CN103258144A - On-line static load modeling method based on data of fault recorder - Google Patents
On-line static load modeling method based on data of fault recorder Download PDFInfo
- Publication number
- CN103258144A CN103258144A CN2013102233251A CN201310223325A CN103258144A CN 103258144 A CN103258144 A CN 103258144A CN 2013102233251 A CN2013102233251 A CN 2013102233251A CN 201310223325 A CN201310223325 A CN 201310223325A CN 103258144 A CN103258144 A CN 103258144A
- Authority
- CN
- China
- Prior art keywords
- data
- static load
- model
- load
- constantly
- 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.)
- Granted
Links
Images
Abstract
The invention relates to an on-line static load modeling method based on data of a fault recorder. The method is characterized in that loads are classified in a mathematical matrix form; load model parameter libraries in different time scales are formed; and model parameters in the different time scales are subjected to on-line correction by a recursive least square method. Data of a certain 110kV line in a typical summer day is adopted for verification, so that the method is simple, convenient and feasible, and can effectively overcome the difficulty of the load time-varying characteristic. The method comprises the steps that classification is performed according to the time characteristic; various corresponding static load model parameter libraries are formed; the static load model parameter libraries are subjected to the on-line correction by the recursive least square method according to corresponding real-time data; and finally a user selects static load models in different time scales according to an application scene.
Description
Technical field
The present invention relates to a kind of online static load modeling method based on the fault oscillograph data.
Background technology
In the trend calculating and the steady-state analysis based on trend of electric system, often use the static model of load.A large amount of emulation show with test findings: the static load model to the trend of electric system calculate, reactive power compensator planning, voltage are stablized, the analysis of frequency stabilization, long-term dynamics process etc. has bigger influence, under critical condition, also might fundamentally change qualitative conclusions.Therefore, be necessary to carry out the research of static load modeling.
Society is along with the appearance of various novel electric power electric equipment now, make part throttle characteristics become and become increasingly complex, and As time goes on, load time variation feature is more obvious, the in-site measurement that U.S. GE company does shows that somewhere load static nature coefficient has changed 20% in 10min.So the modeling method that adopts investigation statistics or steady state test to obtain the static load model can not reflect the time variation feature of load well.Addressing this problem best bet is exactly the online acquisition data, and the online data of carrying out are handled then, obtain each load static nature coefficient constantly.Yet because the restriction of measurement mechanism, this method never effectively is applied in the actual load Modeling Research.Popularizing with fast development of fault wave recording device overcome this suffering, and making the online acquisition data carry out online static load modeling becomes possibility.
At first, on function, transient state process constantly can not only record trouble takes place in modern fault oscillograph, and steady state data that can the continuous recording day-to-day operation, the long-time dynamic process of system's operation, and embeds unified markers; Secondly, in configuration, all 220kV of electrical network and above electric pressure transformer station, part 110kV transformer station have have all installed and used fault oscillograph and have realized networking.In a word, adopt the networking fault oscillograph can realize carrying out the load measurement modeling at each transformer station more than the 110kV electric pressure, effectively solve the dispersed difficult problem of time variation and region in the load modeling.And special-purpose load modeling device is installed relatively, it is short to invest little, instant effect, cycle.Have important significance for theories and engineering practical value so adopt the networking fault oscillograph to carry out the load measurement modeling.
Summary of the invention
Purpose of the present invention is exactly for addressing the above problem, and from the model application point, the modeling thinking in conjunction with the statistics overall approach has provided a kind of online static load modeling method based on the fault oscillograph data.This method adopts the form of matrix on the mathematics that load is classified, and forms the load model parameters storehouse of different time yardstick, and adopts least square method of recursion that the model parameter of variant time scale is carried out online correction.Adopt certain 110kV circuit typical case in summer day data to verify, show that institute's extracting method is simple and feasible, can effectively solve load time variation difficulty.
For achieving the above object, the present invention adopts following technical scheme:
A kind of online static load modeling method based on the fault oscillograph data,
At first classify by the time characteristic, form corresponding all kinds of static load model parameter storehouse;
Adopt all kinds of static load parameter libraries of the online correction of least square method of recursion according to corresponding real time data then;
End user is according to the static load model under the application scenarios selection different time yardstick.
The process of setting up in described all kinds of static load model parameter storehouse is:
(1) load data is handled
The mass data that fault oscillograph collects is carried out analyzing and processing, extract the valid data that are used for the static load modeling;
(2) select the multinomial model structure for use, adopt least square method of recursion to carry out the parameter identification of static load model;
(3) the model parameter database of yardstick of many time makes up
With the proper vector of time as classification, form the load model parameters database of different time yardstick according to the form of matrix on the mathematics: transversely, by different constantly division the in a day, be divided into low-valley interval, morning peak period, mild period, evening peak period; Vertically, per year, season, month, working day and nonworkdays divide;
After classifying by the different time yardstick, carry out corresponding load modeling according to measured data; One group of data of the every collection of fault oscillograph are judged Overload Class under this data sample according to temporal characteristics, adopt the static load model of the online correction respective class of least square method of recursion then, a reserving model parameter identification result;
After different time yardstick load model parameters database made up, the user selected to get final product according to application scenarios.
Described parameter identification process is: handle earlier a batch data that has obtained, obtain the estimated value of θ, come data newly after, more former estimated value is revised, so both can reduce the memory space of computing machine, also can improve computing velocity; For this reason, the steady state data of gathering in real time according to fault oscillograph carries out the static load modeling, and new sampled data constantly refreshes, and adopts least square method of recursion constantly to revise institute's established model, for avoiding the data saturated phenomenon, adopts the recursive algorithm of the memory that fades,
In the formula: x (k), y (k) are respectively k input and the output of observation equation constantly;
Be k estimated parameter constantly; P (k) is the k variance battle array of estimated value constantly; K (k) is k gain coefficient constantly; λ is forgetting factor (0＜λ≤1), makes legacy data " be forgotten " gradually by exponential damping law, thus the effect of outstanding new data, the value of λ is more little, and the speed of " forgeing " is more fast, but the too little meeting of numerical value reduces the precision of identification, and it is big that fluctuation becomes; Generally get λ ∈ [0.8,1], λ=1 is got in off-line identification.
The invention has the beneficial effects as follows: in the past in the load modeling research because the restriction of measurement mechanism and can't effectively solve difficult problems such as time variation and region dispersiveness, the developing into of fault oscillograph solves the above difficult problem condition of providing convenience.The present invention has provided the thinking that the real-time steady state data that adopts the fault oscillograph record carries out online static load modeling, proposed to adopt the form of matrix on the mathematics that load is classified, form the load model parameters storehouse of different time yardstick, and adopt least square method of recursion that the model parameter of variant time scale is carried out online correction.Certain 110kV circuit typical case in summer day data of employing area at sunshine are verified, show that institute's extracting method is simple and feasible.
Description of drawings
Fig. 1 is the meritorious response of least square 1 day data identification comparison diagram;
Fig. 2 is least square method of recursion gained parameter situation of change figure;
Fig. 3 is the total meritorious response comparison diagram of classification identification gained.
Embodiment
The present invention is from the model application point, in conjunction with the modeling thinking of statistics overall approach, provided the thinking that the real-time steady state data that adopts the fault oscillograph record carries out online static load modeling.Proposed to adopt the form of matrix on the mathematics that load is classified, formed the load model parameters storehouse of different time yardstick, and adopted least square method of recursion that the model parameter of variant time scale is carried out online correction.Certain 110kV circuit typical case in summer day data of employing area at sunshine are verified, show that institute's extracting method is simple and feasible, can effectively solve load time variation difficulty.
The mathematical description of 1 static load model
Under steady state conditions, the nonlinear function between load power and terminal voltage and the frequency is called the static model of load.The basic model of describing the static load characteristic mainly contains multinomial model and power function model.
1.1 multinomial model
In the formula, U is virtual voltage; U
0Be base value voltage; P, Q are actual meritorious, reactive power; P
0, Q
0Be base value power, the power when namely voltage and frequency are ratings; A
p, B
p, C
p, A
q, B
q, C
qCoefficient for the load static model; Obviously, each coefficient satisfies in the formula:
1.2 power function model
The citation form of power function model commonly used is as follows:
In the formula, U is virtual voltage; U
0Be base value voltage; P, Q are actual meritorious, reactive power; P
0, Q
0Be base value power, the power when namely voltage and frequency are ratings; p
u, q
uCoefficient for the load static model
2 online static load modeling thinkings
During the static load modeling, traditional statistics overall approach or steady-state test technique are limited to manpower and materials and are difficult to often carry out, make the gained model can't accurate description the time varying characteristic of load.The development of fault oscillograph makes the online acquisition online data carry out modeling becomes possibility, can effectively solve a time variation difficult problem, and basic step is as follows:
(1) load data is handled
The mass data that fault oscillograph collects is carried out analyzing and processing, extract the valid data that are used for the static load modeling.Modern fault oscillograph can be realized stable state, transient state, the dynamic continual overall process data recording of long-time continuous and analytic function, image data is carried out analyzing and processing, set criterion, form the valid data that can be used for the static load modeling, conveniently be directly used in the static load modeling.
(2) load model structure choice
Common static load model structure has the combination of multinomial model and power function model and two kinds of models, along with part throttle characteristics becomes increasingly complex, nonlinear function model structures such as the splines of employing model, neural network model are also arranged, and this paper selects the multinomial model structure for use.
(3) the identification of Model Parameters algorithm is selected
After the load model structure is selected, just need carry out parameter identification according to measurement data.Because fault oscillograph is image data continuously, if one group of new data of every collection is all comprehensively unified identification with legacy data, then calculated amount is big and time-consuming, and the data that fault oscillograph is gathered need periodic refreshing, can not preserve all historical datas, be infeasible so adopt the method for all uniform data identifications.Consider to adopt a kind of recursive algorithm to carry out the modeling of recursion correction static load, only preserve the load model identification result, one group of new data of every collection is revised on former identification result basis.So not only calculated amount is little, operation efficiency is high, and can fully extract the common essential characteristic of all data, obtains describing the comprehensive static load model of all data samples.This paper adopts least square method of recursion to carry out the parameter identification of static load model.
(4) the model parameter database of yardstick of many time makes up
Size and the constituent of load are among the variation constantly, if only adopt a kind of model form that the data of fault oscillograph collection are revised, then the gained load model is the description to all image data, can't embody the time varying characteristic of load, and the descriptive power of model certainly will be not accurate enough.For this reason, the author is in conjunction with the modeling thinking of statistics overall approach, from the model application point, with the proper vector of time as classification, form the load model parameters database of different time yardstick according to the form of matrix on the mathematics: transversely, can be by different constantly division the in a day, as be divided into low-valley interval, morning peak period, mild period, evening peak period etc.; Vertically, can be per year, divisions such as season, month, working day and nonworkdays.Adopt this sorting technique, can obtain the load model of each time period intuitively, make things convenient for the user to select as required.
After classifying by the different time yardstick, just need carry out corresponding load modeling according to measured data.One group of data of the every collection of fault oscillograph are judged Overload Class under this data sample according to temporal characteristics, adopt the static load model of the online correction respective class of least square method of recursion then, a reserving model parameter identification result.Adopt this method can realize real-time update to the load model parameters database, obtain describing the comprehensive static model of correction of such all data samples.
After different time yardstick load model parameters database makes up, need select according to application scenarios, yet owing to adopt the form of matrix, the time of each partitioned matrix is determined, as in one day by per hour carrying out modeling, then just only keep horal model parameter.Adopt this method can obtain the load model of yardstick of many time, thereby adapt to the calculating needs of various different application scenes.
3 least square method of recursion
During the static load modeling, often adopt least square method to all uniform data identifications, when calculating, do not consider the time sequencing between the measurement data, when measurement data is a lot, require computing machine to have very big memory capacity.Yet in actual process, measurement data progressively provides often in chronological order, we can handle a batch data that has obtained earlier, obtain the estimated value of θ, after having come data newly, again former estimated value is revised, so both can have been reduced the memory space of computing machine, also can improve computing velocity.
This paper adopts least square method of recursion to carry out online static load modeling, compare with general least square method, this method does not need a large amount of matrix inversion operation, calculated amount is little, computing velocity is fast, but real-time online is used, and the result of this method result of calculation and batch processing observation data is identical.
The steady state data of gathering in real time according to fault oscillograph carries out the static load modeling, and new sampled data constantly refreshes, and adopts least square method of recursion constantly to revise institute's established model, and for avoiding the data saturated phenomenon, this paper adopts the recursive algorithm of the memory that fades.
In the formula: x (k), y (k) are respectively k input and the output of observation equation constantly, and k is arithmetic number;
Be k estimated parameter constantly; P (k) is the k variance battle array of estimated value constantly; K (k) is k gain coefficient constantly; λ is forgetting factor (0＜λ≤1), makes legacy data " be forgotten " gradually by exponential damping law, thereby gives prominence to the effect of new data, overcomes " data are saturated " phenomenon of least square method of recursion effectively.The value of λ is more little, and the speed of " forgeing " is more fast, but the too little meeting of numerical value reduces the precision of identification, and it is big that fluctuation becomes.Generally get λ ∈ [0.8,1], λ=1 is got in off-line identification.
4 sample calculation analysis
The present invention adopts the data instance of typical case's day outlet whole day of certain 110kV of 220kV transformer station side of area at sunshine in summer to analyze, and gathers one group of data in per five minutes, and data characteristics is only listed preceding 10 groups of data as space is limited shown in table 3.1.
Table 1 load measurement data
Because meritorious and idle expression formula is basic identical, just model parameter is different, and the present invention only carries out the static load modeling with meritorious data instance, and model form adopts polynomial expression II model, and is as follows:
P(k)=a
p+b
pU(k)+c
pU
2(k) （7）
In the formula, P (k) is that k equation output constantly is meritorious, and k is arithmetic number; U (k) is k equation input voltage constantly; a
p, b
p, c
pBe respectively static model parameter to be identified.
For the characterization model quality, adopt following error function:
In the formula, J is the error of calculation, and n represents that sampled data counts, and i is arithmetic number, and P is that model calculates meritorious response, P
mMeritorious for surveying.
1.1.14.1 least square method is unified identification
Adopt these data of day to carry out the static load modeling and simulating, when adopting least square method to all uniform data identifications, gained parameter identification result and error are as shown in table 2, and model calculates meritorious and actual meritorious contrast as shown in Figure 1.
Table 2 least square 1 day data identification result and error
Can obviously be found out by table 2 and Fig. 1, when adopting least square method to all uniform data identifications in a day, because different part throttle characteristics constantly differ bigger in one day, adopt the resulting load model descriptive power of method of this unified identification relatively poor.
1.1.24.2 least square method of recursion identification
Based on these data of day, adopt least square method of recursion to carry out parameter identification, gained parameter identification result and error are as shown in table 3, and identification gained model parameter situation of change is as shown in Figure 2.
Table 3 recursive least-squares 1 day data identification result and error
Contrast table 3 and table 2, adopt least square method of recursion gained identification result with adopting least square method all uniform data identification gained results to be more or less the same as can be seen, basic not variation, the feasibility that adopts least square method of recursion to carry out the static load modeling is described, and least square method of recursion does not need to preserve all data when carrying out modeling, just identification result is handled, efficient height, speed are fast, can conveniently be applied to the online static load modeling based on the fault oscillograph data.
As seen from Figure 2, when adopting least square method of recursion to carry out parameter identification, different gained model parameters constantly change greatly when data volume is fewer, and along with increasing of image data, the Model Distinguish parameter also will tend towards stability.When adopting fault oscillograph to carry out recursion correction modeling, only preserve revised model parameter, along with the fault oscillograph growth of working time, a large amount of image data will make that every class model parameter is more and more stable, and the descriptive power of load model will be more and more accurate.
1.1.34.3 by the time classification model construction
Known by preamble, because the time variation of load, when all data in to a day were unified identification, gained model description ability was relatively poor, can't embody the time varying characteristic of load.Consider by the time characteristic to the load processing of classifying, to eliminate the time variation influence for this reason.This paper classifies a day data by the hour, and the The data least square method of recursion of each hour is carried out modeling, and this load model of day just can be combined by the load model of each hour.At this moment, gained Model Distinguish result and total error are as shown in table 4, and total model calculates meritorious and actual meritorious contrast as shown in Figure 3.
Table 4 classification back identification result and total error
Contrast table 4 and table 2 adopt the error of unifying the identification gained by the meritorious and actual error ratio of gaining merit of gained model calculating after the time classification to be much smaller as can be seen, and the degree of accuracy of gained model is higher; By Fig. 3 also as can be seen, it is more better to adopt the classification back total model of gained to calculate meritorious and actual meritorious fitting effect.
Claims (3)
1. the online static load modeling method based on the fault oscillograph data is characterized in that,
At first classify by the time characteristic, form corresponding all kinds of static load model parameter storehouse;
Adopt all kinds of static load parameter libraries of the online correction of least square method of recursion according to corresponding real time data then;
End user is according to the static load model under the application scenarios selection different time yardstick.
2. the online static load modeling method based on the fault oscillograph data as claimed in claim 1 is characterized in that, the process of setting up in described all kinds of static load model parameter storehouse is:
(1) load data is handled
The mass data that fault oscillograph collects is carried out analyzing and processing, extract the valid data that are used for the static load modeling;
(2) select the multinomial model structure for use, adopt least square method of recursion to carry out the parameter identification of static load model;
(3) the model parameter database of yardstick of many time makes up
With the proper vector of time as classification, form the load model parameters database of different time yardstick according to the form of matrix on the mathematics: transversely, by different constantly division the in a day, be divided into low-valley interval, morning peak period, mild period, evening peak period; Vertically, per year, season, month, working day and nonworkdays divide;
After classifying by the different time yardstick, carry out corresponding load modeling according to measured data; One group of data of the every collection of fault oscillograph are judged Overload Class under this data sample according to temporal characteristics, adopt the static load model of the online correction respective class of least square method of recursion then, a reserving model parameter identification result;
After different time yardstick load model parameters database made up, the user selected to get final product according to application scenarios.
3. the online static load modeling method based on the fault oscillograph data as claimed in claim 1, it is characterized in that, described parameter identification process is: handle a batch data that has obtained earlier, obtain the estimated value of θ, after having come data newly, again former estimated value is revised, so both can have been reduced the memory space of computing machine, also can improve computing velocity; For this reason, the steady state data of gathering in real time according to fault oscillograph carries out the static load modeling, and new sampled data constantly refreshes, and adopts least square method of recursion constantly to revise institute's established model, for avoiding the data saturated phenomenon, adopts the recursive algorithm of the memory that fades,
In the formula: x (k), y (k) are respectively k input and the output of observation equation constantly, and k is arithmetic number;
Be k estimated parameter constantly; P (k) is the k variance battle array of estimated value constantly; K (k) is k gain coefficient constantly; λ is forgetting factor (0＜λ≤1), makes legacy data " be forgotten " gradually by exponential damping law, thereby gives prominence to the effect of new data, overcomes " data are saturated " phenomenon of least square method of recursion effectively; The value of λ is more little, and the speed of " forgeing " is more fast, but the too little meeting of numerical value reduces the precision of identification, and it is big that fluctuation becomes; Get λ ∈ [0.8,1], λ=1 is got in off-line identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310223325.1A CN103258144B (en) | 2013-06-06 | 2013-06-06 | Online static load modeling method based on data of fault recorder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310223325.1A CN103258144B (en) | 2013-06-06 | 2013-06-06 | Online static load modeling method based on data of fault recorder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103258144A true CN103258144A (en) | 2013-08-21 |
| CN103258144B CN103258144B (en) | 2016-10-19 |
Family
ID=48962056
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310223325.1A Expired - Fee Related CN103258144B (en) | 2013-06-06 | 2013-06-06 | Online static load modeling method based on data of fault recorder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103258144B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9829880B2 (en) | 2014-11-20 | 2017-11-28 | General Electric Company | System and method for modelling load in an electrical power network |
| CN111796143A (en) * | 2020-09-10 | 2020-10-20 | 深圳华工能源技术有限公司 | Energy-saving metering method for energy-saving equipment of power distribution and utilization system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030018456A1 (en) * | 2001-06-29 | 2003-01-23 | Sid Browne | Method and system for stress testing simulations of the behavior of financial instruments |
| CN101777765A (en) * | 2010-01-27 | 2010-07-14 | 中国电力科学研究院 | On-line load simulation method of power system |
-
2013
- 2013-06-06 CN CN201310223325.1A patent/CN103258144B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030018456A1 (en) * | 2001-06-29 | 2003-01-23 | Sid Browne | Method and system for stress testing simulations of the behavior of financial instruments |
| CN101777765A (en) * | 2010-01-27 | 2010-07-14 | 中国电力科学研究院 | On-line load simulation method of power system |
Non-Patent Citations (3)
| Title |
|---|
| 张磊: "考虑时变性的电力负荷建模方法研究", 《中国硕士学位论文全文数据库工程科技Ⅱ辑》, no. 4, 15 April 2011 (2011-04-15) * |
| 徐振华: "面向智能电网的广义综合负荷建模方法研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》, no. 4, 15 April 2013 (2013-04-15) * |
| 杨静: "电力系统传输网络与负荷模型辨识及其在电压稳定分析中的应用", 《中国博士学位论文全文数据库工程科技Ⅱ辑》, no. 12, 15 December 2012 (2012-12-15) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9829880B2 (en) | 2014-11-20 | 2017-11-28 | General Electric Company | System and method for modelling load in an electrical power network |
| CN111796143A (en) * | 2020-09-10 | 2020-10-20 | 深圳华工能源技术有限公司 | Energy-saving metering method for energy-saving equipment of power distribution and utilization system |
| CN111796143B (en) * | 2020-09-10 | 2020-12-15 | 深圳华工能源技术有限公司 | Energy-saving metering method for energy-saving equipment of power distribution and utilization system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103258144B (en) | 2016-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070282B (en) | Low-voltage transformer area line loss influence factor analysis method based on comprehensive relevance | |
| CN103020459B (en) | A kind of cognitive method of various dimensions electricity consumption behavior and system | |
| CN108199404B (en) | Spectral clustering cluster division method of high-permeability distributed energy system | |
| CN106019026A (en) | Distribution method based on waveform matching for harmonic voltage responsibility | |
| CN103324841A (en) | Online dynamic load modeling method based on fault recorder data | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN103472430B (en) | Solar simulator irradiation nonuniformity and instability test macro | |
| CN112149873A (en) | Low-voltage transformer area line loss reasonable interval prediction method based on deep learning | |
| CN106405280B (en) | A kind of intelligent substation on-line monitoring parameter trend method for early warning | |
| CN113469488B (en) | Online diagnosis and analysis system for topological structure of power distribution network equipment | |
| CN103258144A (en) | On-line static load modeling method based on data of fault recorder | |
| CN103543637B (en) | A kind of tool environment temperature Analytic modeling method | |
| CN105259795A (en) | Internal impedance parameter expansion method for power battery simulator | |
| CN110098610B (en) | Real-time identification method and system for oscillation leading mode of power system under fault disturbance | |
| CN104200290B (en) | Wind power forecast method | |
| CN115561697A (en) | Intelligent ammeter error analysis method | |
| CN116050636A (en) | Output prediction method, device, equipment and medium of photovoltaic power station | |
| Zhang et al. | Analysis of influencing factors of transmission line loss based on GBDT algorithm | |
| CN114583767A (en) | Data-driven wind power plant frequency modulation response characteristic modeling method and system | |
| CN111898871B (en) | Method, device and system for evaluating data quality of power grid power supply end | |
| Guo et al. | A dynamic equivalence method considering the spatial effect of wind farms | |
| Du et al. | A time sequence data processing method based on random matrix of smart electric meter | |
| CN105808833A (en) | Online parameter identification method of parallel synchronous generator based on multi-data sets | |
| CN110717244A (en) | Data trust degree analysis computer simulation method based on average deviation degree algorithm | |
| CN106779265B (en) | Online state safety detection method based on electric power intelligent mobile terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161019 Termination date: 20210606 |