CN111582557A - Wind power climbing event multi-stage early warning method based on variation function - Google Patents
Wind power climbing event multi-stage early warning method based on variation function Download PDFInfo
- Publication number
- CN111582557A CN111582557A CN202010313374.4A CN202010313374A CN111582557A CN 111582557 A CN111582557 A CN 111582557A CN 202010313374 A CN202010313374 A CN 202010313374A CN 111582557 A CN111582557 A CN 111582557A
- Authority
- CN
- China
- Prior art keywords
- wind power
- climbing event
- time
- early warning
- delta
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000009194 climbing Effects 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013528 artificial neural network Methods 0.000 claims abstract description 12
- 238000010219 correlation analysis Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 210000001061 forehead Anatomy 0.000 claims description 2
- 230000007547 defect Effects 0.000 abstract description 3
- 238000011160 research Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for depicting a wind power climbing event, in particular to a wind power climbing event multi-stage early warning method based on a variation function, which specifically comprises the following steps: firstly, a corresponding relative wind power time sequence is obtained by using the definition of relative wind power. And describing the relative wind power time sequence by using the square root of the variation function, and performing autocorrelation analysis on the obtained wind power real-time change rate time sequence. And according to the analysis result, establishing a prediction model by using a BP neural network, and predicting the sequence. And selecting a multi-level threshold value of the wind power climbing event in a variation depicting mode, and identifying a prediction result of the BP neural network by using the threshold value to realize multi-level early warning of the wind power climbing event. The method makes up the defect that only the first and last points of power are considered in the traditional wind power climbing event definition; the change conditions of all time points in a period of time are comprehensively considered, and the climbing event is more accurately defined and identified.
Description
Technical Field
The invention relates to a method for depicting a wind power climbing event, in particular to a wind power climbing event multi-stage early warning method based on a variation function depicting mode, and the predictability of the prediction of the wind power climbing event is improved.
Background
The uncertain understanding and mastering of wind power is a key basic problem for promoting safe and efficient consumption of large-scale renewable energy and structure transformation of clean energy in China. The power climbing event caused by long-time limit weather is an important expression of wind power uncertainty; although the occurrence probability is small, the prediction and control are difficult, and the method can cause great influence and harm to the safe and stable operation, scheduling planning and real-time control of the power grid. Therefore, it is necessary to perform the prediction work of the wind power climbing event.
At present, the early warning research results of wind power climbing events are more, and Zhangying and the like review and prospect are carried out on the wind power climbing event research; it is a common form to characterize a hill climbing event by the absolute value of the power difference between the first and last time of the time period being greater than a threshold. Ruan Rui et al propose a wind power generation power prediction based on equivalent average wind speed, which carries out mean value processing on the change values at the first and last moments; the combined prediction method based on atom sparse decomposition and back propagation neural network is provided by the tomimen and the like, and the prediction accuracy is improved. However, in the above research, only the first and last time values of a certain period of time are considered for defining and describing the climbing event, the change conditions of all time points within a certain period of time are not considered, and the predictability of the time series obtained by only considering the first and last time change values is poor.
In practice, a wind power climbing event is a random occurrence event, and even a serious wind power climbing event can occur in a short time, so that only a difference value at the first moment and the last moment is selected to define after a fixed time period is set, and all wind power climbing events cannot be considered, so that how to describe the wind power climbing event is more accurate and how to improve the predictability of the climbing event is a problem worthy of deep research.
Disclosure of Invention
In order to improve the predictability of the wind power climbing event and increase the predictable length of the wind power climbing event, the invention provides a wind power climbing event multi-stage early warning method based on a variation function. The method comprises the following steps:
the method comprises the following steps: acquiring a relative wind power curve by utilizing the output power curve of the wind power plant and the corresponding load curve of the power grid;
step two: on the basis of using the definition of an absolute value of the relative wind power variation, quantitatively depicting a wind power climbing event by using a square root of a variation function, and acquiring a time series curve of the real-time variation rate of the wind power after variation;
step three; after the time sequence of the real-time change rate of the wind power is obtained, the self-correlation analysis is carried out on the time sequence, and the predictability of the time sequence is observed;
step four: establishing a predictive model of the wind power climbing event based on the wind power real-time change rate time sequence by utilizing a BP neural network algorithm;
step five: and defining a threshold selecting method of the wind power climbing event according to the absolute value of the relative wind power variation to obtain a multi-level threshold in a variation depicting mode, and effectively identifying a prediction result of the BP neural network by using the threshold to realize multi-level early warning of the wind power climbing event.
The invention has the advantages that:
1) the wind power climbing event is described by using the variation, the predictable length for describing the wind power state sequence is increased, and the predictability of the wind power climbing event is improved.
2) The defect that only the first and last points of power are considered in the traditional wind power climbing event definition is made up; the change conditions of all time points in a period of time are comprehensively considered, and the climbing event is more accurately defined and identified.
3) By utilizing a method of dividing a plurality of thresholds, the multi-stage early warning suitable for climbing of different degrees is realized.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a relative wind power curve;
FIG. 3 is a real-time variation rate curve of wind power;
FIG. 4 is a diagram of the results of the autocorrelation analysis of the time series of the real-time variation rate of the wind power;
FIG. 5 is a partial enlarged view of the autocorrelation analysis result of the time series of the real-time variation rate of the wind power;
FIG. 6 is a diagram of the result of the time-series autocorrelation analysis of the absolute value of the relative wind power variation;
FIG. 7 is a partial enlarged view of the result of the autocorrelation analysis of the absolute time series of the relative wind power variation;
FIG. 8 is a prediction result of the BP neural network model;
fig. 9 shows a wind power climbing event prediction situation after threshold division.
Detailed Description
The first embodiment is as follows:
the wind power climbing event multilevel early warning idea based on the variation function in the embodiment is as follows:
firstly, a corresponding relative wind power time sequence is obtained by using the definition of relative wind power. And describing the relative wind power time sequence by using the square root of the variation function, and performing autocorrelation analysis on the obtained wind power real-time change rate time sequence. And according to the analysis result, establishing a prediction model by using a BP neural network, and predicting the sequence. And selecting a multi-level threshold value of the wind power climbing event in a variation depicting mode, and identifying a prediction result of the BP neural network by using the threshold value to realize multi-level early warning of the wind power climbing event.
The method comprises the following steps: acquiring a relative wind power curve by utilizing the output power curve of the wind power plant and the corresponding load curve of the power grid;
step two: on the basis of using the definition of the absolute value of the variation relative to the wind power, quantitatively depicting the wind power climbing event by using the square root of the variation function, and acquiring a time series curve of the real-time variation rate of the wind power after variation;
step three; after the time sequence of the real-time change rate of the wind power is obtained, the self-correlation analysis is carried out on the time sequence, and the predictability of the time sequence is observed;
step four: establishing a predictive model of the wind power climbing event based on the wind power real-time change rate time sequence by utilizing a BP neural network algorithm;
step five: and defining a threshold selecting method of the climbing event according to the absolute value of the relative wind power variation to obtain a multi-level threshold in a variation depicting mode, and effectively identifying a prediction result of the BP neural network by using the threshold to realize multi-level early warning of the wind power climbing event.
The second embodiment is as follows:
the embodiment further defines the wind power climbing event multi-stage early warning method based on the variation function in the first specific embodiment.
The prediction of the wind power climbing event mainly prevents the wind power climbing event from threatening the stable operation of a power grid. Therefore, a relative wind power curve is selected to predict the wind power climbing event.
The method comprises the following steps: setting the time sequence of the output power curve of the wind power plant as P (t), the time sequence of the load curve of the power grid as Q (t), and setting the time sequence P' (t) relative to the wind power curve as:
P'(t)＝P(t)-Q(t) (1)
the positive and negative peak regulation characteristics of wind power are fully considered relative to the wind power. When the wind power change trend is consistent with the load demand of the power grid, even larger wind power output power change is beneficial to the power grid; when the wind power change trend is opposite to the load demand of the power grid, even smaller wind power output power is harmful to the power grid, so the wind power climbing event using the relative wind power is more accurate.
The invention uses the wind power data of one month of a power plant in Changchun to calculate the relative wind power. The sampling interval of the data points is 1min, and fig. 2 is a relative wind power curve obtained after calculation.
The third concrete implementation mode:
the embodiment further defines the wind power climbing event multi-stage early warning method based on the variation function in the first specific embodiment.
Step two, firstly: the wind power climbing event is defined by using the absolute value of the relative wind power variation, as shown in a formula (2):
|P'(t+Δt)-P'(t)|＞Pval(2)
the change value of the relative wind power represents the difference value of the relative wind power at the first moment and the last moment in the time period from t to t + delta t. When the absolute value of the difference is larger than the set threshold value PvalWhen the wind power climbing event occurs, the wind power climbing event is considered to occur; and when the absolute value of the difference value at the first moment and the last moment is smaller than the set threshold value, the wind power climbing event is not considered to occur.
Step two: the variation function is a tool for measuring the variation of the attribute parameters of the spatial variation along with the distance variation, and can reflect the spatial characteristics of the regionalized variables.
Assuming that P' (t) represents a wind power time series, the variation function of the wind power can be expressed as:
wherein, P' (t)i) Represents the value of the ith point in the time series of the relative wind power curve, P' (t)i+ Δ t) represents the value of the i + Δ t point in the time series of the relative wind power curve, Δ t represents the time interval, and N (Δ t) represents the total amount of data points in the Δ t time.
The variation function refers to the degree of variation of a spatial variable over a distance in a particular direction. Due to this feature of degradation, it can be applied to the time domain. According to the definition in the first step, when a climbing event occurs, the change value of the relative wind power is increased, and the change value of the relative wind power is increased accordingly. The variation alpha (Δ t) can thus be expressed as a hill climbing situation of the power system over a period of time. This value continues to increase when a hill climbing event occurs. The load climbing event can be quantitatively characterized by the parameter of the relative wind power variation alpha (delta t).
However, considering that the square operation in the variation function makes the curve change more steeply, and the predictability is reduced, the method performs the evolution processing on the basis of the variation function, and defines the obtained final variation square root as the real-time variation rate γ (Δ t) of the wind power, as shown in formula (4). The hill climbing event can be effectively identified by setting a reasonable threshold value.
The final description mode based on the variation function carries out square root operation and accumulation operation on the basis of the variation value operation, so that the description sequence is more gentle, and the predictability of the description sequence is improved. And the defect that only the first time point and the last time point can be selected in the change value operation is made up, so that the wind power climbing event is more accurately described.
The time interval delta t selected by the method is 15min, and the wind power climbing condition in the delta t period 15min in the future is predicted. And (3) obtaining a wind power real-time change rate time series curve based on the real relative wind power curve obtained in the step one, wherein the real relative wind power curve is shown in figure 3.
The fourth concrete implementation mode:
the embodiment further defines the wind power climbing event multi-stage early warning method based on the variation function in the first specific embodiment.
Before the actual forecasting is performed, the predictability of the time series needs to be studied first. Let { xt}t＝1:nIs a random time sequence, xtDelayed by k steps of sample xt+kThe autocorrelation coefficient between them indicates the degree of correlation of the two signals. The larger the autocorrelation coefficient is, the stronger dependence relationship exists between the two signals, and the rules hidden in the data can be mined by using a statistical method to realize the prediction of future data. The k steps at this time represent the autocorrelation length, i.e., the best modeling prediction length of the time series.
There are many methods for time series autocorrelation analysis, among which Pearson autocorrelation analysis is a more classical method whose basic principle is as follows:
let { xt}t＝1:nIs a random time sequence, then measure xtDelayed by k steps of sample xt+kThe autocorrelation coefficient ρ (k) of (a) is calculated as follows:
in practice, the random time sequence in the formula (6) is replaced by the actual time sequence in different depicting ways during the specific calculation, k is the delay step length, n is the length of the wind power time sequence in different depicting ways,
is the mean value, x, of the wind power time series under different depicting modest+kIs to time-sequence the original power xtAnd (3) analyzing the predictability of the wind power time sequence under different depicting modes according to the calculation result, wherein gamma (k) is the covariance after delaying k steps, gamma (0) is the covariance when the delay step k is equal to 0, and rho (k) is the autocorrelation coefficient obtained after delaying k steps. In general, the correlation coefficients and their corresponding degrees of correlation are shown in table 1 below:
TABLE 1 correlation coefficient and degree of correlation
In general, the autocorrelation coefficient tends to decrease as the autocorrelation length increases. Moreover, the autocorrelation coefficient is between 0.5 and 0.8, which indicates that strong correlation exists between data, so the threshold value of the autocorrelation coefficient is 0.6 in the invention. Strong dependence exists between signals with autocorrelation coefficients within an autocorrelation time range of more than 0.6, and the time scale is the optimal prediction length. Therefore, in the correlation analysis of the present invention, the calculated predictable length is based on the length of the correlation coefficient 0.6, which is defined as the predictable time series step size.
The obtained real-time wind power change rate sequence is subjected to autocorrelation analysis, the obtained results are shown in fig. 4 and 5, meanwhile, a corresponding time sequence (shown in formula 2) of the absolute value of the relative wind power change amount is selected for autocorrelation analysis, the obtained results are shown in fig. 6 and 7, it can be seen that the predictable length of the real-time wind power change rate sequence is 40min, and the predictable length of the time sequence (shown in formula (2)) of the absolute value of the relative wind power change amount is 6 min. By the method, the predictable length of the climbing event can be obviously improved, and the predictability of the wind power climbing event is improved.
The fifth concrete implementation mode:
the embodiment further defines the wind power climbing event multi-stage early warning method based on the variation function in the first specific embodiment.
Step four: according to the result of the autocorrelation analysis, the time series is found to satisfy a prediction condition. And selecting sequences of the first 1800 time points in one month as a training set, selecting time sequences of the last 100 time points as a prediction set, and establishing a prediction model based on the variation time sequences after evolution by using a BP (back propagation) neural network. The obtained prediction results are shown in fig. 8. The method can be seen that the prediction condition of the series is better, and the prediction requirement of the wind power climbing event is basically met.
The sixth specific implementation mode:
the embodiment further defines the wind power climbing event multi-stage early warning method based on the variation function in the first specific embodiment.
And in the wind power climbing event, the absolute value of the relative wind power variation is defined, and the threshold is selected by using the allowable value of the grid frequency. Let PForehead (forehead)The total installed capacity of the power grid is n%, the wind power station accounts for n% of the total installed capacity, and the allowable deviation of the power supply frequency of the total installed capacity of the power grid above 300 ten thousand kilowatts is 0.2Hz according to the regulation of the national power supply business rules; the allowable deviation of the power supply frequency of the total installed capacity of the power grid below 300 ten thousand kilowatts is 0.5Hz, the difference adjustment coefficient k sigma is generally 4-6%, the power grid frequency requirement in China is 50Hz, and the relative wind power climbing event threshold value can be set as follows:
wherein alpha is 0.2 when the total installed capacity of the power grid is more than 300 ten thousand kilowatts; when the total installed capacity of the power grid is more than 300 ten thousand watts, alpha is 0.5.
And step two, when the wind power climbing event is depicted by using the variation function after the evolution, the wind power climbing event is considered to occur in a certain period of time if the difference between any data point and the data point after the delta t time is greater than a set threshold value due to the fact that the difference represents the sum of squares of the difference between each data point and the data point after the delta t time within the certain period of time delta t. Therefore we can set the threshold γref1(Δt)、γref2(Δt)、γref3(Δ t) and γref4(Δ t) represents the occurrence of a hill climbing event for 1 data point, the occurrence of a hill climbing event for the data point of 1/3, the occurrence of a hill climbing event for the data point of 2/3, and the occurrence of a hill climbing event for all data points, respectively, during the Δ t time period. As shown in equation (8).
When the variation of the relative wind power is square root gamma (delta t)<γref1(delta t), no wind power climbing event occurs in the delta t period, and early warning is not needed; when the relative wind power becomes worse square rootγref1(Δt)<γ(Δt)<γref2(delta t), a small-scale wind power climbing event possibly occurs within a delta t period, and is defined as four-stage early warning; when the relative wind power is changed to gammaref2(Δt)<γ(Δt)<γref3(delta t), a medium-scale wind power climbing event possibly occurs within a delta t period, and is defined as three-level early warning; when the relative wind power is changed to gammaref3(Δt)<γ(Δt)<γref4(delta t), a large-scale wind power climbing event possibly occurs within a delta t period, and secondary early warning is defined; when the relative wind power is changed to gammaref4(Δt)<And when gamma (delta t) is obtained, a large-scale wind power climbing event is certainly generated within the delta t period, and primary early warning is defined. After the threshold value is obtained, the time series prediction result of the step four is identified, and the future wind power climbing event can be accurately predicted by using the actually measured data at the current moment.
In the invention, the actually measured power grid selects 0.2Hz as the allowable deviation of the power supply frequency, the instantaneous maximum value of the power is set as the rated power, and as the capacities of the selected wind power plants are all larger than 300 ten thousand kilowatts, α is 0.2, and the threshold value gamma can be obtainedref1(Δt)＝68.283；γref2(Δt)＝88.153；γref3(Δt)＝176.31；γref4(Δ t) 264.46. The wind power climbing event prediction result divided by the threshold is shown in fig. 9.
The electric power system can make a more reasonable and economic dispatching plan according to the forecasting result of the wind power climbing event, more rotary standby is provided in the time period of forecasting the occurrence of the large power load climbing event, and the safe, efficient and stable operation of the regional electric power system is ensured.
Claims (6)
1. Wind power climbing event multi-stage early warning method based on variation function is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring a relative wind power curve by utilizing the output power curve of the wind power plant and the corresponding load curve of the power grid;
step two: on the basis of using the definition of an absolute value of the relative wind power variation, quantitatively depicting a wind power climbing event by using a square root of a variation function, and acquiring a time series curve of the real-time variation rate of the wind power after variation;
step three: after the time sequence of the real-time change rate of the wind power is obtained, the self-correlation analysis is carried out on the time sequence, and the predictability of the time sequence is observed;
step four: establishing a predictive model of the wind power climbing event based on the wind power real-time change rate time sequence by utilizing a BP neural network algorithm;
step five: and defining a threshold selecting method of the wind power climbing event according to the absolute value of the relative wind power variation to obtain a multi-level threshold in a variation depicting mode, and effectively identifying a prediction result of the BP neural network by using the threshold to realize multi-level early warning of the wind power climbing event.
2. The wind power climbing event multi-stage early warning method based on the variation function as claimed in claim 1, wherein: in the first step, the time sequence P' (t) relative to the wind power curve is:
P'(t)＝P(t)-Q(t) (1)
the time sequence of the output power curve of the wind power plant is P (t), and the time sequence of the load curve of the power grid is Q (t).
3. The wind power climbing event multi-stage early warning method based on the variation function as claimed in claim 1, wherein: and the wind power climbing event in the second step meets the following requirements:
|P'(t+Δt)-P'(t)|＞Pval(2)
wherein P '(t) and P' (t + Deltat) are relative wind power values of t and t + Deltat, respectively, PvalThe set wind power climbing event threshold value is set.
4. The wind power climbing event multi-stage early warning method based on the variation function as claimed in claim 1, wherein: in step two, the variation function is expressed as:
wherein P '(t) represents the time series of the relative wind power curve, P' (t)i) Represents the value of the ith point in the time series of the relative wind power curve, P' (t)i+ Δ t) represents the value of the i + Δ t point in the time sequence of the relative wind power curve, Δ t represents the time interval, and N (Δ t) represents the total amount of data points in the Δ t time;
the real-time change rate gamma (Δ t) of the wind power is expressed as:
5. the wind power climbing event multi-stage early warning method based on the variation function as claimed in claim 1 or 3, wherein: in the fifth step, the threshold value of the wind power climbing event is represented as
Wherein, PForehead (forehead)The total installed capacity of the power grid is represented by n%, the capacity of the wind power plant in the total installed capacity is represented by α being 0.2 when the total installed capacity of the power grid is more than 300 ten thousand kilowatts, and α being 0.5 when the total installed capacity of the power grid is more than 300 ten thousand kilowatts.
6. The wind power climbing event multi-stage early warning method based on the variation function as claimed in claim 5, wherein: in the fifth step, a threshold value gamma is setref1(Δt)、γref2(Δt)、γref3(Δ t) and γref4(Δ t) indicates that a hill climbing event occurs at 1 data point, a hill climbing event occurs at 1/3 data points, a hill climbing event occurs at 2/3 data points, and a hill climbing event occurs at all data points within a Δ t period, respectively, then
N (delta t) represents the total amount of data points in delta t time, and when the real-time change rate gamma of the wind power is (gamma)Δt)<γref1(delta t), no wind power climbing event occurs in the delta t period, and early warning is not needed; when gamma isref1(Δt)<γ(Δt)<γref2(delta t), a small-scale wind power climbing event possibly occurs within a delta t period, and is defined as four-stage early warning; when gamma isref2(Δt)<γ(Δt)<γref3(delta t), a medium-scale wind power climbing event possibly occurs within a delta t period, and is defined as three-level early warning; when gamma isref3(Δt)<γ(Δt)<γref4(delta t), a large-scale wind power climbing event possibly occurs within a delta t period, and secondary early warning is defined; when gamma isref4(Δt)<And when gamma (delta t) is obtained, a large-scale wind power climbing event is certainly generated within the delta t period, and primary early warning is defined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010313374.4A CN111582557A (en) | 2020-04-20 | 2020-04-20 | Wind power climbing event multi-stage early warning method based on variation function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010313374.4A CN111582557A (en) | 2020-04-20 | 2020-04-20 | Wind power climbing event multi-stage early warning method based on variation function |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111582557A true CN111582557A (en) | 2020-08-25 |
Family
ID=72124443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010313374.4A Pending CN111582557A (en) | 2020-04-20 | 2020-04-20 | Wind power climbing event multi-stage early warning method based on variation function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111582557A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949938A (en) * | 2021-03-29 | 2021-06-11 | 哈尔滨工业大学 | Wind power climbing event direct forecasting method for improving training sample class imbalance |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679282A (en) * | 2013-09-30 | 2014-03-26 | 清华大学 | Prediction method for wind power ramp |
| CN106203693A (en) * | 2016-07-05 | 2016-12-07 | 华北电力大学 | A kind of system and method for Power Output for Wind Power Field climbing event prediction |
| CN109033027A (en) * | 2018-08-08 | 2018-12-18 | 长春工程学院 | A kind of high speed fitful wind leads to the prediction technique of climbing event under wind power |
| CN109978222A (en) * | 2019-01-04 | 2019-07-05 | 国网山东省电力公司电力科学研究院 | A kind of wind-powered electricity generation climbing probability of happening prediction technique and system based on Bayesian network |
| CN110071530A (en) * | 2019-05-20 | 2019-07-30 | 中国矿业大学 | The wind-powered electricity generation of electric system containing energy storage climbing coordinated scheduling method based on LSTM |
| CN110264375A (en) * | 2019-06-21 | 2019-09-20 | 哈尔滨工业大学 | A kind of intermittent Multiple Time Scales of wind farm group power output quantify depicting method |
| CN110348637A (en) * | 2019-07-12 | 2019-10-18 | 哈尔滨工业大学 | A kind of consideration field-net factor wind-powered electricity generation climbing event method for early warning |
-
2020
- 2020-04-20 CN CN202010313374.4A patent/CN111582557A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679282A (en) * | 2013-09-30 | 2014-03-26 | 清华大学 | Prediction method for wind power ramp |
| CN106203693A (en) * | 2016-07-05 | 2016-12-07 | 华北电力大学 | A kind of system and method for Power Output for Wind Power Field climbing event prediction |
| CN109033027A (en) * | 2018-08-08 | 2018-12-18 | 长春工程学院 | A kind of high speed fitful wind leads to the prediction technique of climbing event under wind power |
| CN109978222A (en) * | 2019-01-04 | 2019-07-05 | 国网山东省电力公司电力科学研究院 | A kind of wind-powered electricity generation climbing probability of happening prediction technique and system based on Bayesian network |
| CN110071530A (en) * | 2019-05-20 | 2019-07-30 | 中国矿业大学 | The wind-powered electricity generation of electric system containing energy storage climbing coordinated scheduling method based on LSTM |
| CN110264375A (en) * | 2019-06-21 | 2019-09-20 | 哈尔滨工业大学 | A kind of intermittent Multiple Time Scales of wind farm group power output quantify depicting method |
| CN110348637A (en) * | 2019-07-12 | 2019-10-18 | 哈尔滨工业大学 | A kind of consideration field-net factor wind-powered electricity generation climbing event method for early warning |
Non-Patent Citations (2)
| Title |
|---|
| 万杰: "风电场风速特性及不确定性模型研究", 《中国优秀博硕士学位论文全文数据库（博士） 工程科技Ⅱ辑》 * |
| 孙二平 等: "基于单机功率间歇性度量指标的风场发电质量评估策略", 《节能技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949938A (en) * | 2021-03-29 | 2021-06-11 | 哈尔滨工业大学 | Wind power climbing event direct forecasting method for improving training sample class imbalance |
| CN112949938B (en) * | 2021-03-29 | 2022-04-15 | 哈尔滨工业大学 | Wind power climbing event direct forecasting method for improving training sample class imbalance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jamil | Hydroelectricity consumption forecast for Pakistan using ARIMA modeling and supply-demand analysis for the year 2030 | |
| Lei et al. | A review on the forecasting of wind speed and generated power | |
| Gomes et al. | Wind speed and wind power forecasting using statistical models: autoregressive moving average (ARMA) and artificial neural networks (ANN) | |
| CN107704953A (en) | The short-term wind-electricity power probability density Forecasting Methodology of EWT quantile estimate forests | |
| CN104485665A (en) | Dynamic probabilistic power flow (PPF) calculating method considering wind speed predication error temporal-spatial coherence | |
| CN110380444B (en) | Capacity planning method for distributed wind power orderly access to power grid under multiple scenes based on variable structure Copula | |
| CN110348637B (en) | Wind power climbing event early warning method considering field-network factors | |
| CN110197296B (en) | Unit load prediction method based on time series similarity | |
| CN110991725B (en) | RBF ultra-short-term wind power prediction method based on wind speed frequency division and weight matching | |
| Mite-León et al. | Statistical model for the forecast of hydropower production in Ecuador | |
| CN105956708A (en) | Grey correlation time sequence based short-term wind speed forecasting method | |
| Zhang et al. | Ultra-short term wind power prediction model based on modified grey model method for power control in wind farm | |
| CN111582557A (en) | Wind power climbing event multi-stage early warning method based on variation function | |
| Memmel et al. | Forecast of renewable curtailment in distribution grids considering uncertainties | |
| CN105654189B (en) | Icing short-term prediction method based on time series analysis and Kalman filtering algorithm | |
| CN106503828B (en) | Ultra-short term chaotic prediction method for photovoltaic output power | |
| Yang et al. | Probabilistic short-term wind power forecast using componential sparse Bayesian learning | |
| Zhen et al. | Multi-level wavelet decomposition based day-ahead solar irradiance forecasting | |
| CN113946792A (en) | Reservoir group water level control method based on water abandoning probability | |
| Narayana et al. | Adaptive prediction of power fluctuations from a wind turbine at Kalpitiya area in Sri Lanka | |
| Ngo et al. | Ultra-short-term load forecasting using robust exponentially weighted method in distribution networks | |
| Lin et al. | Scenarios analysis method based on wind and photovoltaic power output correlation | |
| Kortmann et al. | From a distributing to a generating network: Assessing PV hosting capacity under uncertainty in distribution grids | |
| CN112633565A (en) | Photovoltaic power aggregation interval prediction method | |
| Aksan et al. | Multivariate Multi-step Forecasting for Cable Pooling Applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200825 |