CN102445660A - Gray Verhulst model-based prediction method of power angle of generator - Google Patents
Gray Verhulst model-based prediction method of power angle of generator Download PDFInfo
- Publication number
- CN102445660A CN102445660A CN2011102950734A CN201110295073A CN102445660A CN 102445660 A CN102445660 A CN 102445660A CN 2011102950734 A CN2011102950734 A CN 2011102950734A CN 201110295073 A CN201110295073 A CN 201110295073A CN 102445660 A CN102445660 A CN 102445660A
- Authority
- CN
- China
- Prior art keywords
- generator
- angle
- network system
- power
- formula
- 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 discloses a gray Verhulst model-based prediction method of a power angle of a generator. According to the prediction method, a gray Verhulst model is introduced into prediction of a power angle of a generator of a grid system as well as the model is modified; two latest data are collected to determine a compensation error and correction is carried out on a predicted value, thereby further improving prediction precision. In addition, the invention also discloses a gray Verhulst model-based determination method for transient instability of a grid system as well as a grid system transient instability determination apparatus applying the method. According to the determination method, real-time identification on disturbance generation is carried out; smoothness prediction of a locus of a power angle after the disturbance is carried out; an instant prediction numerical value is utilized to estimate state quantities of a generator and a whole grid; and it is also determined whether an angular velocity of the generator, an inertia center deviation angle, a maximum angle difference, a homology group clearance angle are greater than threshold values or not; if any of the above-mentioned values exceeds the corresponded threshold value, transient instability early warning is carried out and transient early warning is carried out by a protection device. According to the invention, advantages of high prediction precision and wide applicability are realized.
Description
Technical field
The present invention relates to a kind of generator's power and angle Forecasting Methodology, relate in particular to a kind of generator's power and angle Forecasting Methodology, be applicable to the electrical network that has wide area measurement system (WAMS) based on grey Verhulst model.
Background technology
Along with interconnected, the extensive alternating current-direct current of electric system mixes the continuous access of transmission of electricity and regenerative resource, renewal, more urgent requirement have been proposed for electric power system transient stability early warning and control.The stable control technology based on disturbance event is adopted in conventional defence line, second road, and is with strong points, control rate is fast, reliability is high, but depends on model and parameter, off-line simulation calculating and given in advance fault collection, has significant limitation.Because the widespread use of PMU/WAMS predicts it is the study on power system focus based on the transient state unstability of PMU/WAMS wide area real-time measurement.In fact after the electric system generation catastrophic failure, certainly exist certain Changing Pattern based on the response message of wide area.Before power system transient stability is destroyed,, can be the transient stability real-time estimate and strong foundation is provided with control if can extract and excavate relevant information.
The utilization in succession of phasor measurement unit (PMU) and WAMS (WAMS); Realized real-time monitoring to generator's power and angle relevant informations such as (perhaps being called rotor angle); Guarantee for the running status of observing the whole network provides effective information, also crucial measurement means is provided for the transient stability analysis.But, only, still can't make stability Discrimination in advance according to the historical data at merit angle; Only for power-angle curve forecast analysis in addition; Hold its tendency in advance and exactly, could in regular hour nargin, take measures, keep the stable operation of the whole network.
The forecasting techniques of merit angle track can be divided into two types after the disturbance: one of which is the merit angle track faster than real time simulation method of the depression of order abbreviation Network Based of document one " based on the disturbed trajectory predictions of wide area measurement data and admittance parameter on-line identification " (Automation of Electric Systems was rolled up the 6th page of the 22nd phase in 2007 the 27th) proposition; The 2nd, document two " a kind of power system transient stability quick real-time Forecasting Methodology " (Proceedings of the CSEE was rolled up the 60th page of the 6th phase in 1993 the 13rd) is based on the merit angle track Extrapolation method of curve fitting.Because the latter is fully based on sequence of values analysis and data mining, and do not rely on priori such as electric system parameter, have the better application prospect, receive more concern.On the basis of document two; According to the difference of cubic fitting model, document three " the GPS synchronous clock is used for power system transient stability prediction and control " (Automation of Electric Systems was rolled up the 11st page of the 6th phase in 1998 the 22nd) has proposed the autoregression method of difference, but autoregressive model exists from modifying factor; In the process of prediction; Need a self-control process, and the adjusting duration in early stage can not control, the overall precision of prediction is caused considerable influence.Document four " based on the disturbed trajectory predictions of the electric system of WAMS " (Automation of Electric Systems was rolled up the 27th page of the 23rd phase in 2006 the 30th) proposes the combined triangular function method; The trigonometric function prediction is made up of sinusoidal adding up with cosine term; Because the cycle qualitative change of model self is more suitable for stable trajectory is predicted; For the unstability fault of merit angle monotone variation, in the tracking prediction certain deviation can appear.Document five " based on the fast transient stability forecast method of wide area system " (Automation of Electric Systems was rolled up the 1st page of the 21st phase in 2007 the 31st) adopts the Taylor series method; Obtain the angular velocity track through interpolation; And then predict disturbed merit angle track through continuous moment angular velocity being carried out integration; Its essence is the operation of predicting again integration behind the merit angular data differential, not from improving prediction effect in essence.
Therefore, traditional merit angle trajectory predictions method can not have good universality again when guaranteeing precision of prediction, and its engineering practicability is relatively poor.For this reason, need a kind of new forecasting techniques, desired data is few, length when calculating, for the merit angle track of various characteristics good prediction effect is arranged all.And combine in time reliable transient stability criterion, so that combine with existing wide area measurement system (WAMS) effectively, reach the purpose of electrical power system transient unstability real time discriminating and early warning.
Summary of the invention
Technical matters to be solved by this invention is to overcome existing merit angle trajectory predictions method can not have good universality again when guaranteeing precision of prediction deficiency; Provide a kind of based on generator's power and angle Forecasting Methodology based on grey Verhulst model; It is simple to have algorithm; Precision is high, the long and good characteristics of robustness of predicted time.
Thinking of the present invention is to adopt no inclined to one side grey Verhulst model to carry out the generator's power and angle prediction.Gray model is proposed in nineteen eighty-two by people such as Deng Julong the earliest, through generation, the exploitation to Given information, extracts valuable information, realizes correct description and effective monitoring to the following operation action of system, evolution rule.Verhulst mainly predicts towards approximate S type delta data (increase progressively earlier again and successively decrease).The no inclined to one side grey Verhulst model that was proposed by people such as Wang Zhengxin in 2009 has been eliminated self inherent variability on the master mould basis, accuracy is higher, and universality is better.
Particularly, the present invention adopts following technical scheme:
A kind of generator's power and angle Forecasting Methodology based on grey Verhulst model; Be used to possess the network system of wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time, it is characterized in that this method comprises:
Steps A, the generator's power and angle measurement data of extraction phasor measurement unit in the certain hour window obtain the generator's power and angle time series in this time window;
Step B, the generator's power and angle time series
that obtains with step 1 are as input; Generator's power and angle
according to following grey Verhulst model moment to
is predicted;
is the initial moment of sampling;
is number of samples,
be integer greater than
:
；
Wherein, Gray model parameter
; Through least square method following formula is found the solution and to obtain
In the formula,
.
In order to improve precision of prediction, further, this method also comprises:
Step C, the predicted value
that step B is obtained according to following formula are carried out error compensation, obtain final predicted numerical value
:
Wherein, constantly predicated error that
is
; In the formula;
is the linear coefficient parameter, obtains through computes:
In the formula;
,
are respectively
,
predicated error constantly, obtain through following formula:
，
Preferably, the span of number of samples
is 5-10.
According to invention thinking of the present invention; Also can obtain a kind of network system transient state unstability method of discrimination based on grey Verhulst model; Said network system possesses wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time that this method comprises:
Step 1, according to each generator's power and angle real-time measuring data of phasor measurement unit output, judge that whether network system has the disturbance generation, in this way, then changes step 2;
Step 2, the generator's power and angle time series in back certain hour window takes place in disturbance, exported with phasor measurement unit are as input; Adopt claim 1,2 or 3 said generator's power and angle Forecasting Methodologies that the following merit angle constantly of each generator is predicted, obtain following each generator's power and angle track constantly based on grey Verhulst model;
Step 3, according to following each generator's power and angle track constantly that step 2 obtains, ask for the state parameter of each generator and network system, and judge whether state parameter exceeds predefined normal range, in this way, then be judged to be network system transient state unstability.
Preferably, said each the generator's power and angle real-time measuring data according to phasor measurement unit output of step 1 judges whether network system has disturbance to take place, specifically according to following method:
Gather each generator's power and angle of phasor measurement unit output in real time; Continuous moment angular velocity numerical value is carried out calculus of differences; Obtain angular acceleration time series
, and judge whether following formula is set up, in this way; Judge that then network system has disturbance to take place
In the formula;
is preset sudden change threshold coefficient, and its numerical value is greater than 1;
expression begins the angular acceleration in
individual sampling period from sampling the initial moment
.
The preferred 3-5 of span of said sudden change threshold coefficient
.
Preferably, state parameter described in the step 3 comprises at least a in following each state parameter: angular velocity, inertia off-centring angle, poor, the homology group clearance angle of maximum angle; The calculating of each state parameter and corresponding judgment method are following:
(1) angular velocity:
Following angle track of each generator that prediction is obtained carries out difference, and according to computes angular velocity numerical value:
，
In the formula, subscript
is each generator numbering in the network system;
Judge that whether prediction moment angular velocity satisfies following formula, in this way, then is judged to be network system transient state unstability:
，
In the formula;
is the platform number of generator in the network system;
is the inertia of
platform generator, and
is given angular velocity threshold value;
(2) inertia off-centring angle:
Inertia central angle
according to the computes network system:
In the formula;
is the platform number of generator in the network system, and
is the inertia of
platform generator;
When the inertia off-centring angle of any generator
when satisfying following formula, be judged to be network system transient state unstability:
?，
In the formula,
is the threshold value at predefined inertia off-centring angle;
(3) maximum work angular difference:
Find predicted value minimum and maximum generator in merit angle in the prediction moment network system; When the maximum work angular difference; Poor
of merit angle predicted value that is these two generators then is judged to be network system transient state unstability when satisfying following formula:
In the formula;
and
is respectively the prediction constantly maximal value and the minimum value of generator's power and angle,
be the threshold value of the maximum work angular difference that sets.
(4) homology group clearance angle:
Composite power-angle according to each generator of the computes disturbance moment:
In the formula; Generator's power and angle constantly takes place for disturbance in
; Generator imbalance power constantly takes place for disturbance in
;
is the inertia time constant of generator, and
is the time set value;
The difference computing of adjacent composite power-angle is carried out according to following formula in composite power-angle ordering back, obtains clearance angle
:
In the formula,
;
The clearance angle that aforementioned calculation is obtained carries out sorting operation again, gets place, angle, maximal clearance as hiving off the interval, and the operation of hiving off is divided into the leading crowd and the crowd that lags behind with the generator in the network system;
Carry out the unit polymerization to two groups respectively, the merging formula is:
In the formula;
and
is respectively the leading crowd and the crowd's that lags behind equivalent unit merit angle;
and
is respectively the leading crowd and the crowd's that lags behind generator set; For the inertia of
platform generator among the leading crowd,
is the inertia of
platform generator among the crowd that lags behind
;
Judge the leading crowd and the crowd's that lags behind homology group clearance angle, whether the promptly leading crowd and the crowd's that lags behind equivalent unit merit angle poor satisfies following formula, in this way, then is judged to be network system transient state unstability:
In the formula,
is preset homology group clearance angle threshold value.
Further, this method also comprises:
Step 4, when judging network system transient state unstability, carry out early warning and start the stabilization control device in the network system.Thereby can trigger stable control measure such as cutter, cutting load as early as possible according to result of determination, avoid system sectionalizing or large-area power-cuts.
The thinking according to the present invention; Also can obtain a kind of network system transient state unstability discriminating gear; Said network system possesses wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time that said transient state unstability discriminating gear is connected with said phasor measurement unit signal; Said transient state unstability discriminating gear comprises:
The disturbance identification module; According to each generator's power and angle real-time measuring data of phasor measurement unit output, judge whether network system has disturbance to take place, specifically according to following method: each generator's power and angle of gathering phasor measurement unit output in real time; Continuous moment angular velocity numerical value is carried out calculus of differences; Obtain angular acceleration time series
, and judge whether following formula is set up, in this way; Judge that then network system has disturbance to take place
Where,
is the default threshold factor mutation, its value is greater than 1;
said that since the sampling start time
start first
sampling periods angular acceleration;
The trajectory predictions module; When the disturbance identification module judges that network system has disturbance to take place; The generator's power and angle time series of in back certain hour window takes place in disturbance, exporting with phasor measurement unit is as input; Adopt above-mentioned generator's power and angle Forecasting Methodology that the following merit angle constantly of each generator is predicted, obtain following each generator's power and angle track constantly based on grey Verhulst model;
The transient stability discrimination module; According to following each generator's power and angle track constantly that the trajectory predictions module obtains, ask for the state parameter of each generator and network system, and judge whether state parameter exceeds predefined normal range; In this way, then be judged to be network system transient state unstability;
Transient state early warning module when the transient stability discrimination module is judged network system transient state unstability, is carried out early warning and is started the stabilization control device in the network system.
The present invention has improved the accuracy that predicts the outcome through grey Verhulst model being introduced the merit angle prediction of generator; And confirm compensating error through gathering two up-to-date data, predicted value is revised, thereby further improved precision of prediction.Compare prior art, the present invention has the following advantages:
1. do not have inclined to one side grey Verhulst model and be applicable to the merit angle track fast prediction that comprises under single pendulum unstability, plurality of pendulums unstability, the plurality of pendulums multiple situation in being stabilized in, compare classic method and have better universality.
2. after carrying out error compensation through the last samples data, precision of prediction of the present invention further improves, and it is better to compare the classic method prediction effect, more can accurately reflect the transient characterisitics of system.
3. the present invention is used for the early warning of transient state unstability, and required measurement is few, and calculated amount is little, and is high, simple and reliable for the discrimination of transient state unstability.
Description of drawings
Fig. 1 is the disturbed trajectory predictions principle schematic of generator's power and angle;
Fig. 2 is the overall operation structural drawing of network system transient state unstability discriminating gear of the present invention;
Fig. 3 is the workflow diagram of disturbance identification module in the network system transient state unstability discriminating gear of the present invention;
Fig. 4 is the workflow diagram of trajectory predictions module in the network system transient state unstability discriminating gear of the present invention;
Fig. 5 is the workflow diagram of transient state unstability discrimination module in the network system transient state unstability discriminating gear of the present invention;
Fig. 6 is that the inventive method is applied to IEEE39 node system prediction effect figure (stablizing example);
Fig. 7 is that the inventive method is applied to IEEE39 node system prediction effect figure (unstability example).
Embodiment
Below in conjunction with accompanying drawing technical scheme of the present invention is elaborated:
The stable control technology of the incident that is based on that adopt in conventional defence line, electric system second road; Its advantage is with strong points, reliable; Exceeded the incident scope that it is contained if maximum shortcoming is phylogenetic fault disturbance, it just can not take control measure to avoid system's occurrence of large-area power outage.The present invention is based on the real-time generator angle information of the wide area measurement of PMU/WAMS; Put forth effort to solve system's transient state unstability that the big disturbance beyond the range of control of defence line, two roads causes; Adopt ultra real-time estimate technology recognition system transient state unstability; Trigger stable control measure such as cutter, cutting load as early as possible, avoid system sectionalizing or large-area power-cuts.
Fig. 1 has provided the synoptic diagram of the disturbed trajectory predictions in merit angle.Disturbed trajectory predictions through setting up specific mathematical model, is estimated the merit angle track of (promptly predicting window) in following a period of time through the historical data in the time window of research (being observation window).The key of prediction is that a kind of algorithm of searching is simple, and precision is high, the long and good forecast model of robustness of predicted time.The present invention adopts improved no inclined to one side grey Verhulst model, and metric data is few, calculates simply, for the prediction of plurality of pendulums process medium wave peak and trough very high precision is arranged.
The overall operation structure of network system transient state unstability discriminating gear of the present invention is as shown in Figure 2, and this device comprises:
The disturbance identification module according to each generator's power and angle real-time measuring data of phasor measurement unit output, judges whether network system has disturbance to take place;
The trajectory predictions module; When the disturbance identification module judges that network system has disturbance to take place; The generator's power and angle time series of in back certain hour window takes place in disturbance, exporting with phasor measurement unit is as input; Adopt above-mentioned generator's power and angle Forecasting Methodology that the following merit angle constantly of each generator is predicted, obtain following each generator's power and angle track constantly based on grey Verhulst model;
The transient stability discrimination module; According to following each generator's power and angle track constantly that the trajectory predictions module obtains, ask for the state parameter of each generator and network system, and judge whether state parameter exceeds predefined normal range; In this way, then be judged to be network system transient state unstability;
Transient state early warning module when the transient stability discrimination module is judged network system transient state unstability, is carried out early warning and is started the stabilization control device in the network system.
Below, according to the overall operation structural drawing of Fig. 2, the network system transient state unstability method of discrimination based on grey Verhulst model of the present invention is carried out detailed explanation in conjunction with the course of work of this device.
(1) utilize phasor measurement unit (PMU) to gather each generator data in real time, with synchronous merit angle information input disturbance identification module.The disturbance identification module carries out calculus of differences to the continuous moment angular velocity numerical value of each generator, obtains angular acceleration
:
If saltus step takes place in angular acceleration time series
; Then illustrative system receives big disturbing influence, and the power-balance of generator is destroyed.Used criterion is:
In the formula:
is sudden change threshold coefficient, and the preferred value of the present invention is 3-5.
The workflow of disturbance identification module is as shown in Figure 3.
(2) if disturbance identification module decision-making system receives big disturbance, then get into the trajectory predictions module.
With generator amature angle measurements time series after the disturbance
as the gray model list entries.
is the initial moment of sampling;
is number of samples, and preferred span is 5-10;
is the sampling period, and value is 0.01s.
Do not have inclined to one side grey Verhulst model separate for:
In the formula:
is the gray model parameter;
is the prediction step number from the initial moment meter of sampling,
.
Parameter vector
, find the solution through least square method:
In the formula:
.
Wait for two sampling periods; Gather the merit angle numerical value in up-to-date two moment of typing through PMU, predicated error
is:
In the formula:
is actual measurement merit angle numerical value, and
is tentative prediction merit angle numerical value.
Error to the moment in future is carried out Linear Estimation:
In the formula:
is the linear coefficient parameter.
Obtaining final predicted numerical value
is:
The prediction algorithm flow process of trajectory predictions module is as shown in Figure 4.
(3) on the basis of the ultra real-time estimate of generator angle, call the transient stability discrimination module.According to prediction merit angle numerical value, calculate following state parameter, the analytic system transient stability.
1. angular velocity:
The following angle track of each generator that prediction is obtained carries out difference, estimated angular velocity numerical value.
In the formula:
is each generator numbering of the whole network;
Whether angular velocity is out-of-limit constantly to differentiate prediction according to following formula:
In the formula;
is the platform number of generator in the network system;
is the inertia of
platform generator, and
is given angular velocity threshold value.
2. inertia off-centring angle:
After control center obtains the generator's power and angle predicted data, carry out the estimation of system inertia central angle
according to following formula:
?，
In the formula;
is the platform number of generator in the network system, and
is the inertia of
platform generator;
When the merit angle of any generator and the deviation of inertia central angle, promptly inertia off-centring angle when surpassing threshold value, is judged out-of-limit.
In the formula,
is for setting the threshold value of inertia off-centring.
3. maximum work angular difference:
Through to prediction constantly each generator's power and angle numerical value carry out size ordering, find merit angle value maximum generation machine and merit angle to be worth minimum generator, when the merit angular difference of these two generators during, judge out-of-limitly above threshold value, its expression formula is following:
In the formula;
is worth at the maximum work angle for prediction constantly;
for prediction least work angle value constantly,
be the threshold value of the maximum work angular difference that sets.
4. homology group clearance angle threshold:
Fault takes place to utilize angular velocity and imbalance power constantly, can estimate receiving the generator's power and angle after the disturbance, calculates the composite power-angle of each generator.The computes disturbance is the composite power-angle of each generator constantly:
In the formula; Generator's power and angle constantly takes place for disturbance in
; Generator imbalance power constantly takes place for disturbance in
;
is the inertia time constant of generator;
is the time set value, and preferred value is 100 milliseconds;
The difference computing of adjacent composite power-angle is carried out according to following formula in composite power-angle ordering back, obtains clearance angle
:
In the formula,
;
The clearance angle that aforementioned calculation is obtained carries out sorting operation again, gets place, angle, maximal clearance as hiving off the interval, and the operation of hiving off is divided into the leading crowd and the crowd that lags behind with the generator in the network system:
If the angle, maximal clearance is
; Composite power-angle is the leading crowd of conduct
who hives off more than the interval; That is: if satisfy
, then generator
belongs to
crowd; The conduct hysteresis crowd
who hives off below the interval; That is: if satisfy
, then generator
belongs to
crowd;
Carry out the unit polymerization to two groups respectively, the merging formula is:
In the formula,
and
is respectively A crowd and S crowd's equivalent unit merit angle.
Whether poor (being the homology group clearance angle) at equivalent unit merit angle of judging two groups according to following formula be out-of-limit:
In the formula,
is homology group clearance angle threshold value.
As
and
when satisfying following formula, judge out-of-limit.
Above-mentioned one of four states threshold value criterion if there is wherein any one to surmount its scleronomic constraint, then is regarded as system's transient state unstability.
The workflow of transient state unstability discrimination module is as shown in Figure 5.If this module output result shows the transient state unstability, then call transient state early warning module, send early warning and carry out cutter or the cutting load operation.
Result verification:, use the inventive method IEEE39 node system different faults situation has been carried out simulating, verifying in order to test the accuracy and the validity of institute of the present invention extracting method.
Example 1: it is the head end generation three-phase shortcircuit ground connection of circuit 26-29 that fault is set, and 0 moment fault is sent out, and 0.1s excises fault, promptly stablizes example.(stablizing example)
Example 2: same fault point and fault type are set, and the duration is 0-0.3s.(unstability example)
The observation generator choose 38 with the corresponding motor of 39 nodes, begin to carry out angle the finish time from fault and gather in real time, utilize 10 up-to-date merit angular datas to carry out asking for of model parameter, reform such as take to cease and carry out rolling forecast, the merit angle numerical value behind the calculating 0.25s.The present invention is directed to example 1,2 prediction effect and be drawn on the corresponding moment of Fig. 6 respectively with Fig. 7.Forecasting Methodology of the present invention, trigonometric function prediction, three kinds of forecast models of autoregression prediction are carried out the precision of prediction test, calculate its maximum error and average error, the result is as shown in table 1.
Each method prediction effect of table 1 IEEE39 node example relatively
On above-mentioned base of prediction, 180 ° of threshold value methods of merit angular difference of utilizing transient stability discrimination module of the present invention and industry member generally to adopt are done contrast.Through a large amount of calculated off-line, it is as shown in table 2 to set threshold value.
Table 2 transient stability is differentiated threshold value
| The angular velocity threshold value |
 |
183°/s |
| Inertia off-centring side door sill |
 |
191° |
| Maximum work angular difference threshold | 331° | |
| Homology group clearance angle threshold |
 |
197° |
In example 1 (stablizing example), both phenomenon all do not occur judging by accident.Example 2 (unstability example), through the ultra real-time estimate in merit angle, 180 ° of threshold value methods of merit angular difference are in 0.18s recognition system transient state unstability; The inventive method homology group clearance angle after 0.13s estimation obtains following 0.25s reaches 199.1162 °, surpasses corresponding threshold value, judgement system transient state unstability, and identification is accurately and more quick.
The present invention can be used for the dispatching control center of electrical networks at different levels, based on wide area measurement system, realizes real-time estimate and early warning to the electrical power system transient angle stability.
More than generator angle real-time estimate and the transient state unstability method of discrimination based on gray model provided by the present invention carried out detailed explanation.As far as one of ordinary skill in the art, any conspicuous change of under the prerequisite that does not deviate from connotation of the present invention, it being done all will constitute to infringement of patent right of the present invention, with corresponding legal responsibilities.
Claims (11)
1. generator's power and angle Forecasting Methodology based on grey Verhulst model; Be used to possess the network system of wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time; It is characterized in that this method comprises:
Steps A, the generator's power and angle measurement data of extraction phasor measurement unit in the certain hour window obtain the generator's power and angle time series in this time window;
Step B, the generator's power and angle time series
that obtains with step 1 are as input; Generator's power and angle
according to following grey Verhulst model moment to
is predicted;
is the initial moment of sampling;
is number of samples,
be integer greater than
:
；
Wherein, Gray model parameter
; Through least square method following formula is found the solution and to obtain
，
In the formula,
.
2. like the said generator's power and angle Forecasting Methodology of claim 2, it is characterized in that this method also comprises based on grey Verhulst model:
Step C, the predicted value
that step B is obtained according to following formula are carried out error compensation, obtain final predicted numerical value
:
?，
Wherein, constantly predicated error that
is
; In the formula;
is the linear coefficient parameter, obtains through computes:
In the formula;
,
are respectively
,
predicated error constantly, obtain through following formula:
3. according to claim 1 based on the generator's power and angle Forecasting Methodology of grey Verhulst model; It is characterized in that the span of number of samples
is 5-10.
4. network system transient state unstability method of discrimination based on grey Verhulst model; Said network system possesses wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time; It is characterized in that this method comprises:
Step 1, according to each generator's power and angle real-time measuring data of phasor measurement unit output, judge that whether network system has the disturbance generation, in this way, then changes step 2;
Step 2, the generator's power and angle time series in back certain hour window takes place in disturbance, exported with phasor measurement unit are as input; Adopt claim 1,2 or 3 said generator's power and angle Forecasting Methodologies that the following merit angle constantly of each generator is predicted, obtain following each generator's power and angle track constantly based on grey Verhulst model;
Step 3, according to following each generator's power and angle track constantly that step 2 obtains, ask for the state parameter of each generator and network system, and judge whether state parameter exceeds predefined normal range, in this way, then be judged to be network system transient state unstability.
5. like the said network system transient state unstability method of discrimination of claim 4 based on grey Verhulst model; It is characterized in that; Said each the generator's power and angle real-time measuring data according to phasor measurement unit output of step 1 judges whether network system has disturbance to take place, specifically according to following method:
Gather each generator's power and angle of phasor measurement unit output in real time; Continuous moment angular velocity numerical value is carried out calculus of differences; Obtain angular acceleration time series
; And judge whether following formula is set up; In this way, judge that then network system has disturbance to take place
Where,
mutation is the default threshold coefficient which is greater than 1;
said that since the sampling start time
start first
angular sampling period .
6. like the said network system transient state unstability method of discrimination of claim 5 based on grey Verhulst model; It is characterized in that the span of said sudden change threshold coefficient
is 3-5.
7. like the said network system transient state unstability method of discrimination of claim 4 based on grey Verhulst model; It is characterized in that state parameter described in the step 3 comprises at least a in following each state parameter: angular velocity, inertia off-centring angle, maximum work angular difference, homology group clearance angle; The calculating of each state parameter and corresponding judgment method are following:
(1) angular velocity:
Following angle track of each generator that prediction is obtained carries out difference, and according to computes angular velocity numerical value:
In the formula, subscript
is each generator numbering in the network system;
Judge that whether prediction moment angular velocity satisfies following formula, in this way, then is judged to be network system transient state unstability:
In the formula;
is the platform number of generator in the network system;
is the inertia of
platform generator, and
is given angular velocity threshold value;
(2) inertia off-centring angle:
Inertia central angle
according to the computes network system:
In the formula;
is the platform number of generator in the network system, and
is the inertia of
platform generator;
When the inertia off-centring angle of any generator
when satisfying following formula, be judged to be network system transient state unstability:
?，
In the formula,
is the threshold value at predefined inertia off-centring angle;
(3) maximum work angular difference:
Find predicted value minimum and maximum generator in merit angle in the prediction moment network system; When the maximum work angular difference; Poor
of merit angle predicted value that is these two generators then is judged to be network system transient state unstability when satisfying following formula:
?，
In the formula;
and
is respectively the prediction constantly maximal value and the minimum value of generator's power and angle,
be the threshold value of the maximum work angular difference that sets;
(4) homology group clearance angle:
Composite power-angle according to each generator of the computes disturbance moment:
In the formula; Generator's power and angle constantly takes place for disturbance in
; Generator imbalance power constantly takes place for disturbance in
;
is the inertia time constant of generator, and
is the time set value;
The difference computing of adjacent composite power-angle is carried out according to following formula in composite power-angle ordering back, obtains clearance angle
:
In the formula,
;
The clearance angle that aforementioned calculation is obtained carries out sorting operation again, gets place, angle, maximal clearance as hiving off the interval, and the operation of hiving off is divided into the leading crowd and the crowd that lags behind with the generator in the network system;
Carry out the unit polymerization to two groups respectively, the merging formula is:
?，
In the formula;
and
is respectively the leading crowd and the crowd's that lags behind equivalent unit merit angle;
and
is respectively the leading crowd and the crowd's that lags behind generator set; For the inertia of
platform generator among the leading crowd,
is the inertia of
platform generator among the crowd that lags behind
;
Judge that whether leading crowd and the crowd's that lags behind homology group clearance angle satisfy following formula, in this way, then are judged to be network system transient state unstability:
In the formula,
is preset homology group clearance angle threshold value.
8. like the said network system transient state unstability method of discrimination of claim 7 based on grey Verhulst model; It is characterized in that the value of said time set value
is 100 milliseconds.
9. like each said network system transient state unstability method of discrimination of claim 4-8, it is characterized in that this method also comprises based on grey Verhulst model:
Step 4, when judging network system transient state unstability, carry out early warning and start the stabilization control device in the network system.
10. network system transient state unstability discriminating gear; Said network system possesses wide area measurement system; Said wide area measurement system comprises and is used for phasor measurement unit that the merit angle of this each generator of network system is measured in real time; It is characterized in that said transient state unstability discriminating gear is connected with said phasor measurement unit signal; Said transient state unstability discriminating gear comprises:
The disturbance identification module; Each generator's power and angle real-time measuring data according to phasor measurement unit output; Judge whether network system has disturbance to take place; Specifically according to following method: gather each generator's power and angle of phasor measurement unit output in real time, continuous moment angular velocity numerical value is carried out calculus of differences, obtain angular acceleration time series
; And judge whether following formula is set up; In this way, judge that then network system has disturbance to take place
Where,
mutation is the default threshold coefficient which is greater than 1;
said that since the sampling start time
start first
angular sampling period ;
The trajectory predictions module; When the disturbance identification module judges that network system has disturbance to take place; The generator's power and angle time series of in back certain hour window takes place in disturbance, exporting with phasor measurement unit is as input; Adopt claim 1,2 or 3 said generator's power and angle Forecasting Methodologies that the following merit angle constantly of each generator is predicted, obtain following each generator's power and angle track constantly based on grey Verhulst model;
The transient stability discrimination module; According to following each generator's power and angle track constantly that the trajectory predictions module obtains, ask for the state parameter of each generator and network system, and judge whether state parameter exceeds predefined normal range; In this way, then be judged to be network system transient state unstability;
Transient state early warning module when the transient stability discrimination module is judged network system transient state unstability, is carried out early warning and is started the stabilization control device in the network system.
11., it is characterized in that said state parameter comprises at least a in following each state parameter: angular velocity, inertia off-centring angle, maximum work angular difference, homology group clearance angle like the said network system transient state of claim 10 unstability discriminating gear; The calculating of each state parameter and corresponding judgment method are following:
(1) angular velocity:
Following angle track of each generator that prediction is obtained carries out difference, and according to computes angular velocity numerical value:
In the formula, subscript
is each generator numbering in the network system;
Judge that whether prediction moment angular velocity satisfies following formula, in this way, then is judged to be network system transient state unstability:
In the formula;
is the platform number of generator in the network system;
is the inertia of
platform generator, and
is given angular velocity threshold value;
(2) inertia off-centring angle:
Inertia central angle
according to the computes network system:
?，
In the formula;
is the platform number of generator in the network system, and
is the inertia of
platform generator;
When the inertia off-centring angle of any generator
when satisfying following formula, be judged to be network system transient state unstability:
In the formula,
is the threshold value at predefined inertia off-centring angle;
(3) maximum work angular difference:
Find predicted value minimum and maximum generator in merit angle in the prediction moment network system; When the maximum work angular difference; Poor
of merit angle predicted value that is these two generators then is judged to be network system transient state unstability when satisfying following formula:
In the formula;
and
is respectively the prediction constantly maximal value and the minimum value of generator's power and angle,
be the threshold value of the maximum work angular difference that sets;
(4) homology group clearance angle:
Composite power-angle according to each generator of the computes disturbance moment:
In the formula; Generator's power and angle constantly takes place for disturbance in
; Generator imbalance power constantly takes place for disturbance in
;
is the inertia time constant of generator, and
is the time set value;
The difference computing of adjacent composite power-angle is carried out according to following formula in composite power-angle ordering back, obtains clearance angle
:
In the formula,
;
The clearance angle that aforementioned calculation is obtained carries out sorting operation again, gets place, angle, maximal clearance as hiving off the interval, and the operation of hiving off is divided into the leading crowd and the crowd that lags behind with the generator in the network system;
Carry out the unit polymerization to two groups respectively, the merging formula is:
?，
In the formula;
and
is respectively the leading crowd and the crowd's that lags behind equivalent unit merit angle;
and
is respectively the leading crowd and the crowd's that lags behind generator set; For the inertia of
platform generator among the leading crowd,
is the inertia of
platform generator among the crowd that lags behind
;
Judge the leading crowd and the crowd's that lags behind homology group clearance angle, whether the promptly leading crowd and the crowd's that lags behind equivalent unit merit angle poor satisfies following formula, in this way, then is judged to be network system transient state unstability:
?，
In the formula,
is preset homology group clearance angle threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110295073 CN102445660B (en) | 2011-09-27 | 2011-09-27 | Gray Verhulst model-based prediction method of power angle of generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110295073 CN102445660B (en) | 2011-09-27 | 2011-09-27 | Gray Verhulst model-based prediction method of power angle of generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102445660A true CN102445660A (en) | 2012-05-09 |
| CN102445660B CN102445660B (en) | 2013-09-04 |
Family
ID=46008349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110295073 Expired - Fee Related CN102445660B (en) | 2011-09-27 | 2011-09-27 | Gray Verhulst model-based prediction method of power angle of generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102445660B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103199551A (en) * | 2013-03-13 | 2013-07-10 | 南方电网科学研究院有限责任公司 | Stationing and parameter optimization method of syntony-type fault current limiter |
| CN103995948A (en) * | 2013-07-19 | 2014-08-20 | 南方电网科学研究院有限责任公司 | Oscillation center voltage prediction method based on polynomial model |
| CN106099949A (en) * | 2016-05-15 | 2016-11-09 | 国电南瑞科技股份有限公司 | A kind of improve off-the-line after the islanding control of partition power grid transient stability determine method opportunity |
| CN107918808A (en) * | 2017-11-16 | 2018-04-17 | 淮阴工学院 | Grey forecasting model method based on inverse cosine function and power function transformation and background value Combinatorial Optimization |
| CN108665090A (en) * | 2018-04-06 | 2018-10-16 | 东北电力大学 | Urban distribution network saturation load forecasting method based on principal component analysis Yu Verhulst models |
| CN109073374A (en) * | 2016-04-20 | 2018-12-21 | 赛峰电子与防务公司 | Method for estimating the clearance in electromechanical actuator |
| CN109325880A (en) * | 2018-09-21 | 2019-02-12 | 国网天津市电力公司 | A kind of Mid-long term load forecasting method based on Verhulst-SVM |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7277779B2 (en) * | 2003-06-21 | 2007-10-02 | Abb Research Ltd | Real-time emergency control in power systems |
| CN101540495A (en) * | 2009-04-29 | 2009-09-23 | 西安交通大学 | Loss of synchronism protection system and method based on geometrical characteristics of movement locus of generator |
| CN101794998A (en) * | 2010-02-05 | 2010-08-04 | 湖南大学 | Online transient stability analysis method based on concise expression form of electromagnetic power of single generator in multi-machine power system |
| TW201101209A (en) * | 2009-06-30 | 2011-01-01 | Kun-Li Wen | Grey human-computer interface forecasting system |
| CN102156261A (en) * | 2011-03-23 | 2011-08-17 | 西安交通大学 | Device and method for predicating angle on basis of high-order state information of generator |
-
2011
- 2011-09-27 CN CN 201110295073 patent/CN102445660B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7277779B2 (en) * | 2003-06-21 | 2007-10-02 | Abb Research Ltd | Real-time emergency control in power systems |
| CN101540495A (en) * | 2009-04-29 | 2009-09-23 | 西安交通大学 | Loss of synchronism protection system and method based on geometrical characteristics of movement locus of generator |
| TW201101209A (en) * | 2009-06-30 | 2011-01-01 | Kun-Li Wen | Grey human-computer interface forecasting system |
| CN101794998A (en) * | 2010-02-05 | 2010-08-04 | 湖南大学 | Online transient stability analysis method based on concise expression form of electromagnetic power of single generator in multi-machine power system |
| CN102156261A (en) * | 2011-03-23 | 2011-08-17 | 西安交通大学 | Device and method for predicating angle on basis of high-order state information of generator |
Non-Patent Citations (3)
| Title |
|---|
| 刘兆燕: "基于广域测量信号的电力系统暂态稳定预测及时滞稳定域研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》, no. 07, 15 July 2009 (2009-07-15) * |
| 刘兆燕等: "基于广域测量系统的快速暂态稳定预测方法", 《电力系统自动化》, vol. 31, no. 21, 10 November 2007 (2007-11-10), pages 1 - 4 * |
| 王正新等: "无偏灰色Verhulst模型及其应用", 《系统工程理论与实践》, vol. 29, no. 10, 31 October 2009 (2009-10-31) * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103199551A (en) * | 2013-03-13 | 2013-07-10 | 南方电网科学研究院有限责任公司 | Stationing and parameter optimization method of syntony-type fault current limiter |
| CN103199551B (en) * | 2013-03-13 | 2016-08-31 | 南方电网科学研究院有限责任公司 | Layouting and parameter optimization method of a kind of mode of resonance fault current limiter |
| CN103995948A (en) * | 2013-07-19 | 2014-08-20 | 南方电网科学研究院有限责任公司 | Oscillation center voltage prediction method based on polynomial model |
| CN103995948B (en) * | 2013-07-19 | 2017-11-24 | 南方电网科学研究院有限责任公司 | A kind of voltage of oscillation center Forecasting Methodology based on multinomial model |
| CN109073374A (en) * | 2016-04-20 | 2018-12-21 | 赛峰电子与防务公司 | Method for estimating the clearance in electromechanical actuator |
| CN109073374B (en) * | 2016-04-20 | 2020-09-29 | 赛峰电子与防务公司 | Method for estimating play in an electromechanical actuator |
| CN106099949A (en) * | 2016-05-15 | 2016-11-09 | 国电南瑞科技股份有限公司 | A kind of improve off-the-line after the islanding control of partition power grid transient stability determine method opportunity |
| CN106099949B (en) * | 2016-05-15 | 2019-07-05 | 国电南瑞科技股份有限公司 | The islanding control opportunity of partition power grid transient stability determines method after a kind of raising off-the-line |
| CN107918808A (en) * | 2017-11-16 | 2018-04-17 | 淮阴工学院 | Grey forecasting model method based on inverse cosine function and power function transformation and background value Combinatorial Optimization |
| CN108665090A (en) * | 2018-04-06 | 2018-10-16 | 东北电力大学 | Urban distribution network saturation load forecasting method based on principal component analysis Yu Verhulst models |
| CN108665090B (en) * | 2018-04-06 | 2021-06-18 | 东北电力大学 | Urban power grid saturation load prediction method based on principal component analysis and Verhulst model |
| CN109325880A (en) * | 2018-09-21 | 2019-02-12 | 国网天津市电力公司 | A kind of Mid-long term load forecasting method based on Verhulst-SVM |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102445660B (en) | 2013-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102445660A (en) | Gray Verhulst model-based prediction method of power angle of generator | |
| CN103279639B (en) | Receiving end Network Voltage Stability overall process Situation Assessment based on response and preventing control method | |
| CN103746370B (en) | A kind of wind energy turbine set Reliability Modeling | |
| CN101794998A (en) | Online transient stability analysis method based on concise expression form of electromagnetic power of single generator in multi-machine power system | |
| EP3200316B1 (en) | Electric power system monitoring device and electric power system monitoring system | |
| CN102510071B (en) | Power grid system emergency control method and device | |
| CN102163844B (en) | Method for detecting state of power system based on phasor measurement unit (PMU) | |
| CN108092304B (en) | Electric power system transient stability discrimination method based on phase track and relative kinetic energy change rate | |
| CN103475010B (en) | Transient stabilization control method for interconnection electrical network after impact disturbance | |
| CN103995948B (en) | A kind of voltage of oscillation center Forecasting Methodology based on multinomial model | |
| CN108092319A (en) | A kind of Uncertainty Analysis Method and device of short-term wind-electricity power prediction | |
| CN106229976B (en) | Transient rotor angle stability situation predictor method based on data-driven | |
| CN101917006A (en) | Frequency safety online monitoring and assessment method of power system | |
| CN103618307A (en) | Emergency control method for improving safe stability of power system | |
| CN101819243B (en) | Area information based quiescent voltage unstability prediction method | |
| CN102723710B (en) | Critical cluster identifying method for transient stability analysis of power system | |
| CN102522759B (en) | Method for distinguishing transient state instability of electric power system in real time based on voltage track after disturbance | |
| CN104505827B (en) | Complicated electric power system closed loop control method based on response message | |
| CN107749631B (en) | Unit primary frequency modulation capability pre-estimating method | |
| CN102298115B (en) | Method for online monitoring and analyzing adjustment performance of excitation system based on phasor measurement technology | |
| CN104809512B (en) | A kind of fast evaluation method of electric system difference emergent control measure performance indicator | |
| Deng et al. | A real-time generator-angle prediction method based on the modified grey Verhulst model | |
| CN103488891A (en) | Large grid safe stable prevention and control performance penalty index calculation method based on risks | |
| CN108092272A (en) | A kind of voltage stabilization on-line monitoring method based on the Kalman filtering that fades | |
| CN105896571B (en) | Transient stability based on section interconnection energy predicting promptly cuts machine control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: HUAIAN POWER SUPPLY COMPANY OF JIANGSU ELECTRIC PO Effective date: 20130731 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20130731 Address after: 211000 No. 8 West Buddha Road, Jiangning Development Zone, Jiangsu, Nanjing, China Applicant after: HOHAI University Applicant after: HUAIAN POWER SUPPLY COMPANY OF JIANGSU PROVINCE POWER Co. Applicant after: JIANGSU ELECTRIC POWER Co. Applicant after: State Grid Corporation of China Address before: 211000 No. 8 West Buddha Road, Jiangning Development Zone, Jiangsu, Nanjing, China Applicant before: Hohai University |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130904 Termination date: 20210927 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |