CN107085750A - A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN - Google Patents
A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN Download PDFInfo
- Publication number
- CN107085750A CN107085750A CN201710140823.8A CN201710140823A CN107085750A CN 107085750 A CN107085750 A CN 107085750A CN 201710140823 A CN201710140823 A CN 201710140823A CN 107085750 A CN107085750 A CN 107085750A
- Authority
- CN
- China
- Prior art keywords
- model
- prediction
- data
- ann
- arma
- 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
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Operations Research (AREA)
- Economics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Algebra (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Set up the linear submodels of ARMA the invention discloses a kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN, including step and set up ANN nonlinearities models, so as to obtain mixing mixed model.Compared with prior art, the present invention combines advantages of the ARMA in the advantage and ANN in terms of pull-in time sequences part in terms of Nonlinear Time Series are predicted, influence of the real time data to model parameter is considered during prediction, with reference to ARMA and ANN prediction process, set up real-time dynamic forecast model, it is to avoid the single respective limitation of model.
Description
Technical field
The present invention relates to a kind of equipment fault Forecasting Methodology, more particularly to a kind of mixing dynamic event based on ARMA and ANN
Hinder Forecasting Methodology.
Background technology
Prediction science is a new branch of science, and it is according to historical data and status data, in the guidance of correlation theory and method
Under, analysis and the state of development and trend for inferring research object future, Predicting Technique have been widely used in industry, business at present
The fields such as industry, finance, meteorology.Status predication technology is that, according to equipment operation condition, assessment equipment current state simultaneously predicts future
State.Predicting Technique will can be divided into three classes according to the level of application, precision of prediction and relevant cost of Forecasting Methodology：Based on can
By Forecasting Methodology, the Forecasting Methodology based on data-driven and the Forecasting Methodology based on timeliness physical model that property is theoretical, three kinds of sides
Popularity of the method in engineer applied weakens successively, but precision of prediction is raised successively, relative difficulty and cost also with
Increase.
Existing Predicting Technique has been achieved for larger progress in terms of theoretical research and practical application, still, existing
Forecasting Methodology there is also many limitations, prediction process is larger to the degree of dependence of mathematical modeling, it is impossible to meet complication system
Actual requirement, satisfied result can not be obtained when the mathematical modeling of system is inaccurate.And most forecast models belong to static
Model, lacks self-learning capability, and forecast model keeps immobilizing, do not accounted for by once modeling acquisition, model parameter
Newly-increased influence of the sample to model parameter, for complex equipment prediction generally occur within that Single-step Prediction is inaccurate, multi-step prediction without
The problem of effect.
In current failure prediction method, autoregressive moving-average model is suitable for the linear segment of pull-in time sequence,
And when solving the problems, such as complex nonlinear, error is often very big；And neutral net predict Nonlinear Time Series when effect compared with
It is good, but neutral net shows poor when predicting linear time series.
The content of the invention
To overcome the deficiencies in the prior art, with reference to the respective advantage of ARMA and ANN methods, so that preferably to time series
It is predicted, improves precision of prediction, the present invention proposes a kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN.
The technical proposal of the invention is realized in this way：
A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN, including step
S1：According to the characteristics of sample data, tranquilization data prediction is carried out to sample data, data sequence is generated；
S2：According to the property and AIC criterion of the auto-correlation coefficient of the data sequence and partial correlation coefficient, estimated data sequence
The Autoregressive and moving average order of row, determine the model of data sequence；
S3：Model parameter estimation is carried out according to least square method, the observation and historical juncture for determining current time are observed
The relation of value and white noise sequence；
S4：Verify whether the model reaches precision using the data sequence, if step S2 is otherwise gone back to, until obtaining
Rational arma modeling, and then obtain static multi-step prediction error；
S5：Historical data is substituted into the predictive equation of the ARAM models, the data of subsequent time are obtained；
S6：Repeat step S3-S5 carries out L step predictions, and the data of prediction are added into data sequence；
S7：If prediction loop tests the speed less than prediction data number when carrying out L step predictions, S8 is gone to step；Otherwise, obtain
Predicting the outcome for linear segment, goes to step S9；
S8：Actual observed value is substituted into L step predicted values, as time series, S3 is gone to step, is circulated next time
L step predictions；
S9：The practical static multi-step prediction error, trains ANN model, prediction residual is obtained according to described predict the outcome,
As the time series data of ANN model, repeat step S5-S8 obtains predicting the outcome for non-linear partial；
S10：By predicting the outcome for the linear segment and predicting the outcome for non-linear partial, the prediction of mixed model is obtained
As a result.
Further, in step S10 the predicting the outcome of mixed model=linear segment the+non-linear that predicts the outcome
That divides predicts the outcome.
The beneficial effects of the present invention are compared with prior art, the present invention combines ARMA in pull-in time sequences
The advantage of advantage and ANN in terms of Nonlinear Time Series are predicted in terms of part, considers real time data pair during prediction
The influence of model parameter, with reference to ARMA and ANN prediction process, sets up real-time dynamic forecast model, it is to avoid single model is each
Limitation.
Brief description of the drawings
Fig. 1 is a kind of mixing dynamic fault Forecasting Methodology flow chart based on ARMA and ANN of the present invention；
Fig. 2 is a kind of mixing dynamic fault Forecasting Methodology flow logic figure based on ARMA and ANN of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Because autoregressive moving-average model is suitable for the linear segment of pull-in time sequence, and solving complex nonlinear
During problem, error is often very big.And neutral net effect when predicting Nonlinear Time Series is preferable, but neutral net is pre-
Showed during linear time series poor.So, propose that the mixed model that a kind of ARMA and ANN methods are combined makes it have simultaneously
There are two kinds of respective advantages of model, so as to be preferably predicted to time series, improve precision of prediction.Mixed model bag
Include the linear submodels of ARMA and ANN nonlinearities model two parts.
Refer to Fig. 1, a kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN of the present invention, including step
S1：Tranquilization, data prediction are carried out according to the characteristics of sample data, if data sequence is after processingInstruct
Practice data；
If described time series X=(x1,x2,…,xl), t is current time, and it is carried out to mix dynamic L steps prediction.
Initial time k=t, j=1 (j is prediction loop number of times), N1 is prediction data number.
S2：The determination of Model Identification, i.e. model structure, according to the auto-correlation coefficient (ACF) of rotary speed data sequence and inclined phase
The property and AIC criterion of relation number (PACF) go to estimate Autoregressive n and moving average order m；
Autoregressive moving-average model (ARMA) is a kind of temporal model, not only can reveal that the rule of dynamic data, in advance
Its future value is surveyed, but also the relevant characteristic of system can be studied from many aspects.
For normal state, the time series { x of steady, zero-meant, if xtValue it is not only relevant with the value that its preceding n is walked,
And it is relevant with the excitation that preceding m is walked, then there is general arma modeling by autoregression (AR) model and moving average (MA) model group
Conjunction is formed.
Wherein, n and m are respectively autoregression and moving average exponent number, are abbreviated as ARMA (n, m), if n=0, and this model is
MA models, if m=0, this model is AR models.Real numberReferred to as autoregressive coefficient, real number θiFor moving average coefficient, sequence
{atIt is white noise sequence.
S3：Model parameter estimation is carried out according to least square method, the observation and historical juncture for determining current time are observed
The relation of value and white noise sequence；
During arma modeling is predicted to time series, difference is carried out to time series first, obtains steady random
Sequence, it is then determined that model order, selects suitable model, then estimates model parameter, computation model parameter value, finally
Adaptive test is carried out to model, model application is carried out.
S4：Utilize training dataWhether testing model reaches precision, if meeting, that is, obtains rational arma modeling, from
And static multi-step prediction error e train (t) (namely ANN training data) is obtained, go to step 5；Otherwise, 2 are gone to step；
S5：By historical dataGeneration such as predictive equation, obtains k+1 time datas
Input layer：Input vector X=(x1,x2,…,xl) it is the Condition Monitoring Data of equipment or system, and have passed through certain
Pretreatment, such as noise reduction, normalization.Intermediate layer：Intermediate layer is also known as hidden layer, and it can also be multilayer knot that can be one layer
Structure, passes through wijAnd wjkConnect input layer and output layer.Output layer：Output valve is predicted value, and output layer nodes m is prediction
As a result sum, Yt=(y1,y2,…,yt)。
S6：If k+1-n<Prediction data is added sequence by L, then k=k+1, wherein L to carry out L step mixing dynamic predictions,
3 are gone to step, parameter is reevaluated；Otherwise 7 are gone to step；
Neutral net mainly realizes forecast function by two methods, the first using neutral net as function approximator,
Prediction, the dynamic relationship between second consideration input, output, with the dynamic neural network pair with feedback are fitted to parameter
Parameter is set up dynamic model and is predicted.During being predicted to time series, generally using the neutral net with feedback
It is predicted.
During being predicted based on neural network model, first using Condition Monitoring Data as sample, the rational instruction of selection
Practice, test and analysis sample；Then training pattern is set by network parameter；The network model of training is entered with test sample again
Row test, examines network performance；Finally it is predicted with model and analysis sample.
S7：If jL<N, both during jth time circulation, prediction loop number of times is less than prediction data when carrying out L step mixing dynamic predictions
Number, go to step 8；Otherwise predicted the outcomeGo to step 9；
Assuming that being X=(x for time series input1,x2,…,xl), its really desired output be Yt=(y1,y2,…,
yt).Time series is predicted first with arma modeling, then had：
WhereinFor ARMA predicted values, model order is carried out using AIC criterion, AIC functions are defined as：Make auto-correlation system
NumberThen
Wherein, N is sample size,For the maximum likelihood estimator corresponding to various algorithms.To n, m from low order to high-order
Different values be respectively established, and carry out parameter Estimation, the AIC values of relatively more each model reach minimum model then
For best model, as shown in Equation 3：
Formula 3 is written as form, parameter Estimation is carried out using least square method
Wherein,Calculating solution makes mould
Value of the type when error sum of squares is minimum, that is, seek the minimum of following formula.
To it is upper examination derivation can in the hope of parameter beta estimatorThen have
It can be calculated by above procedure and obtain linear submodel prediction remainder：
S8：L is obtained by step S7, the L step predicted values before actual observed value is replaced, as time series, turn step
Rapid S3, carries out the L step predictions of jth=j+1 circulation；
S9：ANN model is trained with step S4 etrain (t), what ARMA was obtained in step S7 predicts the outcome
Obtain prediction residual YN(t), as ANN time series data, with repeat step S5-S8, the prediction of non-linear partial is obtained
As a result
The remainder Y obtained using formula 6N(t) neural network model is set up, nonlinearities model part is set up：
WhereinTo predict the outcome, wj(j=0,1,2 ..., q) and wij(i=0,1,2 ..., p；J=0,1,2 ..., q) be
The connection weight of neutral net, p, q represent network input layer and the nodes in intermediate layer respectively, usual output layer be 1 be used for into
The step forward prediction of row one, b0And b0jFor bias term, εtFor the predicated error of t, g is the activation primitive of network, is generally used
Logistic function representations, i.e.,：
S10：ByWithObtain predicting the outcome for mixed model：
Composite type 1,6 and 7, obtaining mixed model expression formula is：
Wherein,Finally to predict the outcome,Predicted the outcome for ARMA, YN(t) residual error for being ARMA, and as
The input of ANN model, trains ANN model,Predicted the outcome for ANN model.
Because autoregressive moving-average model is unable to the non-linear partial of pull-in time sequence, so the remainder that formula 9 is obtained
The non-linear component of time series is contained, is combined and can obtain using neural net model establishing remainder, then by two parts result
Higher precision of prediction.
Described above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (2)
1. a kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN, it is characterised in that including step
S1：According to the characteristics of sample data, tranquilization data prediction is carried out to sample data, data sequence is generated；
S2：According to the property and AIC criterion of the auto-correlation coefficient of the data sequence and partial correlation coefficient, estimated data sequence
Autoregressive and moving average order, determine the model of data sequence；
S3：Model parameter estimation is carried out according to least square method, determine the observation at current time and historical juncture observation and
The relation of white noise sequence；
S4：Verify whether the model reaches precision using the data sequence, if otherwise going back to step S2, until obtaining rationally
Arma modeling, and then obtain static multi-step prediction error；
S5：Historical data is substituted into the predictive equation of the ARAM models, the data of subsequent time are obtained；
S6：Repeat step S3-S5 carries out L step predictions, and the data of prediction are added into data sequence；
S7：If prediction loop tests the speed less than prediction data number when carrying out L step predictions, S8 is gone to step；Otherwise, obtain linear
Partial predicts the outcome, and goes to step S9；
S8：Actual observed value is substituted into L step predicted values, as time series, S3 is gone to step, the L steps circulated next time
Prediction；
S9：The practical static multi-step prediction error, trains ANN model, prediction residual is obtained according to described predict the outcome, as
The time series data of ANN model, repeat step S5-S8 obtains predicting the outcome for non-linear partial；
S10：By predicting the outcome for the linear segment and predicting the outcome for non-linear partial, the prediction knot of mixed model is obtained
Really.
2. a kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN as claimed in claim 1, it is characterised in that step
The predicting the outcome of the predicting the outcome of mixed model=linear segment+non-linear partial predicts the outcome in rapid S10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710140823.8A CN107085750A (en) | 2017-03-10 | 2017-03-10 | A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710140823.8A CN107085750A (en) | 2017-03-10 | 2017-03-10 | A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107085750A true CN107085750A (en) | 2017-08-22 |
Family
ID=59615097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710140823.8A Pending CN107085750A (en) | 2017-03-10 | 2017-03-10 | A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107085750A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107483269A (en) * | 2017-09-20 | 2017-12-15 | 程丹秋 | A kind of efficient network apparatus management system |
| CN107590010A (en) * | 2017-08-31 | 2018-01-16 | 西安电子科技大学 | A kind of electromagnetic compatibility Analysis on Fault Diagnosis method based on Dynamic fault tree |
| CN108282360A (en) * | 2017-12-28 | 2018-07-13 | 深圳先进技术研究院 | A kind of fault detection method of shot and long term prediction fusion |
| CN108734341A (en) * | 2018-04-27 | 2018-11-02 | 广东电网有限责任公司 | A kind of self-organizing Electric Load Forecasting reporting method based on time series arma modeling |
| CN109324266A (en) * | 2018-11-21 | 2019-02-12 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of distribution single-phase-to-earth fault analysis method based on deep learning |
| CN109614742A (en) * | 2018-12-25 | 2019-04-12 | 中国海洋大学 | A kind of sea level height duration prediction algorithm |
| CN111190045A (en) * | 2019-12-27 | 2020-05-22 | 国网北京市电力公司 | Voltage abnormity prediction method and device and electronic equipment |
| CN111445006A (en) * | 2020-02-27 | 2020-07-24 | 上海交通大学 | Method and system for predicting number of submission times of developer codes in open source community |
| CN111612226A (en) * | 2020-05-12 | 2020-09-01 | 中国电子科技集团公司电子科学研究院 | Group daily average arrival number prediction method and device based on hybrid model |
| CN112685910A (en) * | 2021-01-11 | 2021-04-20 | 大连理工大学 | Complex equipment power pack fault prediction method based on hybrid prediction model |
| CN113255965A (en) * | 2021-04-26 | 2021-08-13 | 大连海事大学 | Intelligent processing system for prognosis of degradation fault of radar transmitter |
| CN114151147A (en) * | 2021-11-30 | 2022-03-08 | 西安热工研究院有限公司 | Fault early warning method, system, equipment and medium for abnormal rotating speed of steam turbine |
| CN114722876A (en) * | 2022-04-21 | 2022-07-08 | 沈阳顺义科技有限公司 | State prediction and mode classification method and system for rotary machine |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105158598A (en) * | 2015-08-15 | 2015-12-16 | 国家电网公司 | Fault prediction method suitable for power equipment |
| CN105809364A (en) * | 2016-03-22 | 2016-07-27 | 南京邮电大学 | User behavior prediction method based on Markov model |
-
2017
- 2017-03-10 CN CN201710140823.8A patent/CN107085750A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105158598A (en) * | 2015-08-15 | 2015-12-16 | 国家电网公司 | Fault prediction method suitable for power equipment |
| CN105809364A (en) * | 2016-03-22 | 2016-07-27 | 南京邮电大学 | User behavior prediction method based on Markov model |
Non-Patent Citations (1)
| Title |
|---|
| 李向前: "复杂装备故障预测与健康管理关键技术研究", 《中国博士学位论文全文数据库》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590010A (en) * | 2017-08-31 | 2018-01-16 | 西安电子科技大学 | A kind of electromagnetic compatibility Analysis on Fault Diagnosis method based on Dynamic fault tree |
| CN107590010B (en) * | 2017-08-31 | 2019-07-02 | 西安电子科技大学 | A kind of electromagnetic compatibility Analysis on Fault Diagnosis method based on Dynamic fault tree |
| CN107483269A (en) * | 2017-09-20 | 2017-12-15 | 程丹秋 | A kind of efficient network apparatus management system |
| CN108282360B (en) * | 2017-12-28 | 2021-06-18 | 深圳先进技术研究院 | Fault detection method for long-term and short-term prediction fusion |
| CN108282360A (en) * | 2017-12-28 | 2018-07-13 | 深圳先进技术研究院 | A kind of fault detection method of shot and long term prediction fusion |
| CN108734341A (en) * | 2018-04-27 | 2018-11-02 | 广东电网有限责任公司 | A kind of self-organizing Electric Load Forecasting reporting method based on time series arma modeling |
| CN109324266A (en) * | 2018-11-21 | 2019-02-12 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of distribution single-phase-to-earth fault analysis method based on deep learning |
| CN109614742A (en) * | 2018-12-25 | 2019-04-12 | 中国海洋大学 | A kind of sea level height duration prediction algorithm |
| CN111190045A (en) * | 2019-12-27 | 2020-05-22 | 国网北京市电力公司 | Voltage abnormity prediction method and device and electronic equipment |
| CN111445006A (en) * | 2020-02-27 | 2020-07-24 | 上海交通大学 | Method and system for predicting number of submission times of developer codes in open source community |
| CN111612226A (en) * | 2020-05-12 | 2020-09-01 | 中国电子科技集团公司电子科学研究院 | Group daily average arrival number prediction method and device based on hybrid model |
| CN112685910A (en) * | 2021-01-11 | 2021-04-20 | 大连理工大学 | Complex equipment power pack fault prediction method based on hybrid prediction model |
| WO2022147853A1 (en) * | 2021-01-11 | 2022-07-14 | 大连理工大学 | Complex equipment power pack fault prediction method based on hybrid prediction model |
| CN112685910B (en) * | 2021-01-11 | 2022-11-18 | 大连理工大学 | Complex equipment power pack fault prediction method based on hybrid prediction model |
| CN113255965A (en) * | 2021-04-26 | 2021-08-13 | 大连海事大学 | Intelligent processing system for prognosis of degradation fault of radar transmitter |
| CN114151147A (en) * | 2021-11-30 | 2022-03-08 | 西安热工研究院有限公司 | Fault early warning method, system, equipment and medium for abnormal rotating speed of steam turbine |
| CN114151147B (en) * | 2021-11-30 | 2024-04-26 | 西安热工研究院有限公司 | Fault early warning method, system, equipment and medium for abnormal rotating speed of steam turbine |
| CN114722876A (en) * | 2022-04-21 | 2022-07-08 | 沈阳顺义科技有限公司 | State prediction and mode classification method and system for rotary machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107085750A (en) | A kind of mixing dynamic fault Forecasting Methodology based on ARMA and ANN | |
| CN105550323B (en) | Load balance prediction method and prediction analyzer for distributed database | |
| CN107241213A (en) | A kind of web service composition method learnt based on deeply | |
| CN104217258B (en) | A kind of electric load sigma-t Forecasting Methodology | |
| CN107153874A (en) | Water quality prediction method and system | |
| CN108121215B (en) | Process control loops method of evaluating performance and device based on full loop reconstruct emulation | |
| CN105138717A (en) | Transformer state evaluation method by optimizing neural network with dynamic mutation particle swarm | |
| CN108563863B (en) | Energy consumption calculation and scheduling method for urban rail transit system | |
| CN107133181A (en) | A kind of construction method of difference wavelet neural network software fault prediction technology | |
| CN112396187A (en) | Multi-agent reinforcement learning method based on dynamic collaborative map | |
| CN104539601A (en) | Reliability analysis method and system for dynamic network attack process | |
| CN103685014B (en) | Time series predicting model is utilized to strengthen the system and method for router-level topology reliability | |
| CN110543978A (en) | Traffic flow data prediction method and device based on wavelet neural network | |
| CN110837915A (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
| Akbari et al. | Similarity-based error prediction approach for real-time inflow forecasting | |
| CN106845144A (en) | A kind of trend prediction method excavated based on industrial big data | |
| CN106886788A (en) | Single goal emulation Track In Track difficulty detection method based on multi -index decision | |
| Khadr et al. | Integration of data-driven modeling and stochastic modeling for multi-purpose reservoir simulation | |
| CN102123053B (en) | Method for analyzing performance of multi-class closed fork-join queuing network based on horizontal decomposition | |
| CN115293367A (en) | Mixed federal learning method of scheduling model under small sample unbalanced data constraint | |
| Ptitsyna et al. | A neural network approach to overcoming a priori uncertainty in optimal action planning of intelligent information agents for soft architectures of service-oriented systems | |
| CN105260556B (en) | The overhead crane modeling method of hair clip mutation operation RNA genetic algorithm | |
| Fasanghari et al. | Predicting the Success of Projects Using Evolutionary Hybrid Fuzzy Neural Network Method in Early Stages. | |
| Ivanyuk et al. | Forecasting Financial Time Series Based On The Ensemble Method | |
| Wu et al. | Study of software reliability prediction based on GR neural network |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170822 |
|
| WD01 | Invention patent application deemed withdrawn after publication |