Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
As shown in fig. 1, the present embodiment provides a resource scheduling optimization method considering an inter-time payload uncertainty, including the following steps:
step 1: acquiring historical power system statistical data, describing the uncertainty of the net load at a moment point by intervals, and describing two aspects of net load fluctuation quantity and fluctuation speed to obtain an extreme fluctuation trajectory of the net load;
the transition process of the net load between adjacent time points in the conventional model is often simply represented by a connection line of the net load predicted values of two adjacent time points, as shown in fig. 2. In the drawingsDfcst t-1、Dfcst tRespectively representt-1 time andta predicted value of the time payload. The method defaults that if the uncertainty at the time point is effectively coped with, the transition process between the times is naturally satisfied. The description does not take into account the uncertainty of the time-to-time payload fluctuation and cannot guarantee that various possible transition processes of the time-to-time payload are effectively followed by the adjustable units in the system.
The prior art describes the extreme fluctuation trajectory of the net load between moments by the diagonal connecting the end points of the interval of the adjacent moments while describing the uncertainty of the moments by the interval, as shown in fig. 3. It is believed that the system is capable of handling arbitrary waves of time-of-day net load if it can handle these extreme trajectoriesAnd (4) dynamic condition. In FIG. 3Dfcst t、Dmax t、Dmin tRespectively representing the predicted value of the net load at the time point t and the upper limit and the lower limit of the uncertainty interval. Extreme fluctuating trajectories based on diagonal lines in the diagram, i.e. slaveDmin t-1 toDmax tAnd a varying trajectory fromDmax t-1 toDmin tThe change trajectory of (2).
This extreme fluctuation trajectory based on diagonal lines takes into account the uncertainty of the inter-time payload fluctuation, compared to the inter-time payload transition process description from the predicted value to the predicted value. However, the selection of such extreme fluctuation trajectory is independent of the net load limit fluctuation speed and may be too conservative or aggressive. For example, if the net load fluctuation speed corresponding to an extreme trajectory exceeds the net load actual limit fluctuation speed, the trajectory is too aggressive, and the corresponding backup is provided, which may result in waste of resources and decrease the economy of system operation. On the contrary, if the net load fluctuation speed corresponding to a certain extreme trajectory is lower than the actual net load limit fluctuation speed, the trajectory is too conservative, and the spare provided for the trajectory is not enough to deal with the actual net load fluctuation, thereby threatening the safe operation of the system.
Based on the single slot economic scheduling, an extreme fluctuation trajectory based on the net load limit fluctuation speed is given, as shown in fig. 4. The net load on the extreme fluctuation trajectory in the graph is always in the extreme fluctuation speed state. It is believed that as long as the net load varies within the two extreme fluctuation trajectories, both can be effectively dealt with.
But the method is only suitable for the case that the sending point of the extreme fluctuation trajectory is known, and is not suitable for the multi-period optimization problem. In addition, the influence of branch power flow constraint under an extreme fluctuation trajectory is not considered in the model, and the transferability of backup needed for dealing with uncertainty of net load fluctuation cannot be guaranteed.
As one or more embodiments, in step 1, the mining and characterizing the net load extreme fluctuation trajectory from the two aspects of the net load fluctuation quantity and the fluctuation speed by describing the net load uncertainty at the time point by the interval in the embodiment includes:
and carrying out probability statistics on the fluctuation speed of the net load among the moments according to historical statistical data to obtain the probability distribution of the fluctuation speed of the net load among the moments.
Specifically, the historical power system statistics include: and performing data statistics on the upper fluctuation speed and the lower fluctuation speed of the net load at the extreme end of the net load at 24 moments on different typical days in different seasons to obtain a probability curve corresponding to the extreme fluctuation speed of the net load at the moments, and intercepting the extreme fluctuation speed of the net load at the moments under certain confidence coefficient according to different conservative degrees (requirements) of scheduling optimization decision makers for coping with the extreme fluctuation speed as input data.
In this embodiment, a normal distribution is taken as an example, as shown in fig. 5, and in other embodiments, other probability distributions may be used.
And intercepting the probability distribution according to the given confidence coefficient alpha, thereby obtaining the upper and lower fluctuation speeds of the net load limit under the given confidence coefficientUR max AndDR max the calculation is shown below.
Describing the net load uncertainty at a moment point by an interval, and depicting the net load fluctuation between the moments from two angles of fluctuation quantity and fluctuation speed to obtain an extreme fluctuation trajectory, which specifically comprises the following steps:
this embodiment assumes that the fluctuating trajectory of the payload is linear or piecewise linear.
Based on a given extreme fluctuation speed, four extreme fluctuation trajectories can be plotted: upper fluctuation trajectory corresponding to limit fluctuation amountsUpper fluctuation trajectory corresponding to limit fluctuation speeds2, lower fluctuation trajectory corresponding to limit fluctuation speeds3, and a lower fluctuation trajectory corresponding to the number of limit fluctuationss4, the specific information is shown in fig. 6.
If these four extreme fluctuation trajectory systems are all capable of handling, the system can handle net load fluctuations at any time.
When the net load is int-1, the end point on the uncertainty interval of the time is first in the limit speed directionUp wave and down wave at limit speed totWhen the upper end point of the uncertainty interval of the time is positioned, the absolute value of the upward fluctuation is maximum, namely the net load value corresponding to 5 points, so that the upward fluctuation trajectory corresponding to the limit fluctuation quantity can be determineds1. The trajectory pair systemt-1 totThe upper limit of the unit operation capacity between the moments is the most demanding.
When the net load is intThe lower end point of the uncertainty interval at the moment-1 fluctuates upward at the limiting speed and then fluctuates downward at the limiting speed totThe relative magnitude of the upward fluctuation is maximized at the upper end of the uncertainty interval of the time of day, i.e., the difference in net load between point 2 and point 6, from which the upper fluctuation trajectory corresponding to the limit fluctuation speed can be determineds2. The trajectory pair systemt-1 totThe uphill resource demand between moments is the most demanding.
When the net load is intThe end point on the uncertainty interval at the moment-1 fluctuates first downwards at the limiting speed and then upwards at the limiting speed totAt the lower end of the uncertainty interval of the time of day, the relative magnitude of the downward fluctuation is maximized, i.e., the difference in net load between point 1 and point 8, from which the lower fluctuation trajectory corresponding to the limit fluctuation speed can be determineds3. The trajectory pair systemt-1 totThe down hill climbing resource requirements between the moments are most demanding.
When the net load is intThe lower end point of the uncertainty interval at the moment-1 fluctuates downwards at the limiting speed and then upwards at the limiting speedtWhen the uncertainty interval of the time is lower than the end point, the absolute value of the downward fluctuation is the maximum, namely the net load value corresponding to 7 points, so that the downward fluctuation trajectory corresponding to the limit fluctuation quantity can be determineds4. The trajectory pair systemt-1 totThe lower limit of the unit operation capacity between the moments is the most strict.
And 2, step: constructing a resource scheduling optimization model by combining branch power flow constraints under various extreme fluctuation trajectories of net loads with the minimum total cost as an optimization target;
in step 2, the optimization goal of minimizing the total cost is specifically as follows:
the unit combination model considering the uncertainty of the time-interval net load fluctuation specifically expresses the following objective functions:
wherein,
C oper including the unit operating cost, start-up cost and shut-down cost,
C res for up-regulation and down-regulation of system standby cost
;C i,t (
p i,t ,
u i,t )
Is composed ofFirst, the
iThe operation cost function of each unit is expressed by a three-section piecewise linear function; in the formula,
i,
trespectively indexing the unit and the optimization moment;
N G the number of the units;
N T the number of total time segments optimized for a study period;
、
are respectively a unit
iThe start-up and shut-down costs of (1);
p i,t as a unit
iIn that
tThe output of the time period;
、
the prices for standby are up and down respectively;
、
are respectively a unit
iIn the first place
tThe up and down standby number of the time interval;
u i,t as a unit
iIn a period of time
tThe operation state of (1 for operation, 0 for shutdown);
y i,t 、
z i,t indicating machine set
iIn a period of time
tAnd whether to start or stop the binary variable.
The branch flow constraints under the various extreme fluctuation trajectories of the net load comprise: based on predicted value power balance constraint, power balance constraint under any fluctuation scene, climbing constraint considering uncertainty of net load fluctuation between moments, standby constraint, line capacity constraint and other constraints, the specific construction process is as follows:
1) Predicted value based power balance constraints:
equation (5) describes the total power balance of the system based on the predicted value, and equations (6) and (7) are given intAnd predicting the total wind power generation amount and the total expected demand of the load in a time period.
In the formula,
d,
windexes of the load nodes and the fans are respectively;
N D the number of load nodes;
N W the number of the fans;
is a fan
wIn that
tA predicted output power for the time period;
is composed of
tTime interval node
dA predicted value of the load on the system;
for wind power generation
tA predicted output power for the time period;
is composed of
tTotal predicted value of time interval load;
is net loaded at
tA predicted value of the time period.
2) Power balance constraint under arbitrary fluctuation scenarios:
equation (8) describes the power balancing at any time point in the optimization period under different fluctuation trajectories.
In the formula,
is composed of
sUnit under fluctuation scene
iIn that
tWithin a time period
τThe force at that moment;
is composed of
sUnder fluctuating conditions the load is
tWithin a time period
τThe demand of the moment;
is composed of
sWind power generation under fluctuating scene
tWithin a time period
τGenerating capacity at a moment;
is composed of
sNet load under fluctuating scenarios
tWithin a time period
τThe demand of the moment;
Tis the duration of a unit period, in hours.
3) Hill climbing constraints that take into account the uncertainty of the inter-time net load fluctuation:
as only four extreme fluctuation trajectories of the net load in a certain period are met, the system has the capability of coping with all possible fluctuations of the net load in the period, and the robust coping with the fluctuation of the net load in the period is realized.
Thus the index in equation (8)sGet onlys1、s2、s3、s4 four cases are enough.
a) For the tracks1, it needs to ensure that the power balance constraint of any point on the trajectory and the climbing constraint between any two points are satisfied. But since only the trajectory form of the linear variation is considered, only the power balance constraint at points 1, 5 and 3 and the hill climbing constraint from point 1 to point 5 and from point 5 to point 3 need to be considered.
The power balance at point 1 can be expressed as:
in the formula,
to represent
s1 unit under fluctuating scene
iIn that
t-1 output at a time point;
to represent
t-a payload value corresponding to an upper limit of the payload interval at
time 1;
to represent
t-a time of day payload prediction value;
to represent
t-the difference between the upper net load limit at
time 1 and the predicted value, i.e. the amount of fluctuation in net load;
is at the same time
t-1 time of day node
dTo take into account the amount of up-fluctuation of the net load.
The power balance at point 3 is similar to the expression at point 1, and only the subscript in equation (9) is usedt-1 is changed totAnd (4) finishing.
The power balance at point 5 can be expressed as:
wherein,
to represent
s1 unit under fluctuating scene
iIn that
tWithin a time period
τ1
* The force at the time point (i.e., point 5);
to represent
s1 in a fluctuating scene
tWithin a time period
τ1
* Net load value at time point 5;
to represent
tThe net load limit climb speed within a time period;
represent
s1 in a fluctuating scene from
t-the duration of
point 1 to point 5 at
time 1,
can be calculated from the following formula:
wherein,
to represent
tThe net load limit climb-down speed over a period of time. In the formula (11)
The variable is a unique variable and can be conveniently calculated in advance.
The unit ramp constraints from point 1 to point 5 can be expressed as follows:
wherein,UR i andDR i respectively indicating unitsiUp-and down-climb speed of (a).
The unit ramp constraints from point 5 to point 3 may be expressed as follows:
wherein,
to represent
s1 unit under fluctuating scene
iIn that
tThe force applied at the moment.
In addition, in the case of the present invention,sthe output of each generator when the power at points 1, 5 and 3 is balanced under the fluctuation scene 1 is also limited by the capacity thereof, which is specifically expressed as follows:
wherein,
and
respectively indicating units
iLower and upper limits of the output.
b) For the tracks2, the power balance constraint of any point on the trajectory and the climbing constraint between any two points need to be ensured. But since only the trajectory form of the linear variation is considered, only the power balance constraint at points 2, 6 and 3 and the hill climbing constraint from point 2 to point 6 and from point 6 to point 3 need to be considered.
The power balance at point 2 can be expressed as:
in the formula,
represent
s2 Unit under fluctuating scene
iIn that
t-1 output at a time point;
to represent
t-a payload value corresponding to a lower limit of the payload interval at
time 1;
to represent
tThe difference between the lower net load limit at
time 1 and the predicted value, i.e. the amount of fluctuation under net load.
The power at point 3 is balanced ats1, already listed under the fluctuation scene, will not be described again.
The power balance at point 6 can be expressed as:
wherein,
to represent
s2 unit under fluctuating scene
iIn that
tWithin a time period
τ2
* The force at the time point (i.e., point 6);
represent
s2 in a fluctuating scene
tWithin a time period
τ2
* Net load value at
time point 6;
to represent
s2 in a fluctuating scenario from
t-the duration of point 2 to
point 6 at
time 1,
can be calculated from the following formula:
in the formula,
the variable is a unique variable and can be conveniently calculated in advance.
The unit ramp constraints from point 2 to point 6 may be expressed as follows:
the unit ramp constraints from point 6 to point 3 may be expressed as follows:
wherein,
to represent
s2 unit under fluctuating scene
iIn that
tThe force applied at the moment.
sThe output of each generator when the power at points 2, 6 and 3 is balanced under the fluctuation scene is limited by the capacity constraint, and the specific expression is as follows:
c) For the tracks3, the power balance constraint of any point on the trajectory and the climbing constraint between any two points need to be ensured. But since only the trajectory form of the linear variation is considered, only the power balance constraint at points 1, 8 and 4 and the hill climbing constraint from point 1 to point 8 and from point 8 to point 4 need to be considered.
The power at point 1 is balanced ats1 waveThe dynamic scenarios are listed and will not be described in detail.
The power balance at point 4 can be expressed as:
in the formula,
to represent
s3 unit under fluctuating scene
iIn that
tThe force of the time point;
to represent
tThe time corresponds to the payload value of the lower limit of the payload interval.
The power balance at point 8 can be expressed as:
wherein,
to represent
s3 unit under fluctuating scene
iIn that
tWithin a time period
τ3
* The force at the time point (i.e., point 8);
to represent
s3 in a fluctuating scene
tWithin a time period
τ3
* The net load value at time point 8;
to represent
s3 under the fluctuation scene
t-the duration of
point 1 to point 8 at
time 1,
can be calculated from the following formula:
in the formula,
the variable is a unique variable and can be conveniently calculated in advance.
The unit ramp constraints from point 1 to point 8 may be expressed as follows:
wherein,
to represent
s3 unit under fluctuating scene
iIn that
t-force at
time 1.
The unit ramp constraint from point 8 to point 4 can be expressed as follows:
s3 the output of each generator is limited by the capacity constraint when the power at the points 1, 8 and 4 is balanced under the fluctuation scene, and the specific expression is as follows:
d) For the tracks4, the power balance constraint of any point on the trajectory and the climbing constraint between any two points need to be ensured. But since only the trajectory form of the linear variation is considered, only the power balance constraint at points 2, 7 and 4 and the hill climbing constraint from point 2 to point 7 and from point 7 to point 4 need to be considered.
The power at point 2 is balanced ats2, which have already been listed in the fluctuation scenario, are not described in detail.
The power at point 4 is balanced ats3, already listed under the fluctuation scene, are not described in detail.
The power balance at point 7 can be expressed as:
in the formula,
to represent
s4 unit under fluctuating scene
iIn that
tWithin a time period
τ1
* The force at the time point (i.e., point 7);
represent
s4 fluctuation scene
tWithin a time period
τ1
* Net load value at
time point 7;
to represent
s4 under the condition of fluctuation
t-the duration of point 2 to
point 7 at
time 1,
can be calculated from the following formula:
in the formula,
the unique variable can be conveniently calculated in advance.
The unit ramp constraints from point 2 to point 7 may be expressed as follows:
wherein,
to represent
s4 unit under fluctuating scene
iIn that
t-force at
time 1.
The unit ramp constraints from point 7 to point 4 may be expressed as follows:
wherein,
to represent
s4 unit under fluctuating scene
iIn that
tThe force applied at the moment.
sThe output of each generator when the power at the points 2, 7 and 4 is balanced under the condition of 4 fluctuations is limited by the capacity of the generator, and the specific expression is as follows:
4) Standby constraint:
the formula (32) gives the unitiIn thattFor the provision of an up-regulation in the time interval for the handling of a net load fluctuation, equation (33) describes the unitiIn thattAnd the time interval is a down standby for coping with the configuration of the net load fluctuation.
5) And (3) line capacity constraint:
the line capacity constraint based on the expected payload is expressed as follows:
in the formula,
lindexing the line;
as a line
lThe capacity of (a);
、
a power generation load transfer factor.
The line capacity constraint under any fluctuation scenario is expressed as follows:
in the formula,
is composed of
sUnder fluctuating scene
tWithin a time period
τTime line
lThe trend of (2);
as a unit
iIn that
sUnder fluctuating scene
tWithin a time period
τThe output of the machine set at any moment;
to represent
sUnder fluctuating scene
tWithin a time period
τTime node
dThe net load demand of (c).
As long as the equations (35-36) are in extreme fluctuation scenarioss1、s2、 s 3 andsand 4, the requirement is met, so that the power flow transmitted by the line at any time under any fluctuation trajectory can be ensured not to exceed the limit value of the line capacity.
6) Other constraints are:
the proposed model also contains a unit start-stop logic constraint, a minimum start-stop time constraint, an initial state constraint, a unit output power constraint, etc., which are all taken into account in the model.
For the sake of brevity, detailed description is omitted, and references Wang Shibai, han Xueshan, yang Ming, li Benxin and Zhu Xingxu are specifically expressed, power system interval economic dispatch [ J ] taking intermittent characteristics into account, china motor engineering report, 2016,36 (11): 2966-2977.
And step 3: and solving the resource scheduling optimization model to obtain an optimization scheme of resource scheduling.
Similar to the conventional unit combination model, the unit combination model provided in this embodiment, which carefully considers uncertainty of fluctuation of a payload between times, still belongs to a Mixed Integer Linear Programming (MILP) problem, and the solving method is similar to the conventional unit combination solving method.
For example, a conventional commercial solver may be used, or classical heuristic algorithms such as simulated annealing, genetic algorithms, and neural networks may be used.
In order to verify the effectiveness of the model provided by the invention, calculation is carried out by taking an IEEE-RTS single-region system as an example. The IEEE-RTS single-region system comprises 24 nodes, 26 generator sets and 38 transmission lines, wherein 17 nodes are connected with loads, and the peak load is 2550MW. The parameters and load data of 26 generator sets are described in the literature. The unit starting cost and line data reference documents adopt a three-section type to linearize the unit cost in a sectional mode, and 10% of the highest marginal cost of each unit is taken as respective up-regulation and down-regulation standby prices. Six fans are added in the single-area system and are respectively positioned on nodes 3, 10, 13, 14, 15 and 18 of the system, and the capacity of each fan is 100 MW. For convenience, the output power curve of the fan has the same shape as the load demand curve.
The load disturbance interval is assumed to be +/-5% of the predicted node load value, and the wind power disturbance interval is +/-10% of the predicted wind power value. The total disturbance range of a node is the superposition of the load on the node and the wind-power disturbance interval. The limit fluctuation speed of the time-of-day payload is obtained based on historical statistical data. Here, the limit fluctuation speed of the net load between two sets of time instants is given. The first set of diagonal lines, each having a net load fluctuation speed between time points greater than the net load fluctuation amount between time points, are shown by line 1 in fig. 7. The net load limit fluctuation velocities in the second group of time intervals are all 500 MW/h, as shown by line 2 in fig. 7, which is greater than the limit fluctuation velocity corresponding to the diagonal of the net load fluctuation amount in some time intervals and less than the limit fluctuation velocity corresponding to the diagonal of the net load fluctuation amount in some time intervals.
The computer is configured to be a Win10 system, an Intel Core i5-11400k series, a main frequency of 3.0GHz, a memory of 8G and a dual gap of 0.1 percent.
To compare the differences between the different models, consider the following 3 cases:
case a: according to the traditional unit combination model, the form from a predicted value to a predicted value is considered in the time-interval net load transition process, namely the uncertainty of the time-interval net load fluctuation is not considered.
Case B: in the traditional unit combination model, the transition process of the time-interval net load adopts an extreme fluctuation trajectory based on a diagonal line, namely, the uncertainty of the time-interval net load fluctuation is considered.
Case C: the unit combination model adopts four extreme fluctuation trajectories for the time-interval net load fluctuation trajectory. Further subdivision, only scenes with the most severe requirements on climbing resources are considereds2、s3 is referred to as case C1; the model that considers four extreme fluctuation trajectories simultaneously is referred to as case C2.
When the first set of time-interval net load limit fluctuation speed information, i.e., the limit fluctuation speed corresponding to the black curve in fig. 7, is used, the above three cases are optimized, and the various costs are shown in table 1.
TABLE 1 cost comparison of the three models
As can be seen from table 1, the total cost for case B rose from $530273.2 to $538098.9, increasing by 1.47% compared to case a. This is because case B considers the extreme fluctuation of the diagonal-based inter-time payload, whereas case a considers only the inter-time payload transition from predictor to predictor. Under the condition that the net load fluctuation amount at the time point is the same, case B considers the uncertainty of the net load fluctuation between the times, and puts higher requirements on the climbing capacity of the unit, so that the operation cost of the system is increased, and the total cost is increased.
Both case B and case C1 take the net load limit fluctuation speed of time into account, and the net load limit fluctuation speed is selected differently, so that the limit fluctuation speed of case C1 is larger. As can be seen from table 1, the standby cost and the operating cost of case C1 increase, the start-stop cost decreases, and the total cost of case C1 increases by 2.71% compared to case B. This is because case C1 considers a scene of a severe net load fluctuation between times, has a more stringent requirement on the climbing capability of the unit, and needs to start more units and configure more spares, thereby causing the operation base point of the unit to further deviate from the most economical operation point, and causing both the system operation cost and the spare cost to increase. More units are always in an operating state, the number of times of starting and stopping the units is reduced, the starting and stopping cost of the system is reduced to some extent, and the total cost of the system is still obviously increased.
Case C2 is based on case C1, and simultaneously takes into account the extreme fluctuation scenario of the payload between four times, that is, more severe requirements are imposed on the unit climbing capability and the unit commissioning capacity. Therefore, more spare equipment needs to be reserved, the operation base point of the unit is further deviated from the most economic operation point, and the operation cost, the spare cost and the total cost of the unit are obviously increased.
In order to highlight the superiority of the model presented herein, a detailed comparison was developed for the case B and case C1 models, using a second set of time-interval payload limit fluctuation speed information, i.e., the limit fluctuation speed corresponding to curve 2 in fig. 7.
The total cost at each time after optimization is shown in fig. 8.
As can be seen from fig. 8, the single-session cost of case B is greater than the single-session cost of case C1 at some times, while the single-session cost of case B is less than the single-session cost of case C1 at other times. Further analysis was performed on the 6 th and 7 th periods as representative, and the results of the 6 th and 7 th periods are shown in tables 2 and 3, respectively.
TABLE 2 various cost comparisons of the two models at time 6
TABLE 3 various cost comparisons for the two models at time 7
During period 6, the time-to-time payload transitions from 1327.5 MW to 1350.0 MW. Case B takes the diagonal line of net load fluctuation as the extreme fluctuation trajectory, and the extreme upper and lower fluctuation speeds are 253.1 MW/h and 208.1MW/h, respectively, while case C1 has an extreme fluctuation speed of 500 MW/h.
It can be seen that case C1 should cope with the more severe time-interval payload fluctuation, and more spares need to be configured, and the total cost of case C1 during this time interval is higher than that of case B. At this time, the extreme fluctuation trajectory of the payload determined artificially according to the diagonal line in case B cannot cope with some extreme fluctuation scenes of the actual payload, and the standby resources configured by the system are insufficient, thereby threatening the safe operation of the system.
During period 7, the time-to-time payload transitions from 1350.0 MW to 1665.0 MW. Case B takes the diagonal line of net load fluctuation as the extreme fluctuation trajectory, and the extreme upper and lower fluctuation speeds are 574.7 MW/h and 55.38 MW/h, while the extreme fluctuation speed of case C is 500 MW/h. It can be seen that case B needs to cope with the more severe time-interval payload fluctuation, and needs to be configured with more spares, and the total cost of the case B system is higher than that of case C in the time interval. In case B, the limit fluctuation speed corresponding to the extreme trajectory determined manually according to the diagonal line is significantly higher than the actual limit fluctuation speed of the payload, and the spare resource redundancy configured by the system reduces the economic level of the system operation.
In summary, in case B, a diagonal line is used as an extreme fluctuation scene of the net load, which is too conservative or too aggressive, and the uncertainty of the fluctuation of the net load between times cannot be reasonably reflected, so that the climbing capacity in the system is insufficient or excessive, and the safe and economic operation level of the system is reduced. The model provided by the text uses the net load limit fluctuation speed obtained based on historical statistical data, the uncertainty of the net load fluctuation between moments can be described more precisely, the unit climbing capacity required by the net load fluctuation is configured reasonably, and the safe and economic operation of the power system is ensured.
The net load limit fluctuation speed at a time has a remarkable influence on the climbing capability requirement of the unit in the system, and further the total cost of the system is influenced. The trend of the total cost and the spare cost at different net load limit ramp rates is shown in fig. 9.
As can be seen from fig. 9, as the speed of the net load limit fluctuation increases, the system needs to allocate more spare resources to cope with the increase, and the spare cost and the total cost in the system increase. When the net load limit surge speed increases from 520 MW/h to 560 MW/h, the system requires a new set of switches to increase the backup supply to cope with the net load surge, which can cause a significant increase in the total and backup costs.
Line capacity constraints can affect the capacity of the spare transport that a unit within the system configures to account for the uncertainty in the net load fluctuations between times. For this reason, taking case C2 as an example, the case without considering the line capacity constraint under any fluctuation scenario, i.e. constraints (35) - (36), is referred to as case C2 relax Thereafter, case C2 and case C2 were compared relax The result of (1). To analyze the effects of line capacity, the line capacity is multiplied by a scaling factor. As the coefficient gradually increased from 0.7 to 1.0, the total cost of the two models varied as shown in fig. 10.
As can be seen from fig. 10, for any one of the curves, when the line capacity is from 0.7
Gradually increases to 0.8
The total cost of both models is gradually reduced. This reflects the limiting effect of line capacity constraints on the optimal configuration of system resources. When the circuit is onCapacity from 0.8
Gradually increase to 0.9
Case C2
relax The line capacity in the model is in a relaxed state and the total cost of the system no longer changes, while the total cost of case C2 continues to decrease gradually. This reflects that line capacity constraints can affect the spare capacity of units within the system configured to handle the uncertainty of the inter-time payload fluctuation. Comparing the two lines in fig. 10, it can be seen that the total cost of case C2 is always higher than that of case C2
relax The difference of the total cost of the system reflects the limit of the line capacity constraint on the resource optimization configuration under the fluctuation scene. That is, the line capacity constraint in the fluctuation scenario is not considered, the obtained result is too optimistic, and the scheduling result may cause line out-of-limit in some extreme scenarios, and the safe operation of the system is jeopardized.
The calculation times of the proposed model and the existing models are shown in table 4. As can be seen from table 4, the computation times for all models are of the same order of magnitude. The model provided by the invention only upgrades the original climbing constraint, and the complexity of the model is not obviously increased.
TABLE 4 comparison of calculated times for the three models
Aiming at the uncertainty of the net load fluctuation between moments, the invention excavates and describes the extreme fluctuation trajectory of the net load from two angles of the net load fluctuation quantity and the fluctuation speed, and provides a unit combination model for carefully considering the uncertainty of the net load fluctuation between the moments. In the model, the net load uncertainty at a time point is described by intervals, and four net load extreme fluctuation trajectories are described for the net load fluctuation at time. If the system can cope with the four extreme trajectories, the system can guarantee robust coping to the time-interval net load fluctuation. The net load limit fluctuation speed is obtained based on historical statistical data, and the problem that the net load limit fluctuation speed value is too subjective at moments based on diagonal lines is solved. Meanwhile, branch power flow constraints under various net load extreme fluctuation trajectories are also brought into the model, and the standby transferability required by net load fluctuation is ensured. Finally, based on an IEEE-RTS24 node example, the validity of the proposed model is verified.
Example two
The present embodiment provides a resource scheduling optimization system considering an inter-time payload uncertainty, including:
the net load extreme fluctuation trajectory delineating module is used for acquiring historical power system statistical data, describing the net load uncertainty at a moment point in intervals, and delineating the net load extreme fluctuation trajectory from the two aspects of net load fluctuation quantity and fluctuation speed;
the resource scheduling optimization model building module is used for building a resource scheduling optimization model by combining branch power flow constraints under various net load extreme fluctuation trajectories with the minimum total cost as an optimization target;
and the resource scheduling optimization model solving module is used for solving the resource scheduling optimization model to obtain an optimization scheme of resource scheduling.
The method for mining and describing the net load uncertainty at the time point by the interval comprises the following steps of:
carrying out probability statistics on the fluctuation speed of the net load between the moments based on historical power system statistical data to obtain probability distribution of the fluctuation speed of the net load between the moments;
intercepting the probability distribution according to a given confidence coefficient to obtain the upper and lower fluctuation speeds of a net load limit under the given confidence coefficient;
and describing the net load uncertainty at the time point by intervals, and describing an upper fluctuation trajectory and a lower fluctuation trajectory corresponding to the limit fluctuation quantity and an upper fluctuation trajectory and a lower fluctuation trajectory corresponding to the limit fluctuation speed on the basis of the upper fluctuation speed and the lower fluctuation speed of the net load limit and the quantity of the net load fluctuation.
The method for describing the net load uncertainty at the time point by the interval and describing the upper and lower fluctuation loci corresponding to the limit fluctuation quantity and the upper and lower fluctuation loci corresponding to the limit fluctuation speed on the basis of the upper and lower fluctuation speeds of the net load limit and the net load fluctuation quantity comprises the following steps:
when the net load is intThe end point on the uncertainty interval at the moment-1 fluctuates upward at the limiting speed and then fluctuates downward at the limiting speed totWhen the upper end point of the uncertainty interval of the moment is reached, the absolute numerical value of the upward fluctuation is maximum, and an upward fluctuation trajectory corresponding to the limit fluctuation quantity is determined;
when the net load is intThe lower endpoint of the uncertainty interval at time 1 fluctuates upward at a limit speed and then downward at the limit speed totWhen the upper end point of the uncertainty interval of the moment is reached, the relative value of the upward fluctuation is maximum, and an upper fluctuation trajectory corresponding to the limit fluctuation speed is determined;
when the net load is intThe end point on the uncertainty interval at the moment-1 fluctuates first downwards at the limiting speed and then upwards at the limiting speed totWhen the lower end point of the uncertainty interval of the moment is reached, the relative numerical value of downward fluctuation is maximum, and a lower fluctuation trajectory corresponding to the limit fluctuation speed is determined;
when the net load is intThe lower end point of the uncertainty interval at the moment-1 fluctuates downwards at the limiting speed and then upwards at the limiting speedtAnd when the lower end point of the uncertainty interval of the time is reached, the absolute numerical value of the downward fluctuation is maximum, and a lower fluctuation trajectory corresponding to the limit fluctuation quantity is determined.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.