CN103795088B - A kind of hydroenergy storage station Optimization Scheduling quantized based on load curve - Google Patents

Abstract

The invention discloses a kind of hydroenergy storage station Optimization Scheduling quantized based on load curve, it comprises the following steps: first arrange unit maintenance scheduling; Then Load Prediction In Power Systems data are gathered; Set up hydroenergy storage station state transition equation and constraints thereof afterwards; Set up fired power generating unit load curve quantizating index again, be minimised as target function with fired power generating unit load curve quantizating index, use dynamic programming to run hydroenergy storage station and be optimized scheduling.Present invention reduces the degree of fluctuation of the load curve that fired power generating unit is born, improve its power generation load rate, reduce the net coal consumption rate of fired power generating unit.

Description

An Optimal Dispatch Method for Pumped Storage Power Station Based on Load Curve Quantification

technical field

The invention relates to the technology of a pumped storage power station, in particular to an optimal scheduling method for a pumped storage power station based on load curve quantization.

Background technique

The pumped storage power station uses the excess electric energy in the power system to pump the water from the low-elevation reservoir (commonly known as "lower reservoir") to the high-elevation reservoir (commonly known as "upper reservoir") and store it in the form of potential energy. When the system needs electricity, the hydropower station releases water from the upper reservoir to the lower reservoir for power generation. Whether the pumped storage power station is used for frequency regulation, emergency backup or load following, it can provide active power support at a rate that meets the system stability requirements when the grid needs it, so as to maintain the safety and stability of the system operation. In the past ten years, the pumped storage power station has developed rapidly in our country, and its role in the power system is increasing day by day.

The operating cost of the power system mainly involves the coal consumption fuel cost of the thermal power unit, which is affected by the load and change speed of the thermal power unit. The basis for evaluating the fuel cost of thermal power units is concentrated on the load curve it bears.

The load curve of the thermal power unit is formed by connecting the discrete values of the load it bears with a straight line. The quantification of the load curve is divided into two parts, one part is composed of the angle between the load of each dispatching interval and the horizontal axis (indicated by θ), and the other part is composed of the included angle (indicated by ψ) of each dispatching interval, as shown in Figure 1 . Both of these two parts reflect the requirements of the load curve on the output and climbing rate of thermal power units. The larger the included angle, the greater the adjustment required for thermal power units, the higher the corresponding coal consumption rate, and the greater the fuel cost.

From the perspective of power grid dispatching, how to optimize the operation of pumped storage power plants so that they can perform functions that thermal power units do not have or are uneconomical is an urgent problem to be solved. At present, the starting point or objective function of optimal dispatching of pumped storage power plants is the most economical system operation, the lowest power generation operation cost and other economic indicators. The economic evaluation is carried out through the set fuel cost and power loss loss. The difference will affect the optimization result to a certain extent, which needs to be further improved.

Contents of the invention

Purpose of the invention: Aiming at the deficiencies of the existing methods, the purpose of the invention is to provide an optimal scheduling method for pumped storage power plants based on load curve quantization.

Technical scheme: in order to realize the above-mentioned purpose of the invention, the technical scheme adopted in the present invention is as follows:

A pumped storage power station optimization scheduling method based on load curve quantization, characterized in that it includes the following steps:

(1) Arranging unit maintenance plan: determine the number of available thermal power units and the number of pumped storage power generation units; and enter the parameters of available thermal power units and pumped storage power generation units; the parameters include the rated power generation of each thermal power unit, each pumped storage The rated generating power of the generator set of the power station, the pumping-generation cycle efficiency of the pumped storage power station, the given water storage capacity and the maximum storage capacity;

(2) Collect power system load forecast data: the time interval of the power system load forecast data is fixed at Δt , the unit is h, and there are N points in total; the power system load forecast is calculated in the time-power system load forecast value coordinate system Two points adjacent to the data are connected by a straight line to form a power system load curve; the power system load forecast data divides the power system load curve into N dispatching intervals;

(3) Establish the state transition equation of the pumped storage power station:

(1)

In the above formula (1), x _t is the water storage capacity of the upper reservoir of the pumped storage power station at the end of the tth dispatching interval, and the unit is MWh, , whose state set X is:

(2)

In the above formula (2), p _h is the rated capacity of unit h of the pumped storage power station in MW, h = 1, 2, 3... H , H is the number of generating units in the pumped storage power station, and P is the pumped water The greatest common divisor of the capacity of each generating unit of the energy storage power station, the unit is MW, x _max is the maximum storage capacity of the pumped storage power station, the unit is MWh; the number M of the state set elements is determined by the following formula:

;(3)

In the above formula (1), d _t is the decision variable of the tth scheduling interval, and the unit is MW, ; η is the pumping-generation cycle efficiency of the _pumped storage power station when ignoring the influence of natural water; When d _t is greater than zero, it is in the power generation state; when d _t is less than zero, it is in the pumping state, and its state set D is:

;(4)

The constraints of the state transition equation are:

(5)

In the above formula (5), x ₀ is the water storage capacity of the upper reservoir of the pumped storage power station at the beginning of the first dispatching interval, the unit is MWh, and C is the given water storage capacity of the pumped storage power station, the unit is MWh;

(4) Establish the quantitative index Q of thermal power unit load curve:

;(6)

In the formula (6), is the angle between the thermal power unit load curve and the horizontal coordinate axis in the tth scheduling interval, in degrees , Be the angle between the tth dispatch interval and the t- 1 dispatch interval thermal power unit load curve, unit is degree; Described thermal power unit load curve is obtained by subtracting the optimal output curve of the pumped storage power plant from the power system load curve;

(5) Establish the objective function , the dynamic programming method is used to optimize the scheduling of the pumped storage power station. From the first scheduling interval to the Nth scheduling interval, the following sub-steps are executed in sequence:

1) Establish the tth scheduling interval index function g _t ( x _t , d _t ), :

;(7)

2) Establish process indicator function The recursive formula:

, ;(8)

in is the process index function of the tth scheduling interval, the unit is degree;

3) Using the above formula (8) as the objective function, solve the optimal value d _t ^* of the decision variable d _t in the tth scheduling interval, and bring it into the state transition equation (1) of the pumped storage power station In, obtain the optimal value x _t ^* of the state variable x _t of the tth scheduling interval;

4) Substituting the d _t ^* and x _t ^* together into the process index function recursive formula (8) to obtain the process index function of the t-th scheduling interval optimal solution of , to make preparations for the optimal scheduling of the t+1th scheduling interval using the dynamic programming method.

The calculation method of the angle θ _t between the thermal power unit load curve and the horizontal coordinate axis in the tth dispatching interval is:

(9)

in , is the value of the thermal power unit load curve at time t , the unit is MW, t =1,2,3... N, and ;

The calculation method of the included angle ψ _t between the t dispatch interval and the t- 1 dispatch interval thermal power unit load curve is:

(10)

in .

The beneficial effect that produces with above-mentioned technical scheme is:

1. When optimizing dispatching in the present invention, the pumped-storage power station shall bear the changing part of the sequential load curve as far as possible, effectively reducing the fluctuation degree of the load curve borne by the thermal power unit, making the load borne by the thermal power unit more stable, and avoiding frequent and large load curves. Degree of primary and secondary frequency modulation action.

2. Using the present invention to optimize the scheduling of the pumped storage power station to optimize the load curve of the thermal power generation unit, realize the minimum adjustment of the thermal power unit, reduce the coal consumption rate, and save fuel costs.

Description of drawings

Fig. 1 is the optimized scheduling flowchart of the present invention;

Fig. 2 is a schematic diagram of thermal power unit load curve quantitative index of the present invention;

Fig. 3 is the power system load curve and the optimized output curve of the pumped storage power station of the present invention;

Fig. 4 is the thermal power unit load curve after optimal dispatching of the pumped storage power station of the present invention.

detailed description

The present invention will be further described below in conjunction with specific embodiments.

Example:

An optimal scheduling method for pumped storage power plants based on load curve quantification, including the following steps, as shown in Figure 1:

(1) Arranging unit maintenance plan: determine the number of available thermal power units and the number of pumped storage power generation units; and enter the parameters of available thermal power units and pumped storage power generation units; the parameters include the rated power generation of each thermal power unit, each pumped storage The rated generating power of the generator set of the power station, the pumping-generation cycle efficiency of the pumped storage power station, the given water storage capacity and the maximum storage capacity;

(2) Collect power system load forecast data: the time interval of the power system load forecast data is fixed at Δt , the unit is h, and there are N points in total; the power system load forecast is calculated in the time-power system load forecast value coordinate system Two points adjacent to the data are connected by a straight line to form a power system load curve; the power system load forecast data divides the power system load curve into N dispatching intervals;

(3) Establish the state transition equation of the pumped storage power station:

(1)

In the above formula (1), x _t is the water storage capacity of the upper reservoir of the pumped storage power station at the end of the tth dispatching interval, and the unit is MWh, , whose state set X is:

(2)

In the above formula (2), p _h is the rated capacity of unit h of the pumped storage power station in MW, h = 1, 2, 3... H , H is the number of generating units in the pumped storage power station, and P is the pumped water The greatest common divisor of the capacity of each generating unit of the energy storage power station, the unit is MW, x _max is the maximum storage capacity of the pumped storage power station, the unit is MWh; the number M of the state set elements is determined by the following formula:

;(3)

In the above formula (1), d _t is the decision variable of the tth scheduling interval, and the unit is MW, ; η is the pumping-generation cycle efficiency of the _pumped storage power station when ignoring the influence of natural water; When d _t is greater than zero, it is in the power generation state; when d _t is less than zero, it is in the pumping state, and its state set D is:

;(4)

The constraints of the state transition equation are:

(5)

In the above formula (5), x ₀ is the water storage capacity of the upper reservoir of the pumped storage power station at the beginning of the first dispatching interval, the unit is MWh, and C is the given water storage capacity of the pumped storage power station, the unit is MWh;

(4) Establish the quantitative index Q of thermal power unit load curve:

;(6)

As shown in Figure 2, in the formula (6), is the angle between the thermal power unit load curve and the horizontal coordinate axis in the tth scheduling interval, in degrees , Be the angle between the tth dispatch interval and the t- 1 dispatch interval thermal power unit load curve, unit is degree; Described thermal power unit load curve is obtained by subtracting the optimal output curve of the pumped storage power plant from the power system load curve;

(5) Establish the objective function , the dynamic programming method is used to optimize the scheduling of the pumped storage power station. From the first scheduling interval to the Nth scheduling interval, the following sub-steps are executed in sequence:

1) Establish the tth scheduling interval index function g _t ( x _t , d _t ), :

;(7)

2) Establish process indicator function The recursive formula:

, ;(8)

in is the process index function of the tth scheduling interval, the unit is degree;

3) Using the above formula (8) as the objective function, solve the optimal value d _t ^* of the decision variable d _t in the tth scheduling interval, and bring it into the state transition equation (1) of the pumped storage power station In, obtain the optimal value x _t ^* of the state variable x _t of the tth scheduling interval;

4) Substituting the d _t ^* and x _t ^* together into the process index function recursive formula (8) to obtain the process index function of the t-th scheduling interval optimal solution of , to make preparations for the optimal scheduling of the t+1th scheduling interval using the dynamic programming method.

The calculation method of the angle θ _t between the thermal power unit load curve and the horizontal coordinate axis in the tth dispatching interval is:

(9)

in , is the value of the thermal power unit load curve at time t , the unit is MW, t =1,2,3... N, and ;

The calculation method of the included angle ψ _t between the t dispatch interval and the t- 1 dispatch interval thermal power unit load curve is:

(10)

in .

In this embodiment, the power system includes one equivalent thermal power unit with a rated generating power of 60,000 MW; one pumped storage power station with a pumping-generation cycle efficiency of 79.90%, a given water storage capacity of 1000 MWh, and a maximum storage capacity of 12000MWh, 8 generator sets of pumped storage power station, rated generating power of 300MW, total installed capacity of 2400MW.

The power system load forecasting curve in this embodiment is formed from the short-term load forecasting data of the Guangdong State Commission on January 3, 2012. The power system load forecast time interval is It is fixed at 0.25h, that is, 15 minutes, and there are 96 power system load forecast data in one day, as shown in Table 1. In the time-power system load forecast value coordinate system, two adjacent points of the power system load forecast data are connected by a straight line to form a power system load forecast curve. These 96 power system load forecast data divide the power system load curve into N scheduling intervals. Using the method of the present invention to optimize the scheduling of the pumped storage power station, a series of optimal values d _t * are obtained, which are the scheduling schemes of the pumped storage power station at the end of each scheduling interval. The optimization results of this embodiment are shown in FIG. 3 . In order to draw the power system load forecast curve and the optimal output curve of the pumped storage power station in one picture, the power system load forecast value in the figure is shown as being reduced by 10 times.

Fig. 4 is the thermal power unit load curve after optimal dispatching of the pumped storage power station; it can be seen that after the optimal dispatching of the pumped storage power station by the present invention, the fluctuation degree of the thermal power unit load curve is reduced compared with the power system load prediction curve.

Table 2 is the result comparison of whether the present invention is used to optimize the operation of the pumped storage power station. "No benefit storage" means that the pumped storage power station does not participate in the load distribution of the power grid; "after benefit storage experience operation" means that the pumped storage power station is dispatched in the empirical operation mode; Optimum scheduling of energy storage power stations. It can be seen that the optimized adjustment of the pumped storage power station by the present invention reduces the quantitative index of load fluctuation and increases the load rate of power generation. The optimization of these two indicators means that the operating cost of thermal power units is reduced, which has important guiding significance in actual production.

The impact of the thermal power unit load curve on the coal consumption of power supply is not only related to its power generation load rate, but also related to the fluctuation degree of the load curve. Even under the same average load rate, the stable operation of the unit is more conducive to reducing the coal consumption of power supply than the load fluctuation. The quantitative calculation of the influence of power system load curve and thermal power unit load curve fluctuation on power supply coal consumption is carried out below to verify the practicability of the present invention.

For the convenience of calculation, after reducing the load curve value of the thermal power unit by 100 times, it is assumed by a 600MW supercritical thermal power unit, and its power supply coal consumption u ( x ) is shown in Table 3. The second degree polynomial fitting is carried out on the data in Table 3, and the fitting result is:

u ( x )=0.0003 x ² -0.3701 x +422.4429

Let the power system load forecast curve be L ' ₁ ( t ), and the thermal power unit load curve be L ' ₂ ( t ). The formula for calculating coal consumption for power supply in one day is:

Among them, H1 is the coal consumption of power supply _one day before optimal dispatch, and H2 is the coal consumption of power supply one _day after optimal dispatch.

Thus, H ₁ =3.5362×10 ⁹ grams, H ₂ =3.4978×10 ⁹ grams, that is, 38.4 tons of coal can be saved a day after the optimal dispatching of the present invention.

Table 1 Power system load data

load point Load value (MW) load point Load value (MW) 1 38000 49 49000 2 37400 50 46300 3 36900 51 45700 4 36500 52 45600 5 36100 53 45500 6 35700 54 47000 7 35300 55 48600 8 34900 56 51200 9 34600 57 51800 10 34300 58 52400 11 34100 59 52600 12 33800 60 52700 13 33600 61 52800 14 33400 62 53000 15 33200 63 53200 16 33100 64 53400 17 33000 65 53900 18 33000 66 54400 19 33000 67 54900 20 33000 68 55400 twenty one 33100 69 55600 twenty two 33200 70 55700 twenty three 33400 71 54400 twenty four 33700 72 54000 25 34000 73 54700 26 34700 74 55900 27 35400 75 56700 28 36400 76 57000 29 37400 77 56700 30 38900 78 56400 31 40100 79 56000 32 41900 80 55500 33 44900 81 55000 34 49200 82 54700 35 51100 83 54400 36 52400 84 54100 37 53000 85 53700 38 53600 86 53100 39 54000 87 52300 40 54400 88 51500 41 54800 89 50700 42 55200 90 49800 43 55600 91 48700 44 56000 92 47500 45 56300 93 45900 46 56500 94 44300 47 55500 95 42700 48 53500 96 41300

Table 2 Comparison of optimization results

index Savings without favor After the benefit storage experience runs After saving savings Load pulsation quantification index 550.96 538.74 524.64 Generation load rate 81.79% 84.13% 85.96%

Table 3600MW thermal power unit power supply coal consumption

Load x (MW) Coal consumption u ( x ) for power supply (g/kw·h) Load x (MW) Coal consumption u ( x ) for power supply (g/kw·h) 350 330 500 313 400 322 550 310 450 316 600 308

Claims (2)

1. A pumped storage power station optimal scheduling method based on load curve quantization is characterized by comprising the following steps:

(1) arranging a unit maintenance plan: determining the number of available thermal power generating units and the number of generating units of a pumped storage power station; inputting parameters of an available thermal power generating unit and a pumped storage power station generator unit; the parameters comprise rated power generation power of each thermal power generating unit, rated power generation power of each pumped storage power station generator set, pumped storage-power generation cycle efficiency of the pumped storage power station, given water storage capacity and maximum storage capacity;

(2) collecting load prediction data of a power system: the time interval of the power system load prediction data is fixed to deltatThe unit is h, and N points are shared; connecting two adjacent points of the power system load prediction data by using a straight line under a time-power system load prediction value coordinate system to form a power system load curve; the power system load prediction data divides a power system load curve into N scheduling intervals;

(3) establishing a state transition equation of the pumped storage power station:

in the formula (1)x _t The unit of the water storage capacity of the upper reservoir of the pumped storage power station at the end moment of the tth dispatching interval is MWh,the state set X is:

in the formula (2), the compound represented by the formula (2),p _h for pumped storage power stationshThe rated capacity of the unit is MW,h=1,2,3…H，Hthe number of the generator sets of the pumped storage power station,Pthe unit is the maximum common divisor of the capacity of each generator set of the pumped storage power station, MW,x _maxthe unit is the maximum storage capacity of the pumped storage power station and is MWh; the number of state set elements M is determined by:

in the formula (1)d _t A decision variable for the t-th scheduling interval, in MW,；ηwhen the influence of natural water is neglected, the pumping-power generation cycle efficiency of the pumped storage power station is improved; the t-th scheduling interval decision variabled _t The generated energy or pumped-storage power consumption of the pumped-storage power station in the t dispatching interval,d _t when the voltage is more than zero, the generator is in a power generation state;d _t when the value is less than zero, the water pumping state is achieved, and the state set D is as follows:

the constraints of the state transition equation are:

in the formula (5)x ₀The unit of the water storage capacity of an upper reservoir of the pumped storage power station at the starting moment of the 1 st dispatching interval is MWh, and the unit of C is the given water storage capacity of the pumped storage power station and is MWh;

(4) establishing a thermal power generating unit load curve quantization index Q:

in the formula (6), the compound represented by the formula (6),is composed ofT thThe unit of the included angle between the load curve of the thermal power generating unit and the horizontal coordinate axis of each scheduling interval is degree， Is composed ofT thScheduling interval andt-1, scheduling an included angle between load curves of the thermal power generating unit in a section, wherein the unit is degree; the load curve of the thermal power generating unit is obtained by subtracting the optimized output curve of the pumped storage power station from the load curve of the power system;

(5) establishingObjective functionPerforming optimized scheduling on the pumped storage power station by using a dynamic programming method, and sequentially executing the following substeps from a 1 st scheduling interval to an Nth scheduling interval:

1) establishing a t-th scheduling interval index functiong _t (x _t ,d _t )，：

2) Establishing a process indicator functionThe recurrence formula of (c):

whereinThe process index function of the t-th scheduling interval is represented by degree;

3) solving the t-th scheduling interval decision variable by taking the formula (8) as an objective functiond _t Optimum value of (2)d _t ^* The state variable of the t-th dispatching interval is obtained by substituting the state variable into a state transfer equation (1) of the pumped storage power stationx _t Optimum value of (2)x _t ^* ；

4) Will be described ind _t ^* Andx _t ^* substituting the first and second process index function into the recursive formula (8) to obtain the first and second process index functiont procedure index function of scheduling intervalOf (2) an optimal solutionAnd preparing for carrying out optimized dispatching by using a dynamic programming method in the t +1 th dispatching interval.

2. The pumped storage power station optimized dispatching method based on load curve quantization of claim 1 is characterized in that: the first mentionedtIncluded angle between load curve and horizontal coordinate axis of thermal power generating unit in dispatching intervalθ _t The calculation method comprises the following steps:

wherein，Is composed oftThe load curve of the thermal power generating unit is taken at the moment, the unit is MW,t=1,2,3…N，and is；

The first mentionedtScheduling interval andt-1 included angle between load curves of thermal power generating unit in dispatching intervalψ _t The calculation method comprises the following steps:

wherein。