CN107404127B - Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination - Google Patents
Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination Download PDFInfo
- Publication number
- CN107404127B CN107404127B CN201710682313.3A CN201710682313A CN107404127B CN 107404127 B CN107404127 B CN 107404127B CN 201710682313 A CN201710682313 A CN 201710682313A CN 107404127 B CN107404127 B CN 107404127B
- Authority
- CN
- China
- Prior art keywords
- wind power
- short
- time
- interval
- prediction
- 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.)
- Active
Links
Classifications
-
- H02J2103/30—
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
- Wind Motors (AREA)
- Coloring Foods And Improving Nutritive Qualities (AREA)
Abstract
The present invention relates to a kind of wind-powered electricity generation Robust Interval trace scheduling methods that consideration Multiple Time Scales are coordinated.This method binding model PREDICTIVE CONTROL and robust optimize, robust optimization is rolled under the Scheduling Framework of Multiple Time Scales, power interval track boundary and conventional power unit generation schedule can be dissolved by generating wind power plant, when wind power output is all satisfied system security constraint when that can dissolve in the boundary of power interval track, alleviate the system security risk that wind power point prediction in traditional scheduler is inaccurately left, the practical power output of wind power's supervision system Real-time Feedback wind power plant simultaneously, it calculates prediction error and predicted value is corrected, make forecasted future value closer to actual value, it is cut down step by step since wind-powered electricity generation predicts the plan deviation of decision content caused by error, keep optimal planning index more accurate.
Description
Technical Field
The invention relates to the field of operation and control of power systems, in particular to a wind power robust interval trajectory scheduling method considering multi-time scale coordination.
Background
With the gradual increase of the proportion of wind power access in the power system, the uncertainty and randomness of the wind power bring various influences on the power system. As a core component in the operation control of the power system, active scheduling is directly related to the active power balance and frequency stability in the power system, and irreplaceable effects are achieved on the safety, reliability and economic operation of the power system.
The active scheduling operation after wind power access depends on the wind power prediction technology. However, the wind power point prediction still has a large error, and the error between the predicted value and the actual value gradually increases along with the increase of the prediction time. On the premise of high-proportion wind power integration, the reliability of a power generation plan obtained by optimizing a traditional active scheduling method of a wind power point prediction result is reduced, the strong randomness of wind power can cause a wind power plant to possibly deviate from a plan value, an extreme operation mode is generated, and the safety of system operation is threatened; the violent fluctuation of wind power can increase the wind power plant tripping action, so that the abandoned wind rate is increased, and the action of limiting the wind power output can only be carried out after the power system is subjected to an unreliable operation mode, so that the advance prediction cannot be carried out.
At present, the research on the dispatching operation control of wind power-containing systems is more and more intensive at home and abroad, and the research on how to utilize probabilistic information to carry out active dispatching on the power systems is also increased gradually. Random planning and fuzzy planning are applied to system scheduling. And solving the economic dispatching model containing the probability information of the uncertain quantity by constructing the safety constraint in the optimization model into an opportunity constraint according to the probability distribution information of the uncertain quantity by stochastic programming. However, the difficulty of acquiring the probability information of the wind power in practice and the complexity of calculation limit the application of stochastic programming. The fuzzy planning expresses the attitude of a decision maker on uncertain quantity and results caused by the uncertain quantity by setting a membership function, and a satisfactory decision value is obtained by optimizing the maximum membership function, but the fuzzy planning has stronger subjectivity, and most optimization methods are open-loop optimization and lack feedback control as compensation.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a wind power robust interval track scheduling method considering multi-time scale coordination. The method combines model prediction control and robust optimization, rolling robust optimization is carried out under a multi-time-scale scheduling framework, a track limit of a reducible power interval of a wind power plant and a power generation plan of a conventional unit are generated, system safety constraints are met when wind power output is within the track limit of the reducible power interval, the potential safety hazard of a system left by inaccurate prediction of wind power points in traditional scheduling is relieved, meanwhile, a wind power plant monitoring system feeds back actual output of the wind power plant in real time, a predicted value in the future is enabled to be closer to an actual value, and plan deviation of decision quantity caused by wind power prediction errors is gradually reduced, so that optimization plan indexes are enabled to be more accurate.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
a wind power robust interval trajectory scheduling method considering multi-time scale coordination comprises the following steps:
s1, according to the historical wind power actual value (namely the historical wind power field actual output), the historical short-term wind power predicted value and the historical ultra-short-term wind power predicted value, calculating the error range of short-term wind power prediction and the error range of ultra-short-term wind power prediction, and combining the latest short-term wind power predicted value and the latest ultra-short-term wind power predicted value to generate a short-term wind power prediction interval and an ultra-short-term wind power prediction interval as input information of an optimization model;
s2, in the day rolling plan module, based on the short-term wind power prediction interval and the short-term load prediction, performing robust interval rolling optimization by taking the lowest power generation cost of the conventional unit and the minimum upper limit deviation of the short-term wind power prediction interval as a target function and the system safety as a constraint, and calculating the power consumption interval track limit of the wind power plant and the conventional unit power generation plan;
s3, in the real-time adjustment plan module, based on the ultra-short-term wind power prediction interval and the ultra-short-term load prediction, continuously rolling robust optimization by taking the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan obtained by the in-day rolling optimization module as basic values, adjusting the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan, and obtaining a corrected wind power plant absorbable power interval track limit and a corrected conventional unit power generation plan;
s4, in an AGC real-time control module, a non-AGC set tracks and corrects a conventional set power generation plan, a wind power plant adopts a maximum power point tracking mode in a track limit of a power-consumption-capable interval of the corrected wind power plant, the AGC set adjusts a base power value of the AGC set in real time in response to irregular small fluctuation and wind power out-of-limit conditions, the wind power consumption is properly increased when the AGC set is sufficiently reserved, and finally a power generation plan instruction is sent to the wind power plant and the conventional set;
and S5, feeding back the actual output of the wind power plant in real time according to the wind power plant monitoring system, and correcting the input information of the optimization model, thereby improving the accuracy of the scheduling plan in a rolling manner.
In the method, in step S1, the short-term wind power prediction interval is obtained by adding the latest short-term wind power prediction value to the error range of the short-term wind power prediction;
the ultra-short-period wind power prediction interval is obtained by adding the latest ultra-short-period wind power prediction value to the error range of the ultra-short-period wind power prediction.
In the above method, in step S2, the objective function is:
wherein,representing the lower limit and the upper limit of the track limit of the wind power plant acceptable power interval;representing a planning base value of a wind power plant i at the moment t; pj,tRepresenting a power generation plan of a jth conventional unit at the time t; a isj,bj,cjSecondary item system of generating cost of jth conventional unit
Number, first order coefficient and constant coefficient;representing the upper limit of the short-term wind power prediction interval of the wind power plant i at the moment t; lambda [ alpha ]iRepresenting a power prediction upper bound deviation penalty coefficient of the wind power plant i; t represents an optimized time domain; n is a radical ofGRepresenting the number of conventional units, NWRepresenting the number of wind farms.
In the above method, in step S2, the constraint condition includes:
A1) and power balance constraint:
wherein,short-term load prediction representing time t;
B1) and (3) limiting and constraining the output of the conventional unit:
wherein, Pj,t、Respectively representing the lower limit and the upper limit of the output of the jth conventional unit at the time t;
C1) conventional unit climbing rate constraint:
wherein,respectively the maximum downward climbing power and the maximum upward climbing power of the jth conventional unit at the moment t;
D1) rotating standby constraint:
wherein,respectively representing the lower rotation standby and the upper rotation standby of the jth conventional unit at the time t;respectively representing the lower rotation standby requirement and the upper rotation standby requirement of the power system;
E1) and (3) safety restraint of a transmission section:
wherein phi isj-lIs the power generation transfer factor of the jth conventional unit to the section li-lA power generation transfer factor of the wind power plant i to the section l is obtained; fl min、Fl maxRespectively representing the lower current limit and the upper current limit of the section l;
F1) wind power output restraint:
wherein,and the lower limit of the short-term wind power prediction interval of the wind power plant i at the moment t is shown.
In the method, in step S2, in the daily scrolling planning module, the decision value of the future 4h is optimized once every 1h with 15min as the time resolution, and the total of 16 points is obtained, and only the first 4 points are executed each time.
In the above method, in step S3, the objective function of the rolling robust optimization is:
wherein,representing the power generation plan of the jth conventional unit obtained by the optimization of the rolling plan module in the day at the time t;representing the upper limit of the track limit of the digestible power interval of the wind power plant i at the time t, which is obtained by optimizing the day rolling plan module;representing the upper limit of the ultra-short-term wind power prediction interval;respectively representing the lower limit and the upper limit of the track limit of the digestible power interval of the corrected wind power plant i;representing a corrected wind farm planning base value; delta Pj,tRepresenting and correcting the power generation planned value of the conventional unit; t denotes an optimized time domain.
In the above method, in step S3, the constraint conditions of the rolling robust optimization include:
A2) and power balance constraint:
wherein,representing the wind power plant i planning base value obtained by the rolling planning module within the day at the time t,ultra-short term load prediction representing time t;
B2) and (3) limiting and constraining the output of the conventional unit:
C2) conventional unit climbing rate constraint:
D2) rotating standby constraint:
E2) and (3) safety restraint of a transmission section:
F2) wind power output restraint:
G2) and AGC base point adjustment constraint:
wherein, γmin、γmaxRespectively representing a lower limit coefficient and an upper limit coefficient of an AGC unit adjustment margin; pk,tOptimizing the obtained planned output for the kth AGC unit;the capacity of the kth AGC unit.
In the above method, in step S3, the real-time adjustment planning module scrolls every 5min, takes 5min as the time resolution, optimizes the decision value of 1h in the future, and only executes the 1 st point each time, for 12 points.
In the above method, the mathematical expression of step S5 is as follows: after the control is performed at time t, the output values of the object at the future time may be calculated from the prediction model, including the predicted value y of the ith output value at time t +1i,1(t +1| t); each actual output y is measured by t +1i(t +1), the error value e (t +1) is formed and compared with the corresponding predicted value, and the future error is predicted by a weighting method using the error information, and the model-based prediction is compensated by the error information, so that a corrected predicted value can be obtained:
ycor(t+2|t+1)＝ypre(t+2|t+1)+He(t+2|t+1) (16)
wherein, ypre(t +2| t +1) represents a predicted value before correction at the time t +1 to the time t + 2; y iscor(t +2| t +1) represents a predicted value obtained by correcting the t +1 moment to the t +2 moment; e (t +2| t +1) represents an error value predicted based on e (t +1) at the time of t + 1; h is a correction weight value, and H belongs to [0,1 ]]Determined by the accuracy of the error prediction.
The wind power robust interval trajectory scheduling method considering multi-time scale coordination has the following advantages:
the method comprehensively considers the following factors:
1. actual values and predicted values of historical wind power;
2. short-term wind power, short-term load prediction data, ultra-short-term wind power and ultra-short-term load prediction data;
3. the topological structure information of the power system containing wind power is as follows: in the transmission section safety constraint formula (6), section information is related, the sections can also be understood as lines in a popular way, and transfer factors in the section information need to relate to topological structure information of a power system, namely the connection relation of the lines, the position relation of a power plant and a wind power plant, the upper and lower limits of the tidal current of each section and the like;
4. the output limit value, the climbing rate, the reserve capacity and other information of the conventional unit.
The method combines model prediction control and robust optimization, rolling robust optimization is carried out under a multi-time-scale scheduling framework, a track limit of a wind power plant consumable power interval and a conventional unit plan are generated, when wind power output is within the track limit of the wind power plant consumable power interval, safety constraint of an electric power system is met, potential safety hazards of the electric power system left by inaccurate prediction of wind power points in traditional scheduling are relieved, meanwhile, a wind power plant monitoring system feeds back actual output of the wind power plant in real time, prediction errors are calculated and corrected, the future predicted value is closer to an actual value, plan deviation of a decision value caused by the wind power prediction errors is reduced step by step, and optimization plan indexes (namely, issuing plan instructions of an AGC real-time control module to the wind power plant and the conventional unit) are more accurate.
Drawings
Fig. 1 is a flow diagram of a wind power robust interval trajectory scheduling method considering multi-time scale coordination.
Fig. 2 is a schematic diagram of a wind power consumption-capable power interval trajectory limit and a control time scale of a wind power robust interval trajectory scheduling method considering multi-time scale coordination.
Detailed Description
The invention is described in further detail below with reference to figures 1 and 2.
S1, according to the historical wind power actual value (namely the historical wind power field actual output), the historical short-term wind power predicted value and the historical ultra-short-term wind power predicted value, calculating the error range of short-term wind power prediction and the error range of ultra-short-term wind power prediction, combining the latest short-term wind power predicted value and the latest ultra-short-term wind power predicted value, generating a short-term wind power prediction interval and an ultra-short-term wind power prediction interval, namely constructing the wind power prediction interval as the input information of an optimization model. The method comprises the following specific steps:
as shown in fig. 1. The scheduling optimization process is divided into a day rolling plan module, a real-time adjustment plan module and an AGC real-time control module, wherein the day rolling plan module and the real-time adjustment plan module both need wind power prediction data as input, in order to generate a power interval track limit which can be absorbed by a wind power plant, a wind power point predicted value is converted into a wind power interval predicted value (namely a wind power predicted interval), and an error range interval [ -delta, + sigma-sigma ] of each wind power plant is obtained under a certain confidence coefficient by combining error probability statistics of a historical wind power predicted value and an actual value]Combining the predicted value of the future short-term wind power and the predicted value P of the ultra-short-term wind powerwfGenerating a short-term wind power prediction interval and an ultra-short-term wind power prediction interval [ P ]wf-δ,Pwf+σ]。
S2, in the day rolling plan module, based on the short-term wind power prediction interval and the short-term load prediction, performing robust interval rolling optimization by taking the lowest power generation cost of the conventional unit and the minimum deviation of the upper limit of the short-term wind power prediction interval as a target function and the system safety as a constraint, and calculating the power-consumption interval track limit of the wind power plant and the conventional unit power generation plan, specifically as follows:
in the daily rolling planning module, rolling is carried out once every 1h, 15min is taken as time resolution, decision values of 4h in the future are optimized, 16 points are obtained, and only the first 4 points are executed each time. Different from the traditional robust optimization, the uncertain range in the robust optimization model is used as a decision value instead of a given value, namely the decision value is a wind power plant planning base value, a wind power plant can absorb a power interval track limit and a conventional unit power generation plan, and the objective function is as follows:
wherein,representing the lower limit and the upper limit of the track limit of the wind power plant acceptable power interval;representing a planning base value of a wind power plant i at the moment t; pj,tRepresenting a power generation plan of a jth conventional unit at the time t; a isj,bj,cjA quadratic term coefficient, a primary term coefficient and a constant term coefficient of the power generation cost of the jth conventional unit are respectively;representing the upper limit of the short-term wind power prediction interval of the wind power plant i at the moment t; lambda [ alpha ]iRepresenting a power prediction upper bound deviation penalty coefficient of the wind power plant i; t represents an optimized time domain, and T is 16 in the module; n is a radical ofGRepresenting the number of conventional units, NWRepresenting the number of wind farms;
the objective function corresponds to the method, is not the prior art, comprehensively considers the power generation cost and wind power consumption of the conventional unit, can optimize the limit range of the consumable power interval of the wind power plant, can accept the wind power fluctuation in the limit range by the power grid, and can provide reference for a scheduling plan.
The constraint conditions include: the method comprises the following steps of power balance constraint, conventional unit output limit constraint, conventional unit climbing rate constraint, wind power output constraint, rotary standby constraint and transmission section safety constraint. The variables in the economic dispatch optimization model are variable according to specific issues. The variables in the constraint conditions are the variables considered in the method corresponding to the patent, and the constraint conditions can ensure that the result obtained by optimizing the model is within the safe operation range of the power system. The details are as follows:
1) power balance constraint
Wherein,representing the short-term load forecast at time t.
2) Conventional unit output limit constraints
Wherein,P j,t、respectively representing the lower limit and the upper limit of the output of the jth conventional unit at the time t.
3) Conventional unit ramp rate constraint
Wherein,the maximum downward climbing power and the maximum upward climbing power of the jth conventional unit at the moment t are respectively.
4) Rotational back-up restraint
Wherein,respectively representing the lower rotation standby and the upper rotation standby of the jth conventional unit at the time t;respectively representing the lower spinning reserve demand and the upper spinning reserve demand of the power system. The power system herein refers to the entire power system consisting of all conventional units, wind farms, transmission lines, etc., and because the power system has a capacity required for spinning reserve, the spinning reserve capacity referred to herein is in units of the entire power system.
5) Safety restraint of transmission section
Wherein phi isj-lIs the power generation transfer factor of the jth conventional unit to the section li-lA power generation transfer factor of the wind power plant i to the section l is obtained; fl min、Fl maxRespectively representing the lower current limit and the upper current limit of the section l.
6) Wind power output constraint
Wherein,and the lower limit of the short-term wind power prediction interval of the wind power plant i at the time t is represented, the formula represents that the lower limit of the track limit of the wind power plant absorbable power interval should not be lower than the lower limit of the short-term wind power plant power prediction interval, and the upper limit of the track limit of the wind power plant absorbable power interval is not higher than the upper limit of the short-term wind power prediction interval.
S3, in the real-time adjustment plan module, based on the ultra-short-term wind power prediction interval and the ultra-short-term load prediction, continuously rolling robust optimization by taking the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan obtained by the in-day rolling optimization module as basic values, adjusting the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan, and obtaining a corrected wind power plant absorbable power interval track limit and a corrected conventional unit power generation plan as follows:
the decision value of the objective function in the module is used for correcting the track limit of a power-absorbing interval of a wind power plant, correcting the power generation plan of a conventional unit and correcting the plan base value of the wind power plant, AGC unit base point adjustment constraint is added into constraint conditions, enough adjustment margin is reserved for the AGC unit, the real-time adjustment plan module rolls every 5min, 5min is used as time resolution, the decision value of 1h in the future is optimized, 12 points are calculated, and only the 1 st point is executed every time.
As shown in fig. 1. The real-time adjustment plan module mainly adjusts or corrects the wind power plant absorbable power interval track limit obtained by the last optimization and the conventional unit power generation plan according to the ultra-short term prediction data in a shorter time scale, so that the decision value form of the robust interval economic dispatching is changed into delta P, and the objective function is as follows:
wherein,representing the power generation plan of the jth conventional unit obtained by the optimization of the rolling plan module in the day at the time t;representing the upper limit of the track limit of the digestible power interval of the wind power plant i at the time t, which is obtained by optimizing the day rolling plan module;representing the upper limit of the ultra-short-term wind power prediction interval;respectively representing the lower limit and the upper limit of the track limit of the digestible power interval of the corrected wind power plant i;representing a corrected wind farm planning base value; delta Pj,tRepresenting and correcting the power generation planned value of the conventional unit; t represents the optimized time domain, and T is 12 in this block. The objective function here is also the method corresponding to this patent and is not prior art. The decision value of the objective function is the adjustment amount after the last module optimizes the value, the plan is further adjusted, the accuracy and the optimality of the plan are guaranteed, and meanwhile the power generation cost and the wind power consumption of a conventional unit are also considered.
The constraint conditions of the real-time adjustment planning module are as follows:
1) power balance constraint
Wherein,representing the wind power plant i planning base value obtained by the rolling planning module within the day at the time t,indicating an ultra-short term load prediction at time t.
2) Conventional unit output limit constraints
3) Conventional unit ramp rate constraint
4) Rotational back-up restraint
5) Safety restraint of transmission section
6) Wind power output constraint
In order to reserve enough adjustment margin for an AGC real-time control module, an AGC base point adjustment constraint is added into the module, and the formula is (15):
wherein, γmin、γmaxRespectively representing a lower limit coefficient and an upper limit coefficient of an AGC unit adjustment margin; pk,tOptimizing the obtained planned output for the kth AGC unit;the capacity of the kth AGC unit. The formula of the step is used as a constraint condition and added into a real-time adjustment planning module, so that enough margin can be ensured to be reserved for AGC subsequent adjustment.
S4, in the AGC real-time control module, the non-AGC set tracks the power generation plan of the conventional set optimized by the real-time adjustment plan module (namely, the power generation plan of the conventional set is corrected), the wind power plant adopts a maximum power point tracking mode in the track limit of a power-consumption-capable interval of the wind power plant, the AGC set adjusts the power value of a base point of the AGC set in real time according to the irregular small fluctuation and the wind power out-of-limit condition, the wind power consumption is properly increased when the AGC set is sufficiently reserved, and finally a power generation plan instruction is sent to the wind power plant and the conventional set. The method comprises the following specific steps:
the conventional unit comprises an AGC unit and a non-AGC unit, the AGC unit can generally indicate the conventional unit with high force adjustment speed, and the non-AGC unit is low in adjustment speed in the invention, so that the power generation plan is mainly tracked.
In the AGC real-time control module, a non-AGC set tracks and corrects a power generation plan of a conventional set, the AGC set adjusts an output value in real time in response to load fluctuation to maintain the frequency stability of a power system, and when the wind power output is in a corrected wind power plant absorbable power interval track limit, a maximum power point tracking mode is adopted by the wind power plant; and when the wind power output exceeds the upper limit, limiting the wind power output to the upper limit of the track limit of the wind power field acceptable power interval. When the standby of the AGC unit is sufficient, the wind power limit output can be properly reduced, and the wind power can be consumed to the maximum extent.
The reason that the uncertainty of wind power output is large is mainly to prevent exceeding the upper limit, in order to guarantee the power supply requirement, the condition that wind power is zero generated is considered in the installed capacity of the conventional units in the power system, the total installed capacity can meet the load requirement, and when the wind power has output, the installed capacity of part of the conventional units can be used as a spare. Therefore, the lower limit of the wind power output is not limited, but the upper limit generally considers that the maximum wind power output cannot be larger than the predicted value of the wind power, the upper limit and the lower limit of the track limit of the power-acceptable interval of the corrected wind power plant obtained by optimization by the method are within the upper limit and the lower limit of the predicted power (namely a short-term wind power prediction interval and an ultra-short-term wind power prediction interval), the wind power output range can be reduced, and more accurate reference is provided for scheduling and planning.
The schematic diagram of the wind power plant absorbable and reducible power interval track boundary and the coordination of the time scales of the modules are shown in fig. 2.
S5, according to the real-time feedback of the wind power plant actual output of the wind power plant monitoring system, correcting the future input predicted value of the optimization model (mainly correcting the short-term wind power predicted value and the ultra-short-term wind power predicted value, namely correcting P)wfAfter the value is corrected, the upper limit of the short-term wind power prediction interval and the upper limit of the ultra-short-term wind power prediction interval in the formula (1) and the formula (8) are naturally corrected, so that the accuracy of the scheduling plan can be improved by rolling, and the method specifically comprises the following steps:
as shown in fig. 1. The wind power plant monitoring system feeds the monitored actual output of the wind power plant back to the intraday rolling plan module and the real-time adjustment plan module in real time, the prediction result is corrected, the predicted value is closer to the actual value, the optimized decision value is more accurate, and the mathematical expression is as follows:
after the control is performed at time t, the output values of the object at the future time may be calculated from the prediction model, including the predicted value y of the ith output quantity at time t +1i,1(t +1| t). Each actual output y is measured by t +1i(t +1), the error value e (t +1) is formed and compared with the corresponding predicted value, and the future error is predicted by a weighting method using the error information, and the model-based prediction is compensated by the error information, so that a corrected predicted value can be obtained:
ycor(t+2|t+1)＝ypre(t+2|t+1)+He(t+2|t+1) (16)；
wherein, ypre(t +2| t +1) represents the modification of the t +1 time to the t +2 timeA predicted value right before; y iscor(t +2| t +1) represents a predicted value obtained by correcting the t +1 moment to the t +2 moment; e (t +2| t +1) represents an error value predicted based on e (t +1) at the time of t + 1; h is a correction weight value, and H belongs to [0,1 ]]Determined by the accuracy of the error prediction. (if the predicted value is excessively corrected reversely, the subsequent optimization precision is reduced, the selection of H is determined by the accuracy of error prediction, if the error accuracy is high, the correction weight value H can be larger, otherwise, the correction weight value H can be smaller, so that the predicted value can be ensured not to be excessively corrected.)
The model prediction control comprises three links of prediction model, rolling optimization and feedback correction. Step S5 is a feedback correction link, and applying model predictive control to scheduling is one of the innovative points of the present invention, and is not the prior art. The feedback correction can correct the output of the prediction model in real time, and the output of the prediction model is a reference for rolling optimization, so the feedback correction is optimized to form a closed loop, the optimization precision can be improved, and a more accurate plan value can be made.
The method combines model prediction control and robust optimization, is embedded into the dispatching operation of the power system containing wind power, and is obviously different from the prior dispatching technology. The scheduling is hierarchically coordinated on various time scales, and a track limit of a wind power plant consumable power interval is used as a decision value to be optimized, wherein the limit is absent in the prior art, and the planning precision can be improved by thinning layer by layer; meanwhile, the lower rotation standby capacity of the AGC unit is considered, so that the wind power generation is increased under the condition of sufficient standby, the accuracy of wind power prediction can be improved through a feedback correction link in model prediction control, and the accuracy of plan making can be further improved.
Those not described in detail in this specification are within the skill of the art.
Claims (9)
1. A wind power robust interval trajectory scheduling method considering multi-time scale coordination comprises the following steps:
s1, according to the historical wind power actual value, the historical short-term wind power predicted value and the historical ultra-short-term wind power predicted value, calculating the error range of short-term wind power prediction and the error range of ultra-short-term wind power prediction, and combining the latest short-term wind power predicted value and the latest ultra-short-term wind power predicted value to generate a short-term wind power prediction interval and an ultra-short-term wind power prediction interval as input information of an optimization model;
s2, in the day rolling plan module, based on the short-term wind power prediction interval and the short-term load prediction, performing robust interval rolling optimization by taking the lowest power generation cost of the conventional unit and the minimum upper limit deviation of the short-term wind power prediction interval as a target function and the system safety as a constraint, and calculating the power consumption interval track limit of the wind power plant and the conventional unit power generation plan;
s3, in the real-time adjustment plan module, based on the ultra-short-term wind power prediction interval and the ultra-short-term load prediction, continuously rolling robust optimization by taking the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan obtained by the in-day rolling optimization module as basic values, adjusting the track limit of the wind power plant absorbable power interval and the conventional unit power generation plan, and obtaining a corrected wind power plant absorbable power interval track limit and a corrected conventional unit power generation plan;
s4, in an AGC real-time control module, a non-AGC set tracks and corrects a conventional set power generation plan, a wind power plant adopts a maximum power point tracking mode in a track limit of a power-consumption-capable interval of the corrected wind power plant, the AGC set adjusts a base power value of the AGC set in real time in response to irregular small fluctuation and wind power out-of-limit conditions, the wind power consumption is properly increased when the AGC set is sufficiently reserved, and finally a power generation plan instruction is sent to the wind power plant and the conventional set;
and S5, feeding back the actual output of the wind power plant in real time according to the wind power plant monitoring system, and correcting the input information of the optimization model, thereby improving the accuracy of the scheduling plan in a rolling manner.
2. The method of claim 1, wherein: in step S1, the short-term wind power prediction interval is obtained by adding the latest short-term wind power prediction value to the error range of the short-term wind power prediction;
the ultra-short-period wind power prediction interval is obtained by adding the latest ultra-short-period wind power prediction value to the error range of the ultra-short-period wind power prediction.
3. The method of claim 1 or 2, wherein: in step S2, the objective function is:
wherein,representing the lower limit and the upper limit of the track limit of the wind power plant acceptable power interval;representing a planning base value of a wind power plant i at the moment t; pj,tRepresenting a power generation plan of a jth conventional unit at the time t; a isj,bj,cjA quadratic term coefficient, a primary term coefficient and a constant term coefficient of the power generation cost of the jth conventional unit are respectively;representing the upper limit of the short-term wind power prediction interval of the wind power plant i at the moment t; lambda [ alpha ]iRepresenting a power prediction upper bound deviation penalty coefficient of the wind power plant i; t represents an optimized time domain; n is a radical ofGRepresenting the number of conventional units, NWRepresenting the number of wind farms.
4. The method of claim 3, wherein: in step S2, the constraint conditions include:
A1) and power balance constraint:
wherein,short-term load prediction representing time t;
B1) and (3) limiting and constraining the output of the conventional unit:
wherein,P j,t、respectively representing the lower limit and the upper limit of the output of the jth conventional unit at the time t;
C1) conventional unit climbing rate constraint:
wherein,respectively the maximum downward climbing power and the maximum upward climbing power of the jth conventional unit at the moment t;
D1) rotating standby constraint:
wherein,respectively representing the lower rotation standby and the upper rotation standby of the jth conventional unit at the time t;respectively representing the lower rotation standby requirement and the upper rotation standby requirement of the power system;
E1) and (3) safety restraint of a transmission section:
wherein phi isj-lIs the power generation transfer factor of the jth conventional unit to the section li-lA power generation transfer factor of the wind power plant i to the section l is obtained; fl min、Fl maxRespectively representing the lower current limit and the upper current limit of the section l;
F1) wind power output restraint:
wherein,and the lower limit of the short-term wind power prediction interval of the wind power plant i at the moment t is shown.
5. The method of claim 1 or 2, wherein: in step S2, in the daily rolling plan module, the rolling is performed once every 1h, 15min is used as the time resolution, the decision value of 4h in the future is optimized, 16 points are used in total, and only the first 4 points are executed each time.
6. The method of claim 4, wherein: in step S3, the objective function of the rolling robust optimization is:
wherein,representing the power generation plan of the jth conventional unit obtained by the optimization of the rolling plan module in the day at the time t;the method represents the consumable power interval track boundary of the wind power plant i at the time t obtained by optimizing the rolling plan module in the dayAn upper limit;representing the upper limit of the ultra-short-term wind power prediction interval;respectively representing the lower limit and the upper limit of the track limit of the digestible power interval of the corrected wind power plant i;representing a corrected wind farm planning base value; delta Pj,tRepresenting and correcting the power generation planned value of the conventional unit; t denotes an optimized time domain.
7. The method of claim 6, wherein: in step S3, the constraint conditions of the rolling robust optimization include:
A2) and power balance constraint:
wherein,representing the wind power plant i planning base value obtained by the rolling planning module within the day at the time t,ultra-short term load prediction representing time t;
B2) and (3) limiting and constraining the output of the conventional unit:
C2) conventional unit climbing rate constraint:
D2) rotating standby constraint:
E2) and (3) safety restraint of a transmission section:
F2) wind power output restraint:
G2) and AGC base point adjustment constraint:
wherein, γmin、γmaxRespectively representing a lower limit coefficient and an upper limit coefficient of an AGC unit adjustment margin; pk,tOptimizing the obtained planned output for the kth AGC unit;for the capacity of the kth AGC unit,represents the power generation plan, delta P, of the jth conventional unit obtained by the optimization of the rolling plan module in the day at the time t-1j,t-1Represents the corrected conventional unit power generation planned value at the time t-1,and the lower limit of the ultra-short-term wind power prediction interval of the wind power plant i at the time t is shown.
8. The method of claim 1 or 2, wherein: in step S3, the real-time adjustment planning module rolls every 5min, takes 5min as the time resolution, optimizes the decision value of 1h in the future, and has 12 points, and only executes the 1 st point each time.
9. The method of claim 1 or 2, wherein: the mathematical expression of step S5 is as follows: after the control is performed at time t, the output values of the object at the future time may be calculated from the prediction model, including the predicted value y of the ith output value at time t +1i,1(t +1| t); measuring actual outputs y by t +1i(t +1), the error value e (t +1) is formed and compared with the corresponding predicted value, and the future error is predicted by a weighting method using the error information, and the model-based prediction is compensated by the error information, so that a corrected predicted value can be obtained:
ycor(t+2|t+1)＝ypre(t+2|t+1)+He(t+2|t+1) (16)；
wherein, ypre(t +2| t +1) represents a predicted value before correction at the time t +1 to the time t + 2; y iscor(t +2| t +1) represents a predicted value obtained by correcting the t +1 moment to the t +2 moment; e (t +2| t +1) represents an error value vector predicted based on e (t +1) at the time of t + 1; h is a correction weight value, and H belongs to [0,1 ]]Determined by the accuracy of the error prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710682313.3A CN107404127B (en) | 2017-08-10 | 2017-08-10 | Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710682313.3A CN107404127B (en) | 2017-08-10 | 2017-08-10 | Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107404127A CN107404127A (en) | 2017-11-28 |
| CN107404127B true CN107404127B (en) | 2019-12-03 |
Family
ID=60396580
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710682313.3A Active CN107404127B (en) | 2017-08-10 | 2017-08-10 | Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107404127B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108063453B (en) * | 2017-12-04 | 2021-07-13 | 国网辽宁省电力有限公司 | A full-time multivariate optimal scheduling method for power grids to improve wind power capacity |
| CN107994571A (en) * | 2017-12-04 | 2018-05-04 | 国网山东省电力公司德州供电公司 | A kind of Regional Energy net Multiple Time Scales management method of the parameter containing Full Fuzzy |
| CN108090632B (en) * | 2018-01-23 | 2020-06-02 | 南方电网科学研究院有限责任公司 | New energy grid-connected power system multi-time scale scheduling method based on robust optimization |
| CN108306287A (en) * | 2018-02-09 | 2018-07-20 | 清华大学 | A kind of small interference unstability Risk Identification Method and system of the electric system containing wind-powered electricity generation |
| CN109787279B (en) * | 2019-01-14 | 2022-05-17 | 三峡大学 | Interval quantity solving method for uncertain scheduling problem of wind power grid-connected system |
| CN109742799B (en) * | 2019-01-24 | 2022-07-01 | 福州大学 | D-U space mixed multi-attribute wind power plant installed capacity interval decision method |
| CN110458344B (en) * | 2019-07-26 | 2022-05-06 | 湘潭大学 | Ultrashort-term wind power prediction method adaptive to time resolution |
| CN113969873B (en) * | 2021-10-18 | 2023-04-07 | 国网湖南省电力有限公司 | Ultra-short term output prediction method and system for wind turbine generator and storage medium |
| CN115207913A (en) * | 2022-07-27 | 2022-10-18 | 中国华能集团清洁能源技术研究院有限公司 | Method and system for correcting ultra-short-term power prediction data of wind storage combined operation |
| JP7756667B2 (en) * | 2023-01-13 | 2025-10-20 | 株式会社日立製作所 | Power operation system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104181895A (en) * | 2014-08-28 | 2014-12-03 | 国家电网公司 | Strategy for optimizing short-term and ultra-short-term coordination rolling schedules adapting to access of new energy resources |
| CN104933516A (en) * | 2015-05-27 | 2015-09-23 | 华南理工大学 | Multi-time-scale power system robustness scheduling system design method |
-
2017
- 2017-08-10 CN CN201710682313.3A patent/CN107404127B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104181895A (en) * | 2014-08-28 | 2014-12-03 | 国家电网公司 | Strategy for optimizing short-term and ultra-short-term coordination rolling schedules adapting to access of new energy resources |
| CN104933516A (en) * | 2015-05-27 | 2015-09-23 | 华南理工大学 | Multi-time-scale power system robustness scheduling system design method |
Non-Patent Citations (2)
| Title |
|---|
| Short-term wind power Prediction based on Combined Grey-Markov Model;CHEN Sheng et al.;《2011 International Conference on Advanced Power System Automation and Protection》;20120412;第3卷;第1705-1711页 * |
| 多时间尺度的电力系统鲁棒调度研究;范刘洋等;《电网技术》;20170531;第41卷(第5期);第1576-1582页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107404127A (en) | 2017-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107404127B (en) | Wind Power Robust Interval Trajectory Scheduling Method Considering Multi-Time Scale Coordination | |
| Huang et al. | A control strategy based on deep reinforcement learning under the combined wind-solar storage system | |
| CN103259285B (en) | Method for optimizing short running of electric power system comprising large-scale wind power | |
| CN105046395B (en) | An intraday rolling plan formulation method for power system with multiple types of new energy | |
| CN112868157A (en) | Method for feeding electric power into an electric supply network and wind park | |
| CN104915737A (en) | Coordinating economic dispatching method for power system containing large-scale wind power based on risk control | |
| US20160211664A1 (en) | Dispatch Controller for an Energy System | |
| CN108054771B (en) | An energy storage system charge and discharge control method and system | |
| US20220407310A1 (en) | Intelligent energy management system for distributed energy resources and energy storage systems using machine learning | |
| CN104181895A (en) | Strategy for optimizing short-term and ultra-short-term coordination rolling schedules adapting to access of new energy resources | |
| CN107203855B (en) | Robust two-layer optimization model with real-time dispatching of wind power system and its transformation equivalent method | |
| CN106655279B (en) | The grid-connected active power real-time control method in power plant of meter and electricity trading program | |
| Yang et al. | Wind power cluster ultra-short-term prediction error correction method based on the load peak and valley characteristics | |
| CN107800153B (en) | A rolling robust scheduling method of electric heat energy for electric heat storage and wind power consumption | |
| CN112398115B (en) | A multi-time scale thermal power-photovoltaic-pumped storage combined optimal dispatching scheme based on improved model predictive control | |
| CN117767349A (en) | Optimized dispatching and coordinated control method and system for wind, solar, and new energy storage power plants | |
| CN115765004A (en) | AGC real-time scheduling method, system and equipment for flywheel-thermal power system | |
| CN108964121B (en) | Wind, light and water real-time control method considering water and power planning and power target in day before water and power | |
| CN108090632A (en) | Multi-time scale scheduling method for new energy grid-connected power system based on robust optimization | |
| CN112510703A (en) | Multi-energy access power grid optimized scheduling method considering coal consumption curve correction | |
| CN120355174A (en) | Balanced scheduling system for energy management and balanced scheduling method thereof | |
| CN114336592A (en) | Wind power plant AGC control method based on model predictive control | |
| CN107947246B (en) | Wind power generation index distribution and increased power evaluation method considering frequency modulation and increased power | |
| CN106251074B (en) | A method for establishing a coordinated grid ahead dispatch model and its calculation method | |
| Ji et al. | Research on short-term optimization and scheduling of multi-energy complementary systems based on forecast scenario dynamic correction |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Ye Lin Inventor after: Zhang Cihang Inventor before: Ye Lin Inventor before: Zhang Cihang |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |