility of finding a better subset of features to establish the coal consumption prediction model.
The problem of obtaining the optimal key feature set from the original operating data of the thermal power plant generally has high dimension, complex nonlinearity and strong coupling, a large number of features cause errors in the model construction process, and the calculation cost of the prediction model is obviously increased. Therefore, it is critical to identify key features to use as input variables for the predictive model. Finding missing relationships between features and constructing potentially better features is a process known as feature construction, which may also lead to the creation of many unnecessary features. One approach that is often used to address this problem is feature selection, which involves selecting important features and eliminating non-important features. Feature selection may enhance the machine learning predictive model by reducing time complexity, improving accuracy, and reducing risk of overfitting.
In summary, the existing research method cannot fully utilize thermal power unit data and accurately predict the coal consumption level of the thermal power unit, so that an intelligent thermal power unit coal consumption prediction method based on feature construction and feature selection is provided for the problems.
Disclosure of Invention
The embodiment provides an intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, and the prediction coal consumption of the thermal power generating unit is obtained through calculation, so that the method has the characteristics of objectivity, fairness, accuracy and easiness.
According to one aspect of the application, an intelligent prediction method for coal consumption of a thermal power generating unit based on feature construction and feature selection is provided, and the intelligent prediction method comprises the following steps:
step one, for history monitoring data x= (X) collected by a sensor mounted on a thermal power generating unit 1 ,...,X d ) Clustering analysis is carried out by using a K-means algorithm to divide different thermal power unit basic working conditions, wherein X is as follows j ＝(x 1,j ,…,x n,j ) T Is a column vector representing the jth feature in X, X i,j Is the ith sample of the jth feature, y= (Y) 1 ,...,y n ) T Is the coal consumption of the thermal power generating unit corresponding to the monitoring data X, wherein y i The coal consumption value representing the i-th sample, i=1, 2,..n, j=1, 2,..d, n is the total number of samples and d is the total number of features;
firstly, setting the number of clusters as k, and then randomly selecting k observation values as initial cluster centers; secondly, after measuring the distance between each observation point and each cluster center, designating the cluster closest to each observation point; then, recalculating the clustering center of each category and updating the clustering center result; repeating the process until the number of iterations reaches a predetermined threshold m;
for the values of different cluster numbers k, evaluating the optimal cluster number by using a contour coefficient; calculating contour coefficients corresponding to all different clustering quantities k, and selecting the corresponding clustering quantity with the largest contour coefficient as a final clustering quantity;
step two, recording data of various characteristics in the thermal power generating unit system come from various sensors placed at various system positions; however, the measurements of these features may not be obtained immediately, which means that there may be corresponding delay or advance times within the different parameters and between the features and the corresponding coal consumption; therefore, in order to create an accurate intelligent prediction model of the coal consumption of the thermal power generating unit, we use feature construction to reflect the delay or advance time phenomenon;
the recorded data of each feature is a time series, and we consider one time interval at a time to describe the delay or advance effect of one time point of each feature; we consider the delay or advance time interval for a total of l, respectively, because each feature can build an additional 2l features, the number of features increases from d to 2ld+d;
step three, the high-dimensional feature set can have a plurality of adverse effects on the regression machine learning model, such as increasing the time complexity of the model, reducing the accuracy of the model, increasing the risk of overfitting and the like, and we use genetic algorithm to eliminate nonsensical redundant features so as to realize the refinement and simplification of feature subsets;
first encoding, each possible subset of the original feature set may be represented as a chromosome, and each gene in the chromosome indicates a respective feature; when the value of the gene is assigned 1, the corresponding feature is selected to represent this feature subset, and when the value of the gene is assigned 0, the corresponding feature is not selected to represent this feature subset;
then randomly generating a given number of chromosomes according to the coding rule to form an initial population; each chromosome is evaluated using its root mean square error as a fitness function; generating a new generation chromosome population based on the selection, crossover and mutation operations of the genetic algorithm;
repeating the previous operation until the designated iteration number M is reached, and outputting the optimal chromosome in the last generation population, wherein the feature subset corresponding to the chromosome is the result of feature selection;
step four, using the feature subset screened in the step three as independent variables through a bagging tree model in Matlab 2021b software, using coal consumption as the dependent variables, and using a 5-cross verification method to establish a bagging tree regression prediction model;
and carrying the sample into the regression prediction model to obtain the coal consumption prediction of the thermal power generating unit corresponding to the sample for the subsequent random newly arrived thermal power generating unit monitoring data sample.
The application has the advantages that:
compared with the prior art, the intelligent prediction method for the coal consumption of the thermal power generating unit based on the feature construction and the feature selection acquires the potentially better features to construct a regression prediction model based on the feature construction and the feature selection, and the prediction value of the regression prediction model can accurately predict the coal consumption of the thermal power generating unit.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the application, and that other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of an intelligent prediction method of the present application;
FIG. 2 is a graph showing the actual coal consumption of 80000 samples used in the present application;
FIG. 3 is a graph comparing predicted and actual coal consumption errors using the method of the present application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
Examples
Referring to fig. 1-3, an intelligent prediction method for coal consumption of a thermal power generating unit based on feature construction and feature selection comprises the following steps:
step one, first, the data used in the present application is from a power plant in china, comprising 80000 observation samples of 35 monitored features, the 80000 sample true coal consumption curve being shown in fig. 2.
The selectable range of the number k of categories in the cluster analysis is set to 2 to 300. For each number of categories within the selectable range, a number of observation samples equal to the number of categories is randomly selected as an initial cluster center. After measuring the distance between each observation point and each cluster center, designating the cluster closest to each observation point, recalculating the cluster center of each category now and updating the cluster center result. This process is repeated until the number of iterations reaches a predetermined threshold m of 100.
For all possible values in the selectable range of the number of categories in the cluster analysis, the most ideal cluster number is estimated by using the contour coefficient, and the calculation formula of the contour coefficient is as followsWhere a (j) is the average of the distances from the observation sample j to all other observations in the cluster to which the observation belongs, b (j) is the average of the distances from the observation sample j to all other observations in the cluster to which the observation does not belong, the distance metric index is selected as the Euclidean distance, and the corresponding cluster number in which the profile coefficient is the largest is selected as the final cluster number.
We repeat 10 experiments for each given number of clusters, since the initial center position affects the final clustering result, we treat the best result of the 10 experiments as the final clustering result for that particular cluster number. When the number of cluster groups is 4, the maximum value of the profile coefficient is obtained, and we choose class 2 as an example to perform the subsequent steps, and class 2 includes 25802 observation samples.
In the second step, in the feature construction method, the value of the phase number l considering the delay or advance time interval is set to be 3, and then the feature number is increased from the original 35 to 2×35×3+35=245.
Step three, encoding first, each possible subset of the original feature set may be represented as a chromosome, and each gene in the chromosome indicates a corresponding feature. When the value of the gene is assigned 1, the corresponding feature is selected to represent this feature subset, and when the value of the gene is assigned 0, the corresponding feature is not selected to represent this feature subset.
An initial population containing 100 chromosomes is then randomly generated according to the encoding rules. Each chromosome is evaluated using its root mean square error as a fitness function. The calculation formula is thatWherein y is i And->The actual value of the coal consumption and the predicted value of the coal consumption are respectively.
The selection operator of the genetic algorithm is set to use a proportional selection operator, i.e. the probability that an individual is selected is proportional to its fitness function value, and assuming that the fitness of individual i in a population containing 100 chromosomes is fi, the probability that individual i is selected and inherited to the next generation population isThe crossover operator of the genetic algorithm is set to a simple single point crossover and the crossover probability is set to 1. The mutation probability of the mutation operator of the genetic algorithm is set to 0.05.
And repeating the previous step until the designated iteration number M=50 is reached, and outputting the optimal chromosome in the last generation population, wherein the feature subset corresponding to the chromosome is the feature selection result.
And step four, using the feature subset screened in the step three as independent variables through a bagging tree model in Matlab 2021b software, using coal consumption as the dependent variables, and using a 5-cross verification method to establish a bagging tree regression prediction model. Clicking an APP button; clicking a regression learner button; clicking a new session button; clicking a data set variable drop-down arrow to select a data set of a model to be built, wherein the format is that each column represents a feature, each row represents an observation sample, and coal consumption is put in the last column; clicking a start session button; clicking a pull-down button on the model category module, and clicking to select a bagging tree model; and clicking training to obtain a regression prediction model.
And clicking the derived model, namely using a yfit=trainedsmodel.predictfcn (x) function to monitor a data sample of a thermal power unit which is newly arrived at any time later, and bringing the sample into the x position of the model to obtain the coal consumption prediction of the thermal power unit corresponding to the sample.
The above description is only of the preferred embodiments of the present application and is not intended to limit the present application, but various modifications and variations can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (10)
1. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection is characterized by comprising the following steps of: the intelligent prediction method comprises the following steps:
step one, collecting historical monitoring data of a thermal power unit through a sensor arranged on the thermal power unit, and performing cluster analysis on the data by using a K-means algorithm to divide different basic working conditions of the thermal power unit;
step two, under the condition that the recorded data of various features in the thermal power generating unit system possibly have delay or advance between the features and between corresponding coal consumption, the feature construction is used for reflecting the delay or advance time phenomenon so as to obtain more potential quality subsets;
removing nonsensical redundant features by using a genetic algorithm to refine and simplify feature subsets, wherein the nonsensical redundant features are used for coping with a high-dimensional feature set to generate a plurality of adverse effects on a regression machine learning model;
and step four, establishing a bagging tree regression prediction model by using the feature subset screened in the step three as independent variable coal consumption as a dependent variable, and predicting the coal consumption of the thermal power unit in real time according to the regression prediction model for subsequent real-time monitoring data.
2. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the K-means cluster analysis in the first step is as follows:
for history monitoring data x= (X) collected by sensors mounted on thermal power generating unit 1 ,...,X d ) Clustering the data by using a K-means algorithm to divide different thermal power unit basic working conditions, wherein X is as follows j ＝(x 1,j ,…,x n,j ) T Is a column vector representing the jth feature in X, X i,j Is the ith sample of the jth feature, y= (Y) 1 ,...,y n ) T Is the corresponding coal consumption, wherein y i The coal consumption value representing the i-th sample, i=1, 2,..n, j=1, 2,..d, n is the total number of samples and d is the total number of features; setting the designated clustering number as k, and then randomly selecting k observation values as initial clustering centers; then, after measuring the distance between each observation point and each cluster center, designating the cluster closest to each observation point; then re-calculating the clustering center of each category and updating the clustering center result; repeating the process until the number of iterations reaches a predetermined threshold m; for the values of different cluster numbers k, evaluating the optimal cluster number by using a contour coefficient; and calculating contour coefficients corresponding to all different clustering numbers k.
3. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: and finally, selecting the corresponding cluster number with the largest contour coefficient as the final cluster number.
4. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the features in the second step are constructed as follows:
the effect of delay or advance at a point in time is characterized by a time interval for each feature.
5. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the feature construction considers a total of i delay or advance time intervals, respectively, and each feature can construct an additional 2 i features.
6. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the number of features increases from d to 2ld+d.
7. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the feature selection, extraction and simplification in the third step are as follows:
using genetic algorithms to effect feature selection; first encoding, each possible subset of the original feature set may be represented as a chromosome, and each gene in the chromosome indicates a respective feature; when the value of the gene is assigned 1, the corresponding feature is selected to represent this feature subset, and when the value of the gene is assigned 0, the corresponding feature is not selected to represent this feature subset; then randomly generating a given number of chromosomes according to the coding rule to form an initial population; each chromosome is evaluated using its root mean square error as a fitness function; generating a new generation chromosome population based on the selection, crossover and mutation operations of the genetic algorithm; repeating the previous step until the designated iteration number M is reached, and outputting the optimal chromosome in the last generation population.
8. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: and the feature subset corresponding to the optimal chromosome is the result of feature selection.
9. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the regression prediction model in the fourth step is as follows:
the feature subset screened in the step three is used as an independent variable, coal consumption is used as a dependent variable, and a 5-cross verification method is used for establishing a bagging tree regression prediction model through a bagging tree model; and carrying the sample into the regression prediction model to obtain the coal consumption prediction of the thermal power generating unit corresponding to the sample for the subsequent random newly arrived thermal power generating unit monitoring data sample.
10. The intelligent prediction method for the coal consumption of the thermal power generating unit based on feature construction and feature selection, which is characterized by comprising the following steps of: the bagging tree model is provided by Matlab 2021b software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310617519.3A CN116629432A (en) | 2023-05-29 | 2023-05-29 | Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310617519.3A CN116629432A (en) | 2023-05-29 | 2023-05-29 | Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116629432A true CN116629432A (en) | 2023-08-22 |
Family
ID=87612979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310617519.3A Pending CN116629432A (en) | 2023-05-29 | 2023-05-29 | Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116629432A (en) |
-
2023
- 2023-05-29 CN CN202310617519.3A patent/CN116629432A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Madhukumar et al. | Regression model-based short-term load forecasting for university campus load | |
| CN115796393B (en) | Energy management optimization method, system and storage medium based on multi-energy interaction | |
| TWI540533B (en) | Method of short-term wind power generation forecasting | |
| Kolhe et al. | GA-ANN for short-term wind energy prediction | |
| CN114511132A (en) | Photovoltaic output short-term prediction method and prediction system | |
| CN115471014A (en) | Power grid enterprise electricity carbon consumption prediction method based on deep neural network | |
| CN116599151A (en) | Source network storage safety management method based on multi-source data | |
| Pinson | Introducing distributed learning approaches in wind power forecasting | |
| CN117013527A (en) | Distributed photovoltaic power generation power prediction method | |
| Bao et al. | Wind turbine condition monitoring based on improved active learning strategy and KNN algorithm | |
| CN114611757A (en) | Electric power system short-term load prediction method based on genetic algorithm and improved depth residual error network | |
| Buonanno et al. | Comprehensive method for modeling uncertainties of solar irradiance for PV power generation in smart grids | |
| Zhang et al. | Joint forecasting of regional wind and solar power based on attention neural network | |
| CN112508278A (en) | Multi-connected system load prediction method based on evidence regression multi-model | |
| CN116632875A (en) | Hybrid energy storage planning method and system for power distribution network | |
| Su et al. | Opportunistic maintenance optimisation for offshore wind farm with considering random wind speed | |
| Catalão et al. | Short-term wind power forecasting using a hybrid evolutionary intelligent approach | |
| CN116629432A (en) | Intelligent prediction method for coal consumption of thermal power generating unit based on feature construction and feature selection | |
| Xu et al. | Research on Load Forecasting Based on CNN-LSTM Hybrid Deep Learning Model | |
| CN113901622A (en) | Genetic algorithm optimization-based SVM power distribution network topology identification method | |
| KR102292748B1 (en) | Solar cell power plant and control system, and its method of operating thereof | |
| CN114066068A (en) | Short-term power load prediction method, device, equipment and storage medium | |
| Liu et al. | Ultra-short-term wind power forecasting based on stacking model | |
| Wang et al. | Reliability evaluation of power-generating systems including time-dependent sources based on binary particle swarm optimization | |
| Ghaffari et al. | Optimal economic load dispatch based on wind energy and risk constrains through an intelligent algorithm |
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 |