CN109510203A - A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting - Google Patents
A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting Download PDFInfo
- Publication number
- CN109510203A CN109510203A CN201811594173.5A CN201811594173A CN109510203A CN 109510203 A CN109510203 A CN 109510203A CN 201811594173 A CN201811594173 A CN 201811594173A CN 109510203 A CN109510203 A CN 109510203A
- Authority
- CN
- China
- Prior art keywords
- power supply
- distribution network
- load
- power distribution
- power
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The power distribution network evaluation of power supply capability method based on Mid-long term load forecasting that the present invention relates to a kind of, to according to distribution web area internal loading variation tendency, obtain the power supply bottleneck that limitation actual load increases, comprising the following steps: 1) construct the power distribution network power supply capacity prediction model based on Mid-long term load forecasting；2) power distribution network power supply capacity prediction model is solved to obtain the power supply capacity of current power distribution network；3) the power supply nargin of current power distribution network is obtained according to power supply capacity.Compared with prior art, the present invention has many advantages, such as that assessment is accurate, comprehensively true.
Description
Technical field
The present invention relates to power distribution networks for electrical domain, powers more particularly, to a kind of power distribution network based on Mid-long term load forecasting
Capability assessment method.
Background technique
Mid-long term load forecasting is the premise and basis of distribution network planning, and distribution network planning requires to predict future load
Total amount, predict the position that future load increases again, i.e. Mid-long Term Load Prediction of Total and Spatial Load Forecasting, they for
Reasonably distribution network planning all has important directive significance, and prediction result is regional electric power development speed, power construction rule
The reliable foundations of offers such as mould, power industry layout and energy resources balance；The reliability of its prediction result and accuracy are straight
Connect the daily life of the development and the people that influence national economy.
But there is no a kind of power distribution network evaluation of power supply capability methods suitable for Mid-long term load forecasting now.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind based on medium-term and long-term negative
The power distribution network evaluation of power supply capability method of lotus prediction.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting, to be born according in distribution web area
Lotus variation tendency obtains the power supply bottleneck that limitation actual load increases, comprising the following steps:
1) the power distribution network power supply capacity prediction model based on Mid-long term load forecasting is constructed；
2) power distribution network power supply capacity prediction model is solved to obtain the power supply capacity of current power distribution network；
3) the power supply nargin of current power distribution network is obtained according to power supply capacity.
The step 1) specifically includes the following steps:
11) the optimum prediction model of each load point is determined；
For each of power distribution network load point, using a variety of individual event prediction models and multiple combinations prediction model to it
Historical data is verified, and selects the smallest prediction model of average absolute value error as the optimum prediction model of the load point；
12) the optimum prediction model of all load points is overlapped, obtains the Additive Model of whole load points；
13) Load flow calculation is carried out using Additive Model, obtains the power supply capacity of power distribution network.
The step 11) specifically includes the following steps:
111) it selects u kind Single model and v kind built-up pattern to verify respectively the historical data of load point i, obtains u
+ v prediction models, select the smallest prediction model P of average absolute value errorLik0(t) as the optimum prediction mould at load point i
Type PLi(t), then have:
PLi(t)=PLik0(t) k=1,2 ..., u+v
Net load optimum prediction model if containing distributed generation resource at load point i, at the load point are as follows:
P′Li(t)=PLik0(t)-PNi
Wherein, PNiFor the nominal output of distributed generation resource at load point i.
112) the optimum prediction model of whole load points is overlapped, obtains the Additive Model of whole load points, then has:
In the step 112), using Additive Model as power distribution network power supply capacity prediction model, then power distribution network is for electric energy
The objective function of power prediction model are as follows:
Wherein, PLFor the maximum burden with power that power distribution network is capable of supply that, NfFor load points, PLiFor having at load point i
Workload.
In the step 112), the constraint condition of power distribution network power supply capacity prediction model includes:
Power-balance constraint:
Aiz=Iy
Power supply point power output bound constraint:
The constraint of node voltage bound:
The power-carrying of line and transformer constrains:
Wherein, A is node/branch incidence matrix, izFor branch current column vector, IyFor node Injection Current column vector,
gd、The respectively power output and its bound of power supply point d, Uj、Respectively the voltage of node j and
Its bound, il、The respectively electric current and its upper limit of route l, ST、The respectively power and its upper limit of transformer T.
In the step 2), Load flow calculation acquisition critical point is repeated by constantly increasing system loading, i.e., currently
The peak load that system can supply.
The acquisition methods of critical point in the step 2) specifically:
Successively call the Additive Model P ' of whole load point net loadsL(t) Load flow calculation is carried out until system occurs for the first time
It is out-of-limit, remember t=t at this time0, then the power supply capacity P of systemXIn PL(t0- 1) and PL(t0) between, it is sought using split-half method full
The critical point data of sufficient required precision, given accuracy ε enable a=(t0- 1), b=t0, the midpoint of section [a, b] is set as c, uses
P′L(c) Load flow calculation is carried out, if system is not out-of-limit, taking [c, b] is new section, is otherwise taken [a, c], and above step is repeated
Until meeting required precision, i.e. b-a < ε, then P is exportedL(a) value is the power supply capacity of current power distribution network.
In the step 3), the power supply nargin of system is the difference of current power supply capacity and current loads, is changed as power grid
Good foundation.
Compared with prior art, the invention has the following advantages that
The present invention is respectively adopted a variety of prediction models by the historical data to different load point and verifies, by it is a variety of most
Good prediction model is overlapped to obtain respectively the folded of the Additive Model of power distribution network whole load point and whole load point net loads
Add model, establish the mathematical model of power supply capacity calculating as the growth pattern of load using Additive Model and combines split-half method
The power supply capacity of power distribution network is calculated, the power supply that limitation actual load increases can be sequentially found by calculating power supply capacity repeatedly
Bottleneck, sample result show this method assessment result more can it is accurate, can comprehensively reflect the true operating condition of power distribution network.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention.
Fig. 2 is the power distribution network power supply capacity calculation process based on load prediction.
Fig. 3 is the model library of Mid-long term load forecasting software package.
Fig. 4 is that critical point solves schematic diagram.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment
As shown in Figure 1, the present invention provides a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting, use
Power grid is improved with realizing, the present invention is respectively adopted a variety of prediction models by the historical data to different load point and tests
Card, obtains the optimum prediction model of each load point and the optimum prediction model of each load point net load, then by two kinds
Optimum prediction model is overlapped to obtain respectively the Additive Model of power distribution network whole load point and whole load point net loads
Additive Model finally establishes the mathematical model of power supply capacity calculating using obtained Additive Model as the growth pattern of load
And the power supply capacity of power distribution network is calculated in conjunction with split-half method.Limitation actual negative can be sequentially found by calculating power supply capacity repeatedly
The power supply bottleneck that lotus increases, this has important directive significance to the operation of actual electric network and the optimization of network structure.
1) Mid-long term load forecasting model is constructed:
Mid-long Term Load has statistical law, while being influenced again by many random factors and having uncertainty.No
Variation characteristic with load has significant difference with the difference of load nature of electricity consumed, for example resident load and Commercial Load have warp
Normal year growth and apparent seasonal fluctuation feature, and industrial load seasonal fluctuation is little.These features cause pre-
The diversity of model is surveyed, to adapt to the load prediction of different regions, different periods.Therefore, actual prediction work there is choosing
The problem of selecting appropriate prediction model, this is that have a critical step, can be transported simultaneously using several prediction models when necessary
It calculates, to compare selection.
(1) individual event prediction model
Individual event prediction model described in this chapter is to refer to that the load under ordinary meaning is pre- for combination forecasting
Survey model.What it reflected is the general features of empirical documentation internal structure, is not fully coincide with the specific structure of the data.In advance
The materialization for surveying model is exactly predictor formula, and using the available numerical value for having similar structure to observation of predictor formula, this is just
It is predicted value.
There are many type of individual event prediction model, and common also has tens kinds, only provide the individual event prediction used herein here
Model: econometric model, Gradual regression analysis model, grey family molding, grey increasing order evaluation model, grey exponential smoothing model, ash
Color moving average model, Fuzzy Cluster Model, Estimates of Fuzzy Linear Regression Model, Fuzzy Exponential smoothing model.
(2) combination forecasting
Combination forecasting is to be added several obtained prediction results of individual event prediction model according to weight appropriate
Basic principle: a kind of prediction model of weight average assuming that during actual prediction, has k kind prediction side to certain prediction object f
Method, wherein being f using predicted value of the s kind method to the t periodst(s=1,2 ..., k) constitutes one using this k predicted value
To the final prediction result of f, i.e. f=Y (f1t,f2t,…,fkt), if the weight W=[w of various methods1,w2,…,wk]T, then
Combination forecasting can indicate are as follows:
The key problem of combination forecasting is the determination of weight coefficient, determines that weight will obtain using different methods
Different combination forecastings.Common method has the power methods such as equal weight averages method, variance-covariance method, optimal weighted method, recurrence
And recursive variance counting backward technique etc..
Combination forecasting can utilize to a certain extent the useful information of multiple individual event prediction models, while reduce it
Respective forecasting risk, most cases can obtain more preferably prediction result, but also have exception, actual prediction work
In should be verified simultaneously using individual event prediction model and combination forecasting, then select optimum prediction model.
2) mathematical model that power distribution network power supply capacity calculates
The power supply capacity of power distribution network refers to can supply most under conditions of power distribution network meets power constraint and voltage constraint
Big load.Therefore, it is necessary to the objective functions of optimization to indicate are as follows:
In formula: PLThe maximum burden with power being capable of supply that for power distribution network；NfFor load points；PLi
For the burden with power at load point i.
Constraint condition includes power-balance constraint, power supply point power output bound constraint, the constraint of node voltage bound, route
It is constrained with the power-carrying of transformer, it may be assumed that
Aiz=Iy (4)
In formula: A is node/branch incidence matrix；izFor branch current column vector；IyFor node Injection Current column vector；
gd、The respectively power output and its bound of power supply point d；Uj、Respectively the voltage of node j and its
Bound；il、The respectively electric current and its upper limit of route road l；ST、The respectively power and its upper limit of transformer T.
3) solution of model
The basic ideas that power distribution network power supply capacity computation model based on load prediction solves are: first against power distribution network
Each load point selects optimum prediction model；Secondly the optimum prediction model of all load points is overlapped, obtains whole
The Additive Model of load point；Load flow calculation finally is carried out using Additive Model, acquires the power supply capacity of power distribution network.
(1) the optimum prediction model of single load point
Since the load configuration of different load point is different, their load character often has bigger difference, and therefore, one negative
The optimum prediction model of lotus point may not be able to be suitble to another load point simultaneously.
In order to select optimum prediction model for each load point, by engineering common Single model and combination in practice
Model development is at software package, which is integrated with 9 kinds of Single models and 5 kinds of built-up patterns, as shown in Figure 3.
If sharing N in systemfA load point, then for load point i (1≤i≤Nf), the selecting party of optimum prediction model
Method is as follows:
Historical data selection u (1≤u≤9) kind Single model and v (1≤v≤5) kind built-up pattern difference to load point i
It is verified, is obtained u+v prediction model (k=1,2 ..., u+v), the average absolute value of more each model prediction data is missed
Difference, if model PLik0(t) average absolute value error is minimum, then the optimum prediction model of load point i is are as follows:
PLi(t)=PLik0(t) (9)
In formula: PLiIt (t) is the optimum prediction model at load point i.
If containing distributed generation resource at load point i, the optimum prediction model of net load can be indicated at the load point are as follows:
P′Li(t)=PLik0(t)-PNi (10)
In formula: P 'LiIt (t) is the optimum prediction model of net load at load point i, PNiFor distributed generation resource at load point i
Nominal output.
(2) Additive Model of whole load points
In order to facilitate calculating, the optimum prediction model of each load point can be overlapped, obtain the folded of whole load points
Add model:
In formula: PLIt (t) is the Additive Model of whole load points.
Above formula is the objective function for needing to optimize in power supply capacity computation model, can be written as follow form:
Therefore it may only be necessary to calculate the maximum value t of the load growth time limitmaxThe power supply capacity P of power grid can be obtainedL
(tmax)。
In addition, the optimum prediction model of each load point net load is overlapped, available whole load point is born only
The Additive Model of lotus:
In formula: P 'LIt (t) is the Additive Model of whole load point net loads.
(3) determination of critical point
According to definition, the power supply capacity of power distribution network is equal to what power distribution network when just having a constraint to work in model was supplied
Load sum.The most direct acquiring method of this critical point is repeated power flow method, and basic thought is by constantly increasing system
Load, and Load flow calculation is repeated to determine peak load that system can supply.
As shown in figure 4, setting PXFor the power supply capacity of power grid, PLIt (t) is the Additive Model of whole load points, then critical point
The method of determination can be stated are as follows: successively call P 'L(t) (t=1,2,3 ...) carry out Load flow calculation until system occur for the first time it is out-of-limit,
Remember t=t at this time0, then the power supply capacity P of systemXIt is necessarily in PL(t0- 1) and PL(t0) between, it next can use split-half method
The critical point data for meeting required precision is sought, given accuracy ε enables a=(t0- 1), b=t0, the midpoint in section [a, b] is asked to set
For c, P ' is calledL(c) carry out Load flow calculation, taken if system is not out-of-limit [c, b] be new section, otherwise take [a, c], repeat with
Upper step exports P until meeting required precision, i.e. b-a < εL(a) value is the power supply capacity of current power distribution network.
As shown in Fig. 2, the detailed step that power distribution network power supply capacity calculates is as follows:
(1) the optimum prediction model of each load point is determined.For each of power distribution network load point, using a variety of lists
Item prediction model and multiple combinations prediction model verify its historical data, select the smallest prediction of average absolute value error
Optimum prediction model of the model as the load point.
(2) Additive Model of whole load points is determined.The optimum prediction model of each load point is overlapped, is matched
The Additive Model P of power grid whole load pointL(t)。
(3) Additive Model of whole load point net loads is determined.By the optimum prediction model of each load point net load into
Row superposition, obtains the Additive Model P ' of power distribution network whole load point net loadL(t)。
(4) system load flow calculates.Successively call P 'L(t) (t=1,2,3 ...) carries out Load flow calculation, until system first time
Occur it is out-of-limit until, note t=t at this time0。
(5) critical point is determined.Given accuracy ε, enables a=t0- 1, b=t0, construct closed interval [a, b] and sought with split-half method
Meet the critical point of required precision.
(6) system power supply capability analysis.Export the value P of Additive Model corresponding to critical pointL(a) to get the confession of system
Electric energy power.Power supply capacity and the difference of current loads are the power supply nargin of system.Successively ignore the system weakness acquired again
Secondary to carry out calculating the weak link sequence and its corresponding time limit that can be obtained that limitation actual load increases, obtained result can
Using the foundation as system improvement.
The factor for influencing power distribution network power supply capacity mainly includes two parts, and first part is that substation's factor mainly includes becoming
Station capacity, substation's quantity, main transformer capacity, main transformer quantity, second part are that network contact factor mainly includes interconnection
Quantity, interconnection capacity etc..
In order to reflect the expansible power supply capacity of power distribution network, defining η indicates expansible power supply capacity ratio:
The numerical value of η reflects the power supply capacity of system, in order to make system power supply ability maximize, according to following table to rack into
Row optimization, as shown in table 1.
The range and meaning of the expansible power supply capacity ratio of table 1
Claims (8)
1. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting, which comprises the following steps:
1) the power distribution network power supply capacity prediction model based on Mid-long term load forecasting is constructed；
2) power distribution network power supply capacity prediction model is solved to obtain the power supply capacity of current power distribution network；
3) the power supply nargin of current power distribution network is obtained according to power supply capacity.
2. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 1, special
Sign is, the step 1) specifically includes the following steps:
11) the optimum prediction model of each load point is determined；
For each of power distribution network load point, using a variety of individual event prediction models and multiple combinations prediction model to its history
Data are verified, and select the smallest prediction model of average absolute value error as the optimum prediction model of the load point；
12) the optimum prediction model of all load points is overlapped, obtains the Additive Model of whole load points；
13) Load flow calculation is carried out using Additive Model, obtains the power supply capacity of power distribution network.
3. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 1, special
Sign is, the step 11) specifically includes the following steps:
111) it selects u kind Single model and v kind built-up pattern to verify respectively the historical data of load point i, obtains u+v
Prediction model selects the smallest prediction model P of average absolute value errorLik0(t) as the optimum prediction model P at load point iLi
(t), then have:
PLi(t)=PLik0(t) k=1,2 ..., u+v
Net load optimum prediction model if containing distributed generation resource at load point i, at the load point are as follows:
P′Li(t)=PLik0(t)-PNi
Wherein, PNiFor the nominal output of distributed generation resource at load point i.
112) the optimum prediction model of whole load points is overlapped, obtains the Additive Model of whole load points, then has:
4. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 3, special
Sign is, in the step 112), using Additive Model as power distribution network power supply capacity prediction model, then and power distribution network power supply capacity
The objective function of prediction model are as follows:
Wherein, PLFor the maximum burden with power that power distribution network is capable of supply that, NfFor load points, PLiIt is active negative at load point i
Lotus.
5. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 4, special
Sign is, in the step 112), the constraint condition of power distribution network power supply capacity prediction model includes:
Power-balance constraint:
Aiz=Iy
Power supply point power output bound constraint:
The constraint of node voltage bound:
The power-carrying of line and transformer constrains:
Wherein, A is node/branch incidence matrix, izFor branch current column vector, IyFor node Injection Current column vector, gd、The respectively power output and its bound of power supply point d, Uj、The respectively voltage of node j and thereon
Lower limit, il、The respectively electric current and its upper limit of route l, ST、The respectively power and its upper limit of transformer T.
6. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 5, special
Sign is, in the step 2), Load flow calculation acquisition critical point, i.e., current system is repeated by constantly increasing system loading
The peak load that system can be supplied.
7. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 6, special
Sign is, the acquisition methods of critical point in the step 2) specifically:
Successively call the Additive Model P of whole load point net loadsL' (t) carry out Load flow calculation until system occur for the first time it is out-of-limit,
Remember t=t at this time0, then the power supply capacity P of systemXIn PL(t0- 1) and PL(t0) between, it seeks meeting precision using split-half method
It is required that critical point data, given accuracy ε enables a=(t0- 1), b=t0, the midpoint of section [a, b] is set as c, using P 'L(c) into
Row Load flow calculation, if system is not out-of-limit, taking [c, b] is new section, is otherwise taken [a, c], repeats above step until meeting
Required precision, i.e. b-a < ε, then export PL(a) value is the power supply capacity of current power distribution network.
8. a kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting according to claim 1, special
Sign is, in the step 3), the power supply nargin of system is the difference of current power supply capacity and current loads.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811594173.5A CN109510203A (en) | 2018-12-25 | 2018-12-25 | A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811594173.5A CN109510203A (en) | 2018-12-25 | 2018-12-25 | A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109510203A true CN109510203A (en) | 2019-03-22 |
Family
ID=65754680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811594173.5A Pending CN109510203A (en) | 2018-12-25 | 2018-12-25 | A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109510203A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993360A (en) * | 2019-03-29 | 2019-07-09 | 国网山东省电力公司费县供电公司 | A kind of prediction technique and device of power data |
-
2018
- 2018-12-25 CN CN201811594173.5A patent/CN109510203A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993360A (en) * | 2019-03-29 | 2019-07-09 | 国网山东省电力公司费县供电公司 | A kind of prediction technique and device of power data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Reliability evaluation of power system considering time of use electricity pricing | |
| Macedo et al. | Demand side management using artificial neural networks in a smart grid environment | |
| CN101179195B (en) | Power distribution network planning scheme assistant decision system | |
| Zhang et al. | Three-stage hierarchically-coordinated voltage/var control based on PV inverters considering distribution network voltage stability | |
| Neimane | On development planning of electricity distribution networks | |
| Lisowski et al. | PLC-enabled low voltage distribution network topology monitoring | |
| Krok et al. | A coordinated optimization approach to Volt/VAr control for large power distribution networks | |
| CN101888087A (en) | Method for realizing distributed super-short-term area load forecasting in distribution network terminal | |
| CN107169631B (en) | Active power distribution network transformer substation planning method based on improved weighted Voronoi diagram | |
| Valgaev et al. | Building power demand forecasting using K-nearest neighbours model–practical application in Smart City Demo Aspern project | |
| Leal et al. | Distribution transformer losses evaluation: A new analytical methodology and artificial neural network approach | |
| CN110363334A (en) | Grid-connected grid line loss prediction technique based on Grey Neural Network Model | |
| CN112184028A (en) | Substation engineering dynamic site selection planning method based on harmony search algorithm | |
| Chang et al. | Distribution transformer load modeling using load research data | |
| CN113807589A (en) | Rolling optimization method and device for operator energy storage based on model prediction | |
| He et al. | Forecasting power grid states for regional energy markets with deep neural networks | |
| CN108346009A (en) | A kind of power generation configuration method and device based on user model self study | |
| Hosseini et al. | Day-ahead scheduling for economic dispatch of combined heat and power with uncertain demand response | |
| Velasco et al. | Day-ahead base, intermediate, and peak load forecasting using k-means and artificial neural networks | |
| Kim et al. | Economical energy storage systems scheduling based on load forecasting using deep learning | |
| CN109510203A (en) | A kind of power distribution network evaluation of power supply capability method based on Mid-long term load forecasting | |
| Šemić et al. | Distributed generation allocation in low voltage distribution network using artificial neural network | |
| CN117595398A (en) | Power system flexibility optimization method and device | |
| Sauhats et al. | Stochastic planning of distribution lines | |
| Deng et al. | Medium-term rolling load forecasting based on seasonal decomposition and long short-term memory 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |