CN109872252A - A kind of electricity provider integrated evaluating method based on MATLAB algorithm - Google Patents
A kind of electricity provider integrated evaluating method based on MATLAB algorithm Download PDFInfo
- Publication number
- CN109872252A CN109872252A CN201910171882.0A CN201910171882A CN109872252A CN 109872252 A CN109872252 A CN 109872252A CN 201910171882 A CN201910171882 A CN 201910171882A CN 109872252 A CN109872252 A CN 109872252A
- Authority
- CN
- China
- Prior art keywords
- electricity consumption
- data
- supplier
- electricity
- history
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 129
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 title claims description 27
- 238000004519 manufacturing process Methods 0.000 claims abstract description 17
- 239000000463 material Substances 0.000 claims abstract description 14
- 230000009466 transformation Effects 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000011156 evaluation Methods 0.000 claims abstract description 11
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 4
- 238000009412 basement excavation Methods 0.000 claims description 4
- 238000005065 mining Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000006978 adaptation Effects 0.000 claims description 2
- 230000008901 benefit Effects 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims description 2
- 238000003825 pressing Methods 0.000 claims description 2
- 238000007405 data analysis Methods 0.000 claims 1
- 238000012549 training Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000003869 coulometry Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of electricity provider evaluation method based on MATLAB algorithm, using the corresponding goods and materials type of collection supplier's contract and history electricity consumption information, certain amount data are chosen as sample, by wavelet transformation by electricity consumption data separating be trend and fluctuation, establish supplier's analysis of electric power consumption model, according to the reasonable interval of the monthly electricity consumption of model analysis supplier, in conjunction with the monthly practical electricity consumption of supplier, whether the production capacity so as to more objective and accurate evaluation supplier meets the requirements.
Description
Technical field
The present invention relates to a kind of electricity provider evaluation method, specifically a kind of electricity provider based on MATLAB algorithm
Evaluation method.
Background technique
The Supplier Number of Power Material company is numerous, and for enterprise practical production capacity information, supplier will not active reporting
To Materials Company, it is insufficient that there are part supplier production capacities, or even production task is contracted out to other suppliers, improves object
Provide the risk of supply.In practical work process, Materials Company also can not audit and supervise to the production scene of all suppliers
Control, but there is an urgent need to for supplier's production capacity, subcontract the problems such as carry out early warning.
Summary of the invention
The quality of material appraisal procedure based on machine learning that the object of the present invention is to provide a kind of.This method is public using goods and materials
The supplier's relevant information that can be collected into is taken charge of, the reasonable electricity consumption section of supplier is predicted by MATLAB algorithm, thus into
The production capacity early warning of row science.
The purpose of the present invention is achieved through the following technical solutions:
A kind of electricity provider integrated evaluating method based on MATLAB algorithm, it is characterised in that: this method is using collection
Supplier's contract goods and materials type and history electricity consumption information are chosen certain amount data as sample, will be used by wavelet transformation
Electricity data is separated into trend and fluctuation, establishes supplier's analysis of electric power consumption model, according to the monthly electricity consumption of model analysis supplier
The reasonable interval of amount, in conjunction with the monthly practical electricity consumption of supplier, whether the production capacity of objective and accurate evaluation supplier is conformed to
It asks.
Specifically comprise the following steps:
S1 is collected for answering trade company number: being collected for answering quotient's electricity consumer information by vendor class；
S2 extracts history electricity consumption: pressing the monthly electricity consumption information of history that supplier is extracted at family number；
S3 establishes electricity demand forecasting model: the electricity consumption information of collection is standardized, wavelet transformation is recycled,
Stable electricity consumption trend is excavated, using filtering technique, abnormal electricity consumption information is removed, is finally built using BP network model
Vertical electricity demand forecasting model, according to the reasonable interval of history electricity demand forecasting time month electricity consumption；
S4 assesses supplier's production capacity: being compared according to the electricity consumption section of prediction and practical electricity consumption, is produced to supplier
It can be carried out assessment.
It is specific as follows to establish electricity demand forecasting model:
First supplier's history electricity consumption data are standardized and data separating treatment, then after excavating history abnormal data
Predict supplier time month electricity consumption data, electricity demand forecasting model is write in last set month electricity consumption early warning section；
1) data normalization is handled
Since electricity consumption data amount diversity ratio is larger between single supplier's history moon, to improve the convenience that data are analyzed,
Carry out data normalization processing；
2) electricity consumption data separating
The monthly electricity consumption data of the history of supplier can be regarded as one group of signal, using Fourier transformation, by this group of signal
Trend and random fluctuation separated；Wavelet transformation is recycled, the most stable electricity consumption trend inside data sequence is excavated
Out；Data and initial data trend and fluctuation situation after standardization are almost the same, and the data after standardization are certain
Cut down the influence of electricity consumption difference between Nian Yunian in degree；
3) history Outlier mining
Excavation for history abnormal data, using the common filtering technique of Data processing --- low order differential denoising come
Realize the target；
4) time month electricity consumption data are predicted
In the selection of prediction technique, BP neural network is selected to be predicted, utilizes Neural Network Self-learning and adaptation
Function establishes the non-parametric model of electricity consumption sequence；
5) time month electricity consumption section is set
By sampling method again, early warning section is set；
6) electricity demand forecasting model is write
Programming is carried out to model using MATLAB, is carried out more using monthly electricity consumption of the model to each supplier
Newly.
The present invention utilizes a large amount of history electricity consumption data, predicts time month electricity consumption；It is deleted not when creating prediction model
Stable electricity consumption information and abnormal electricity consumption information；Prediction result is an electricity consumption section.
The present invention creates quality of material assessment models, selects the artificial neural network algorithm of back transfer, constructs reversed pass
System network model.The historical sample information of collection is divided into two parts of training data and verify data.Use training data pair
Model is trained, and obtains Model Weight parameter and Dynamic gene.Test is carried out to model training result using verify data to test
Card, until verification result meets the information collected, to obtain the non-linear relation of quality of material and influence factor quality inspection.Most
Afterwards using the model and model parameter having verified that, the quality level of material is predicted.
The model parameter difference of different suppliers, the qualitative factor of different materials are larger, in order to improve the standard of prediction
Exactness, in training pattern, it is desirable that be trained respectively according to different types of supplier.
The present invention chooses the conduct of certain amount data using supplier's contract goods and materials type and history electricity consumption information is collected
Sample, by wavelet transformation by electricity consumption data separating be trend and fluctuation, supplier's analysis of electric power consumption model is established, according to mould
The reasonable interval of the monthly electricity consumption of type analysis supplier, in conjunction with the monthly practical electricity consumption of supplier, so as to more objective standard
Whether the production capacity of true evaluation supplier meets the requirements.
Detailed description of the invention
Fig. 1 is the electricity provider overall merit flow chart based on MATLAB algorithm.
Fig. 2 is BP neural network model structure schematic diagram.
Fig. 3 is electricity consumption early warning interval algorithm schematic diagram.
Specific embodiment
A kind of electricity provider integrated evaluating method based on MATLAB algorithm, using collection supplier's contract goods and materials type
With history electricity consumption information, choose certain amount data be used as sample, by wavelet transformation by electricity consumption data separating be trend
And fluctuation, supplier's analysis of electric power consumption model is established, according to the reasonable interval of the monthly electricity consumption of model analysis supplier, in conjunction with confession
The monthly practical electricity consumption of quotient is answered, whether the production capacity of objective and accurate evaluation supplier meets the requirements.
Step 1: it is collected for answering quotient's electricity consumer information
At present according to application demand, supplier is divided into steel tower class, Switch, cement pole class, bin class, cable, change
Nine seed types such as depressor class, fitting class, cable protection tubing, other classes.
Step 2: supplier's history electricity consumption information is extracted
According to supply trade company's information, the monthly electricity consumption information of supplier's history is pulled.Select representative supply
Quotient carries out coulometric analysis.
Step 3: electricity demand forecasting model is generated according to historical data
First supplier's history electricity consumption data are standardized and data separating treatment, then after excavating history abnormal data
Predict supplier time month electricity consumption data, electricity demand forecasting model is write in last set month electricity consumption early warning section.
1, data normalization is handled
Since electricity consumption data amount diversity ratio is larger between single supplier's history moon, to improve the convenience that data are analyzed,
Data normalization processing is carried out, calculation method is as follows:
2, electricity consumption data separating
The monthly electricity consumption data of the history of supplier can be regarded as one group of signal, using Fourier transformation, by this group of signal
Trend and random fluctuation separated.Wavelet transformation is recycled, the most stable electricity consumption trend inside data sequence is excavated
Out.Data after standardization and initial data trend and fluctuation situation are almost the same, and the data after standardizing can be
Cut down the influence of electricity consumption difference between Nian Yunian to a certain extent.
3, history Outlier mining
Excavation for history abnormal data, using the common filtering technique of Data processing --- low order differential denoising come
Realize the target.
Low order differential Denoising Algorithm analyzes the exceptional value in electricity consumption data, and basic principle is as follows:
The time series to be studied is set as P (t), and preceding 3 months data are normal values, if follow-up data meets item
Part:
Then illustrate that data P (t) is abnormal point numerical.
The actual conditions excavated according to electricity consumption abnormal point numerical, it is only necessary to consider the case where electricity consumption declines suddenly, because
This optimizes above-mentioned condition are as follows:
It selects preceding 3 months data to be positive constant value (i.e. n=3), can effectively exclude the influence of seasonal factor；γ is early warning
The factor can be adjusted according to supplier's actual history situation, its value is bigger, and the abnormal point of excavation is fewer, in example
Show that γ selection is that comparison is reasonable between 1~3 in verifying.Design parameter can be supplied after model foundation by other
The truthful data of quotient carries out verification determination.
4, time month electricity consumption data are predicted
In the selection of prediction technique, BP neural network is selected to be predicted, advantage is can use neural network certainly
Study and the function of adapting to establish the non-parametric model of electricity consumption sequence.
BP neural network model structure includes input layer (input), hidden layer (hide layer) and output layer (output
Layer) (as shown in Figure 2).
During carrying out secondary month electricity demand forecasting, for the reasonability for ensuring prediction result, the exception that will first excavate is needed
Point data carries out serial mean replacement.The wave of mean value replacement will be carried out by 3 groups of trend datas of wavelet decomposition twice and 1 group
Dynamic data carry out neural metwork training respectively.
In the selection of input layer, hidden layer and output layer, it is contemplated that season and year influence, and every group of data select 6 numbers
According to as neuron；Comprehensively consider trained pressure and training effect, hidden layer number is set as 3；Output layer is set as 1, by four groups of numbers
According to output layer it is comprehensive, as secondary month electricity demand forecasting value.
After prediction result is formed, the performance of prediction is evaluated by setting up evaluation index, evaluation index includes AMAPE
And VAR, the relative accuracy of the former prediction, the latter represent the stability of prediction, calculation method formula is as follows:
5, time month electricity consumption section is set
The setting in early warning section, it can be understood as deviation range of the predicted value under certain confidence interval, in statistics
It usually assume that deviation Gaussian distributed obtains deviation range, but this section setting means and actual conditions are often
Deviation.Alves da Sillva and Moulin propose a kind of method of forecast interval unrelated with data distribution --- again
Sampling method, reliability with higher.
Electricity consumption early warning interval algorithm is as shown in Figure 3:
Because six continuous data can train to obtain a trained values in the training process of neural network, training
The electricity consumption of the moon is all known where value, by the difference of neural metwork training value and actual value, available every group of training
The deviation of value constitutes a new biased sequence.
Biased sequence is sorted out by numerical values recited, after for statistical analysis, can learn the predicted value category of time month electricity consumption
In the Probability p of each deviation class.If deviation class sjInclude njThe sample of a random error, then:
It, can be with determination deviation section [x_1, x_2], the section after giving a level of significance α (general α takes 0.2)
Probability comprising true electricity consumption magnitude is 1-2 α, and algorithm is as follows:
1) x_1 is calculated, so that p (x≤x_1)=α；
2) x_2 is calculated, so that p (x≤x_2)=1- α.
If predicted value is θ, it can be deduced that early warning section is [θ+x_1, θ+x_2].
6, electricity demand forecasting model is write
In order to be updated using monthly electricity consumption of the model to each supplier, we use MATLAB to model
Carry out programming.
Step 4: model evaluation supplier production capacity is utilized
Using the above-mentioned computation model being proved to be successful, the monthly electricity consumption of each supplier and electricity consumption section are predicted, and
In conjunction with the practical electricity consumption of supplier, assess the production capacity of supplier, and to the production activity of electricity consumption exception supplier carry out with
Track finds risk of honouring an agreement in time.
Claims (4)
1. a kind of electricity provider integrated evaluating method based on MATLAB algorithm, it is characterised in that: this method is used and is collected for
It answers quotient's contract goods and materials type and history electricity consumption information, chooses certain amount data as sample, by wavelet transformation by electricity consumption
Measuring data separating is trend and fluctuation, supplier's analysis of electric power consumption model is established, according to the monthly electricity consumption of model analysis supplier
Reasonable interval, in conjunction with the monthly practical electricity consumption of supplier, whether the production capacity of objective and accurate evaluation supplier meets the requirements.
2. the electricity provider integrated evaluating method according to claim 1 based on MATLAB algorithm, it is characterised in that: packet
Include following steps:
S1 is collected for answering trade company number: being collected for answering quotient's electricity consumer information by vendor class；
S2 extracts history electricity consumption: pressing the monthly electricity consumption information of history that supplier is extracted at family number；
S3 establishes electricity demand forecasting model: the electricity consumption information of collection being standardized, wavelet transformation is recycled, is dug
Stable electricity consumption trend is excavated, using filtering technique, removes abnormal electricity consumption information, is finally established using BP network model
Electricity demand forecasting model, according to the reasonable interval of history electricity demand forecasting time month electricity consumption；
S4 assesses supplier's production capacity: be compared according to the electricity consumption section of prediction and practical electricity consumption, to supplier's production capacity into
Row assessment.
3. the electricity provider integrated evaluating method according to claim 2 based on MATLAB algorithm, it is characterised in that: build
Vertical electricity demand forecasting model is specific as follows:
First supplier's history electricity consumption data are standardized and data separating treatment, then are predicted after excavating history abnormal data
Supplier time month electricity consumption data, last set month electricity consumption early warning section, write electricity demand forecasting model；
1) data normalization is handled
Since electricity consumption data amount diversity ratio is larger between single supplier's history moon, for the convenience for improving data analysis, carry out
Data normalization processing；
2) electricity consumption data separating
The monthly electricity consumption data of the history of supplier can be regarded as one group of signal, using Fourier transformation, by becoming for this group of signal
Gesture and random fluctuation are separated；Wavelet transformation is recycled, the most stable electricity consumption trend inside data sequence is excavated；
Data and initial data trend and fluctuation situation after standardization are almost the same, and the data after standardization are to a certain extent
Cut down the influence of electricity consumption difference between Nian Yunian；
3) history Outlier mining
Excavation for history abnormal data, using the common filtering technique of Data processing --- low order differential denoises to realize
The target；
4) time month electricity consumption data are predicted
In the selection of prediction technique, selects BP neural network to be predicted, utilize the function of Neural Network Self-learning and adaptation
Establish the non-parametric model of electricity consumption sequence；
5) time month electricity consumption section is set
By sampling method again, early warning section is set；
6) electricity demand forecasting model is write
Programming is carried out to model using MATLAB, is updated using monthly electricity consumption of the model to each supplier.
4. the electricity provider integrated evaluating method according to claim 1 based on MATLAB algorithm, it is characterised in that: benefit
With a large amount of history electricity consumption data, time month electricity consumption is predicted；Unstable electricity consumption information is deleted when creating prediction model
With abnormal electricity consumption information；Prediction result is an electricity consumption section.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910171882.0A CN109872252A (en) | 2019-03-07 | 2019-03-07 | A kind of electricity provider integrated evaluating method based on MATLAB algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910171882.0A CN109872252A (en) | 2019-03-07 | 2019-03-07 | A kind of electricity provider integrated evaluating method based on MATLAB algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109872252A true CN109872252A (en) | 2019-06-11 |
Family
ID=66920064
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910171882.0A Pending CN109872252A (en) | 2019-03-07 | 2019-03-07 | A kind of electricity provider integrated evaluating method based on MATLAB algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872252A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222044A (en) * | 2019-06-13 | 2019-09-10 | 国网浙江省电力有限公司 | A kind of supervisory control of substation method |
| CN112070302A (en) * | 2020-09-08 | 2020-12-11 | 北京中宸泓昌科技有限公司 | Early warning method and system for power utilization safety of power grid |
| CN112381442A (en) * | 2020-11-25 | 2021-02-19 | 兰和科技（深圳）有限公司 | Campus security information management system based on artificial intelligence technology |
| CN113935568A (en) * | 2021-08-30 | 2022-01-14 | 国网江苏省电力有限公司物资分公司 | Auxiliary decision-making method for making purchasing strategy in productivity recovery stage |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504619A (en) * | 2015-01-05 | 2015-04-08 | 国家电网公司 | Temperature/ economic growth factor considered monthly total electricity consumption predication method |
| CN107506938A (en) * | 2017-09-05 | 2017-12-22 | 江苏电力信息技术有限公司 | A kind of quality of material appraisal procedure based on machine learning |
| CN108197011A (en) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | A kind of single index prediction and method for early warning based on artificial intelligence big data platform |
-
2019
- 2019-03-07 CN CN201910171882.0A patent/CN109872252A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504619A (en) * | 2015-01-05 | 2015-04-08 | 国家电网公司 | Temperature/ economic growth factor considered monthly total electricity consumption predication method |
| CN107506938A (en) * | 2017-09-05 | 2017-12-22 | 江苏电力信息技术有限公司 | A kind of quality of material appraisal procedure based on machine learning |
| CN108197011A (en) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | A kind of single index prediction and method for early warning based on artificial intelligence big data platform |
Non-Patent Citations (2)
| Title |
|---|
| 井伟 等: ""基于机器学习的用电量-供应商产能预警研究"", 《企业改革与管理》 * |
| 殷玮珺 等: ""供应商产能预警系统的功能设计"", 《科技创新与应用》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222044A (en) * | 2019-06-13 | 2019-09-10 | 国网浙江省电力有限公司 | A kind of supervisory control of substation method |
| CN110222044B (en) * | 2019-06-13 | 2021-07-23 | 国网浙江省电力有限公司 | Transformer substation monitoring method |
| CN112070302A (en) * | 2020-09-08 | 2020-12-11 | 北京中宸泓昌科技有限公司 | Early warning method and system for power utilization safety of power grid |
| CN112381442A (en) * | 2020-11-25 | 2021-02-19 | 兰和科技（深圳）有限公司 | Campus security information management system based on artificial intelligence technology |
| CN112381442B (en) * | 2020-11-25 | 2021-07-20 | 兰和科技（深圳）有限公司 | Campus security information management system based on artificial intelligence technology |
| CN113935568A (en) * | 2021-08-30 | 2022-01-14 | 国网江苏省电力有限公司物资分公司 | Auxiliary decision-making method for making purchasing strategy in productivity recovery stage |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| CN109872252A (en) | A kind of electricity provider integrated evaluating method based on MATLAB algorithm | |
| De Silva et al. | A data mining framework for electricity consumption analysis from meter data | |
| CN106022518B (en) | A kind of piping failure probability forecasting method based on BP neural network | |
| CN110298574B (en) | Convolutional neural network-based electricity user payment risk rating method | |
| CN104835103A (en) | Mobile network health evaluation method based on neural network and fuzzy comprehensive evaluation | |
| Mohammad et al. | Energy load forecasting model based on deep neural networks for smart grids | |
| Kim et al. | Factor selection for delay analysis using knowledge discovery in databases | |
| CN104572449A (en) | Automatic test method based on case library | |
| CN102498445A (en) | Supervised fault learning using rule-generated samples for machine condition monitoring | |
| CN116681187B (en) | Enterprise carbon quota prediction method based on enterprise operation data | |
| CN105005711A (en) | Method and device for acquiring statistical line loss | |
| CN104574141A (en) | Service influence degree analysis method | |
| CN110991985A (en) | Project progress data processing method | |
| Zong | Maintenance cost and residual value prediction of heavy construction equipment | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| Pan | The use of genetic programming for the construction of a financial management model in an enterprise | |
| CN114154866A (en) | Marketing enterprise financial risk early warning method and system | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN110941646A (en) | Project progress control system based on 10KV project three-rate data mining | |
| CN112348220A (en) | Credit risk assessment prediction method and system based on enterprise behavior pattern | |
| CN115130924A (en) | Microgrid power equipment asset evaluation method and system under source grid storage background | |
| CN113435494A (en) | Low-voltage resident user abnormal electricity utilization identification method and simulation system | |
| CN113627585A (en) | Single well index prediction method and system based on long-time and short-time memory neural network | |
| Tenfjord et al. | The value of interpretable machine learning in wind power prediction: an emperical study using shapley addidative explanations to interpret a complex wind power prediction model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190611 |