CN112417764A - K nearest neighbor regression prediction method for boiler special equipment steam flow prediction - Google Patents

K nearest neighbor regression prediction method for boiler special equipment steam flow prediction Download PDF

Info

Publication number
CN112417764A
CN112417764A CN202011392931.2A CN202011392931A CN112417764A CN 112417764 A CN112417764 A CN 112417764A CN 202011392931 A CN202011392931 A CN 202011392931A CN 112417764 A CN112417764 A CN 112417764A
Authority
CN
China
Prior art keywords
data
steam
value
prediction
historical
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
Application number
CN202011392931.2A
Other languages
Chinese (zh)
Inventor
陈华
曾亚辉
顾娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Special Equipment Safety Supervision Inspection Institute of Jiangsu Province
Original Assignee
Special Equipment Safety Supervision Inspection Institute of Jiangsu Province
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Special Equipment Safety Supervision Inspection Institute of Jiangsu Province filed Critical Special Equipment Safety Supervision Inspection Institute of Jiangsu Province
Priority to CN202011392931.2A priority Critical patent/CN112417764A/en
Publication of CN112417764A publication Critical patent/CN112417764A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • 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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Fluid Mechanics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Optimization (AREA)
  • Algebra (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a K nearest neighbor regression prediction method for predicting steam flow of special boiler equipment, which is used for improving the accuracy and efficiency of steam flow prediction. The method comprises the following steps: 1. acquiring historical steam data and characteristic data, and preprocessing the data; 2. according to the correlation, the variables with low correlation are removed, the influence of interference variables on a predicted value is reduced, and dimension explosion is prevented; 3. dividing the historical data after the irrelevant variables are removed into a training set, a verification set and a test set, accelerating the determination of K neighbors through a KD tree algorithm, and further determining an optimal K value; 4. removing interference variables from the new data and predicting steam quantity according to the determined K value; 5. and periodically adding the data with low similarity to the historical data after new data is collected into the data set, and repeating the steps to achieve a better prediction effect. The method improves the accuracy of the steam quantity prediction of special equipment of the boiler, and provides decision assistance for safe and effective production.

Description

K nearest neighbor regression prediction method for boiler special equipment steam flow prediction
Technical Field
The invention relates to the field of data prediction, in particular to a K nearest neighbor regression prediction method for predicting steam quantity of special boiler equipment.
Background
The special equipment refers to boiler, pressure vessel, pressure pipeline, elevator and other equipment with great life safety and danger. Wherein the boiler refers to a pressure-bearing device which utilizes various fuels, electricity or other energy sources and passes through a designated volume. The steam volume of the boiler is an important index for measuring the stability and the production effectiveness of the boiler, and the accuracy of steam volume prediction can effectively improve the efficiency of the boiler and promote safe and efficient production.
Generally, factors involved in boiler steam prediction include primary air flow, feedwater flow, secondary air temperature, feedwater pressure, bed temperature of the boiler, bed pressure of the boiler, furnace temperature, and the like. The acquisition frequency of the boiler is acquired according to a fixed-length time sequence, and sometimes the steam quantity needs to be predicted according to acquired time sequence data and related characteristics to carry out detection or early warning. And the supply of each link can be adjusted in the model according to the required steam quantity, so that the appropriate supply quantity is found, and the adjustment risk in the actual production is reduced.
The regression analysis method is relatively mature in technical application, clear in model structure and suitable for being applied to actual production practice. The boiler steam prediction has a serious multiple collinearity problem among the collected data characteristics, and although the multiple collinearity problem is a very common problem in the process of actually applying the model, the serious collinearity problem may cause the traditional linear regression model to have very unstable performance. To effectively solve the instability, a ridge regression method is often used, which can reduce the influence of multiple collinearity to some extent, but has a higher requirement on the form of the historical data if a higher prediction accuracy is required. The K nearest neighbor regression can well utilize historical data, has high prediction accuracy, avoids instability caused by multiple collinearity, and reduces the accuracy of the K nearest neighbor regression by an interference variable with low prediction correlation. In addition, how to determine the K value has higher accuracy, and how to avoid the efficiency reduction of K nearest neighbor regression caused by dimension explosion and overlarge data amount is a problem to be faced when the K nearest neighbor regression prediction method is used for boiler special equipment steam prediction.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims at the characteristic data characteristics and the historical data of the steam quantity acquired by special boiler equipment, and establishes a K nearest neighbor regression prediction method for predicting the steam quantity of the special boiler equipment so as to improve the accuracy and efficiency of the steam quantity prediction.
The technical scheme is as follows: a K nearest neighbor regression prediction method for boiler special equipment steam flow prediction is characterized by comprising the following three aspects: the first aspect is how to determine the value of K, thereby better utilizing the existing historical data for the prediction of new observed data; the second aspect is to reduce the influence of some interference variables, even weak correlation variables, on the predicted value, and avoid performance degradation caused by dimension disaster; a third aspect is how to avoid the reduction in efficiency of K-nearest neighbor regression due to the search space being too large due to the data set being too large. The technical scheme of the invention provides the following steps:
step 1, acquiring historical steam data and characteristic data (including various sensor characteristics and instrument data), and preprocessing the data;
step 2, eliminating variables with low correlation according to the correlation, reducing the influence of interference variables on predicted values, and preventing dimension explosion;
step 3, dividing the historical data from which the irrelevant variables are removed in the step 2 into a training set, a verification set and a test set, accelerating the determination of K neighbors through a KD tree algorithm, and further determining an optimal K value;
step 4, screening out relevant variables of the new data according to the step 2, and predicting steam quantity according to the K value determined in the step 3;
and 5, after new data are collected regularly, adding the data with low similarity with the historical data into the data set, and returning to the step 1 again to carry out the process.
Steps 1 to 3 belong to an offline model updating part, and step 4 belongs to an online prediction part. The overall flow chart of the invention is shown in fig. 1, the flow chart of the off-line model is shown in fig. 2, and the flow chart of the on-line prediction is shown in fig. 3.
Further, in step 1, the collected d-feature set is V ═ V1,V2,...,VdD ═ X ═ historical steam data collected1,X2,...,Xm}，Y＝{y1,y2,...,ymContains m pieces of history data, each piece of data contains d-dimensional characteristics, namely Xi＝{xi1,xi2,...,xid}. The preprocessing of the features comprises missing value filling, outlier processing, normalization processing and the like.
Further, in step 2, the ith feature V is calculatediCoefficient of correlation r (V) with Yi,Y)，
Figure BDA0002811509810000021
Wherein, Cov (V)iY) is ViCovariance with Y, Var (V)i) Is ViVar (Y) is the variance of Y;
determining a threshold value z according to the obtained d r values, and removing variables | r | < z; the determination mode of the threshold z can be manually set, and the optimal threshold size can also be found by dividing a training set and a test set; in statistics, variables with the absolute value of the correlation coefficient smaller than 0.3 can be generally considered as weakly correlated variables, and the threshold value can be set to 0.3 here, and can also be dynamically adjusted according to the characteristics of actual data.
Further, in step 3, the history data D ' ═ { X ' ═ X ' of the feature removed in step 2 is set to be in step 3.1,X'2,...,X'm}，Y＝{y1,y2,...,ymDivide into training set Train { X }train,YtrainThe verification set validations { X }valid,YvalidAnd Test set Test { X }test,Ytest}. Set of features is V ═ V'1,V'2,...,V's}(s≤d)；
For training set { Xtrain,YtrainConstructing a KD tree, calculating the variance of each feature from the features of the training set, and obtaining the feature V 'with the maximum variance'kAs root node, take V'kTaking the median as a dividing point, dividing a sample smaller than the median into a left sub-tree, and otherwise dividing into a right sub-tree; generating a KD tree by recursion by using the same method; the KD tree can accelerate the search of K neighbors, improve the search efficiency and reduce the complexity;
taking different K values, for each K value, searching the verification set { X through a KD treevalid,YvalidEvery sample on }
Figure BDA0002811509810000031
The distance metric may use the euclidean distance, the manhattan distance, or the minkowski distance. The neighbor distance is calculated using the euclidean distance as follows:
Figure BDA0002811509810000032
wherein
Figure BDA0002811509810000033
Is the ith sample on the validation set,
Figure BDA0002811509810000034
are samples searched through the KD-tree on the training set,
Figure BDA0002811509810000035
is on the verification set
Figure BDA0002811509810000036
The value of the v-th dimension is,
Figure BDA0002811509810000037
is on the test set
Figure BDA0002811509810000038
The value of the v-th dimension of (a);
by measurement to obtain
Figure BDA0002811509810000039
The K neighbors of (a) are as follows:
Figure BDA00028115098100000310
obtained by distance weighted summation
Figure BDA00028115098100000311
Predicted value of (2)
Figure BDA00028115098100000312
The calculation formula is as follows:
Figure BDA00028115098100000313
wherein, wiIn order to be the distance weight,
Figure BDA00028115098100000314
is composed of
Figure BDA00028115098100000315
A corresponding steam value;
for each on the verification set
Figure BDA00028115098100000316
Predicted value of (2)
Figure BDA00028115098100000317
Calculating the square of the prediction error, and further calculating the square error (MSE) of the verification set; repeating the process by selecting different K values; and obtaining the optimal K value to be selected by selecting the minimum square error.
Further, in step 4, new sensor data or meter data is collected (or a value of the dependent variable is assumed as input based on data of the required steam amount); removing irrelevant variables according to the step 2; calculating K neighbors in the KD tree in the step 3 according to the determined K value; and setting weight according to the distance, and carrying out weighted summation to obtain the predicted value of the new data. To improve accuracy, the KD-tree for new data may also be reconstructed with all of the historical data.
Furthermore, in step 5, since the KD tree creation requires a certain time, new data can be periodically collected and then added to the historical data for periodic offline updating; in the selection of new data, the new data can be selected according to the similarity calculation, and the data with lower similarity to the historical data is added into the data set, so that the phenomenon that the KD tree is too large due to too large data set, so that the online search calculation consumption is too large is avoided, and the data set is enriched, so that the prediction can have better effect.
Has the advantages that: the method has the obvious advantages that the invention provides the method for predicting the steam of the special boiler equipment by K nearest neighbor regression, so as to improve the prediction effect, provide a more accurate steam prediction value for production, and provide an index for high-efficiency production. The invention has simple and practical model and higher efficiency, accords with the historical data shape characteristics of the steam of special equipment of the boiler, and has certain practical value.
Drawings
FIG. 1 is an overall flow chart of the present invention.
FIG. 2 is a flow chart of the K-nearest neighbor regression algorithm off-line training in the present invention.
FIG. 3 is a flow chart of the K-nearest neighbor regression algorithm for on-line prediction in the present invention.
FIG. 4 is an exemplary graph of severe multiple collinearity (VIF value >10) for multiple signatures of a boiler specialty.
FIG. 5 is an illustration of a thermodynamic diagram of boiler plant characteristic correlations.
Detailed Description
The invention is further illustrated below with reference to specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention.
The embodiment of the invention relates to a method for predicting steam volume of special equipment of a boiler by using K neighbor regression, and the implementation flow is shown in figure 1. The method comprises the following specific steps:
step 1, acquiring historical steam data and characteristic data (including various sensor characteristics and instrument data), and preprocessing the data, wherein the preprocessing specifically comprises the following steps:
step 11, collecting historical characteristic data and steam data and storing the historical characteristic data and the steam data in a database; in a further embodiment, the set of collected d-features is V ═ V1,V2,...,VdD ═ X ═ historical steam data collected1,X2,...,Xm}，Y＝{y1,y2,...,ymContains m pieces of history data, each piece of data contains d-dimensional characteristics, namely Xi＝{xi1,xi2,...,xid}。
And step 12, carrying out preprocessing such as missing value filling, outlier processing, normalization processing and the like on each column of characteristics.
Step 2, rejecting variables with low correlation according to the correlation, reducing the influence of interference variables on predicted values, and preventing dimension explosion, specifically comprising:
step 21, calculating the correlation r between each row of characteristics and the steam quantity; in a further embodiment, the ith feature V is calculatediCoefficient of correlation r (V) with Yi,Y)：
Figure BDA0002811509810000051
Wherein, Cov (V)iY) is ViCovariance with Y, Var (V)i) Is ViVar (Y) is the variance of Y;
determining a threshold value z according to the obtained d r values;
and step 22, eliminating the features with the | r | < 0.3.
Step 3, dividing the historical data from which the irrelevant variables are removed in the step 2 into a training set, a verification set and a test set, accelerating the determination of K neighbors through a KD tree algorithm, and further determining an optimal K value, wherein the method specifically comprises the following steps:
step 31, extracting the historical data with the variables removed from the database to divide the historical data into a training set, a verification set and a test setCollecting; in a further embodiment, the history data D '═ { X' ═ X 'of the feature culled in step 2 is'1,X'2,...,X'm}，Y＝{y1,y2,...,ymDivide into training set Train { X }train,YtrainThe verification set validations { X }valid,YvalidAnd Test set Test { X }test,Ytest}. Set of features V ═ V1',V2',...,Vs'}(s≤d)。
Step 32, constructing a KD tree according to the training set; in a further embodiment, for the training set { Xtrain,YtrainBuilding KD tree, calculating variance of each feature from features of training set, and calculating feature V with maximum variancek' As root node, take VkTaking the median of' as a dividing point, dividing samples smaller than the median into a left sub-tree, and otherwise dividing into a right sub-tree; generating a KD tree by recursion by using the same method;
and step 33, making K equal to 5, searching K most similar historical data in the KD tree for each piece of data in the verification set, and selecting an euclidean distance. Calculating weights of the K neighbors according to the distance between the K neighbors and the prediction points, and carrying out weighted summation on steam values of the K neighbors according to the weights to obtain steam quantity values of the prediction points; calculating the error between the predicted value and the true value of each piece of data, and further calculating the overall mean square error of the verification set;
step 34, making K equal to 6:12, repeating 3.3, and selecting an optimal K value according to the minimum value of the mean square error;
and step 35, testing and checking effects on the test set according to the optimal K value.
Step 4, screening out relevant variables for the new data according to the step 2, and predicting the steam quantity according to the K value determined in the step 3, wherein the method specifically comprises the following steps:
step 41, collecting the feature data on line, and forming an array by the feature data according to the feature sequence in the database;
42, removing the features confirmed in the step 22;
step 43, searching the optimal K neighbor in the constructed KD tree according to the data in the step 42;
and step 44, calculating the steam predicted value of the new data according to the weighted summation of the steam values of the K neighbors, wherein the weight setting mode is the same as that in the step 33.
And 5, after collecting new data regularly, adding the data with low similarity to the historical data into the data set, and returning to the step 1 again to perform the process, wherein the process specifically comprises the following steps:
step 51, periodically summarizing new data;
step 52, calculating the similarity of the new data and the historical data;
step 53, adding the data with lower similarity into the data table of 1.1;
and step 54, returning to the step 1 and repeating the process.
Steps 1-3 belong to an offline model updating part, and step 4 belongs to an online prediction part. The overall flow chart of the invention is shown in fig. 1, the flow chart of the off-line model is shown in fig. 2, and the flow chart of the on-line prediction is shown in fig. 3.
The embodiment provides a specific method, steps and parameters for performing steam prediction on special boiler equipment by k-nearest neighbor regression, so that the prediction effect is improved, a relatively accurate steam prediction value can be provided for production, and an index can be provided for efficient production. In addition, the implementation is only a preferred specific implementation of the present invention, and the setting of the parameters needs to be adjusted according to specific variables and data in the specific implementation process, so as to achieve a better practical effect.

Claims (6)

1. A K nearest neighbor regression prediction method for boiler special equipment steam flow prediction is characterized by comprising the following steps:
step S1, acquiring historical steam data and characteristic data including characteristic data of each sensor and instrument data, and preprocessing the data;
step S2, eliminating variables with low correlation according to the correlation, reducing the influence of interference variables on predicted values, and preventing dimension explosion;
step S3: dividing the historical data from which the irrelevant variables are removed in the step S2 into a training set, a verification set and a test set, accelerating the determination of K neighbors through a KD tree algorithm, and further determining an optimal K value;
step S4: screening out relevant variables for the new data according to the step S2, and predicting the steam quantity according to the K value determined in the step S3;
step S5: and after new data are collected regularly, adding the data with low similarity to the historical data into the data set, and returning to the step 1 again to carry out the process.
2. The method for predicting the steam flow of the special equipment of the boiler based on the K-nearest neighbor regression as claimed in claim 1, wherein said step S1 further comprises:
step S101, collecting historical characteristic data and steam data and storing the historical characteristic data and the steam data in a database;
and step S102, carrying out preprocessing such as missing value filling, outlier processing, normalization processing and the like on each column of characteristics.
3. The method for predicting the steam flow of the special equipment of the boiler based on the K-nearest neighbor regression as claimed in claim 1, wherein said step S2 further comprises:
step S201, calculating the correlation r between each row of characteristics and the steam quantity;
and step S202, eliminating the features with the correlation | r | < 0.3.
4. The method for predicting the steam flow of the special equipment of the boiler based on the K-nearest neighbor regression as claimed in claim 1, wherein said step S3 further comprises:
step S301, taking out historical data with variables removed from a database to divide the historical data into a training set, a verification set and a test set;
s302, constructing a KD tree according to the training set;
step S303, letting K = 5, searching K most similar historical data in the KD tree for each data in the verification set, and selecting Euclidean distance from the K most similar historical data; calculating weights of the K neighbors according to the distance between the K neighbors and the prediction points, and carrying out weighted summation on steam values of the K neighbors according to the weights to obtain steam quantity values of the prediction points; calculating the error between the predicted value and the true value of each piece of data, and further calculating the overall mean square error of the verification set;
step S304, letting K = 6:12, repeating step S303, and selecting an optimal K value according to the minimum value of the mean square error;
and S305, testing and checking effects on the test set according to the optimal K value.
5. The method for predicting the steam flow of the special equipment of the boiler based on the K-nearest neighbor regression as claimed in claim 1, wherein said step S4 further comprises:
s401, collecting feature data on line, and forming arrays of the feature data according to the feature sequence in a database;
step S402, removing the characteristics confirmed in the step S202;
step S403, searching the optimal K neighbor in the constructed KD tree according to the data in the step S402;
and S404, calculating a steam predicted value of the new data according to the weighted summation of the steam values of the K neighbors, wherein the weight setting mode is the same as that in S303.
6. The method for predicting the steam flow of the special equipment of the boiler based on the K-nearest neighbor regression as claimed in claim 1, wherein said step S5 further comprises:
step S501, new data are gathered periodically;
step S502, calculating the similarity of the new data and the historical data;
step S503, adding the data with low similarity into the data table of the step S101;
step S504 returns to step S1, and the above-described flow is repeated.
CN202011392931.2A 2020-12-01 2020-12-01 K nearest neighbor regression prediction method for boiler special equipment steam flow prediction Pending CN112417764A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011392931.2A CN112417764A (en) 2020-12-01 2020-12-01 K nearest neighbor regression prediction method for boiler special equipment steam flow prediction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011392931.2A CN112417764A (en) 2020-12-01 2020-12-01 K nearest neighbor regression prediction method for boiler special equipment steam flow prediction

Publications (1)

Publication Number Publication Date
CN112417764A true CN112417764A (en) 2021-02-26

Family

ID=74829680

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011392931.2A Pending CN112417764A (en) 2020-12-01 2020-12-01 K nearest neighbor regression prediction method for boiler special equipment steam flow prediction

Country Status (1)

Country Link
CN (1) CN112417764A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112906307A (en) * 2021-03-24 2021-06-04 青岛洪锦智慧能源技术有限公司 Steam yield prediction method of steam boiler based on data mining
CN113282124A (en) * 2021-07-20 2021-08-20 深圳市佳运通电子有限公司 Control method of heating furnace gas flow, temperature hosting method and system
CN113779754A (en) * 2021-08-02 2021-12-10 张家港宏昌钢板有限公司 Method and system for analyzing influence factors of blast furnace, electronic device and computer-readable storage medium
CN113934185A (en) * 2021-10-09 2022-01-14 北京远舢智能科技有限公司 Control method and device for improving process quality and electronic equipment

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415885A (en) * 2018-02-08 2018-08-17 武汉蓝泰源信息技术有限公司 The real-time bus passenger flow prediction technique returned based on neighbour
CN109034879A (en) * 2018-07-06 2018-12-18 东华大学 A kind of cloud computing based on k neighbour's regression algorithm is bidded example price expectation method
CN110222416A (en) * 2019-06-05 2019-09-10 重庆邮电大学 A kind of industrial steam amount prediction technique based on big data
CN110378064A (en) * 2019-07-29 2019-10-25 安徽工业大学 A method of the prediction dead stock column temperature of blast furnace furnace core
CN110852511A (en) * 2019-11-13 2020-02-28 苏文电能科技股份有限公司 Method for predicting short-term electricity consumption of residents based on weather temperature and k-neighborhood theory
CN111768034A (en) * 2020-06-29 2020-10-13 上海积成能源科技有限公司 Method for interpolating and supplementing missing value based on neighbor algorithm in power load prediction

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415885A (en) * 2018-02-08 2018-08-17 武汉蓝泰源信息技术有限公司 The real-time bus passenger flow prediction technique returned based on neighbour
CN109034879A (en) * 2018-07-06 2018-12-18 东华大学 A kind of cloud computing based on k neighbour's regression algorithm is bidded example price expectation method
CN110222416A (en) * 2019-06-05 2019-09-10 重庆邮电大学 A kind of industrial steam amount prediction technique based on big data
CN110378064A (en) * 2019-07-29 2019-10-25 安徽工业大学 A method of the prediction dead stock column temperature of blast furnace furnace core
CN110852511A (en) * 2019-11-13 2020-02-28 苏文电能科技股份有限公司 Method for predicting short-term electricity consumption of residents based on weather temperature and k-neighborhood theory
CN111768034A (en) * 2020-06-29 2020-10-13 上海积成能源科技有限公司 Method for interpolating and supplementing missing value based on neighbor algorithm in power load prediction

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
唐振浩等: "带有特征选取电站锅炉燃烧效率建模", 《哈尔滨理工大学学报》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112906307A (en) * 2021-03-24 2021-06-04 青岛洪锦智慧能源技术有限公司 Steam yield prediction method of steam boiler based on data mining
CN113282124A (en) * 2021-07-20 2021-08-20 深圳市佳运通电子有限公司 Control method of heating furnace gas flow, temperature hosting method and system
CN113282124B (en) * 2021-07-20 2021-09-21 深圳市佳运通电子有限公司 Control method of heating furnace gas flow, temperature hosting method and system
CN113779754A (en) * 2021-08-02 2021-12-10 张家港宏昌钢板有限公司 Method and system for analyzing influence factors of blast furnace, electronic device and computer-readable storage medium
CN113934185A (en) * 2021-10-09 2022-01-14 北京远舢智能科技有限公司 Control method and device for improving process quality and electronic equipment

Similar Documents

Publication Publication Date Title
CN112417764A (en) K nearest neighbor regression prediction method for boiler special equipment steam flow prediction
CN109992921B (en) On-line soft measurement method and system for thermal efficiency of boiler of coal-fired power plant
CN112699913A (en) Transformer area household variable relation abnormity diagnosis method and device
CN106055918A (en) Power system load data identification and recovery method
CN113570138B (en) Method and device for predicting residual service life of equipment of time convolution network
CN110570030A (en) Wind power cluster power interval prediction method and system based on deep learning
CN110766200A (en) Method for predicting generating power of wind turbine generator based on K-means mean clustering
CN109066651B (en) Method for calculating limit transmission power of wind power-load scene
CN113515049A (en) Operation regulation and control system and method for gas-steam combined cycle generator set
CN109523077B (en) Wind power prediction method
CN108830417A (en) A kind of residential energy consumption prediction technique and system based on ARMA and regression analysis
CN116186624A (en) Boiler assessment method and system based on artificial intelligence
CN118194607B (en) Dynamic modeling method and system for boiler unit data
CN109921426A (en) Wind-electricity integration system probability load flow calculation method based on CV-KDE
CN114970766A (en) Power station unit operation parameter reference value obtaining method based on linear fitting
CN111142488B (en) Process industrial optimization method based on working condition comparison driving
CN108664000A (en) A kind of alumina producing evaporation process steady state detecting method for use and system
CN112765746B (en) Turbine blade top gas-thermal performance uncertainty quantification system based on polynomial chaos
CN116029433A (en) Energy efficiency reference value judging method, system, equipment and medium based on grey prediction
CN116384740A (en) Pipe network leakage risk assessment method
CN112348696B (en) BP neural network-based heating unit peak regulation upper limit evaluation method and system
CN115081899A (en) Online self-adaptive obtaining method for power station unit operation parameter reference value
CN115017818A (en) Power plant flue gas oxygen content intelligent prediction method based on attention mechanism and multilayer LSTM
CN113408076A (en) Small sample mechanical residual life prediction method based on support vector machine model
Kler et al. High-speed mathematical models of cogeneration steam turbines: optimization of operation at heat and power plants

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