CN114039378B - Wind-fire combined dispatching method and system capable of interrupting load and storage medium - Google Patents

Abstract

The present invention discloses a wind-storage-fire joint dispatching method and system and storage medium for interruptible loads. The dispatching method includes the following steps: S1, wind power prediction error analysis; S2, construction of a wind-storage-fire joint dispatching model: constructing an inner model with the goal of minimizing the standard deviation of wind-storage joint operation fluctuations, and constructing an outer model with the maximum goal of the daily revenue of the power generation enterprise under the full absorption of wind power for interruptible loads; S3, substituting the wind-storage joint operation output obtained by the inner model into the outer model to obtain the output of each unit that meets the system's rotating reserve and the dispatching plan for the interruptible load to participate in the system. The present invention solves the problem of insufficient rotating reserve capacity of traditional units caused by large-scale wind power grid connection by using pumped storage units and traditional units at the power supply end to provide rotating reserve and using the interruptible load at the load end as equivalent to rotating reserve after the increase in wind power grid connection.

Description

Wind-storage-fire joint dispatching method and system for interruptible load, and storage medium

Technical Field

The present invention relates to the technical field of electric energy dispatching, and in particular to a wind-storage-fire joint dispatching method and system for interruptible loads, and a storage medium.

Background Art

At present, in order to achieve the goal of "carbon peak and carbon neutrality", the power system has built a new power system with new energy as the main body. Wind power and other new energy are the main new energy sources, and their proportion in my country's energy structure will increase significantly, bringing severe challenges to the safe and stable operation of the power grid. Pumped storage power stations are important regulating power sources for the power system. They are of great significance to ensuring the safety of the power grid, promoting the consumption of new energy, and promoting the green and low-carbon transformation of energy. They can alleviate the impact of wind power energy on the safety and stability of the power grid to a certain extent. However, relying solely on pumped storage cannot meet the safety and reliability of wind power and other new energy sources connected to the grid.

Therefore, it is necessary to design a new wind-storage-fire joint dispatching method to improve the safety and reliability of grid connection of new energy such as wind power.

Summary of the invention

The purpose of the present invention is to provide a wind-storage-fire joint scheduling method and system, and a storage medium for interruptible loads. The wind-storage-fire joint scheduling method fully considers the uncertainty of wind power. First, traditional units and pumped storage units are used to meet the rotating reserve demand caused by wind power prediction errors and loads. When the requirements cannot be met from the power supply end, the interruptible load is considered to be equivalent to the rotating reserve from the load end to meet the system reliability requirements, thereby effectively improving the safety and reliability of the grid connection of new energy such as wind power.

The present invention is achieved through the following technical solutions:

The wind-storage-fire joint dispatching method for interruptible loads comprises the following steps:

S1. Wind power forecast error analysis:

S11. Fitting wind power output probability density function based on Beta distribution;

S12, performing per-unit normalization processing on the total capacity P _we of the wind farm installed capacity, the predicted wind power output power P _w and the real value P of the actual output power to obtain the normalized predicted wind power output power p _{wn and} the actual wind power output power p;

S13, constructing a wind power prediction error analysis model under a certain confidence level α based on the wind power output probability density function constructed in step S1 and the normalized wind power predicted output power p _wn and wind power actual output power p obtained in step S12;

S2. Construct wind-storage-fire joint dispatch model:

Based on the average grid-connected output of wind-storage combined units and the grid-connected output P _wh,t of the wind-storage combined unit in period t, and an inner model is constructed with the goal of minimizing the standard deviation of fluctuations in the wind-storage combined operation. The inner model is used to determine the pumping or generating power of the pumped storage unit;

Based on the wind-storage-thermal combined grid connection income _Cwhf , as well as the fuel cost of thermal power units _Cfuel , the environmental cost of thermal power units _Cenvir , the spinning reserve cost _CR and the interruptible load incentive cost _Cload given by the power generation enterprise to the user, the outer model is constructed with the daily income of the power generation enterprise under the full absorption of wind power of the interruptible load as the maximum goal;

S3. Substitute the wind-storage joint operation output obtained by the inner model into the outer model, combine the constraints of the inner model and the outer model, wherein, based on the wind power prediction error analysis model constructed in step S13, the upper spinning reserve constraint and the lower spinning reserve constraint of the outer model are transformed into deterministic constraints with a confidence level of α through chance constraint planning, and the particle swarm algorithm is used to solve the outer model to obtain the output of each unit that meets the system spinning reserve and the scheduling plan of the interruptible load participating in the system.

This method takes into account the uncertainty of wind power. First, traditional units and pumped storage units are used to meet the spinning reserve demand caused by wind power prediction errors and loads. When the requirements cannot be met from the power supply end, the interruptible load is considered to be equivalent to the spinning reserve from the load end to meet the system reliability requirements. BP neural network is used to predict wind power, the wind power output is obtained by using the relationship between wind speed and power function, the probability density function of wind power output is fitted using the Beta function, and the wind power prediction error is represented by the probability method in mathematics. The inner and outer two-layer model is nested for solution, that is, the inner layer wind power grid-connected minimum fluctuation model is established first to determine the pumping or power generation power of the pumped storage unit, and then the outer layer is established to take into account the interruptible load and the system spinning reserve cost. The maximum target model of the daily profit of the wind-storage-fire joint operation is established. The random variables in the wind-storage-fire joint operation model are processed by chance constrained programming, and the constraint conditions are processed by the penalty function method. The solution method adopts the improved particle swarm algorithm, which effectively improves the safety and reliability of the grid connection of new energy such as wind power.

Furthermore, in step S1, the probability density function of wind power output is as shown in formula (1):

In formula (1), p is a random variable according to the Beta distribution function, which represents the actual wind power output, α and β represent parameters, and B(α, β) represents the Beta function. The Beta function is shown in formula (2):

The parameters α and β in the above formula (2) are related to the parameter variance σ ² and the mean _μ . α and β are expressed by formula (3) and formula (4) respectively:

Furthermore, in step S1, the wind power prediction error analysis model includes the following two cases:

(1) At a certain confidence level, the predicted wind power is greater than the actual wind power, which can be expressed in terms of probability as shown in formula (5):

(2) At a certain confidence level, the predicted wind power is less than the actual wind power, which can be expressed in terms of probability as shown in formula (6):

In formula (5) and formula (6), a and b are the lower limit and upper limit of the probability interval estimation of wind power prediction error under a certain confidence level α, respectively. The relationship between a and b and the confidence level α is shown in formula (7) and formula (8), respectively:

F _cdf (a)=(1-α)F _cdf (P _wn,t ) (7)

F _cdf (b)＝α+(1-α)F _cdf (P _wn,t ) (8)

In equations (7) and (8), F _cdf (·) is the cumulative probability distribution, and its relationship with the probability density function is:

Among them, the calculation models of the normalized wind power output prediction value P _wn and the actual value P of wind power output power are shown in equations (9) and (10) respectively:

In equations (9) and (10), _Pwe is the total capacity of the wind farm installed capacity, _Pw and P are the true values of the predicted wind power output power and the actual output power, respectively.

Furthermore, in step S2, the inner layer model is as shown in formula (11):

In formula (11), P _wh,t is the grid-connected output of the wind-storage combined unit in period t, T represents a dispatching cycle of 24 hours, is the average grid-connected output of the wind-storage combined unit, The calculation model of is shown in formula (12):

In formula (12), T represents one scheduling period, which is 24 hours. The calculation model of P _wh,t is shown in formula (13) and formula (14):

P _wh,t =P _w,t -P _p,t u _p,t +P _g,t u _g,t (13)

In equations (13) and (14), P _wh,t is the grid-connected output of the wind-storage combined unit in period t; P _p,t and P _g,t are the pumping power and generating power in period t, respectively; and are the wind-storage combined operation and grid-connected output of the pumped-storage unit in power generation and pumping states during period t; P _w,t and are the predicted power and average predicted power of wind power in period t respectively; _up,t and u _g,t are whether the pumped storage unit is in the pumping or generating state respectively; the state representation of _up,t and u _g,t is shown in formula (15):

In formula (15), 1 means in this state, and 0 means not in the corresponding state. Formula (15) indicates that the pumped storage unit cannot be in the pumping and power generation states at the same time;

in, The calculation model of is shown in formula (16):

Furthermore, in step S2, the calculation model of the outer layer model is shown in formula (17):

C _max =C _whf -(C _fuel +C _envir +C _R +C _load ) (17)

Among them, the calculation model of wind-storage-thermal combined grid-connected income _Cwhf is shown in formula (18):

In formula (18), is the active power of the interruptible load, P _wh λ _wh is the income of the wind-storage joint operation, P _i,t is the output of thermal power unit i in period t; λ _G is the unit price of thermal power on the grid, T represents a dispatching cycle, which is 24h;

The calculation model of the fuel cost C _fuel of thermal power units is shown in formula (19):

In formula (19), N represents the number of thermal power units; a _i , b _i , c _i are the fuel cost coefficients of thermal power unit i; T represents a scheduling cycle, which is 24 hours; P _i,t is the output of thermal power unit i in period t; is the active power of the interruptible load;

The calculation model of the environmental cost C _envir of thermal power units is shown in formula (20):

In formula (20), λ _envir,c and λ _envir,s represent the environmental cost coefficients of CO ₂ and SO ₂ generated by thermal power generation, respectively; α _c,i , β _c,i , γ _c,i represent the CO ₂ emission coefficient of thermal power unit i; α _s,i , β _s,i , γ _s,i represent the SO ₂ emission coefficient of thermal power unit i; P _i,t is the output of thermal power unit i in period t; is the active power of the interruptible load; u _i,t is the start and stop status of thermal power unit i in period t, 1 means start, 0 means stop;

The calculation model of spinning reserve cost _CR is shown in formula (21):

In formula (21), N represents the number of thermal power units; T represents a scheduling cycle, which is 24 hours; and They represent the upper and lower spinning reserve demands of the system in period t respectively; λ _up and λ _dn represent the upper and lower spinning reserve demand cost coefficients respectively; is the active power of the interruptible load;

The calculation model of interruptible load incentive cost C _load is shown in formula (22):

In formula (22), M represents the number of interruptible load nodes, is the economic compensation for the interruptible load m during period t.

Furthermore, in step S3, the constraints of the inner model include wind-storage combined grid-connected output constraints, upper reservoir energy storage constraints, pumped storage unit power constraints, pumped storage unit power constraints and wind turbine unit predicted power constraints.

Furthermore, in step S3, the constraints of the outer model include system power balance constraints, system power flow constraints, 0-line transmission power constraints, system node voltage constraints, thermal power unit output constraints, thermal power unit ramp constraints and spinning reserve constraints.

Furthermore, the spinning reserve constraints include:

The upper spinning reserve constraint based on chance-constrained programming is shown in equations (23) to (25):

The lower spinning reserve constraint based on chance-constrained programming is shown in equations (26) to (28):

In formula (23) to formula (28), N represents the number of thermal power units, and are the upper and lower spinning reserve capacities provided by thermal power unit i in period t respectively; and All are the reserve demand capacity of the system load during period t; and are the upper and lower spinning reserves provided by the pumped storage unit during period _t ; α is the confidence level; T ₁₀ is the spinning reserve response time; is the active power of the interruptible load, p is the normalized actual wind power output power, P _wh,t is the grid-connected output of the wind-storage combined unit in period t, P _i ^max and P _i ^min are the maximum and minimum output of thermal power unit i respectively, ri _,up and ri _,dn are the upward climbing power and downward climbing power of unit i per unit time respectively; E _max is the minimum energy storage of the upper reservoir; η _g and η _p are the power generation efficiency and pumping efficiency of the pumped storage unit respectively, E _t is the energy storage of the upper reservoir in period t, P _i,t is the output of thermal power unit i in period t, P _p,t is the pumping power of the pumped storage unit in period t, and P _p ^max is the maximum pumping power of the pumped storage unit.

A system for wind-storage-fire combined scheduling of interruptible loads, the system comprising a processor, and the processor is configured to execute the above-mentioned wind-storage-fire combined scheduling method.

A computer-readable storage medium stores instructions, which are used to enable the processor to execute the above-mentioned wind-storage-fire joint scheduling method when executed by the processor.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The present invention fully considers the uncertainty of wind power. First, it uses traditional units and pumped storage units to meet the spinning reserve demand caused by wind power prediction errors and loads. When the requirements cannot be met from the power supply end, it considers converting the interruptible load from the load end into spinning reserve to meet the system reliability requirements, which effectively improves the safety and reliability of the grid connection of new energy such as wind power.

2. In order to ensure the reliability of the power grid, the present invention configures a pumped storage power station at the power supply end and uses interruptible loads at the load end to ensure that the system has sufficient spinning reserve capacity to cope with the increase in system spinning reserve demand caused by wind power prediction errors.

3. The present invention can ensure the absorption of wind power grid-connected, reduce the consumption of thermal power units, protect the environment and increase the economic benefits of power generation companies.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

FIG1 is a schematic diagram of the modified IEEE30 node of the present invention;

Figure 2 is a diagram of wind power prediction and load prediction;

Figure 3 is a graph showing active power curves during wind and storage combined operation;

Figure 4 is a comparison chart of upper spinning reserve and demand capacity;

FIG5 is a comparison chart of the spinning reserve demand capacity at each confidence level;

Figure 6 is a schematic diagram of the unit output plan for each period when the confidence level is 0.9.

DETAILED DESCRIPTION

In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary implementation modes of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

Embodiment 1:

The wind-storage-fire joint dispatching method for interruptible loads comprises the following steps:

S1. Wind power forecast error analysis:

Beta distribution is selected to fit the probability density function of wind power output. Beta has two advantages: first, the value range of the independent variable that obeys the Beta distribution object is [0,1], which is consistent with the value of the wind power output after normalization; second, when the two shape parameters of the Beta distribution are different, the wind power output curves in different situations can be described.

The probability density function of Beta distributed wind power output can be expressed as follows.

Where p is a random variable with Beta distribution function, which represents the actual wind power output, α and β represent parameters, and B(α,β) represents Beta function.

The parameters α and β in the above formula (2) are related to the parameter variance σ ² and mean μ, and can be expressed by the following formula.

Since the value range of the Beta distribution object is [0,1], when analyzing and solving the wind power prediction data, it is necessary to normalize the predicted value P _wn and the actual value P of the wind power output power.

Where _pwn and p are the normalized predicted wind power output power and actual wind power output power, _Pw and P are the true values of the predicted wind power output power and actual wind power output power, and _Pwe is the total capacity of the wind farm installed capacity.

Wind power forecast error analysis includes the following two cases:

(1) At a certain confidence level, the predicted wind power is greater than the actual wind power, which can be expressed in the following probabilistic way.

(2) At a certain confidence level, the predicted wind power is less than the actual wind power, which can be expressed in the following probabilistic way.

Where a and b are the lower and upper limits of the probability interval estimate of wind power forecast error at a certain confidence level α.

a and b have the following relationship with the confidence level α.

F _cdf (a)＝(1-α)F _cdf (P _wn,t ) (9)

F _cdf (b)＝α+(1-α)F _cdf (P _wn,t ) (10)

Where F _cdf (·) is the cumulative probability distribution, and its relationship with the probability density function is:

S2. Construction of wind-storage-fire joint dispatch model for interruptible loads:

The inner model establishes the minimum standard deviation target for the combined operation of wind and storage to obtain the combined operation output of wind and storage, and substitutes it into the outer model to meet the maximum objective function of the power generation enterprise's daily profit under the full absorption of wind power taking into account the interruptible load, thereby obtaining the output of each unit that meets the system's rotating reserve and the dispatch plan for the interruptible load to participate in the system.

Inner objective function: Minimize the standard deviation of wind-storage combined operation fluctuations. The smaller the standard deviation of wind-storage combined operation fluctuations, the smoother the wind-storage combined grid-connected power.

Where, F _whvv is the standard deviation of fluctuation of wind-storage combined operation; It is the average grid-connected output of the wind-storage combined unit.

P _wh,t =P _w,t -P _p,t u _p,t +P _g,t u _g,t (13)

Where, P _wh,t is the grid-connected output of the wind-storage combined unit in period t; P _p,t and P _g,t are the pumping power and generating power in period t, respectively; and are the wind-storage combined operation and grid-connected output of the pumped-storage unit in power generation and pumping states during period t; P _w,t and are the predicted power and average predicted power of wind power in period t respectively; _up,t and _g,t respectively represent whether the pumped storage unit is in the pumping or generating state, 1 indicates it is in this state, and 0 indicates it is not in the corresponding state; T represents one scheduling cycle, i.e. 24h; Formula (15) indicates that the pumped storage unit cannot be in the pumping and generating state at the same time.

Inner Constraints

Wind-storage combined operation grid-connected output constraints under pumping and power generation states of pumped storage units

In the formula, and They are respectively the minimum and maximum values of the grid-connected output of the wind-storage units in joint operation.

Upper reservoir energy storage constraints

E _min ≤E _t ≤E _max (19)

Where E _max and E _min are the maximum and minimum energy storage, respectively; η _g and η _p are the power generation efficiency and pumping efficiency of the pumped storage unit, respectively.

Power Constraints of Pumped Storage Units

Where P _g ^max and P _p ^max are the maximum power generation capacity and pumping capacity of the pumped storage unit, respectively; E _t is the energy storage in the reservoir during period t; Δt is the length of a scheduling period, i.e., 1h.

Wind turbine prediction power constraints

0≤P _w,t ≤P _we (23)

Where P _we is the rated output value of the wind turbine installed capacity.

Outer objective function: includes the benefits of wind-storage-firepower combined grid connection, the fuel cost of thermal power units, the environmental cost of thermal power units, the spinning reserve cost and the interruptible load incentive cost given by power generation enterprises to users.

C _max =C _whf -(C _fuel +C _envir +C _R +C _load ) (24)

(1) Wind-storage-thermal combined grid connection income C _whf

In the formula, is the active power of the interruptible load.

(2) Fuel cost of thermal power units including interruptible load

Where N represents the number of thermal power units; a _i , b _i , c _i are the fuel cost coefficients of thermal power unit i.

(3) Environmental costs of thermal power units including interruptible loads

In the formula, λ _envir,c and λ _envir,s represent the environmental cost coefficients of _CO2 and _SO2 generated by thermal power generation, respectively; α _c,i , β _c,i , and γ _c,i represent the _CO2 emission coefficients of thermal power unit i, respectively; α _s,i , β _s,i , and γ _s,i represent the _SO2 emission coefficients of thermal power unit i, respectively.

(4) Spinning reserve costs provided by thermal power units including interruptible loads

In the formula, and They represent the upper and lower spinning reserve demands of the system in period t respectively; λ _up and λ _dn represent the upper and lower spinning reserve demand cost coefficients respectively.

(5) Response costs provided by power generation companies to interruptible loads

Outer Constraints

(1) System power balance constraints

Where P _load,t is the load power during period t, and P _loss,t is the system active power loss during period t.

(2) System power flow constraints

Where, _Pe,t and Qe _,t are the active power and reactive power injected into node e by the power supply during period t, respectively; _Ve,t and Vf _,t are the voltage amplitudes of node s and node j during period t, respectively; _Gef and _Bef are the conductance and susceptance of system node e and node f, respectively; _θef is the voltage phase difference between system node e and node f.

(3)0 Line Transmission Power Constraint

Where P _ef,t is the active power transmitted on line ef during period t; and They are respectively the upper and lower limits of the active power transmitted on line ef.

(4) System node voltage constraints

In the formula, and are the lower and upper limits of the voltage amplitude at node f respectively.

(5) Output constraints of thermal power units

u _i,t P _i ^min ≤P _i,t ≤u _i,t P _i ^max (35)

^Where _Pimax and _Pimin are the maximum and minimum output values of thermal power unit i ^, respectively.

(6) Thermal power unit climbing constraints

Where _ri,up and ri _,dn are the upward climbing power and downward climbing power of unit i per unit time, respectively, and Δt represents the scheduling unit time period of 1h.

(7) Spinning reserve constraints

Up-spinning reserve constraints based on chance-constrained programming

Lower Spinning Reserve Constraints Based on Chance Constrained Programming

In the formula, and are the upper and lower spinning reserve capacities provided by thermal power unit i in period t respectively; and Both are the reserve demand capacity of the system load during period t, generally 5% of the total load; and They are the upper and lower spinning reserves provided by the pumped storage unit during period _t ; α is the confidence level; T ₁₀ is the spinning reserve response time, which is 10 minutes.

S3. Processing of random variables:

The normalized actual wind power output is a random variable. Through chance constrained programming, the upper and lower spinning reserve constraints are transformed into deterministic constraints with a confidence level of α.

Convert equations (37) and (40) into equations (43) and (44).

Where _Pwn,t is the normalized predicted output; a and b are the lower and upper limits of the corresponding interval when the wind power output meets the confidence level α.

S4. Model solution method

The model studies the multi-peak economic dispatch optimization problem, so the variable weight particle swarm algorithm that can better balance the global search and local search is selected. The formula is as follows:

Where w(k) is the inertia weight updated in the kth iteration, w _max and w _min are the maximum and minimum values of the inertia weight during the iteration, respectively; K _max is the maximum number of algorithm iterations.

This embodiment first uses the Beta function to fit the probability density function of wind power output, and uses the probability method in mathematics to represent the wind power prediction error. Considering that the wind power prediction error is the basis for the dispatch of the wind power system, and considering the wind power prediction error will lead to an increase in the system's spinning reserve, the power supply end uses pumped storage units and traditional units to provide spinning reserve, and after the increase in wind power grid-connected capacity, the load end can interrupt the load equivalent to the spinning reserve to solve the problem of insufficient spinning reserve capacity of traditional units caused by large-scale wind power grid connection.

To verify the effectiveness of the inner and outer two-layer model described in this embodiment, the example uses the modified IEEE30-node 6-machine system for simulation verification as shown in Figure 1, and the wind-storage combined unit is connected at node 7. The parameters of the thermal power unit are shown in Table 1, and the wind power prediction and load prediction are shown in Figure 2. The rated capacity of the wind farm is 100MW, the maximum power generation of pumped storage is 40MW, the maximum pumping power is 60MW, the power generation efficiency of the reversible turbine is 0.8, the pumping efficiency is 0.85, and the initial energy storage of the upstream reservoir is 50MWh. For ease of calculation, the power generation of wind power in 24 periods meets the Beta probability density function with parameters α=2.767 and β=2.517. The impact of thermal power units on the environment is quantified by the environmental costs of _CO2 and _SO2 . The environmental cost coefficients are both 3.5 yuan/kg, and the upper and lower rotating reserve cost coefficients are 140 yuan/MWh and 80 yuan/MWh respectively. The on-grid electricity price for wind-storage-thermal combined operation is: C _t = 540 yuan/M Wh, 0≤t＜8, C _t = 1038.4 yuan/M Wh, 8≤t＜22, C _t = 540 yuan/M Wh, 22≤t＜24, and the pumping fee for pumped storage is: C _{p, t} = 0.25C _t , 1≤t≤24. The population of the improved particle swarm algorithm is 100 individuals, the maximum number of iterations is 80 times, the learning factors c ₁ and c ₂ in the algorithm are both 2, and the maximum value w _max and minimum value w _min of the variable inertia weight are 0.9 and 0.3 respectively.

Table 1 Thermal power unit parameters

When the wind power grid-connected capacity is 100MW, the active power curves of each type during wind-storage combined operation are as shown in Figure 3:

As can be seen from Figure 3, the pumped storage energy storage is 50MW at the beginning and end of the dispatch cycle. The entire wind-storage joint operation process reflects the time-space transfer characteristics of pumped storage. During the wind-storage joint dispatch process, the total wind abandonment is 0MWh, and the standard deviation of wind power grid-connected output is 14MW. In the wind power grid without pumped storage, the total wind abandonment is 47MWh, and the standard deviation of wind power grid-connected output is 19.7MW.

During the dispatch cycle, thermal power units and pumped storage can provide a large lower spinning reserve capacity to meet the lower spinning reserve demand capacity caused by load and wind power uncertainty. During the dispatch cycle, the comparison of the upper spinning reserve provided by the system and the upper spinning reserve demand capacity of the system at different confidence levels is shown in Figure 4.

The upper spinning reserve provided by the combined operation of wind, storage and thermal power meets the system upper spinning reserve demand capacity at all confidence levels. When the system contains pumped storage units, the system reliability is high.

Figure 5 shows the upper and lower spinning reserve demand capacities at five confidence levels of 1, 0.95, 0.9, 0.85, and 0.8:

At this time, the wind-fire joint operation benefits and wind-storage-fire joint operation benefits at five confidence levels are shown in Table 2:

Table 2 Comparison of system benefits with and without pumped storage units at different confidence levels

The addition of pumped storage can not only ensure the reliability of the system, but also reduce coal burning, alleviate environmental pollution and improve the economic benefits of the system.

When the confidence level is 0.9, the thermal power units are arranged to output. At this time, the power generation profit is large and the reserve demand meets the requirements.

The output strategy of the thermal power unit when the confidence level is 0.9 is shown in Figure 6:

The results show that when the load is low in periods 1-7 and 21-24, unit 4 is at the lower boundary and does not choose to shut down because the system's spinning reserve demand is large and the thermal power unit needs to provide the system with upper spinning reserve to increase system reliability. Units 5 and 6 are basically in full output during periods 9-20 due to the large system load.

As the amount of wind power connected to the grid increases, the pumped storage capacity of a certain site will be limited due to actual conditions and cannot continue to increase to meet the needs of larger wind power capacity. When the capacity of wind power connected to the grid increases, it will be considered that only thermal power units and pumped storage power stations are not enough to solve the problem of reserve capacity. From the load side, it will be considered to use interruptible loads to meet the problem of insufficient spinning reserve caused by the increase in wind power connected to the grid. The maximum interruption capacity of the interruptible load is considered to be 15MW.

The reserve capacity of interruptible load corresponding to five confidence levels of 1, 0.95, 0.9, 0.85 and 0.8 under different wind power grid-connected capacities is shown in Table 3.

Table 3 Interruptible load reserve capacity for different wind power grid-connected quantities

When the wind power grid-connected capacity is 100MW, the spinning reserve provided by pumped storage and thermal power units meets the spinning reserve demand capacity considering wind power forecast errors at all confidence levels, and no interruptible load is required to participate in system dispatch. When the wind power grid-connected capacity is 150MW and the confidence level is 1, the system's interruptible load capacity of 15MW is fully involved in the system dispatch.

The upper spinning reserve demand for wind power grid-connected capacity ranging from 100MW to 150MW at different confidence levels is shown in Table 4.

Table 4 The required capacity of upper spinning reserve for different wind power grid-connected quantities

Under the same amount of wind power connected to the grid, the higher the confidence level, the greater the upper spinning reserve demand capacity that the system needs to meet. The confidence level reflects the reliability of the system in meeting the upper spinning reserve, that is, the higher the reliability, the greater the upper spinning reserve capacity that needs to be met. Under the same confidence level, the increase in wind power grid-connected capacity will increase the system's upper spinning reserve demand capacity, because the increase in wind power grid-connected capacity will increase the prediction error of wind power grid-connected capacity, that is, the system's upper spinning reserve demand capacity will increase.

The economic benefits of power generation companies under different wind power grid-connected amounts are shown in Table 5.

With the same wind power grid-connected capacity, the lower the confidence level, the more revenue the power generation company will earn. When the interruptible load is required to participate in the system dispatch to meet the spinning reserve demand capacity, the difference in revenue of the power generation company is mainly the interruptible load cost provided by the power generation company to the user and the spinning reserve cost of the system. When the wind power grid-connected capacity increases, the difference in revenue of the power generation company is mainly due to the reduction in fuel costs of thermal power units and the reduction in environmental pollution costs caused by nitrogen and sulfides.

Table 5 Profits of power generation companies with different wind power grid-connected amounts

In summary:

The dispatching method described in this embodiment verifies through calculation examples that under the condition of large-scale grid connection of wind power, in order to ensure the reliability of the power grid, a pumped storage power station is configured at the power supply end and an interruptible load is used at the load end to ensure that the system has sufficient spinning reserve capacity to cope with the increase in system spinning reserve demand caused by wind power forecast errors. This solution can ensure the absorption of wind power grid connection, reduce the consumption of thermal power units, protect the environment and increase the economic benefits of power generation companies.

The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims (8)

1. A wind-storage-fire joint dispatching method for interruptible loads, characterized in that it comprises the following steps: S1. Wind power forecast error analysis: S11. Fitting wind power output probability density function based on Beta distribution; S12, performing per-unit normalization processing on the total capacity P _we of the wind farm installed capacity, the predicted wind power output power P _w and the real value P of the actual output power to obtain the normalized predicted wind power output power p _{wn and} the actual wind power output power p; S13, constructing a wind power prediction error analysis model under a certain confidence level α based on the wind power output probability density function constructed in step S1 and the normalized wind power predicted output power p _wn and wind power actual output power p obtained in step S12; S2. Constructing a wind-fire joint dispatching model: Based on the average grid-connected output of wind-storage combined units and the grid-connected output P _wh,t of the wind-storage combined unit in period t, and an inner model is constructed with the goal of minimizing the standard deviation of fluctuations in the wind-storage combined operation. The inner model is used to determine the pumping or generating power of the pumped storage unit; Based on the wind-storage-thermal combined grid connection income _Cwhf , as well as the fuel cost of thermal power units _Cfuel , the environmental cost of thermal power units _Cenvir , the spinning reserve cost _CR and the interruptible load incentive cost _Cload given by the power generation enterprise to the user, the outer model is constructed with the daily income of the power generation enterprise under the full absorption of wind power of the interruptible load as the maximum goal; The inner layer model is shown in formula (11): In formula (11), P _wh,t is the grid-connected output of the wind-storage combined unit in period t, T represents a dispatching cycle of 24 hours, is the average grid-connected output of the wind-storage combined unit, The calculation model of is shown in formula (12): In formula (12), T represents one scheduling period, which is 24 hours. The calculation model of P _wh,t is shown in formula (13) and formula (14): P _wh,t =P _w,t -P _p,t u _p,t +P _g,t u _g,t (13) In equations (13) and (14), P _wh,t is the grid-connected output of the wind-storage combined unit in period t; P _p,t and P _g,t are the pumping power and generating power in period t, respectively; and are the wind-storage combined operation and grid-connected output of the pumped-storage unit in power generation and pumping states during period t; P _w,t and are the predicted power and average predicted power of wind power in period t respectively; _up,t and u _g,t are whether the pumped storage unit is in the pumping or generating state respectively; the state representation of _up,t and u _g,t is shown in formula (15): In formula (15), 1 means in this state, and 0 means not in the corresponding state. Formula (15) indicates that the pumped storage unit cannot be in the pumping and power generation states at the same time; in, The calculation model of is shown in formula (16): The calculation model of the outer layer model is shown in formula (17): C _max =C _whf -(C _fuel +C _envir +C _R +C _load ) (17) Among them, the calculation model of wind-storage-thermal combined grid-connected income _Cwhf is shown in formula (18): In formula (18), is the active power of the interruptible load, P _wh λ _wh is the income of the wind-storage joint operation, P _i,t is the output of thermal power unit i in period t; λ _G is the unit price of thermal power on the grid, T represents a dispatching cycle, which is 24h; The calculation model of the fuel cost C _fuel of thermal power units is shown in formula (19): In formula (19), N represents the number of thermal power units; a _i , b _i , c _i are the fuel cost coefficients of thermal power unit i; T represents a scheduling cycle, which is 24 hours; P _i,t is the output of thermal power unit i in period t; is the active power of the interruptible load; The calculation model of the environmental cost C _envir of thermal power units is shown in formula (20): In formula (20), λ _envir,c and λ _envir,s represent the environmental cost coefficients of CO ₂ and SO ₂ generated by thermal power generation, respectively; α _c,i , β _c,i , γ _c,i represent the CO ₂ emission coefficient of thermal power unit i; α _s,i , β _s,i , γ _s,i represent the SO ₂ emission coefficient of thermal power unit i; P _i,t is the output of thermal power unit i in period t; is the active power of the interruptible load; u _i,t is the start and stop status of thermal power unit i in period t, 1 means start, 0 means stop; The calculation model of spinning reserve cost _CR is shown in formula (21): In formula (21), N represents the number of thermal power units; T represents a scheduling cycle, which is 24 hours; and They represent the upper and lower spinning reserve demands of the system in period t respectively; λ _up and λ _dn represent the upper and lower spinning reserve demand cost coefficients respectively; is the active power of the interruptible load; The calculation model of interruptible load incentive cost C _load is shown in formula (22): In formula (22), M represents the number of interruptible load nodes, is the economic compensation for the interruptible load m during period t; S3. Substitute the wind-storage joint operation output obtained by the inner model into the outer model, combine the constraints of the inner model and the outer model, wherein, based on the wind power prediction error analysis model constructed in step S13, the upper spinning reserve constraint and the lower spinning reserve constraint of the outer model are transformed into deterministic constraints with a confidence level of α through chance constraint planning, and the particle swarm algorithm is used to solve the outer model to obtain the output of each unit that meets the system spinning reserve and the scheduling plan of the interruptible load participating in the system. 2. The wind-storage-fire combined dispatching method for interruptible loads according to claim 1 is characterized in that, in step S1, the probability density function of wind power output is as shown in formula (1): In formula (1), p is a random variable according to the Beta distribution function, which represents the actual wind power output, α and β represent parameters, and B(α, β) represents the Beta function. The Beta function is shown in formula (2): The parameters α and β in the above formula (2) are related to the parameter variance σ ² and the mean μ. α and β are expressed by formula (3) and formula (4) respectively: 3. The wind-storage-fired joint dispatching method for interruptible loads according to claim 1 is characterized in that, in step S1, the wind power prediction error analysis model includes the following two cases: (1) At a certain confidence level, the predicted wind power is greater than the actual wind power, which can be expressed in terms of probability as shown in formula (5): (2) At a certain confidence level, the predicted wind power is less than the actual wind power, which can be expressed in terms of probability as shown in formula (6): In formula (5) and formula (6), a and b are the lower limit and upper limit of the probability interval estimation of wind power prediction error under a certain confidence level α, respectively. The relationship between a and b and the confidence level α is shown in formula (7) and formula (8), respectively: F _cdf (a)=(1-α)F _cdf (P _wn,t ) (7) F _cdf (b)＝α+(1-α)F _cdf (P _wn,t ) (8) In equations (7) and (8), F _cdf (·) is the cumulative probability distribution, and its relationship with the probability density function is: Among them, the calculation models of the normalized wind power output prediction value P _wn and the actual value P of wind power output power are shown in equations (9) and (10) respectively: In equations (9) and (10), _Pwe is the total capacity of the wind farm installed capacity, _Pw and P are the true values of the predicted wind power output power and the actual output power, respectively. 4. The wind-storage-fire joint scheduling method for interruptible loads according to claim 1 is characterized in that, in step S3, the constraints of the inner model include the wind-storage joint grid-connected output constraints, the upper reservoir energy storage constraints, the power constraints of the pumped storage units, the power constraints of the pumped storage units and the predicted power constraints of the wind turbine units. 5. The wind-storage-fire combined scheduling method for interruptible loads according to claim 1 is characterized in that, in step S3, the constraints of the outer model include system power balance constraints, system flow constraints, 0-line transmission power constraints, system node voltage constraints, output constraints of thermal power units, thermal power unit climbing constraints and rotating standby constraints. 6. The wind-storage-fired joint dispatching method for interruptible loads according to claim 5, characterized in that the spinning reserve constraints include: The upper spinning reserve constraint based on chance-constrained programming is shown in equations (23) to (25): The lower spinning reserve constraint based on chance-constrained programming is shown in equations (26) to (28): In formula (23) to formula (28), N represents the number of thermal power units, and are the upper and lower spinning reserve capacities provided by thermal power unit i in period t respectively; and All are the reserve demand capacity of the system load during period t; and are the upper and lower spinning reserves provided by the pumped storage unit during period t; α is the confidence level; T ₁₀ is the spinning reserve response time; is the active power of the interruptible load, p is the normalized actual wind power output, P _wh,t is the grid-connected output of the wind-storage combined unit in period t, P _i ^max and P _i ^min are the maximum and minimum output of thermal power unit i, _ri,up and ri _,dn are the upward climbing power and downward climbing power of unit i per unit time, respectively; E _max is the minimum energy storage of the upper reservoir; η _g and η _p are the power generation efficiency and pumping efficiency of the pumped storage unit, respectively, E _t is the energy storage of the upper reservoir in period t, P _i,t is the output of thermal power unit i in period t, P _p,t is the pumping power of the pumped storage unit in period t, It is the maximum pumping power of the pumped storage unit. 7. A system for wind-storage-fire combined scheduling of interruptible loads, characterized in that the system comprises a processor, and the processor is configured to execute the wind-storage-fire combined scheduling method according to any one of claims 1-6. 8. A computer-readable storage medium, characterized in that the storage medium stores instructions, which are used to enable the processor to execute the wind-storage-fire combined scheduling method described in any one of claims 1-6 when executed by the processor.