CN105988974A - Load curve convergence optimal inflection point recognition method - Google Patents
Load curve convergence optimal inflection point recognition method Download PDFInfo
- Publication number
- CN105988974A CN105988974A CN201510039045.4A CN201510039045A CN105988974A CN 105988974 A CN105988974 A CN 105988974A CN 201510039045 A CN201510039045 A CN 201510039045A CN 105988974 A CN105988974 A CN 105988974A
- Authority
- CN
- China
- Prior art keywords
- load curve
- data
- inflection point
- flex point
- load
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A load curve convergence optimal inflection point recognition method includes the following steps: S1, data collection; S2, data pretreatment including S2-1 data filling and S2-2 data correction through a wavelet de-noising method; S3, achievement of approximate fitting of a load curve based on a data pretreatment technique; and S4, performing mathematical treatment on the obtained fitting daily load curve to determine an optimal inflection point. After pretreatment is performed on the original data through data filling and the wavelet de-noising method, Matlab software is called to perform approximate fitting on polynomial, and in this way, a smooth continuous function format is obtained; an inflection point definition of the mathematics field is applied to the power load curve, the inflection point of the load curve can be rapidly found, and then power dispatching workers can timely determine the present inflection point and can make ready for corresponding dispatching works, and the stability and the reliability of power grid operation can be improved.
Description
Technical field
The present invention relates to the identification technology of optimum flex point, be specifically related to the recognition methods of the convergent optimum flex point of a kind of load curve.
Background technology
Load curve is the time dependent curve of all kinds of electric loads in power system, is the power system foundation that carries out economic load dispatching and systems organization.Along with the popularization in the range of nationwide integrated power grid of the intelligent grid dispatching technique, transregional data sharing transprovincially has been carried out, this just provides reliable data basis for most optimum distribution of resources on a large scale, and interconnection plan is one of key link realizing transregional most optimum distribution of resources transprovincially under existing operation plan pattern.Therefore, flex point for electric load curve identifies the existence being possible not only to help the timely flex point of power scheduling personnel accurately, carry out and dispatch preparation accordingly, the planning of suitable interconnection can also be formulated for grid company simultaneously, determine that interconnection shape provides certain reference, it is favorably improved the economic benefit of grid company and the reliability of power supply, improves the utilization ratio of the energy.But, Study of recognition currently for electric load curve flex point is the most less, most research concentrates on the foundation of interconnection planning model, the planning and designing of interconnection and the aspect such as operation control and available transfer capability, the most also need the recognition methods studying the convergent optimum flex point of a kind of load curve badly.
Summary of the invention
In order to overcome disappearance and the deficiency of above-mentioned prior art, the present invention provides the recognition methods of the convergent optimum flex point of a kind of load curve, can the flex point of quantitative judge load curve, be conducive to instructing management and running personnel to formulate corresponding strick precaution corrective measure and interconnection plan, thus improve the security reliability of operation of power networks.
In order to realize foregoing invention purpose, the present invention adopts the following technical scheme that:
The recognition methods of the convergent optimum flex point of a kind of load curve is provided, said method comprising the steps of:
S1. data collection
Collect invention and intend implementing the historical load data of area every 15 minutes of every day (totally 96 data points every day).
S2. data prediction
The data collected in step S1 are carried out pretreatment, and this step comprises following sub-step:
S2-1. Supplementing Data
For the related data collected, have loss and gaps and omissions unavoidably, for the data of disappearance, carry out completion by the average of synchronization historical data.
S2-2. Wavelet Denoising Method is used to carry out data correction
For abnormal data, analyse whether it is abnormity point.In order to ensure the correct identification to load curve entirety flex point, exceptional value to be processed, call Matlab wavelet analysis module and load data is done monolayer wavelet decomposition, carry out low-pass filtering, then do low frequency signal reduction, i.e. can get the data after Wavelet Denoising Method.Concrete flow process is as shown in Figure 1.
S3. based on Data Preprocessing Technology, it is achieved the approximate fits of load curve
With the data set (i.e. every 15min every day totally 96 time points) on discrete point and known functional value (i.e. power load charge values corresponding to each 15min) on this point set, construct an analytical function, make it as close possible to given functional value on former discrete point, call polyfit function method of least square in Matlab and carry out the fitting of a polynomial of daily load curve.
S4. the matching daily load curve obtained is carried out Mathematical treatment, determine optimum flex point
The matching daily load curve obtained in step S3 is made Mathematical treatment, and the second dervative obtaining curve is the point of 0, it is judged that the symbol of the second dervative of these both sides, and the point that symbol changes is defined as the optimum flex point of load curve.
Accompanying drawing explanation
Fig. 1 data correction Wavelet Denoising Method flow chart
The identification process figure of the convergent optimum flex point of Fig. 2 load curve
Detailed description of the invention
As shown in Figure 2, the embodiment of the recognition methods of the convergent optimum flex point of a kind of load curve that the present invention proposes specifically comprises the following steps that
(1) data collection
Collect invention and intend implementing the historical load data of area every 15 minutes of every day (totally 96 data points every day).
(2) data prediction
The data collected in step S1 are carried out pretreatment, and this step comprises following sub-step
1) Supplementing Data
For the related data collected, have loss and gaps and omissions unavoidably, for the data of disappearance, carry out completion by the average of synchronization historical data.
2) Wavelet Denoising Method is used to carry out data correction
For abnormal data, analyse whether it is abnormity point.In order to ensure the correct identification to load curve entirety flex point, exceptional value processed, to call Matlab wavelet analysis module and load data is done monolayer wavelet decomposition, after carrying out low-pass filtering, HFS is removed, then remaining low frequency part is reduced, i.e. obtain the data after denoising.Concrete flow process is as shown in Figure 1.
(3) based on Data Preprocessing Technology, it is achieved the approximate fits of daily load curve.
With the data set (i.e. every 15min every day totally 96 time points) on discrete point and known functional value (i.e. power load charge values corresponding to each 15min) on this point set, construct an analytical function, make it as close possible to given functional value on former discrete point, with method of least square, daily load curve is carried out approximate fits.Calling polyfit function in Matlab and carry out the fitting of a polynomial of load, the function called is:
P=polyfit (x, y, n)
[p, s]=polyfit (x, y, n) (1)
According to the load data of every day, input time data corresponding for every 15min as x value, input the load value of corresponding time point respectively as y value, recently enter matched curve high-order term 6 as n value, obtain the approximate fits polynomial curve of corresponding daily load curve.
(4) the matching daily load curve obtained is carried out Mathematical treatment, determine optimum flex point.
The matching daily load curve obtained in step (3) being made Mathematical treatment, obtains the point that curvilinear function second dervative is 0, it is judged that the symbol of the second dervative of these both sides, the point that symbol changes is defined as the optimum flex point of load curve.
The recognition methods of the convergent optimum flex point of a kind of load curve that the present invention proposes first passes through Supplementing Data and Wavelet Denoising Method and initial data has been carried out pretreatment, then the data call Matlab software after processing carries out polynomial approximate fits, thus can obtain the continuous function form smoothed, finally the flex point of art of mathematics is defined and be applied on electric load curve, the flex point of load curve can be found out fast and effectively, power scheduling personnel are helped to judge the existence of flex point in time, carry out and dispatch preparation accordingly, the planning of suitable interconnection can also be formulated for grid company simultaneously, determine that interconnection shape provides certain reference, it is favorably improved the economic benefit of grid company and the reliability of power supply, improve the utilization ratio of the energy.
Claims (3)
1. the recognition methods of the convergent optimum flex point of load curve, it is characterised in that comprise the steps of:
S1. data collection；
S2. data prediction；
Described data prediction comprises the steps:
S2-1 Supplementing Data；
S2-2. Wavelet Denoising Method is used to carry out data correction；
S3. based on Data Preprocessing Technology, it is achieved the approximate fits of load curve
S4. the matching daily load curve obtained is carried out Mathematical treatment, determine optimum flex point.
The recognition methods of the convergent optimum flex point of load curve the most according to claim 1, it is characterised in that
The load data collected is smoothed, for entering one by the Wavelet Denoising Method mode of using described in S2
The approximate fits of the load curve of step is prepared.
The very Short-Term Load Forecasting Method of consideration impact load disturbance the most according to claim 1, it is special
Levy and be that the flex point of art of mathematics being defined described in S4 is applied on electric load curve, Ke Yiyou
Effect quickly finds out the flex point of load curve, has good practical value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039045.4A CN105988974A (en) | 2015-01-27 | 2015-01-27 | Load curve convergence optimal inflection point recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039045.4A CN105988974A (en) | 2015-01-27 | 2015-01-27 | Load curve convergence optimal inflection point recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105988974A true CN105988974A (en) | 2016-10-05 |
Family
ID=57035536
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510039045.4A Pending CN105988974A (en) | 2015-01-27 | 2015-01-27 | Load curve convergence optimal inflection point recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105988974A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110242589A (en) * | 2019-06-25 | 2019-09-17 | 江苏大学 | A kind of centrifugal pump performance fitting modification method |
| CN110821850A (en) * | 2019-12-18 | 2020-02-21 | 江苏国泉泵业制造有限公司 | Centrifugal pump test data correction method |
| CN111539654A (en) * | 2020-05-26 | 2020-08-14 | 国网湖南省电力有限公司 | Big data-based impact type electric power large customer identification method |
-
2015
- 2015-01-27 CN CN201510039045.4A patent/CN105988974A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110242589A (en) * | 2019-06-25 | 2019-09-17 | 江苏大学 | A kind of centrifugal pump performance fitting modification method |
| CN110821850A (en) * | 2019-12-18 | 2020-02-21 | 江苏国泉泵业制造有限公司 | Centrifugal pump test data correction method |
| CN111539654A (en) * | 2020-05-26 | 2020-08-14 | 国网湖南省电力有限公司 | Big data-based impact type electric power large customer identification method |
| CN111539654B (en) * | 2020-05-26 | 2023-04-25 | 国网湖南省电力有限公司 | Impact type power large customer identification method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105988974A (en) | Load curve convergence optimal inflection point recognition method | |
| Min et al. | A trust region interior point algorithm for optimal power flow problems | |
| CN103049431B (en) | ICD (IED Capability Description) inspection method based on objectified module semantics | |
| CN104899457A (en) | Satellite data prediction method based on improved unbiased GM (1, 1) model | |
| CN102692558A (en) | Monitoring and analyzing system for electricity data and realization method thereof | |
| CN105447082A (en) | Distributed clustering method for mass load curves | |
| CN111950197A (en) | Distribution network attack and fault acquisition and analysis system based on artificial intelligence semantics | |
| CN115882455B (en) | Distributed photovoltaic power generation prediction method, system and terminal | |
| CN109324264B (en) | Identification method and device for abnormal value of power distribution network line impedance data | |
| CN110994666A (en) | Order reduction method for reducing impedance model to RLC circuit at resonance frequency | |
| CN113659712A (en) | Power equipment big data label operation platform and method | |
| CN103984985A (en) | Method for optimizing customized furniture serial batch scheduling | |
| Feng et al. | Auxiliary model-based stochastic gradient algorithm for multivariable output error systems | |
| WO2022111741A1 (en) | System and method for load identification in collaboration with cloud end | |
| CN102542028B (en) | Information iterative classification method of smart grid on basis of automatic control theory | |
| CN106570618A (en) | Cluster analysis and neutral network-based load simultaneous factor prediction method | |
| CN107248118A (en) | Data digging method, device and system | |
| CN106684865A (en) | Reactive daily curve bad data identification and correction method | |
| CN102609513A (en) | Multi-data-source database system for condition monitoring of electric transmission and transformation equipment | |
| Han | Research on automation control technology of electrical and electronic equipment combined with integrated wireless module | |
| CN108197750A (en) | A kind of wind power combination forecasting method based on LMD-Elman | |
| CN102999648B (en) | Design method and design system of artificial electromagnetic material unit structure | |
| CN109473985A (en) | One kind being based on BP neural network smart grid distribution method | |
| CN117743810B (en) | Intended understanding analysis method for electroencephalogram signal monitoring | |
| CN103353978A (en) | Method for compiling medium and long term electric energy production plan |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161005 |