CN113837463A - Nuclear power station system operation trend prediction method based on improved random forest - Google Patents

Nuclear power station system operation trend prediction method based on improved random forest Download PDF

Info

Publication number
CN113837463A
CN113837463A CN202111107117.6A CN202111107117A CN113837463A CN 113837463 A CN113837463 A CN 113837463A CN 202111107117 A CN202111107117 A CN 202111107117A CN 113837463 A CN113837463 A CN 113837463A
Authority
CN
China
Prior art keywords
prediction
random forest
data
slope
follows
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.)
Granted
Application number
CN202111107117.6A
Other languages
Chinese (zh)
Other versions
CN113837463B (en
Inventor
刘永阔
单龙飞
艾鑫
黄学颖
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.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
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 Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN202111107117.6A priority Critical patent/CN113837463B/en
Publication of CN113837463A publication Critical patent/CN113837463A/en
Application granted granted Critical
Publication of CN113837463B publication Critical patent/CN113837463B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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
    • G06Q50/06—Energy 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)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Evolutionary Computation (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Water Supply & Treatment (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method for predicting the operation trend of a nuclear power station system based on improved random forests, which comprises the following steps: (1) predicting the single step of the random forest; (2) a random forest multi-step prediction step; (3) and (4) multi-step prediction correction. The random forest method used by the invention is accurate in single-step prediction and poor in multi-step prediction effect, because the random forest can predict the change trend of a curve during prediction, but is insensitive to data change, the error of the random forest is likely to become larger when the data is increased or decreased and becomes slow, and the slope prediction considers the change of the slope in a short time and is sensitive to the change of the data in the short time, so that the prediction can be carried out in a weighted average mode of the random forest prediction and the slope prediction, the method can effectively improve the accuracy of the multi-step prediction and improve the usability of the multi-step prediction of the random forest.

Description

Nuclear power station system operation trend prediction method based on improved random forest
Technical Field
The invention relates to the field of prediction of on-line operation trends of nuclear power plant systems, in particular to a prediction method of operation trends of a nuclear power plant system based on an improved random forest.
Background
In the industrial field, safety is important, and especially in the nuclear power field with multiple accidents and high harmfulness, the safety is particularly important. The operation trend prediction method uses the historical data of the nuclear power station operation to predict the future operation trend of the nuclear power station, and corresponding measures are taken before the fault occurs, so that the further deterioration of the accident can be avoided. Therefore, the research on the operation trend prediction method has important significance for improving the safety of the nuclear power station.
At present, various methods are applied to the nuclear power plant for trend prediction. Such as: traditional statistical time series methods (autoregressive prediction method, moving average prediction method, autoregressive moving average prediction method), grey prediction method, neural network prediction method (such as BP neural network, etc.), and some machine learning prediction methods (such as support vector machine, random forest, etc.). The gray prediction method and the neural network prediction method are widely applied. The gray prediction method is suitable for short-term load prediction, especially in the absence of data samples. The BP neural network in the neural network prediction method is most applied, but the BP neural network has strong dependence on initial parameters and is easy to fall into a local minimum value, the convergence speed is low, and the prediction effect needs to be improved. In particular, in the real-time prediction process, each prediction needs to be retrained, and in order to avoid local minimum, a parameter optimization algorithm needs to be optimized, so that a large amount of resources need to be occupied. Random forests, while also commonly used for running trend prediction, have poor prediction accuracy in multi-step prediction processes.
Disclosure of Invention
The invention aims to provide a nuclear power station system operation trend prediction method based on an improved random forest, which aims to solve the problem that a multi-step prediction error of an operation trend prediction method is large.
The purpose of the invention is realized as follows:
a nuclear power station system operation trend prediction method based on improved random forests comprises the following steps:
step 1: predicting the single step of the random forest;
step 1.1: constructing a random forest model according to historical data of current parameters;
step 1.2: the next time stamped data is brought into the random forest. In the decision tree, the timestamp traverses the whole decision tree from the root node to the leaf node according to the rule when the decision tree is constructed, and the value of the leaf node is used as the predicted value of the decision tree;
step 1.3: repeating the step 1.2, the timestamp traverses each decision tree, each decision tree generates a predicted value, and the weighted average of all the predicted values is the final predicted value
Step 2: a random forest multi-step prediction step;
the time series multi-step prediction is based on the historical data X before the current time of the parameter and the predicted value p of each stepiTo predict the next data of the parameter, the following steps are carried out:
step 2.1: predicting data p at the next moment according to the random forest single step prediction in the step 1i；
Step 2.2: using the data predicted in step 1.1 and the historical data as new training data [ Xi]Training is performed by random forests. And repeating the random forest single step prediction to predict the data of the next moment.
Step 2.3: and (5) repeating the step 1.2 and the step 1.3 to realize multi-step prediction.
And step 3: multi-step prediction correction;
correcting the slope of the curve, and predicting the next moment data according to the current moment data and the slope of the curve at the next moment:
setting the current time as t, and setting n historical data of the current time as follows:
x＝{xt-n+1,xt-n+2,......,xt}
and (3) performing a least square method on each m pieces of historical data in the formula to obtain n-m +1 slopes in total:
k＝{kt-(n-m+2),kt-(n-m+3),......,kt}
the slope rate of change can be calculated from the slope k in the equation:
Figure BDA0003272869510000021
can be determined by the rate of change of slope in the above equation
Figure BDA0003272869510000022
To predict the slope of the next steps:
Figure BDA0003272869510000023
the prediction result of the random forest in the future n steps is set as follows:
X＝{xt+1,xt+2,......,xt+n}
the prediction result of the slope prediction in the future n steps is as follows:
Y＝{yt+1,yt+2,......yt+n}
the accuracy of the random forest is worse along with the increase of the prediction step number, the weighting weight corresponding to the result is reduced, and the weight range is verified to be as follows through experiments: [0.7,0.3], and the corresponding slope prediction weight value variation range is as follows: [0.3,0.7]. Therefore, the weight corresponding to each step of the n-step prediction random forest is linearly divided into:
w＝{0.7,...,n·(0.7-0.3)/15,...,0.3}
the corresponding slope prediction weight value change range is 1-w, and the final prediction result is as follows:
res＝{xt+1·w1+yt+1·(1-w1),......,xt+n·wn+yt+n·(1-wn)}
compared with the prior art, the invention has the beneficial effects that:
the random forest method used by the invention is accurate in single-step prediction and poor in multi-step prediction effect, because the random forest can predict the change trend of a curve during prediction, but is insensitive to data change, the error of the random forest is likely to become larger when the data is increased or decreased and becomes slow, and the slope prediction considers the change of the slope in a short time and is sensitive to the change of the data in the short time, so that the prediction can be carried out in a weighted average mode of the random forest prediction and the slope prediction, the method can effectively improve the accuracy of the multi-step prediction and improve the usability of the multi-step prediction of the random forest.
Drawings
FIG. 1 is a flow chart of the improved random forest prediction of the present invention;
FIG. 2 is a diagram of the prediction of the trend of the improved random forest single step operation of the present invention;
FIG. 3 is a diagram of the prediction of the multi-step operation trend of the improved random forest according to the present invention;
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The implementation of the invention comprises the following steps:
step 1: random forest single step prediction step
The time series single step prediction is to predict the data of the next moment of the parameter according to the historical data before the current time of the parameter, and comprises the following steps:
1) and constructing a random forest model according to the historical data of the current parameters.
2) The next time stamped data is brought into the random forest. In the decision tree, the timestamp traverses the whole decision tree from the root node to the leaf node according to the rule when the decision tree is constructed, and the value of the leaf node is used as the predicted value of the decision tree.
3) And repeating the step 2), traversing each decision tree by the timestamp, generating a predicted value by each decision tree, and taking the weighted average of all the predicted values as the final predicted value.
Step 2: random forest multi-step prediction step
The time series multi-step prediction is based on the historical data X before the current time of the parameter and the predicted value p of each stepiTo predict the next data of the parameter, the following steps are carried out:
1) predicting data p at the next moment according to the random forest single step prediction in the step 1i。
2) Using the predicted data and historical data in 1) as new training data [ X ],...pi]Training is performed by random forests. And repeating the random forest single step prediction to predict the data of the next moment.
3) And repeating the steps 1) and 2) to realize multi-step prediction.
And step 3: multi-step predictive correction
Because the deviation of the random forest multi-step prediction data and the actual data is large, the curve slope is adopted for correction. The change of the slope of the curve can reflect the change trend of the curve to a certain extent, and if the change of the slope of the curve at the next moment is known, the data at the next moment can be predicted through the data at the current moment and the slope of the curve at the next moment.
Setting the current time as t, and setting n historical data of the current time as follows:
x＝{xt-n+1,xt-n+2,......,xt} (1)
performing a least square method on each m pieces of historical data in the step (3) to obtain n-m +1 slopes in total:
k＝{kt-(n-m+2),kt-(n-m+3),......,kt} (2)
the slope change rate can be calculated by the slope k in (4):
Figure BDA0003272869510000041
can be measured by the rate of change of slope in (5)
Figure BDA0003272869510000042
To predict the slope of the next steps:
Figure BDA0003272869510000043
the random forest can predict the change trend of the curve during prediction, but the random forest is insensitive to data change, the error of the random forest may become larger when the data is increased or decreased, and the slope prediction considers the change of the slope in a short time and is sensitive to the change of the data in the short time, so the prediction can be performed in a weighted average mode of the random forest prediction and the slope prediction.
The prediction result of the random forest in the future n steps is set as follows:
X＝{xt+1,xt+2,......,xt+n} (5)
the prediction result of the slope prediction in the future n steps is as follows:
Y＝{yt+1,yt+2,......yt+n} (6)
the accuracy of the random forest is worse along with the increase of the prediction step number, the weighting weight corresponding to the result is reduced, and the weight range is verified to be as follows through experiments: [0.7,0.3], and the corresponding slope prediction weight value variation range is as follows: [0.3,0.7]. Therefore, the weight corresponding to each step of the n-step prediction random forest is linearly divided into:
w＝{0.7,...,n·(0.7-0.3)/15,...,0.3} (7)
the corresponding slope prediction weight value change range is 1-w, and the final prediction result is as follows:
res＝{xt+1·w1+yt+1·(1-w1),......,xt+n·wn+yt+n·(1-wn)} (8)
the program of the invention is compiled by Python3.6 language with Pycharm as a development platform, and has the main functions of:
when the system is connected, inputting real-time operation data of the nuclear power station, intercepting historical data with a certain length by using a sliding window as training data, training and improving a random forest prediction model by using the training data, and then predicting the operation trend of the nuclear power station system by using the trained model. The predicted results are shown in the figure together with the actual operation results.
The technical scheme of the invention is as follows:
1. the operating data of the pressurized water reactor nuclear power plant is obtained by using PCTran simulation software, and simulation data of a steam generator heat transfer pipe rupture accident is used as verification data.
2. An improved random forest prediction model is constructed according to the flow chart shown in fig. 1, and the flow is as follows:
a. selecting the step number i to be predicted;
b. selecting current k pieces of history data X ═ { X ] using a sliding windown-k+1,...,xnAnd the predicted value p of the next momentiCombined into a training number
According to { xn-k+1,...,xn,pi}；
c. Training the random forest and the slope prediction by using the training data in the step b, and outputting a prediction result of the next moment;
d. weighting the prediction results of the random forest and slope prediction in the step c and outputting a final prediction result pi；
f. And judging whether the number n of the current prediction step is less than i, if so, returning i predicted values, and otherwise, skipping to the step b to continue prediction.
(3) For the prediction process of (2), if the prediction step number i is 1, the prediction process is a single-step prediction, and the prediction result of the steam production of the steam generator is shown in fig. 2. If i >1 is the multi-step prediction, the results for steam generator steam production are shown in fig. 3, for example, i-15.

Claims (1)

1. A nuclear power station system operation trend prediction method based on improved random forests is characterized by comprising the following steps: the method comprises the following steps:
step 1: predicting the single step of the random forest;
step 1.1: constructing a random forest model according to historical data of current parameters;
step 1.2: the next time stamped data is brought into the random forest. In the decision tree, the timestamp traverses the whole decision tree from the root node to the leaf node according to the rule when the decision tree is constructed, and the value of the leaf node is used as the predicted value of the decision tree;
step 1.3: repeating the step 1.2, the timestamp traverses each decision tree, each decision tree generates a predicted value, and the weighted average of all the predicted values is the final predicted value
Step 2: a random forest multi-step prediction step;
the time series multi-step prediction is based on the history of the parameter before the current timeData X and predicted value p of each step predictioniTo predict the next data of the parameter, the following steps are carried out:
step 2.1: predicting data p at the next moment according to the random forest single step prediction in the step 1i；
Step 2.2: using the data predicted in step 1.1 and the historical data as new training data [ Xi]Training is performed by random forests. And repeating the random forest single step prediction to predict the data of the next moment.
Step 2.3: and (5) repeating the step 1.2 and the step 1.3 to realize multi-step prediction.
And step 3: multi-step prediction correction;
correcting the slope of the curve, and predicting the next moment data according to the current moment data and the slope of the curve at the next moment:
setting the current time as t, and setting n historical data of the current time as follows:
x＝{xt-n+1,xt-n+2,......,xt}
and (3) performing a least square method on each m pieces of historical data in the formula to obtain n-m +1 slopes in total:
k＝{kt-(n-m+2),kt-(n-m+3),......,kt}
the slope rate of change can be calculated from the slope k in the equation:
Figure FDA0003272869500000011
can be determined by the rate of change of slope in the above equation
Figure FDA0003272869500000012
To predict the slope of the next steps:
Figure FDA0003272869500000013
the prediction result of the random forest in the future n steps is set as follows:
X＝{xt+1,xt+2,......,xt+n}
the prediction result of the slope prediction in the future n steps is as follows:
Y＝{yt+1,yt+2,......yt+n}
the accuracy of the random forest is worse along with the increase of the prediction step number, the weighting weight corresponding to the result is reduced, and the weight range is verified to be as follows through experiments: [0.7,0.3], and the corresponding slope prediction weight value variation range is as follows: [0.3,0.7]. Therefore, the weight corresponding to each step of the n-step prediction random forest is linearly divided into:
w＝{0.7,...,n·(0.7-0.3)/15,...,0.3}
the corresponding slope prediction weight value change range is 1-w, and the final prediction result is as follows:
res＝{xt+1·w1+yt+1·(1-w1),......,xt+n·wn+yt+n·(1-wn)} 。
CN202111107117.6A 2021-09-22 2021-09-22 Nuclear power station system operation trend prediction method based on improved random forest Active CN113837463B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111107117.6A CN113837463B (en) 2021-09-22 2021-09-22 Nuclear power station system operation trend prediction method based on improved random forest

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111107117.6A CN113837463B (en) 2021-09-22 2021-09-22 Nuclear power station system operation trend prediction method based on improved random forest

Publications (2)

Publication Number Publication Date
CN113837463A true CN113837463A (en) 2021-12-24
CN113837463B CN113837463B (en) 2024-06-07

Family

ID=78960400

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111107117.6A Active CN113837463B (en) 2021-09-22 2021-09-22 Nuclear power station system operation trend prediction method based on improved random forest

Country Status (1)

Country Link
CN (1) CN113837463B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510858A (en) * 2009-03-24 2009-08-19 山东大学 Channel long-range forecast method based on slope correction
DE102013225997A1 (en) * 2013-12-16 2015-06-18 Robert Bosch Gmbh A method of determining a model value from a random forest model
CN109635309A (en) * 2018-10-17 2019-04-16 广州地理研究所 A kind of surface temperature space NO emissions reduction method
CN110717610A (en) * 2018-07-11 2020-01-21 华北电力大学（保定） Wind power prediction method based on data mining
AU2020100709A4 (en) * 2020-05-05 2020-06-11 Bao, Yuhang Mr A method of prediction model based on random forest algorithm
KR20200128966A (en) * 2019-05-07 2020-11-17 박해경 Severe Drought Area Prediction Model based on Random Forest using Satellite Image & Topography Data
CN112686713A (en) * 2021-01-21 2021-04-20 海澜智云科技有限公司 Clothing sales prediction method, system and equipment based on artificial intelligence
CN112862142A (en) * 2019-11-28 2021-05-28 新奥数能科技有限公司 Load and price prediction and correction method
CN113031031A (en) * 2021-03-09 2021-06-25 南京航空航天大学 Weighting positioning method based on GNSS signal accurate classification in urban canyon

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510858A (en) * 2009-03-24 2009-08-19 山东大学 Channel long-range forecast method based on slope correction
DE102013225997A1 (en) * 2013-12-16 2015-06-18 Robert Bosch Gmbh A method of determining a model value from a random forest model
CN110717610A (en) * 2018-07-11 2020-01-21 华北电力大学（保定） Wind power prediction method based on data mining
CN109635309A (en) * 2018-10-17 2019-04-16 广州地理研究所 A kind of surface temperature space NO emissions reduction method
KR20200128966A (en) * 2019-05-07 2020-11-17 박해경 Severe Drought Area Prediction Model based on Random Forest using Satellite Image & Topography Data
CN112862142A (en) * 2019-11-28 2021-05-28 新奥数能科技有限公司 Load and price prediction and correction method
AU2020100709A4 (en) * 2020-05-05 2020-06-11 Bao, Yuhang Mr A method of prediction model based on random forest algorithm
CN112686713A (en) * 2021-01-21 2021-04-20 海澜智云科技有限公司 Clothing sales prediction method, system and equipment based on artificial intelligence
CN113031031A (en) * 2021-03-09 2021-06-25 南京航空航天大学 Weighting positioning method based on GNSS signal accurate classification in urban canyon

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王兴德: "现代管理决策的计算机方法", 中国财政经济出版社, pages: 437 - 439 *

Also Published As

Publication number Publication date
CN113837463B (en) 2024-06-07

Similar Documents

Publication Publication Date Title
CN104794534B (en) A kind of power grid security Tendency Prediction method based on improvement deep learning model
CN109214708B (en) Electric power system risk assessment method based on cross entropy theory optimization support vector machine
GB2600213A (en) Plate performance prediction method and apparatus
Chen et al. Interpretable time-adaptive transient stability assessment based on dual-stage attention mechanism
CN111784061B (en) Training method, device and equipment for power grid engineering cost prediction model
CN112947672B (en) Maximum power point tracking method and device for photovoltaic cell
CN106773700B (en) The Multi model Predictive Controllers of genetic algorithm optimization
Sundarrajan et al. Open-loop control co-design of floating offshore wind turbines using linear parameter-varying models
CN113964884A (en) Power grid active frequency regulation and control method based on deep reinforcement learning
CN104865827A (en) Oil pumping unit oil extraction optimization method based on multi-working-condition model
CN116599050A (en) Photovoltaic prediction method and related device based on self-attention mechanism
CN113988481A (en) Wind power prediction method based on dynamic matrix prediction control
CN112615364A (en) Novel wide-area intelligent cooperative control method for power grid stability control device
CN115912367A (en) Intelligent generation method for operation mode of power system based on deep reinforcement learning
CN112819224B (en) Unit output prediction and confidence evaluation method based on deep learning fusion model
Xiangsong et al. PID controller design based on radial basis function neural networks for the steam generator level control
CN114326395A (en) Intelligent generator set control model online updating method based on working condition judgment
CN111950811B (en) Regional photovoltaic power prediction method and system based on double-layer artificial neural network
CN113837463B (en) Nuclear power station system operation trend prediction method based on improved random forest
CN113836819B (en) Bed temperature prediction method based on time sequence attention
CN114819391B (en) Photovoltaic power generation power prediction method and system based on historical data set time span optimization
CN116484675A (en) Crack propagation life prediction method and system for ship engine blade
CN112651628A (en) Power system transient stability evaluation method based on capsule neural network
CN112667957A (en) Intelligent electric energy meter failure rate prediction method based on deep neural network
CN110631003A (en) Reheated steam temperature adjusting method based on hierarchical scheduling multi-model predictive control

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant