CN105279707B - A kind of random production analog method considering load and wind-powered electricity generation temporal characteristics - Google Patents

Abstract

The invention discloses a stochastic production simulation method that comprehensively considers load and wind power timing characteristics, and is characterized in that it is aimed at the timing intermittency, anti-peak regulation and non-schedulability of the output of wind turbines, and the fact that traditional stochastic production simulation mostly ignores load According to the timing characteristics, four unit models are proposed according to the different operation modes of the generating units, and the stochastic production simulation is carried out. The invention can quickly and reasonably study the influence of wind power timing characteristics and various operating factors on random production simulation under different operating states of the generator set, thereby establishing a more comprehensive mathematical model and improving calculation efficiency.

Description

A Stochastic Production Simulation Method Considering Load and Wind Power Sequential Characteristics

technical field

The invention relates to the field of stochastic production simulation of power systems, in particular to a stochastic production simulation method that comprehensively considers load and wind power timing characteristics.

Background technique

Power system stochastic production simulation is an important method and tool for power system planning and operation optimization. It is an algorithm that calculates the power generation of each power plant, the production cost of the power system, and the reliability index of each power plant under the optimal operation mode by optimizing the production of the generating set and taking into account the randomness of the generating set and the electric load. It is often used in power system power planning, reliability assessment, unit combination and power market. At present, the methods of stochastic production simulation in power system mainly include simulation method and analytical method. The simulation model is simple and adaptable, but the calculation results have obvious statistical properties, and the calculation efficiency is not high. The analytical method is to establish a mathematical model of the power system under certain simplified assumptions. It has a clear physical concept, an accurate mathematical model, and high calculation efficiency. However, the traditional model considers fewer factors and is difficult to describe the various operating states of the unit. Therefore, it is of great significance to establish a unit model that can consider various operating factors for studying the influence of various operating factors.

The most important thing in stochastic production simulation of power system including wind power is to determine the optimal load position of wind power on the load curve. At present, there are two main ways to deal with wind power: negative load model and multi-state unit model. Equivalenting wind turbines to conventional multi-state unit models can show the probability of wind turbine output, but due to the loss of time series information, it is difficult to accurately describe the impact of wind power generation anti-peaking characteristics; wind power is treated as a negative load, directly Subtracting from the load has certain feasibility for small-scale wind power grid connection. With the increase of wind power scale, it is of great significance to establish a large-scale wind power grid-connected model.

Contents of the invention

In order to avoid the shortcomings of the above-mentioned prior art, the present invention provides a stochastic production simulation method that comprehensively considers load and wind power timing characteristics, in order to quickly and reasonably study wind power timing characteristics and generating units in different operating states The impact of various operating factors on random production simulation can establish a more comprehensive mathematical model and improve calculation efficiency.

The present invention adopts following technical scheme for reaching above-mentioned purpose of the invention:

The present invention comprehensively considers the stochastic production simulation method of load and wind power sequence characteristics, is applied in the power system, and is characterized in that it is carried out according to the following steps:

Step 1. Obtain the generator set data, wind power resource data and time-series load data of the power system in the research period T, and obtain the original time-series load curve from the time-series load data, and record it as Load1;

Step 2. Establish four unit models, including: base load unit model, start-stop peak-shaving unit model, pressure load peak-shaving unit model, and two-stage start-stop peak-shaving unit model;

Step 3. Preliminarily correct the original time-series load curve Load1 from the generator set data and time-series load data to obtain a preliminary corrected time-series load curve, which is denoted as Load2;

Step 4. Set the wind power correction load times as n, and initialize n=1;

Step 5. Obtain the forced output of the heating unit from the data of the generating set, and modify the preliminary correction time-series load curve Load2 according to the wind power resource data and the forced output of the heating unit to obtain the nth wind power acceptance W _n and the N time sequence net load curve Load3 _n ;

Step 6, converting the nth time series net load curve Load3 _n into the nth equivalent continuous load curve, and calculating the initial cumulative probability at the load value X on the nth equivalent continuous load curve P _0,n (X), initial accumulative frequency F _0,n (X), initial low battery expected value EENS _0,n (X) and initial maximum growth rate

Step 7. Determine the initial spinning reserve capacity of the power system according to the operating requirements of the power system and the initial ramp capacity value

Step 8. Obtain the total number of times I is put into operation from the data of the generating set; set the number of times of putting into operation as i, and initialize i=1;

Step 9. According to the constraints of the power system and aiming at the lowest power generation cost, judge whether it is necessary to re-correct the wind power. If necessary, assign n+1 to n and return to step 5 for execution. If not, determine the i-th The generator set number put into operation for the second time;

Step 10, according to the type of the generator set put into operation for the ith time and the condition of the base load, determine the unit model to which the generator set put into operation for the ith time belongs;

When i=1, from the initial cumulative probability P _0,n (X) and the initial cumulative frequency F _0,n (X) to obtain the i-th commissioning generator set, the power system needs the generating set to be put into operation transfer rate and transfer rates that do not require unit commissioning

When i≠1, the i-th cumulative probability P _i-1,n (X) and the i-1 cumulative frequency F _i-1,n (X) after the i-1th generator set is put into operation can be used to obtain the i-th When the generator set is put into operation, the power system requires the transfer rate of the generator set to be put into operation and transfer rates that do not require unit commissioning

Step 11. Correct and obtain the spinning reserve capacity of the power system after the ith generator set is put into operation according to the number of the generator set put into operation for the ith time and climbing capacity

Step 12, converting the unit model to which the generating unit put into operation for the ith time belongs to an equivalent two-state unit model or an equivalent three-state unit model;

Step 13. Perform stochastic production simulation on the equivalent two-state unit model or equivalent three-state unit model, and obtain the cumulative probability P _i,n (X) and cumulative frequency F _i,n of the generator set put into operation for the ith time. (X), the expected value of insufficient power EENS _i,n (X) and the maximum growth rate

Step 14. Accumulate the total operating capacity of the generator after the i-th operation, and judge whether i<I is true; if it is true, assign i+1 to i; and return to step 9; otherwise, it means that the random production simulation is completed .

The stochastic production simulation method of the present invention, which comprehensively considers load and wind power sequence characteristics, is also characterized by:

The generating sets described in step 1 include thermal power generating sets, hydroelectric generating sets and nuclear power generating sets; the generating set data includes the type, number, capacity, upper and lower limits of output, mean time between failures (MTTF), mean time to repair (MTTR) and consumption The amount of slight increase rate; the wind power resource data includes the output of wind turbines and the corresponding time period; the time series load data is the hourly load arranged in chronological order.

In step 2, four unit models are established according to the different operating states of the unit during the stochastic production simulation process, the condition with base load and the capacity of the segmented unit;

The base load unit model: represents a non-segmented continuous operation unit operating at the base load with rated capacity;

The start-stop peak-shaving unit model: represents a non-segmented intermittent operation unit whose rated capacity is above the base load for peak-shaving operation;

The model of the pressure load peak-shaving unit: represents a segmented continuous operation unit in which the first segment operates at the base load with the minimum technical output, and the second segment performs peak-shaving operation above the base load;

The two-segment start-stop peak-shaving unit model: represents a segmented intermittent operation unit whose two segments are both above the base load for peak-shaving operation.

The nth wind power acceptance W _n mentioned in step 5 is obtained according to the following steps:

Step a, set the variable t, and initialize t=1; record the load value of the preliminary corrected time series load curve Load2 at time t as Load2t;

Step b, judging whether n=1 is established, if established, then go to step c; otherwise, go to step d;

Step c, judging whether time t is satisfied: Load2t-forced output >= wind turbine generating capacity;

If it is satisfied, then let the wind power acceptance at time t=the power generation of wind turbines; otherwise, let the wind power acceptance at time t=Load2t-forced output; go to step f;

Step d, taking λ as the wind abandonment ratio; judging whether the time t is satisfied: Load2t-(forced output+λ×wind power installed capacity×n)>=wind turbine generating capacity;

If it is satisfied, set the wind power acceptance amount at time t = wind power generating capacity; otherwise, go to step e;

Step e. Determine whether Load2t-(forced output+λ×wind power installed capacity×n)>0 is true, and if it is true, then set wind power acceptance at time t=Load2t-(forced output+λ×wind power installed capacity×n); otherwise , so that wind power acceptance at time t = wind power generating capacity;

Step f, judging whether t=T is true, if true, it means that the nth wind power acceptance amount W _n is obtained; otherwise, assign t+1 to t, and return to step b.

Step 9 is to judge whether it is necessary to re-correct the wind power according to the following steps:

Step (1), judging whether i>1 and the total operating capacity after the i-th putting into operation of the generator is greater than the maximum load of the power system is true, if true, select the unit or unit group with the smallest consumption micro-increase rate put into operation, and go to step (7); otherwise, go to step (2);

Step (2), judging whether there are forced output units or unit segments among all the units that have not been put into operation, if so, select the forced output unit or unit segment with the smallest consumption micro-increase rate to put into operation, and go to step (7); Otherwise, go to step (3);

Step (3), judging whether the spinning reserve constraint and the climbing constraint condition of the power system are satisfied, if satisfied, then go to step (5); otherwise, go to step (4);

Step (4), judging whether there is the first section of the sectioned unit among all the units not put into operation, if it exists, select the first section with the smallest consumption micro-increase rate to put into operation, and go to step (7) ; Otherwise, it means that wind power needs to be re-corrected;

Step (5), find the unit or unit segment with the smallest consumption micro-increase rate among all the units not put into operation, and judge whether the unit or unit segment is the second subsection of the segmented unit; if so, go to step ( 6); otherwise, go to step (7);

Step (6), pre-correcting the spinning reserve capacity and ramping capacity of the power system, and judging whether the pre-corrected spinning reserve constraint and ramping constraint conditions are satisfied, and if so, put into the second Segmentation, and go to step (7); otherwise, turn to step (5) under the condition that the minimum specific consumption unit is the second subsection of the subsection unit;

Step (7), indicating that wind power does not need to be re-corrected.

The constraints of the power system described in step 9 include the spinning reserve constraints and ramp constraints shown in equations (1) and (2):

In formula (1) and formula (2): Indicates the spinning reserve capacity of the power system after the generator set is put into operation for the ith time; SPR ₀ is the spinning reserve capacity demand of the system; Indicates the load-increasing capacity of the power system after the generator set is put into operation for the ith time; t ₀ is the time required to put in the spinning reserve capacity; Rotate reserve capacity requirements for incidents; is the demand for load growth spinning reserve capacity; X _i is the operating point of the generator set put into operation for the ith time on the equivalent continuous load curve; Indicates the maximum load-increasing rate of the generator set put into operation for the _ith time at the point Xi; Δt represents the ramp-up time.

In step 10, the transfer rate of the unit that needs to be put into operation and transfer rates that do not require unit commissioning Calculate according to formula (3) and formula (4):

In formula (3) and formula (4): F _i-1,n (X _i ) represents the cumulative probability of load value X=X _i after the i-1th generator set is put into operation; P _i-1,n ( X _i ) represents the cumulative frequency at the load value X=X _i after the generator set is put into operation for the i-1th time.

In step 11, modify the spinning reserve capacity as follows and climbing capacity

(a) If the generating unit put into operation for the i-th time is the first section of the segmented unit, and its capacity is C _d ⁱ , then use formula (5) and formula (6) to make corrections:

(b) If the generating unit put into operation for the i-th time is the second section of the subsection unit, and its capacity is X _d ⁱ , then use formula (7) and formula (8) to make corrections:

(c) If the generator set put into operation for the i-th time is a non-segmented unit, use formula (9) and formula (10) to make corrections:

In formula (5) - formula (10), C _e ⁱ represents the rated capacity of the generator set put into operation for the ith time; Indicates the spinning reserve capacity of the power system after the generator set is put into operation for the i-1th time; Indicates the climbing capacity of the system after the i-1th generator set is put into operation.

Cumulative probability P _i,n (X), cumulative frequency F _i,n (X), expected battery deficit EENS _i,n (X) and maximum growth rate as described in step 13 is obtained as follows:

1) If the generating unit put into operation for the i-th time is a non-segmented unit or the first segment of a segmented unit, the cumulative probability P _i,n ( X), cumulative frequency F _i,n (X) and expected value of insufficient power EENS _i,n (X):

P _i,n (X)=P _i-1,n (X)·(1-FOR ⁱ )+P _i-1,n (XC ⁱ )·FOR ⁱ (12)

F _i,n (X)＝F _i-1,n (X)·(1-FOR ⁱ )+F _i-1,n (XC ⁱ )·FOR ⁱ +u _e ⁱ ·FOR ⁱ ·[P _{i- 1,n} (XC ⁱ )-P _i-1,n (X)] (13)

EENS _i,n (X)＝EENS _i-1,n (X)·(1-FOR ⁱ )+EENS _i-1,n (XC ⁱ )·FOR ⁱ (14)

In formula (12), formula (13) and formula (14): X represents the load value on the equivalent continuous load curve; P _i,n (X), F _i,n (X) and EENS _i,n (X) respectively represent the cumulative probability, cumulative frequency and expected value of insufficient electricity on the equivalent continuous load curve of the i-th unit after it is put into operation; P _i-1,n (X), F _i-1,n (X) and EENS ⁱ _-1,n (X) represent the cumulative probability, cumulative frequency, and expected value of electricity shortage on the equivalent continuous load curve of the i-1 unit after it is put into operation, respectively; The probability that the output of the equivalent two-state unit of the generator set put into operation for the i time is zero; u _e ⁱ represents the equivalent repair rate of the equivalent two-state unit; C ⁱ represents the generator set or unit group put into operation for the i time The capacity of the segment, if the generator set put into operation for the i-th time is a non-segmented unit, then C ⁱ =C _e ⁱ , if the generator set put into operation for the i-th time is the first segment of a segmented unit, then C ⁱ = C _d ⁱ .

If the generator set put into operation for the i-th time is the second segment of the segmented unit, use formula (15), formula (16) and formula (17) to perform deconvolution operation to obtain the cumulative probability after deconvolution operation P ₁ ⁱ (X), cumulative frequency F ₁ ⁱ (X) and expected value of battery shortage EENS ₁ ⁱ (X): Reuse formula (18), formula (19) and formula (20) to obtain cumulative probability P _i,n (X), cumulative frequency F _i,n (X) and expected value of insufficient power EENS _i,n (X):

P ₁ ⁱ (X)=[P _i-1,n (X)-P ₁ ⁱ (XC _d ⁱ )·FOR ⁱ ]/(1-FOR ⁱ ) (15)

F ₁ ⁱ (X)＝[F _i-1.n (X)-F ₁ ⁱ (XC _d ⁱ )·FOR ⁱ -u _e ⁱ ·FOR ⁱ ·(P ₁ ⁱ (XC _d ⁱ )-P ₁ ⁱ (X))]/(1-FOR ⁱ ) (16)

EENS ₁ ⁱ (X)＝[EENS _i-1,n (X)-EENS ₁ ⁱ (XC _d ⁱ )]/(1-FOR ⁱ ) (17)

P _i,n (X)＝P ₁ ⁱ (X)·AV ⁱ +P ₁ ⁱ (XX _d ⁱ )·DFOR ⁱ +P ₁ ⁱ (XC _e ⁱ )·FOR ⁱ (18)

F _i,n (X)＝F ₁ ⁱ (X)·AV ⁱ +F ₁ ⁱ (XX _d ⁱ )·DFOR ⁱ +F ₁ ⁱ (XC _e ⁱ )·FOR ⁱ +AV ⁱ ·λ _T ⁱ ·[ P ₁ ⁱ (XC _e ⁱ )-P ₁ ⁱ (X)]+AV ⁱ ·ρ _- ^i,n ·[P ₁ ⁱ (XX _d ⁱ )-P ₁ ⁱ (X)]+DFOR ⁱ ·λ _D ⁱ ·[P ₁ ⁱ (XC _e ⁱ )-P ₁ ⁱ (XX _d ⁱ )]

(19)

EENS _i,n (X)＝EENS ₁ ⁱ (X) AV ⁱ +EENS ₁ ⁱ (XX _d ⁱ ) DFOR ⁱ +EENS ₁ ⁱ (XC _e ⁱ ) FOR ⁱ (20)

In formula (15) - formula (20): AV ⁱ , DFOR ⁱ , FOR ⁱ respectively represent the rated output probability, derated output probability and zero output probability of the equivalent three-state unit; λ _T ⁱ and λ _D ⁱ represent The forced outage rates of the rated state and the derated state in the equivalent three-state unit, P ₁ ⁱ (X), F ₁ ⁱ (X), and EENS ₁ ⁱ (X) represent the equivalent Cumulative probability, cumulative frequency and expected value of power shortage at the load value X on the continuous load curve.

Compared with the prior art, the beneficial effects of the present invention are reflected in:

1. By establishing a stochastic production simulation method that comprehensively considers load and wind power timing characteristics, the present invention fully considers the influence of wind power timing characteristics and various operating factors on stochastic production simulation under different operating states of the unit. A method for efficient and comprehensive stochastic production simulation of power systems. Different from the traditional random production simulation unit model, the present invention determines the unit model according to the type of unit in operation and the load position, and the same unit model changes due to different load positions, effectively reflecting the operation conditions of all units; in addition, the present invention can effectively Considering the system backup and ramp constraints, as well as the influence of factors such as unit minimum start and stop restrictions, startup failure, and ramp time constraints, the calculation speed is improved, which has important engineering application value.

2. The present invention adopts the time series load curve and extracts the relevant information of the time series load, which can effectively reflect the influence of the load time series change on the entire stochastic production simulation process.

3. The present invention establishes four kinds of multi-state unit models according to the different operating states (such as derating, intermittent, continuous) of the unit in the stochastic production simulation process, the load position (whether with base load) and the sub-unit units, which can be considered The start-up failure rate of the unit, the minimum start-stop time limit and the ramp rate limit fully consider the influence of the unit's own structure and various operating factors, and improve the rationality of the mathematical model of the analytical method.

4. The present invention focuses on the probability of forced outage during the power generation period of the unit, and uses the conditional forced outage rate CFOR of the unit failure under the condition that the system needs the unit to be put into operation to carry out the convolution operation, wherein the conditional forced outage rate CFOR is not only related to the already put into operation The outage characteristics of the operating units are also related to the load frequency transfer characteristics. Therefore, the dynamic calculation of the conditional forced outage rate CFOR can effectively remove the influence of faults on the convolution calculation when the system does not require the units to be put into operation.

5. The present invention focuses on anti-peak regulation and non-schedulability of wind power, and proposes a method of curtailing wind power in proportion to valley load when large-scale wind power is connected to the grid, fully retaining the timing characteristics of wind power output.

6. The present invention provides the system's spinning reserve and climbing constraint conditions, and checks each time the unit or unit is put into operation, ensuring that the spinning reserve and climbing constraints can be met at all times, improving the reliability of the system .

7. The present invention converts the four multi-state unit models into an equivalent two-state unit model and an equivalent three-state unit model, which simplifies the convolution operation of the unit model and improves calculation efficiency.

Description of drawings

Fig. 1 is the modeling flow chart of stochastic production simulation method comprehensively considering load and wind power timing characteristics in the present invention;

Fig. 2 is a two-state unit model (model 1) state space diagram of the base load unit in the present invention;

Fig. 3 is a five-state unit model (model 2) state space diagram of the start-stop peak-shaving unit model in the present invention;

Fig. 4 is the state space diagram of the four-state unit model (model 3) of the medium pressure load peaking unit model of the present invention;

Fig. 5 is a seven-state unit model (model 4) state space diagram of two-stage start-stop peak-shaving unit model among the present invention;

Fig. 6 is a schematic diagram of the load-shaving method of storage capacity hydropower peak-shaving successively in the present invention;

Fig. 7 is equivalent continuous load curve figure among the present invention;

Fig. 8 is a schematic diagram of wind power abandonment in valley load in the present invention;

Fig. 9 is a state space diagram of an equivalent two-state unit model in the present invention;

Fig. 10 is a state space diagram of an equivalent three-state unit model in the present invention.

Detailed ways

As shown in Figure 1, in this embodiment, the stochastic production simulation method that comprehensively considers the load and the timing characteristics of wind power is carried out as follows:

Step 1. Obtain the generator set data, wind power resource data and time-series load data of the power system in the research period T, and obtain the original time-series load curve from the time-series load data, which is recorded as Load1;

The generating sets in step 1 include thermal power generating sets, hydroelectric generating sets and nuclear power generating sets; generating set data includes the type of generating sets (segmentation), number, capacity, upper and lower limits of output, mean time between failures MTTF, mean time to repair MTTR And the slight increase rate of consumption; the wind power resource data includes the output of wind turbines and the corresponding time period; the time series load data is the hourly load arranged in chronological order. When calculating all numerical values containing time information, it is necessary to ensure that the unit is unified (for example, the unit of hour can be unified).

Step 2. Establish four unit models, including: base load unit model, start-stop peak-shaving unit model, pressure load peak-shaving unit model, and two-stage start-stop peak-shaving unit model;

Step 2 is based on the different operating states of the unit during the stochastic production simulation process (such as derating, intermittent, continuous), the condition with base load (whether with base load) and the capacity of the segmented unit (assuming that the capacity of the first segment of the segmented unit is equal to derating state output) to establish four unit models;

The base load unit model (two-state model) is shown in Figure 2: it represents a non-segmented continuous operation unit operating at the base load with rated capacity; its outage probability is the same as that of the traditional production simulation model, and the forced Calculation of the outage rate FOR (that is, the zero state probability of the unit).

The start-stop peak-shaving unit model (five-state model) is shown in Figure 3: it represents a non-segmented intermittent operation unit whose rated capacity is above the base load for peak-shaving operation; it needs to be converted into an equivalent two-state unit model for calculation .

The model of the pressure load peak-shaving unit (four-state model) is shown in Figure 4: it means that the first segment operates at the base load with the minimum technical output, and the second segment performs peak-shaving operation above the base load for segmental continuous operation Unit; when the system requires the unit to be put into operation, it needs to be transformed into an equivalent three-state unit model for calculation.

The two-stage start-stop peak-shaving unit model (seven-state model) is shown in Figure 5: it represents a segmented intermittent operation unit whose two segments are both above the base load for peak-shaving operation. It also needs to be transformed into an equivalent three-state unit model for calculation.

Step 3. Preliminarily correct the original time-series load curve Load1 based on the generator set data and time-series load data, and obtain the preliminary corrected time-series load curve, which is denoted as Load2;

The preliminary correction of the original time-series load curve Load1 in step 3 is mainly: the hourly load data arranged in chronological order, and the nuclear power and runoff hydropower that bear the base load are first processed according to the negative load model; for the system with a small proportion of hydropower, then press Sequential load shedding method arranges storage capacity hydropower for peak shaving to reduce the peak-valley difference of the load curve; if hydropower accounts for a large proportion, the storage capacity hydropower unit seeks the best working position on the load curve and performs convolution operation.

Successive load shedding method: get the maximum load according to the daily load diagram; subtract the maximum allowable output of the unit from the maximum load to determine the initial working position of the unit; as shown in Figure 6, if the area of the shaded part of the unit is greater than the given value , the working position will be raised; otherwise, the working position will be lowered. Let the load above the working position be carried by the unit, and take the maximum allowable output value for the output that is greater than the maximum allowable output period of the unit.

As shown in Figure 6, the step size of raising or lowering the working position is Δx=(EE _g )/T, T is the simulation time length, generally 24h, when |EE _g |≤ε meets the iteration requirements, ε is the set specified accuracy.

Step 4. Set the number of wind power correction loads as n, and initialize n=1; if wind power does not need to perform multiple valley load curtailment, then n=1 will not change; if wind power requires multiple valley load curtailment, then Every time Guhe abandons wind, the value of n will increase by 1, but since the number of times of abandoning wind cannot be determined in advance, the maximum value of n cannot be known in advance, and can only be determined after the entire random production process is completed (note What is interesting is that as long as the number of wind power corrections changes, all units need to be rescheduled to be put into operation, so the parameters of the units and the parameters of the system constraints are the values after the nth wind power correction load).

Step 5. Obtain the forced output of the heating unit from the data of the generating unit, and modify the preliminary correction time-series load curve Load2 according to the wind power resource data and the forced output of the heating unit (that is, the sum of the minimum technical output of all units), and obtain the nth wind power acceptance W _n and n time sequence net load curve Load3 _n ;

The nth wind power acceptance W _n in step 5 is obtained by the following steps:

Step a, set the variable t, and initialize t=1; record the load value of the preliminary corrected time series load curve Load2 at time t as Load2t;

Step b, judging whether n=1 is established, if established, then go to step c; otherwise, go to step d;

Step c, judging whether time t is satisfied: Load2t-forced output >= wind turbine generating capacity;

If it is satisfied, then let the wind power acceptance at time t=the power generation of wind turbines; otherwise, let the wind power acceptance at time t=Load2t-forced output; go to step f;

Step d, taking λ as the wind abandonment ratio; judging whether the time t is satisfied: Load2t-(forced output+λ×wind power installed capacity×n)>=wind turbine generating capacity;

If it is satisfied, set the wind power acceptance amount at time t = wind power generating capacity; otherwise, go to step e;

Step e. Determine whether Load2t-(forced output+λ×wind power installed capacity×n)>0 is true, and if it is true, then set wind power acceptance at time t=Load2t-(forced output+λ×wind power installed capacity×n); otherwise , so that wind power acceptance at time t = wind power generating capacity;

Step f, judging whether t=T is true, if true, it means that the nth wind power acceptance amount W _n is obtained; otherwise, assign t+1 to t, and return to step b.

Taking T=24 as an example, as shown in Figure 7, when n=1, after the initial valley load curtailment, the initial curtailed wind volume and the initial wind power acceptance capacity W ₁ in the period T can be obtained; when n>1, After several times of abandoning the wind in Guhe until the constraint conditions are met, the amount of abandoned wind and wind power acceptance W _n in the period T can be finally obtained;

Step 6. Convert the nth time series net load curve Load3 _n into the nth equivalent continuous load curve, and calculate the initial cumulative probability P _0,n at the load value X on the nth equivalent continuous load curve (X), initial cumulative frequency F _0,n (X), initial expected value of insufficient power EENS _0,n (X) and initial maximum growth rate

The load value in step 6 is the initial cumulative probability P _0,n (X) at X, the initial cumulative frequency F _0,n (X), the initial expected value of insufficient electricity EENS _0,n (X) and the initial maximum growth rate It is obtained according to the following steps:

As shown in Figure 8, take the continuous 24h data from 12:00 noon on a certain day:

Initial cumulative probability P _0,n (X), cumulative frequency F _0,n (X) and maximum load growth rate The calculation formula is as follows:

In formula (1)-formula (3): P _t (X), F _t (X) and is the accumulative probability, accumulative frequency and growth rate at the load value X at each moment, and its calculation formula is as follows:

In Formula (4)-Formula (6): x _t , x _t+1 are the load values at time t and t+1.

The expected value of the initial electricity shortage EENS _0,n (X) is the area of the shaded part in Figure 8; the load value X in the present invention represents a series of load points, and its value starts from 0 to ((maximum load+total capacity of the unit)/ΔX+1 ) · ΔX, the step size is ΔX (generally take the greatest common divisor of unit capacity).

Step 7. Determine the initial spinning reserve capacity of the power system (after the nth wind power correction load) according to the operating requirements of the power system and the initial ramp capacity value And climbing capacity demand, emergency reserve capacity demand and load growth reserve capacity demand;

Step 8. Obtain the total operation times I from the generator set data; set the number of operation units as i, and initialize i=1;

The total number of times of commissioning in step 8 I = the number of segmented units × 2 + the number of non-segmented units.

Step 9. According to the constraints of the power system and aiming at the lowest power generation cost, judge whether it is necessary to re-correct the wind power. If necessary, assign n+1 to n and return to step 5 for execution. If not, determine the i-th The generator set number put into operation for the second time;

Step 9 is to judge whether it is necessary to re-correct the wind power according to the following steps:

Step (1), judging whether i>1 and the total operating capacity of the generator after the i-th operation is greater than the maximum load of the power system is true, if true, select the unit or unit segment with the smallest consumption micro-increase rate Put into operation, and go to step (7); otherwise, go to step (2);

Step (2), judging whether there are forced output units or unit segments among all the units that have not been put into operation, if so, select the forced output unit or unit segment with the smallest consumption micro-increase rate to put into operation, and go to step (7); Otherwise, go to step (3);

Step (3), judging whether the spinning reserve constraint and the climbing constraint condition of the power system are satisfied, if satisfied, then go to step (5); otherwise, go to step (4);

Step (4), judging whether there is the first section of the sectioned unit among all the units not put into operation, if it exists, select the first section with the smallest consumption micro-increase rate to put into operation, and go to step (7) ; Otherwise, it means that wind power needs to be re-corrected;

Step (5), find the unit or unit segment with the smallest consumption micro-increase rate among all the units not put into operation, and judge whether the unit or unit segment is the second subsection of the segmented unit; if so, go to step ( 6); otherwise, go to step (7);

Step (6), pre-correct the spinning reserve capacity and ramping capacity of the power system, and judge whether the pre-corrected spinning reserve constraints and ramping constraints are satisfied, and if so, put into the second segment of the segmented unit , and go to step (7); otherwise, go to step (5) under the condition that the minimum specific consumption unit is excluded as the second section of the subsection unit;

Step (7), indicating that wind power does not need to be re-corrected.

The constraints of the power system in step 9 include spinning reserve constraints and ramp constraints as shown in equations (7) and (8):

In formula (7) and formula (8): Indicates the spinning reserve capacity of the power system after the i-th putting into operation of the generator set (after the nth wind power correction load). When no conventional generating set is put into operation, its value is the capacity of the cold standby set that can be put into operation quickly; SPR ₀ is the system Spinning reserve capacity requirement of , usually a given value; Indicates (after the nth wind power correction load) the increased load capacity of the power system after the ith generator set is put into operation. Make corrections; t ₀ is the time required to invest in spinning reserve capacity; For the accident spinning reserve capacity demand, that is, the accident spinning reserve capacity required by the system All arrive within t ₀ hours (if t ₀ =0.2h, it is required that the system can provide the required spinning reserve capacity value within 12 minutes); Spin reserve capacity demand for load growth, the average load increase rate required by load reserve is The right side of the inequality in equation (8) as a whole represents the climbing capacity requirement of the system, because It can be seen that its value is a variable; X _i is the operating point of the generator set put into operation for the ith time on the equivalent continuous load curve; Indicates (after the nth wind power correction load) the maximum load increase rate of the generator set put into operation for the _ith time at the operating point Xi; ).

In addition to considering the system spinning reserve constraints and ramp constraints, the relationship between the total operating capacity and the maximum load, the random failure of the unit, the unit startup failure probability P _S , the minimum start-stop time limit of the unit and the limit of the unit ramp rate TR (Take the start-stop peak-shaving unit model as an example, that is, the ramp-up time required from zero output state to full output state).

Step 10, according to the type of the generator set put into operation for the ith time and the condition of the base load, determine the unit model to which the generator set put into operation for the ith time belongs;

When i=1, it is obtained from the initial cumulative probability P _0,n (X) and initial cumulative frequency F _0,n (X) (after the nth wind power correction load) that the power system needs a generating set when it is put into operation for the first time Shipping transfer rate and transfer rates that do not require unit commissioning

When i≠1, it is obtained from the cumulative probability P _i-1,n (X) and the cumulative frequency F _i-1,n (X) of the i-1th time after the generator set is put into operation for the i-1th time (the nth After the wind power load is corrected for the second time), the power system needs the transfer rate of the unit to be put into operation when the generator set is put into operation for the ith time and transfer rates that do not require unit commissioning

The transfer rate that requires the unit to be put into operation in step 10 and transfer rates that do not require unit commissioning Calculate according to formula (9) and formula (10):

In formulas (9) and (10): F _i-1,n (X _i ) represents (after the n-th wind power correction load) the cumulative load value X=X _i after the i-1th generator set is put into operation Probability; P _i-1,n (X _i ) represents (after the nth wind power correction load) the cumulative frequency at the load value X=X _i after the i-1th generator set is put into operation.

Set the minimum continuous running time of the generator set and minimum economic downtime Assuming that the outage duration and uptime of the entire system obey the transfer rate that the power system needs to put the unit into operation (also known as the "demand rate") and transfer rates that do not require units to be commissioned (Also known as the "dispossession rate") exponential distribution, so:

In formula (11) and formula (12): p _u ⁱ is the running time of the generator set put into operation for the ith time greater than the minimum continuous running time The probability of ; p _d ⁱ is the outage time of the generator set put into operation for the ith time is greater than the minimum economic downtime The probability.

The start-stop limit constraints of the unit: the demand rate in the start-stop peak-shaving unit and the two-stage start-stop peak-shaving unit model needs to be and does not require Correct as follows:

The revised demand rate at this time and does not require Discounted on the original basis, the obtained unit model is the state space diagram of the unit during the entire stochastic production simulation process after considering the minimum start-up and shutdown restrictions of the unit, and the probability of each part is the position of the unit during the entire production simulation process. Probability of the state; corrected demand rate It represents the actual transfer rate of the unit from the standby state to the ramp state after adding the on-off time limit.

Step 11. According to the generator set number put into operation for the ith time, the spinning reserve capacity of the power system after the generator set put into operation for the ith time is corrected to obtain (after the nth wind power correction load) and climbing capacity

In step 11, the spinning reserve capacity is corrected as follows and climbing capacity

(a) If the generator set put into operation for the ith time is the first segment of the segmented unit, and its capacity is C _d ⁱ , at this time, the first segment of the unit is put into operation, and the second segment is in the spinning standby state, and the reserve capacity increases amount, the climbing capacity also increases, and the increased climbing capacity is at most 2 stages of capacity, then use formula (15) and formula (16) to correct:

(b) If the generator set put into operation for the ith time is the second segment of the segmented unit, its capacity is X _d ⁱ , and the second segment exits the spinning reserve state, the reserve capacity decreases, and the ramp capacity also decreases accordingly. The reduced The climbing capacity is at most 2 stages of capacity, then use formula (17) and formula (18) to modify:

(c) If the generator set put into operation for the i-th time is a non-segmented unit, and the operation of the unit has no effect on the reserve capacity and ramp capacity of the system, then use formula (19) and formula (20) to make corrections:

In formula (15) - formula (20), C _e ⁱ represents the rated capacity of the generating set put into operation for the ith time; Indicates (after the nth wind power correction load) the spinning reserve capacity of the power system after the i-1th generator set is put into operation; Indicates (after the nth wind power correction load) the climbing capacity of the system after the i-1th generator set is put into operation.

Step 12, converting the unit model of the generating unit put into operation for the ith time into an equivalent two-state unit model, as shown in Figure 9, or an equivalent three-state unit model, as shown in Figure 10;

The base load unit model and the start-stop peak-shaving unit model need to be transformed into an equivalent two-state unit model, and the pressure-load peak-shaving unit model and the two-stage start-stop peak-shaving unit model need to be transformed into an equivalent three-state unit model; the equivalent The state probabilities for a two-state unit model or an equivalent three-state unit model are calculated as follows:

1) The base-load unit model (two-state model) can be directly calculated using the forced outage rate FOR (ie, the zero-state probability of the unit). which is:

CFOR=FOR (21)

2) The start-stop peak-shaving unit model (five-state model) needs to be transformed into an equivalent two-state unit model for calculation. The outage probability is calculated using the conditional forced outage rate CFOR, which generally refers to the probability of unit failure when the system requires the unit to be put into operation; it should be noted that the ramp time constraint is considered in Figure 3. The probability that the output of the unit is zero in the case of operation, the climbing state in the figure is the state of the unit just started, and its output is zero. Therefore:

In formula (22): P3, P4, and P5 represent the probabilities of states 3, 4, and 5 in the unit model in the figure, respectively.

3) The model of the pressure load peak-shaving unit (four-state model) needs to be transformed into an equivalent three-state unit model for calculation when the system requires the unit to be put into operation. The three states of the equivalent three states are: the unit rated output state probability AV, the unit derated output state probability DFOR and the unit zero state probability FOR, and the corresponding calculation formula is as follows:

In formula (23): P1, P2, P3, and P4 represent the probabilities of states 1, 2, 3, and 4 in the unit model in the figure, respectively.

4) The two-stage start-stop peak-shaving unit model (seven-state model) is also equivalent to a three-state unit for calculation; in Figure 4, the first stage of climbing is the moment of just starting, and the output is approximately zero, and the third stage of climbing is just from the start. When the derating starts to increase the output, the output is approximately the derated output, and the equivalent three-state state probability is calculated as follows:

In formula (24): P3, P4, P5, P6, and P7 represent the probabilities of states 3, 4, 5, 6, and 7 in the unit model in the figure, respectively.

Step 13. Perform stochastic production simulation on the equivalent two-state unit model or the equivalent three-state unit model, and obtain the cumulative probability P _i,n (X) after the i-th operation of the generating unit (after the n-th wind power correction load) , cumulative frequency F _i,n (X), expected value of insufficient power EENS _i,n (X) and maximum growth rate And calculate the power generation, power generation cost and cold and hot start times of the generator set put into operation for the ith time;

Cumulative probability P _i,n (X), cumulative frequency F _i,n (X), expected value of battery shortage EENS _i,n (X) and maximum growth rate in step 13 is obtained as follows:

1) If the generating unit put into operation for the i-th time is a non-segmented unit or the first segment of a segmented unit, the cumulative probability P _i,n ( X), cumulative frequency F _i,n (X) and expected value of insufficient power EENS _i,n (X):

P _i,n (X)=P _i-1,n (X)·(1-FOR ⁱ )+P _i-1,n (XC ⁱ )·FOR ⁱ (26)

F _i,n (X)＝F _i-1,n (X)·(1-FOR ⁱ )+F _i-1,n (XC ⁱ )·FOR ⁱ +u _e ⁱ ·FOR ⁱ ·[P _{i- 1,n} (XC ⁱ )-P _i-1,n (X)] (27)

EENS _i,n (X)＝EENS _i-1,n (X)·(1-FOR ⁱ )+EENS _i-1,n (XC ⁱ )·FOR ⁱ (28)

In formula (26), formula (27) and formula (28): X represents the load value on the equivalent continuous load curve, with ΔX (generally taking the greatest common divisor of the capacity of all generator sets) as the interval, X=0,ΔX ,2·ΔX...(maximum load+P _i (X)/ΔX+1)·ΔX; P _i,n (X), F _i,n (X) and EENS _i,n (X) represent respectively (n After the first wind power correction load) the cumulative probability, cumulative frequency and expected value of power shortage on the equivalent continuous load curve after the i-th unit is put into operation at the load value X; P _i-1,n (X), F _{i-1, n} (X) and EENS _i-1,n (X) respectively represent (after the n-th wind power correction load) the cumulative probability and cumulative frequency and power shortage expectation; FOR ⁱ represents the probability that the output of the equivalent two-state unit of the i-th generator set is zero; u _e ⁱ represents the equivalent repair rate of the equivalent two-state unit; C ⁱ represents the i-th The capacity of the generator set or unit section in operation, if the generator set put into operation for the i-th time is a non-segmented unit, then C ⁱ =C _e ⁱ , if the generator set put into operation for the i-th time is the first One segment, then C ⁱ =C _d ⁱ .

If the generator set put into operation for the i-th time is the second segment of the segmented unit, use formula (29), formula (30) and formula (31) to perform deconvolution operation to obtain the cumulative probability after deconvolution operation P ₁ ⁱ (X), cumulative frequency F ₁ ⁱ (X) and expected value of battery shortage EENS ₁ ⁱ (X): Reuse formula (32), formula (33) and formula (34) to obtain cumulative probability P _i,n (X), cumulative frequency F _i,n (X) and expected value of insufficient power EENS _i,n (X):

P ₁ ⁱ (X)=[P _i-1,n (X)-P ₁ ⁱ (XC _d ⁱ )·FOR ⁱ ]/(1-FOR ⁱ ) (29)

F ₁ ⁱ (X)＝[F _i-1.n (X)-F ₁ ⁱ (XC _d ⁱ )·FOR ⁱ -u _e ⁱ ·FOR ⁱ ·(P ₁ ⁱ (XC _d ⁱ )-P ₁ ⁱ (X))]/(1-FOR ⁱ ) (30)

EENS ₁ ⁱ (X)＝[EENS _i-1,n (X)-EENS ₁ ⁱ (XC _d ⁱ )]/(1-FOR ⁱ ) (31)

P _i,n (X)＝P ₁ ⁱ (X)·AV ⁱ +P ₁ ⁱ (XX _d ⁱ )·DFOR ⁱ +P ₁ ⁱ (XC _e ⁱ )·FOR ⁱ (32)

F _i,n (X)＝F ₁ ⁱ (X)·AV ⁱ +F ₁ ⁱ (XX _d ⁱ )·DFOR ⁱ +F ₁ ⁱ (XC _e ⁱ )·FOR ⁱ +AV ⁱ ·λ _T ⁱ ·[ P ₁ ⁱ (XC _e ⁱ )-P ₁ ⁱ (X)]+AV ⁱ ·ρ _- ^i,n ·[P ₁ ⁱ (XX _d ⁱ )-P ₁ ⁱ (X)]+DFOR ⁱ ·λ _D ⁱ ·[P ₁ ⁱ (XC _e ⁱ )-P ₁ ⁱ (XX _d ⁱ )]

(33)

EENS _i,n (X)＝EENS ₁ ⁱ (X) AV ⁱ +EENS ₁ ⁱ (XX _d ⁱ ) DFOR ⁱ +EENS ₁ ⁱ (XC _e ⁱ ) FOR ⁱ (34)

In Equation (29)—Equation (34): AV ⁱ , DFOR ⁱ , FOR ⁱ respectively represent the rated output probability, derated output probability and zero output probability of the equivalent three-state unit; λ _T ⁱ and λ _D ⁱ represent the equivalent The forced outage rates of the rated state and the derated state in the three-state unit, P ₁ ⁱ (X), F ₁ ⁱ (X), and EENS ₁ ⁱ (X) represent the deconvoluted equivalent continuous load curve The load value is the cumulative probability, cumulative frequency and expected value of insufficient power at X.

Step 14. Accumulate the total operating capacity of the generator after the i-th operation, and judge whether i<I is true; if it is true, assign i+1 to i; and return to step 9; otherwise, it means that the random production simulation is completed .

In the final stochastic production simulation results, the wind power acceptance and wind abandonment rate can be calculated and processed, and the number of cold and hot starts, start-up costs and cold and hot start-up costs in the production simulation cycle, as well as reliability indicators EENS and LOLP can be calculated; in addition, It is also possible to study the influence of unit constraints and system constraints on stochastic production simulation.

Claims (9)

1. a kind of random production analog method considering load and wind-powered electricity generation temporal characteristics is applied in electric system, It is characterized in that the method carries out as follows：

Step 1, generating set data, wind power resources data and the sequential load data for obtaining electric system in T research cycle, by The sequential load data obtains original temporal load curve, and is denoted as Load1；

Step 2 establishes four kinds of unit models, including：Base lotus unit model, start and stop regulating units model, pressure load regulating units Model and two segmentation start and stop regulating units models；

Step 3, by the generating set data and sequential load data, tentatively correct the original temporal load curve Load1, Preliminary amendment sequential load curve is obtained, Load2 is denoted as；

Step 4 sets wind-powered electricity generation modified load number as n, and initializes n=1；

Step 5 is obtained the forced output of thermal power plant unit by the generating set data, according to the wind power resources data and heat supply machine Group, which is forced, contributes, and corrects the preliminary amendment sequential load curve Load2, obtains n-th wind-powered electricity generation receiving amount W_nWith n-th sequential Net load curve Load3_n；

Step 6, by the n-th sequential net load curve Load3_nIt is converted into n-th equivalent load duration curve, and seeks institute It states on n-th equivalent load duration curve, load value is the initial build probability P at X_0,n(X), initial build frequency F_0,n(X)、 Initial quantity of electricity deficiency desired value EENS_0,n(X) and initial maximum rate of rise

Step 7, the initial rotation spare capacity that the electric system is determined according to the operation demand of the electric systemWith Initial climbing capability value

Step 8, always put into operation number I by generating set data acquisition；The unit number that puts into operation is set as i, and initializes i=1；

Step 9, the constraints according to electric system judge whether to need to correct wind again using minimum cost of electricity-generating as target Electricity, if desired, n+1 is then assigned to n, and return to step 5 executes, if not needing, it is determined that the generating set that ith puts into operation Number；

The situation of step 10, the generating set type and tape base lotus that are put into operation according to the ith, determines the power generation that ith puts into operation Unit model belonging to unit；

As i=1, by the initial build probability P_0,n(X) and the initial build frequency F_0,n(X) ith is obtained to put into operation hair The electric system needs the rate of transform that unit puts into operation when motor groupWith the rate of transform for not needing unit and putting into operation

As i ≠ 1, by the cumulative probability P after (i-1)-th generating set that puts into operation_i-1,n(X) and (i-1)-th cumulative frequency F_i-1,n(X) Obtain ith put into operation generating set when the electric system rate of transform that needs unit to put into operationWith do not need what unit put into operation The rate of transform

Step 11, the generator group number to be put into operation according to the ith are corrected and obtain ith and put into operation the electric power after generating set The spinning reserve capacity of systemWith climbing capacity

Step 12, convert the unit model belonging to generating set that the ith puts into operation to equivalent two states unit model or Equivalent three-state machine group model；

The equivalent two states unit model or equivalent three-state machine group model are carried out Stochastic Production Simulation by step 13, are obtained Ith puts into operation the cumulative probability P after generating set_i,n(X), cumulative frequency F_i,n(X), expected loss of energy EENS_i,n(X) and Maximum growth rate

Step 14, accumulative ith put into operation the capacity that always puts into operation after generator, and judge whether i < I are true；If so, then by i+ 1 is assigned to i；And return to step 9；Otherwise, it indicates to complete Stochastic Production Simulation.

2. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature Be generating set described in step 1 include thermoelectricity generating set, hydroelectricity generator group and nuclear power generator group；The generating set Data include the type of unit, number of units, capacity, output bound, mean free error time MTTF, mean repair time MTTR with And consumption tiny increment；The wind power resources data include Wind turbines output and its corresponding period；The sequential load data To be sequentially arranged by hour load.

3. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature It is according to the different operating statuses of unit, the situation of tape base lotus during Stochastic Production Simulation and to be segmented unit in step 2 to be Capacity establishes four kinds of unit models；

The base lotus unit model：It indicates one and continuous operation unit is not segmented in basic load operation with rated capacity；

The start and stop regulating units model：Indicate the not sectional intermittent for doing peaking operation more than base lotus with a rated capacity fortune Row unit；

The pressure load regulating units model：Indicate that one first segmentation is contributed with minimum technology in basic load operation, the second segmentation The zonal cooling operating unit of peaking operation is done more than base lotus；

The two segmentations start and stop regulating units model：Between indicating that the segmentation of peaking operation is done in two segmentations more than base lotus It has a rest operating unit.

4. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature It is the receiving amount of n-th wind-powered electricity generation described in step 5 W_nIt obtains as follows：

Step a, variable t is set, and initializes t=1；The preliminary sequential load curve Load2 that corrects is remembered in the load value of t moment For Load2t；

Step b, judge whether n=1 is true, if so, then go to step c；Otherwise, d is gone to step；

Step c, judge whether t moment meets：Load2t-, which is forced, to contribute>=Wind turbines generated energy；

If satisfied, then enabling wind-powered electricity generation receiving amount=Wind turbines generated energy of t moment；Otherwise, enable t moment wind-powered electricity generation receiving amount= Load2t-, which is forced, to contribute；Go to step f；

Step d, it is to abandon wind ratio with λ；Judge whether t moment meets：Load2t- (forced output+λ × installed capacity of wind-driven power × n)>=Wind turbines generated energy；

If satisfied, then enabling t moment wind-powered electricity generation receiving amount=Wind turbines generated energy；Otherwise, e is gone to step；

Step e, judge Load2t- (forcing output+λ × installed capacity of wind-driven power × n)>Whether 0 is true, if so, then enable t moment Wind-powered electricity generation receiving amount=Load2t- (forces output+λ × installed capacity of wind-driven power × n)；Otherwise, t moment wind-powered electricity generation receiving amount=wind-powered electricity generation is enabled Unit generation amount；

Step f, judge whether t=T is true, if so, it then indicates to obtain n-th wind-powered electricity generation receiving amount W_n；Otherwise, t+1 is assigned to T, and return to step b.

5. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature Be in step 9 it is to judge whether to need to correct wind-powered electricity generation again as follows：

Step (1) judges that i > 1 and the ith capacity that always puts into operation after generator that puts into operation are more than the peak load of the electric system It is whether true, if so, it then selects the unit of consumption tiny increment minimum or machine set of segmentation to put into operation, and goes to step (7)；It is no Then, it goes to step (2)；

Whether step (2) judges to also have in all units not put into operation and forces output unit or machine set of segmentation, if so, then choosing The forced output unit or machine set of segmentation of consumption tiny increment minimum put into operation, and go to step (7)；Otherwise, it goes to step (3)；

Step (3) judges whether the spinning reserve constraint for meeting the electric system and Climing constant condition, if satisfied, then turning Step (5)；Otherwise, it goes to step (4)；

Step (4) judges the first segmentation with the presence or absence of segmentation unit in all units not put into operation, and if it exists, then chooses consumption The first minimum segmentation of amount tiny increment puts into operation, and goes to step (7)；Otherwise, it indicates to need to correct wind-powered electricity generation again；

Step (5), the unit or machine set of segmentation for searching consumption tiny increment minimum in all units not put into operation, judge the unit or Whether machine set of segmentation is the second segmentation for being segmented unit；If so, going to step (6)；Otherwise, it goes to step (7)；

Step (6) carries out pre-corrected to the spinning reserve capacity and climbing capacity of the electric system, and judges whether to meet pre- Revised spinning reserve constraint and Climing constant condition if satisfied, then putting into the second segmentation of segmentation unit, and are gone to step (7)；Otherwise, then it is gone to step (5) in the case where it is the second segmentation for being segmented unit to exclude minimum specific consumption unit；

Step (7), expression need not correct wind-powered electricity generation again.

6. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature The constraints for being electric system described in step 9 includes spinning reserve constraint and Climing constant as shown in formula (1) and formula (2) 's：

In formula (1) and formula (2)：Indicate that ith puts into operation the spinning reserve capacity of the electric system after generating set；SPR₀ For the spinning reserve capacity demand of system；Indicate that the ith increasing lotus of the electric system after generating set of putting into operation is held Amount；t₀For the time for requiring spinning reserve capacity to put into；For accident spinning reserve capacity demand；For load growth Spinning reserve capacity demand；X_iGenerating set the putting into operation a little on equivalent load duration curve put into operation for ith；Table Show generating set that ith puts into operation in the point X that puts into operation_iMost increase lotus rate at place；Δ t indicates the climbing time.

7. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature It is in step 10, it is described to need the rate of transform that unit puts into operationWith the rate of transform for not needing unit and putting into operationBy the formula of utilization (3) It is calculated with formula (4)：

In formula (3) and formula (4)：F_i-1,n(X_i) indicate (i-1)-th generating set afterload value X=X that puts into operation_iThe cumulative probability at place； P_i-1,n(X_i) indicate (i-1)-th generating set afterload value X=X that puts into operation_iThe cumulative frequency at place.

8. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature It is to correct the spinning reserve capacity by following situation in step 11With climbing capacity

If (a) generating set that ith puts into operation is the first segmentation for being segmented unit, capacity C_d ⁱ, then formula (5) and formula are utilized (6) it is modified：

If (b) generating set that ith puts into operation is the second segmentation for being segmented unit, capacity X_d ⁱ, then formula (7) and formula are utilized (8) it is modified：

If (c) generating set that ith puts into operation is not to be segmented unit, it is modified using formula (9) and formula (10)：

In formula (5)-formula (10), C_e ⁱIndicate the rated capacity for the generating set that ith puts into operation；It indicates to put into operation for (i-1)-th time The spinning reserve capacity of electric system after generating set；The climbing of system after (i-1)-th generating set that puts into operation of expression Capacity.

9. the random production analog method according to claim 1 for considering load and wind-powered electricity generation temporal characteristics, feature It is the cumulative probability P described in step 13_i,n(X), cumulative frequency F_i,n(X), expected loss of energy EENS_i,n(X) and most it increases Long rateIt is to be obtained by following situation：

1) if the generating set that ith puts into operation is the first segmentation for not being segmented unit or being segmented unit, formula (12), formula are utilized (13) and formula (14) obtains cumulative probability P_i,n(X), cumulative frequency F_i,n(X) and expected loss of energy EENS_i,n(X)：

P_i,n(X)=P_i-1,n(X)·(1-FORⁱ)+P_i-1,n(X-Cⁱ)·FORⁱ (12)

F_i,n(X)=F_i-1,n(X)·(1-FORⁱ)+F_i-1,n(X-Cⁱ)·FORⁱ+u_e ⁱ·FORⁱ·[P_i-1,n(X-Cⁱ)-P_i-1,n(X)] (13)

EENS_i,n(X)=EENS_i-1,n(X)·(1-FORⁱ)+EENS_i-1,n(X-Cⁱ)·FORⁱ (14)

In formula (12), formula (13) and formula (14)：X indicates the load value on the equivalent load duration curve；P_i,n(X)、F_i,n(X) And EENS_i,n(X) indicate that load value is the cumulative probability at the places X, tires out on equivalent load duration curve after ith unit puts into operation respectively Product frequency and expected loss of energy；P_i-1,n(X)、F_i-1,n(X) and EENS_i-1,n(X) after indicating that (i-1)-th unit puts into operation respectively Load value is cumulative probability, cumulative frequency and the expected loss of energy at X on equivalent load duration curve；FORⁱIndicate i-th The probability that the equivalent two states unit output of the secondary generating set to put into operation is zero；u_e ⁱIndicate the equivalent of the equivalent two states unit Repair rate；CⁱIndicate the capacity of generating set or machine set of segmentation that ith puts into operation, if the generating set that ith puts into operation be regardless of Section unit, then Cⁱ=C_e ⁱIf the generating set that ith puts into operation is the first segmentation for being segmented unit, Cⁱ=C_d ⁱ；

If the generating set that ith puts into operation is the second segmentation for being segmented unit, carried out using formula (15), formula (16) and formula (17) De-convolution operation obtains the cumulative probability P after de-convolution operation₁ ⁱ(X), cumulative frequency F₁ ⁱ(X) and expected loss of energy EENS₁ ⁱ(X)：Recycling formula (18), formula (19) and formula (20) obtain cumulative probability P_i,n(X), cumulative frequency F_i,n(X) and electricity Insufficient desired value EENS_i,n(X)：

P₁ ⁱ(X)=[P_i-1,n(X)-P₁ ⁱ(X-C_d ⁱ)·FORⁱ]/(1-FORⁱ) (15)

F₁ ⁱ(X)=[F_i-1.n(X)-F₁ ⁱ(X-C_d ⁱ)·FORⁱ-u_e ⁱ·FORⁱ·(P₁ ⁱ(X-C_d ⁱ)-P₁ ⁱ(X))]/(1-FORⁱ) (16)

EENS₁ ⁱ(X)=[EENS_i-1,n(X)-EENS₁ ⁱ(X-C_d ⁱ)]/(1-FORⁱ) (17)

P_i,n(X)=P₁ ⁱ(X)·AVⁱ+P₁ ⁱ(X-X_d ⁱ)·DFORⁱ+P₁ ⁱ(X-C_e ⁱ)·FORⁱ (18)

F_i,n(X)=F₁ ⁱ(X)·AVⁱ+F₁ ⁱ(X-X_d ⁱ)·DFORⁱ+F₁ ⁱ(X-C_e ⁱ)·FORⁱ+AVⁱ·λ_T ⁱ·[P₁ ⁱ(X-C_e ⁱ)-P₁ ⁱ (X)]+AVⁱ·ρ-^i,n·[P₁ ⁱ(X-X_d ⁱ)-P₁ ⁱ(X)]+DFORⁱ·λ_D ⁱ·[P₁ ⁱ(X-C_e ⁱ)-P₁ ⁱ(X-X_d ⁱ)]

(19)

EENS_i,n(X)=EENS₁ ⁱ(X)·AVⁱ+EENS₁ ⁱ(X-X_d ⁱ)·DFORⁱ+EENS₁ ⁱ(X-C_e ⁱ)·FORⁱ (20)

In formula (15)-formula (20)：C_e ⁱIndicate the rated capacity for the generating set that ith puts into operation, C_d ⁱIndicate the hair that ith puts into operation Motor group is the capacity for the first segmentation for being segmented unit；AVⁱ, DFORⁱ, FORⁱIndicate that the equivalent three condition unit is specified respectively Output probability, derated output probability and zero output probability；λ_T ⁱAnd λ_D ⁱRated condition in the equivalent three condition unit is indicated respectively With the forced outage rate of drop volume state, P₁ ⁱ(X), F₁ ⁱ(X), EENS₁ ⁱ(X) it indicates persistently to bear equivalent after one section of deconvolution respectively Load value is cumulative probability, cumulative frequency and the expected loss of energy at X on lotus curve.