CN109409614A - A kind of Methods of electric load forecasting based on BR neural network - Google Patents
A kind of Methods of electric load forecasting based on BR neural network Download PDFInfo
- Publication number
- CN109409614A CN109409614A CN201811369445.1A CN201811369445A CN109409614A CN 109409614 A CN109409614 A CN 109409614A CN 201811369445 A CN201811369445 A CN 201811369445A CN 109409614 A CN109409614 A CN 109409614A
- Authority
- CN
- China
- Prior art keywords
- network
- neural network
- parameter
- formula
- error
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000005611 electricity Effects 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000005457 optimization Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of Methods of electric load forecasting based on BR neural network, comprising the following steps: obtains electricity historical data, the crucial factor that analyzing influence electricity consumption increases；Determine BP neural network structure；Overall error F (W) is set to reach minimum with BP algorithm training network；Calculate actual parameter number γ；The new estimated value of hyper parameter α and β are calculated using bayes method；Repeat it is above-mentioned until reaching the required accuracy, thus complete Bayes's canonical optimization neural network foundation；The new crucial factor for influencing electricity consumption and increasing of input, obtains whole society's electric load situation of the period.The invention has the advantages that bayes method is used in the modeling process of neural network, the training performance function of neural network is corrected to improve its generalization ability by regularization method, fast convergence rate can obtain smaller training error.
Description
Technical field
The present invention relates to a kind of Methods of electric load forecasting based on BR neural network.
Background technique
Electric system is collectively constituted by power network, power consumer, task be provided incessantly to users it is economical,
Reliably, the electric energy met the quality standard, meets the needs of each type load, provides power for socio-economic development.Due to electric power
Production is with using having particularity, i.e. electric energy is difficult to largely store, and demand of all types of user to electric power changes constantly,
This requires system power generations should be at any time with the variation dynamic equilibrium of system loading, i.e. system will play equipment energy to the maximum extent
Power makes whole system keep stablizing and efficiently running, to meet the needs of users.Otherwise, it just will affect the quality for electricity consumption,
Even jeopardize the safety and stablization of system, this.Therefore, Load Prediction In Power Systems technology has developed, and is everything
Smooth premise and basis.Traditional mathematical model is described with ready-made mathematic(al) representation, is had and is calculated
Measure small, fireballing advantage, but there is also many flaws and limitation simultaneously, for example, do not have self study, adaptive ability,
The robustness of forecasting system does not ensure.In particular with China's expanding economy, the structure of electric system is increasingly sophisticated, electricity
The feature of the non-linear of power load variations, time variation and uncertainty is more obvious, is difficult to establish a suitable mathematical model
Clearly to express the relationship between load and the variable for influencing load.And non-mathematical model prediction method neural network based, it is
The deficiency for solving mathematical model method provides new thinking.BP neural network can solve complex nonlinear function approximation problem,
There is apparent superiority when processing missing values and nonlinear problem, but there is local convergences, convergence rate mistake for standard BP algorithm
The problems such as slow and overfitting.
Summary of the invention
The purpose of the present invention is to provide a kind of Methods of electric load forecasting based on BR neural network, energy
It is enough effectively to solve the problems, such as that existing prediction technique local convergence, convergence rate be excessively slow and overfitting.
In order to solve the above-mentioned technical problem, the present invention is achieved by the following technical solutions: one kind being based on Bayes just
Then change the Methods of electric load forecasting of neural network, comprising the following steps:
(1) electricity historical data, the crucial factor that analyzing influence electricity consumption increases are obtained；
(2) it determines BP neural network structure, initial value is assigned to network parameter according to prior distribution, initializes hyper parameter α and β；
Remember neural network training model training sample D=(xi,ti), i=1,2, L, n, n be training sample sum, W be network parameter to
Amount, under conditions of giving network structure H and network parameter W, the error function E of networkDTake the quadratic sum of error:
F () is the reality output of network in formula, and t is the output quantity of neural network；
In order to improve the generalization ability of BP neural network, by regularization method plus power attenuation term after error function
EW:
M is network parameter sum in formula, and then overall error function may be defined as:
F (W)=β ED+αEW
(3) overall error F (W) is made to reach minimum with BP algorithm training network:
By calculating the output valve of each layer neuron, connection weight is corrected using gradient descent method, each time connection weight
Correction amount it is directly proportional to the gradient of error function, from input layer back transfer to each layer, by initial weight and corresponding adjustment
Amount is added, and calculates new weight, and so circulation is until each layer error sum of squares reaches setting value；
(4) actual parameter number γ is calculated:
If weight F (W) corresponding when being minimized is WMP, by F (W) in WMPNeighbouring Taylor expansion, ignores high-order term,
It obtains:
In formula,Indicate F (W) in WMPThe Hessian matrix of point；It is calculated to improve
Speed is further simplified Hessian matrix using Gauss-Newton approximatioss, obtains
In formula, J indicates EDIn point WMPThe Jacobian matrix at place；Actual parameter number is calculated according to the following formula
In formula, m indicates network parameter sum.γ indicates how many parameter acts as in terms of reducing overall error function in network
With value range is [0, m]；
(5) the new estimated value of hyper parameter α and β are calculated using bayes method:
According to maximum likelihood principle, is found out using following formula and meet the maximum α and β of likelihood function p (D | α, β, H) to get arriving
Optimal hyper parameter:
(6) step 3-5 is repeated until reaching the required accuracy, is built to complete Bayes's canonical optimization neural network
It is vertical；
(7) the new crucial factor for influencing electricity consumption and increasing of input, obtains whole society's electric load situation of the period.
Compared with prior art, the invention has the advantages that bayes method is used in the modeling process of neural network, lead to
The training performance function of regularization method amendment neural network is crossed to improve its generalization ability, fast convergence rate can obtain more
Small training error.
Detailed description of the invention
Fig. 1 is the relationship and once linear equation model curve of GDP and electricity consumption；
Fig. 2 is the relationship and once linear equation model curve of gross fixed assets investment and electricity consumption；
Fig. 3 is the relationship and once linear equation model curve of per capita income and electricity consumption；
Fig. 4 is BR neural network training error change curve；
Fig. 5 is electricity consumption actual value compared with predicted value and error line.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings.Below with reference to
The embodiment of attached drawing description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
A kind of Methods of electric load forecasting based on BR neural network, comprising the following steps:
(1) electricity historical data, the crucial factor that analyzing influence electricity consumption increases are obtained；
(2) it determines BP neural network structure, initial value is assigned to network parameter according to prior distribution, initializes hyper parameter α and β；
Remember neural network training model training sample D=(xi,ti), i=1,2, L, n, n be training sample sum, W be network parameter to
Amount, under conditions of giving network structure H and network parameter W, the error function E of networkDTake the quadratic sum of error:
F () is the reality output of network in formula, and t is the output quantity of neural network；
In order to improve the generalization ability of BP neural network, by regularization method plus power attenuation term after error function
EW:
M is network parameter sum in formula, and then overall error function may be defined as:
F (W)=β ED+αEW
(3) overall error F (W) is made to reach minimum with BP algorithm training network:
By calculating the output valve of each layer neuron, connection weight is corrected using gradient descent method, each time connection weight
Correction amount it is directly proportional to the gradient of error function, from input layer back transfer to each layer, by initial weight and corresponding adjustment
Amount is added, and calculates new weight, and so circulation is until each layer error sum of squares reaches setting value；
(4) actual parameter number γ is calculated:
If weight F (W) corresponding when being minimized is WMP, by F (W) in WMPNeighbouring Taylor expansion, ignores high-order term,
It obtains:
In formula,Indicate F (W) in WMPThe Hessian matrix of point；It is calculated to improve
Speed is further simplified Hessian matrix using Gauss-Newton approximatioss, obtains
In formula, J indicates EDIn point WMPThe Jacobian matrix at place；Actual parameter number is calculated according to the following formula
In formula, m indicates network parameter sum.γ indicates how many parameter acts as in terms of reducing overall error function in network
With value range is [0, m]；
(5) the new estimated value of hyper parameter α and β are calculated using bayes method:
According to maximum likelihood principle, is found out using following formula and meet the maximum α and β of likelihood function p (D | α, β, H) to get arriving
Optimal hyper parameter:
(6) step 3-5 is repeated until reaching the required accuracy, is built to complete Bayes's canonical optimization neural network
It is vertical；
(7) the new crucial factor for influencing electricity consumption and increasing of input, obtains whole society's electric load situation of the period.
Economic society data and Analyzing Total Electricity Consumption number with the 1986-2005 obtained from somewhere statistics bureau and power office
For, it is related to Analyzing Total Electricity Consumption that GDP, gross fixed assets investment, per capita income are calculated separately out using MATLAB
Coefficient are as follows: 0.9733,0.9741,0.9430, once linear equation model curve is as shown in Figs. 1-3.It follows that these three because
There are substantial connections between element and Analyzing Total Electricity Consumption.
BP neural network input layer number is set as 3, middle layer node number is 40, and output layer number of nodes is 1, regularization
Hyper parameter initial value is α=0 and β=1, using above-mentioned GDP, gross fixed assets investment, these three amounts of per capita income as the mind
Established mind is trained and emulated to input sample through network, the output sample using Analyzing Total Electricity Consumption as the neural network
Through network.Fig. 4 gives neural metwork training error change curve, when training is to 400 step, network training convergence, and internetworking
Expected requirement can be reached.Fig. 5 describes Analyzing Total Electricity Consumption actual value and predicted value, they are corresponded to substantially coincide, error line
It is swung near 0.In conclusion the convergence rate of the neural network is very fast due to using Bayesian regularization optimization algorithm,
And the training error of very little can be obtained, it is fully able to meet the needs of decision.
From the point of view of the comparison of prediction data and real data, the Optimized BP Neural Network estimated performance is preferable, relative error
It is smaller, it is fully able to meet the needs of decision, and predetermined speed is fast, it is easy to operate.
Above is only a specific embodiment of the present invention, but technical characteristic of the invention is not limited thereto, Ren Heben
Within the field of the present invention, made changes or modifications all cover within the scope of the patent of the present invention the technical staff in field.
Claims (1)
1. a kind of Methods of electric load forecasting based on BR neural network, it is characterised in that: the following steps are included:
(1) electricity historical data, the crucial factor that analyzing influence electricity consumption increases are obtained；
(2) it determines BP neural network structure, initial value is assigned to network parameter according to prior distribution, initializes hyper parameter α and β；Note mind
Through network training model training sample D=(xi,ti), i=1,2, L, n, n are training sample sum, and W is network parameter vector, are given
Under conditions of determining network structure H and network parameter W, the error function E of networkDTake the quadratic sum of error:
F () is the reality output of network in formula, and t is the output quantity of neural network；
In order to improve the generalization ability of BP neural network, by regularization method plus power attenuation term E after error functionW:
M is network parameter sum in formula, and then overall error function may be defined as:
F (W)=β ED+αEW
(3) overall error F (W) is made to reach minimum with BP algorithm training network:
By calculating the output valve of each layer neuron, connection weight is corrected using gradient descent method, connection weight is repaired each time
Positive quantity is directly proportional to the gradient of error function, from input layer back transfer to each layer, by initial weight and corresponding adjustment amount phase
Add, calculate new weight, so circulation is until each layer error sum of squares reaches setting value；
(4) actual parameter number γ is calculated:
If weight F (W) corresponding when being minimized is WMP, by F (W) in WMPNeighbouring Taylor expansion, ignores high-order term, obtains:
ZF≈(2π)m/2(det(▽2F(WMP))-1)1/2×exp(-F(WMP))
In formula, ▽2F(WMP)=β ▽2ED+α▽2EWIndicate F (W) in WMPThe Hessian matrix of point；In order to improve calculating speed,
Hessian matrix is further simplified using Gauss-Newton approximatioss, obtains ▽2F(WMP)≈2βJTJ+2αIm
In formula, J indicates EDIn point WMPThe Jacobian matrix at place；Actual parameter number γ=m-2 α is calculated according to the following formulaMPtr[▽2F
(WMP)]-1
In formula, m indicates network parameter sum.γ indicates how many parameter works in terms of reducing overall error function in network,
Its value range is [0, m]；
(5) the new estimated value of hyper parameter α and β are calculated using bayes method:
According to maximum likelihood principle, is found out using following formula and meet the maximum α and β of likelihood function p (D | α, β, H) to get to optimal
Hyper parameter:
(6) step 3-5 is repeated until reaching the required accuracy, to complete the foundation of Bayes's canonical optimization neural network；
(7) the new crucial factor for influencing electricity consumption and increasing of input, obtains whole society's electric load situation of the period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811369445.1A CN109409614A (en) | 2018-11-16 | 2018-11-16 | A kind of Methods of electric load forecasting based on BR neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811369445.1A CN109409614A (en) | 2018-11-16 | 2018-11-16 | A kind of Methods of electric load forecasting based on BR neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109409614A true CN109409614A (en) | 2019-03-01 |
Family
ID=65473656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811369445.1A Pending CN109409614A (en) | 2018-11-16 | 2018-11-16 | A kind of Methods of electric load forecasting based on BR neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409614A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110969285A (en) * | 2019-10-29 | 2020-04-07 | 京东方科技集团股份有限公司 | Prediction model training method, prediction device, prediction equipment and medium |
| CN110991722A (en) * | 2019-11-26 | 2020-04-10 | 广东电网有限责任公司 | Power load prediction method |
| CN111415010A (en) * | 2020-03-20 | 2020-07-14 | 广东电网有限责任公司阳江供电局 | Bayesian neural network-based wind turbine generator parameter identification method |
| CN111796514A (en) * | 2019-04-09 | 2020-10-20 | 罗伯特·博世有限公司 | Controlling and monitoring a physical system based on a trained bayesian neural network |
| CN112488399A (en) * | 2020-12-04 | 2021-03-12 | 国网冀北电力有限公司计量中心 | Power load prediction method and device |
| CN112613637A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | Method and device for processing charging load |
| CN113065704A (en) * | 2021-03-31 | 2021-07-02 | 贵州电网有限责任公司 | Hyper-parameter optimization and post-processing method of non-invasive load decomposition model |
| CN113240025A (en) * | 2021-05-19 | 2021-08-10 | 电子科技大学 | Image classification method based on Bayesian neural network weight constraint |
| CN113866204A (en) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | Bayesian regularization-based soil heavy metal quantitative analysis method |
| CN114584358A (en) * | 2022-02-25 | 2022-06-03 | 安捷光通科技成都有限公司 | Intelligent network security system, device and storage medium based on Bayesian regularization |
| US11592491B2 (en) | 2019-12-23 | 2023-02-28 | Samsung Electronics Co., Ltd. | Battery state estimation method and apparatus |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103295081A (en) * | 2013-07-02 | 2013-09-11 | 上海电机学院 | Electrical power system load prediction method based on back propagation (BP) neural network |
| CN103425743A (en) * | 2013-07-17 | 2013-12-04 | 上海金自天正信息技术有限公司 | Steam pipe network prediction system based on Bayesian neural network algorithm |
| CN107749833A (en) * | 2017-10-24 | 2018-03-02 | 重庆工程职业技术学院 | A kind of self-adapted pre-distortion algorithm for merging polyteny multiplier |
-
2018
- 2018-11-16 CN CN201811369445.1A patent/CN109409614A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103295081A (en) * | 2013-07-02 | 2013-09-11 | 上海电机学院 | Electrical power system load prediction method based on back propagation (BP) neural network |
| CN103425743A (en) * | 2013-07-17 | 2013-12-04 | 上海金自天正信息技术有限公司 | Steam pipe network prediction system based on Bayesian neural network algorithm |
| CN107749833A (en) * | 2017-10-24 | 2018-03-02 | 重庆工程职业技术学院 | A kind of self-adapted pre-distortion algorithm for merging polyteny multiplier |
Non-Patent Citations (2)
| Title |
|---|
| 冯居易;郭晔;: "基于LM算法的石油期货价格预测研究", 技术经济与管理研究 * |
| 李慧良: "基于贝叶斯神经网络的短期负荷预测应用研究", 广东电力 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111796514A (en) * | 2019-04-09 | 2020-10-20 | 罗伯特·博世有限公司 | Controlling and monitoring a physical system based on a trained bayesian neural network |
| CN110969285A (en) * | 2019-10-29 | 2020-04-07 | 京东方科技集团股份有限公司 | Prediction model training method, prediction device, prediction equipment and medium |
| CN110969285B (en) * | 2019-10-29 | 2023-04-07 | 京东方科技集团股份有限公司 | Prediction model training method, prediction device, prediction equipment and medium |
| CN110991722A (en) * | 2019-11-26 | 2020-04-10 | 广东电网有限责任公司 | Power load prediction method |
| US11782094B2 (en) | 2019-12-23 | 2023-10-10 | Samsung Electronics Co., Ltd. | Battery state estimation method and apparatus |
| US11592491B2 (en) | 2019-12-23 | 2023-02-28 | Samsung Electronics Co., Ltd. | Battery state estimation method and apparatus |
| CN111415010B (en) * | 2020-03-20 | 2022-11-22 | 广东电网有限责任公司阳江供电局 | Bayesian neural network-based wind turbine generator parameter identification method |
| CN111415010A (en) * | 2020-03-20 | 2020-07-14 | 广东电网有限责任公司阳江供电局 | Bayesian neural network-based wind turbine generator parameter identification method |
| CN112613637A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | Method and device for processing charging load |
| CN112488399A (en) * | 2020-12-04 | 2021-03-12 | 国网冀北电力有限公司计量中心 | Power load prediction method and device |
| CN112488399B (en) * | 2020-12-04 | 2024-04-23 | 国网冀北电力有限公司计量中心 | Power load prediction method and device |
| CN113065704A (en) * | 2021-03-31 | 2021-07-02 | 贵州电网有限责任公司 | Hyper-parameter optimization and post-processing method of non-invasive load decomposition model |
| CN113240025B (en) * | 2021-05-19 | 2022-08-12 | 电子科技大学 | Image classification method based on Bayesian neural network weight constraint |
| CN113240025A (en) * | 2021-05-19 | 2021-08-10 | 电子科技大学 | Image classification method based on Bayesian neural network weight constraint |
| CN113866204A (en) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | Bayesian regularization-based soil heavy metal quantitative analysis method |
| CN114584358A (en) * | 2022-02-25 | 2022-06-03 | 安捷光通科技成都有限公司 | Intelligent network security system, device and storage medium based on Bayesian regularization |
| CN114584358B (en) * | 2022-02-25 | 2023-10-13 | 安捷光通科技成都有限公司 | Intelligent network security system, device and storage medium based on Bayesian regularization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409614A (en) | A kind of Methods of electric load forecasting based on BR neural network | |
| US11409347B2 (en) | Method, system and storage medium for predicting power load probability density based on deep learning | |
| CN110984062B (en) | Simulation scheduling method for large-scale reservoir group of watershed main and branch flows | |
| Wei et al. | Model-free adaptive optimal control for unknown nonlinear multiplayer nonzero-sum game | |
| CN110504676B (en) | Power distribution network state estimation method based on APSO-BP | |
| CN109861202A (en) | A kind of flexible interconnection distribution networks dynamic optimization dispatching method and system | |
| CN107565553A (en) | A kind of power distribution network robust dynamic state estimator method based on UKF | |
| CN109685265A (en) | A kind of prediction technique of power-system short-term electric load | |
| CN112990500B (en) | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis | |
| CN108197725A (en) | Water supply network node water requirement check method based on water requirement prior information | |
| CN104701858A (en) | Reactive voltage control method considering dynamic reactive power reserves of partitions | |
| CN111626003B (en) | Heating system heat load layered prediction method, storage medium and prediction equipment | |
| CN116629461B (en) | Distributed optimization method, system, equipment and storage medium for active power distribution network | |
| CN111415068A (en) | Power distribution decision modeling method based on relevance of transformation measures and loss load index | |
| CN104834975A (en) | Power network load factor prediction method based on intelligent algorithm optimization combination | |
| CN112113146A (en) | Synchronous self-adaptive check method for roughness coefficient and node water demand of water supply pipe network pipeline | |
| CN114971090A (en) | Electric heating load prediction method, system, equipment and medium | |
| CN109599866B (en) | Prediction-assisted power system state estimation method | |
| CN114777192A (en) | Secondary network heat supply autonomous optimization regulation and control method based on data association and deep learning | |
| CN104834216A (en) | Binomial-based wireless sensor network trust management method | |
| CN108459570B (en) | Irrigation water distribution intelligent control system and method based on generation of confrontation network architecture | |
| CN103489037B (en) | A kind of Forecasting Methodology that can power generating wind resource | |
| CN108108837A (en) | A kind of area new energy power supply structure optimization Forecasting Methodology and system | |
| CN104537224B (en) | Multi-state System Reliability analysis method and system based on adaptive learning algorithm | |
| CN114995106A (en) | PID self-tuning method, device and equipment based on improved wavelet 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190301 |
|
| RJ01 | Rejection of invention patent application after publication |