US-12620807-B2 - Coordinated optimization peak shaving method for plurality of power supplies based on fluctuation characteristics of renewable energy

US12620807B2US 12620807 B2US12620807 B2US 12620807B2US-12620807-B2

Abstract

Disclosed is a coordinated peak shaving optimization method for a plurality of power supplies based on a fluctuation characteristic of a renewable energy source, including the following steps: s1: determining a weekly generated electricity quantity of hydropower based on an available capacity and a storage capacity of an electricity quantity; s2: predicting a renewable energy power generation curve and a load curve of a system weekly; s3: determining a start point of peak shaving of the hydropower based on an external transmission curve, the renewable energy generation curve, and the load curve of the system and a generating capacity of the hydropower; s4: determining a weekly peak shaving demand of the system; and s5: establishing an optimization model with a maximum peak shaving demand. The present disclosure proposes a reasonable arrangement for peak shaving, so as to resolve an accommodation problem caused by large-scale access of a renewable energy source.

Inventors

Qiang Zhou
Chengjia Bao
Jinping Zhang
Dingmei Wang
Jianmei Zhang
Jin Li
Zhenzhen Zhang
Sheng Wang
Yongrui Zhang
Yanqi ZHANG
Pengfei Gao
Kun Ding
QINGQUAN LV
Long Zhao
Ruixiao Zhang
Yanhong Ma
Guodong Wu

Assignees

STATE GRID GANSU ELECTRIC POWER RESEARCH INSTITUTE
STATE GRID GANSU ELECTRIC POWER COMPANY
STATE GRID QINGHAI ELECTRIC POWER RESEARCH INSTITUTE
STATE GRID QINGHAI ELECTRIC POWER COMPANY
STATE GRID CORPORATION OF CHINA

Dates

Publication Date: 20260505
Application Date: 20210610
Priority Date: 20200729

Claims (5)

1 . A coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy, comprising the following steps: s1: determining a weekly generated electricity quantity of hydropower based on an available capacity and a storage capacity of an electricity quantity; s2: predicting a renewable energy power generation curve and a load curve of a system weekly; s3: determining a start point of peak shaving of the hydropower based on an external transmission curve, the renewable energy power generation curve, and the load curve of the system and a generating capacity of the hydropower; s4: determining a weekly peak shaving demand of the system; s5: establishing an optimization model with a maximum peak shaving demand; s6: calculating a system optimization model based on a branch and bound algorithm to determine a unit combination scheme of the system; and s7: calculating a peak shaving capability of the system and a margin of the peak shaving capability in a peak period of wind power generation, and controlling a thermal power generator to start up or shut down based on the unit combination scheme of the system.
2 . The coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy according to claim 1 , wherein the determining a start point of peak shaving of the hydropower in step s3 comprises the following steps: a1: determining a basic operating power P B of the hydropower, and determining, based on a capacity limit of a reservoir, that a limit of a planned weekly electricity quantity generated is [E min E max ], wherein a difference between E min and E max is controlled to be between 2% and 5%; when the P B is calculated, a generated electricity quantity of the hydropower is calculated according to a formula E H =(E min +E max )/2 (1); a weekly load curve in a regional power grid is denoted as P l (t), an external transmission power curve is denoted as P T (t), renewable energy power generation is denoted as P R (t); and in this case, an equivalent load of the system is: P l E ( t ) = P l ( t ) + P T ( t ) - P R ( t ) , ( 2 ) the generated electricity quantity of the hydropower meets: ∫ 0 T [ P l E ( t ) - P B ] ⁢ dt = E H , ( 3 ) and a following constraint is met: max ⁢ { P l E ( t ) } - P B ≤ R H max ; ( 4 ) a2: performing initialization based on historical data and operating experience to make P B = P B 0 , and setting an error parameter ε; a3: testing whether an inequality max ⁢ { P l E ( t ) } - P B ≤ R H max ( 4 ) true, and if the inequality does not hold true, increasing P B until the inequality holds true; a4: calculating an integral ∫ 0 T [ P l F ( t ) - P B ] ⁢ d ⁢ t according to a formula ∫ 0 T [ P l E ( t ) - P B ] ⁢ dt = E H ; and ⁢ if ( 3 ) ❘ "\[LeftBracketingBar]" ∫ 0 T [ P l E ( t ) - P B ] ⁢ dt - E H ❘ "\[RightBracketingBar]" > ε , and ( 5 ) ∫ 0 T [ P l E ( t ) - P B ] ⁢ dt > E H , ( 6 ) setting P B =P B −ΔP (7), and on the contrary, setting P B =P B +ΔP (8); and a5: repeating step a4 until ❘ "\[LeftBracketingBar]" ∫ 0 T [ P l E ( t ) - P B ] ⁢ dt - E H ❘ "\[RightBracketingBar]" ≤ ε , ( 9 ) and recording a current value of P B .
3 . The coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy according to claim 1 , wherein an equation condition of an active power balance satisfied by active power outputs, loads, and active power losses of all generators in a power grid at any time for the peak shaving demand in step s4 is as follows: ∑ i = 1 n P G ⁢ i - ∑ i = 1 m P L ⁢ i - Δ ⁢ P Σ = 0 ( 10 ) wherein when a startup mode of thermal power is determined, in order to ensure a maximum power output of the thermal power and reliable power supply of a load when the hydropower is not started, a power P B of the start point of the peak shaving of the hydropower is used to replace a total equivalent load ∑ i = 1 m P L ⁢ i , wherein i represents a total quantity of generators, m represents a total quantity of loads, and P Li represents an active power of an i th load, ∑ i = 1 n P G ⁢ i represents a sum of power outputs of all the generators in the power grid, and a power output P Gi of each generator meets upper and lower limit constraints shown in an equation (11): P Gi min ≤ P Gi ≤ P Gi max ( 11 ) wherein ΔP Σ represents an active power loss of the power grid; and P Gi min and P Gi max respectively represent lower and upper limits of a power output of an i th generator; when an equivalent load of the system is reduced from P B to P min in a peak period of renewable energy power generation, the active power balance is written as follows: ∑ i = 1 n P G ⁢ i ′ - ( P B - Δ ⁢ P L ) - Δ ⁢ P Σ = 0 ( 12 ) wherein ΔP L represents a variable of the equivalent load, which is a positive value; and P Gi ′ represents a changed power of the i th generator; a following formula is obtained by subtracting the equation (10) from an equation (12): ∑ i = 1 n P G ⁢ i ′ - ∑ i = 1 n P G ⁢ i = - Δ ⁢ P L = - R c ; ( 13 ) and in this case, ΔP L represents the variable of the equivalent load, and −ΔP L is always a negative value; −R c is defined as a negative peak shaving capacity of a conventional generator; and in the formula (13), a negative peak shaving capability of the conventional generator is a capability of the conventional generator in reducing an active power output when an equivalent active load in the power grid decreases from the P B to the P min , wherein the negative peak shaving capacity is equal to a difference between an output after adjustment and an output before the adjustment, which is always negative.
4 . The coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy according to claim 1 , wherein the establishing an optimization model with a maximum peak shaving demand in step s5 comprises the following steps: b1: setting a maximum negative peak shaving capacity of a thermal power generator, wherein the maximum negative peak shaving capacity of the thermal power generator depends on a given lower limit parameter ∑ n ∑ i = 1 P Gimin ′ of a power output of a generator combination; and in a power grid, there are a plurality of generator combinations meeting a given load requirement, and different generator combinations have different output lower limits; b2: finding, through optimization, a minimum value of the ∑ n ∑ i = 1 P Gimin ′ meeting a load demand, so as to provide a negative peak shaving capacity and increase an accommodation proportion of renewable energy; b3: determining that there are N generators participating in the peak shaving in a regional power grid, and a power output of an i th generator is P Gi whose value range is defined as [P G min P G max ], wherein P G min represents a lower limit of the power output of the i th generator, and P G max represents an upper limit of the power output of the i th generator; defining an N-dimensional vector to represent a generator combination, namely, C=[c 1 , c 2 , . . . c i , . . . c n ], wherein C i represents a state of the i th generator, and the corresponding generator does not participate in power generation when the state of the i th generator is 0, or has provided a certain power output to the power grid when the state of the i th generator is 1; defining a vector P G min =[P G1 min , . . . P Gi min , . . . P Gn min ] and a vector P G max =[P G1 max , . . . P Gi max , . . . P Gn max ] and then obtaining R c max ( C ) = P B + Δ ⁢ P Σ - C T ⁢ P G ⁢ min ; ( 14 ) and using Ω to represent a value space of C, wherein in this space, a quantity of states of the generator combination is 2 n-1 , and a purpose of the generator combination is to search for the generator state vector C in the state space Ω to maximize a peak shaving capacity, so a corresponding objective function is written as a formula max ⁢ R c max ( C ) = P B + Δ ⁢ P Σ - C T ⁢ P G ⁢ min ; ( 15 ) b4: performing summarization according to step b3 to obtain { C T ⁢ P G ⁢ min ≤ P B + Δ ⁢ P Σ ≤ C T ⁢ P G ⁢ max ( a ) P Gmin ≤ P G ⁢ i ≤ P Gmax ( b ) c i ∈ [ 0 , 1 ] ( c ) ; ( 16 ) b5: the equivalent load changes within a range of L=[P min P B ] when the start point P B of the peak shaving of the hydropower is met, and when the equivalent load of the system continues to decrease, a limit of the negative peak shaving capacity of the system continues to decrease, for ensuring adequacy of a system capacity in the case of a minimum load, namely, P min −C T P G min ≥R th (17), wherein R th represents a minimum peak shaving demand based on a load and a renewable energy prediction error; and b6: obtaining a final optimization model of a thermal power generator combination based on the peak shaving of the hydropower, namely, { max ⁢ R c max ( C ) = P B ′ - C T ⁢ P Gmin ( a ) s . t . C T ⁢ P Gmin ≤ P B + Δ ⁢ P Σ ≤ C T ⁢ P Gmax ( b ) P Gmin ≤ P G ⁢ i ≤ P Gmax ( c ) P min - C T ⁢ P Gmin ≥ R th ( d ) c i ∈ [ 0 , 1 ] ( e ) . ( 18 )
5 . The coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy according to claim 1 , wherein the branch and bound algorithm in step s6 is as follows: C1: placing an original problem P0 into a to-be-resolved problem set L, and setting a target value z*=∞ and a solution variable x*=Ø; C2: determining whether the L is null, wherein if the L is null, the current operation is stopped; or if the L is not null, ILP(k) is selected from the L according to a certain policy, and the ILP(k) is deleted from the L; C3: solving a linear relaxation problem LP(k) of the ILP(k), wherein if the LP(k) has no feasible solution, step C2 is performed; otherwise, z LP k is set to represent an objective function value of the LP(k), wherein x LP k represents the corresponding solution of the LP(k); and C4: if z LP k ≥z*, performing step C2, wherein if x LP k does not meet an integer constraint, step C3 is performed; otherwise, z*=z LP k and x*=x LP k are set, and step C2 is performed.

Description

TECHNICAL FIELD The present disclosure relates to a peak shaving optimization method, and in particular, to a coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy, and belongs to the technical field of electrical engineering. BACKGROUND Large-scale access of a renewable energy source will bring great difficulties to dispatching of a power grid. Although centralized transmission of electric energy can alleviate peak shaving pressure of a local unit to a certain extent, an external transmission curve generally shows a similar characteristic to a load curve, and a power output of a renewable energy source will drop from a maximum power output to zero or rise from zero to a maximum power output within a day. These two factors make a local regional power grid still face an arduous peak shaving task after large-scale access of the renewable energy source to the local regional power grid. Therefore, in order to maximize accommodation of the renewable energy source while tracking a given external transmission curve, it is necessary to reasonably regulate a unit in the regional power grid. The present disclosure comprehensively considers various power supplies in the regional power grid, proposes a weekly coordinated peak shaving strategy of the power supplies, and establishes a unit combination optimization model for a thermal power generator to meet a peak shaving demand under a penetration rate of a high and renewable energy, so as to meet a requirement for safe system operation while tracking the external transmission curve and maximizing the accommodation of the renewable energy. SUMMARY The present disclosure is intended to provide a coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy, so as to resolve the problems described in the background. To achieve the above objective, the present disclosure provides the following technical solutions: A coordinated optimization peak shaving method for a plurality of power supplies based on fluctuation characteristics of renewable energy includes the following steps: s1: determining a weekly generated electricity quantity of hydropower based on an available capacity and a storage capacity of an electricity quantity;s2: predicting a renewable energy power generation curve and a load curve of a system weekly;s3: determining a start point of peak shaving of the hydropower based on an external transmission curve, the renewable energy power generation curve, and the load curve of the system and a generating capacity of the hydropower;s4: determining a weekly peak shaving demand of the system;s5: establishing an optimization model with a maximum peak shaving demand;s6: calculating a system optimization model based on a branch and bound algorithm, and determining a unit combination scheme of the system; ands7: calculating a peak shaving capability of the system and a margin of the peak shaving capability in a peak period of wind power generation. As a preferred technical solution of the present disclosure, the determining a start point of peak shaving of the hydropower in step s3 includes: a1: determining a basic operating power PB of the hydropower, and determining, based on a capacity limit of a reservoir, that a limit of a planned weekly electricity quantity generated [Emin Emax], where a difference between Emin and Emax is controlled to be between 2% and 5%; when the PB is calculated, a generated electricity quantity of the hydropower is calculated according to a formula EH=(Emin+Emax)/2;(1) a weekly load curve in a regional power grid is denoted as Pl(t), an external transmission power curve is denoted as PT(t), renewable energy power generation is denoted as PR(t); and in this case, an equivalent load of the system is: PlE(t)=Pl(t)+PT(t)-PR(t),(2) the generated electricity quantity of the hydropower meets: ∫0T[PlE(t)-PB]⁢d⁢t=EH,(3) and a following constraint is met: max⁢{PlE(t)}-PB≤RHmax;(4)a2: performing initialization based on historical data and operating experience to make PB=PB0, and setting an error parameter ε; a3: testing whether an inequality max⁢{PlE(t)}-PB≤RHmax(4) holds true, and if the inequality does not hold true, increasing PB until the inequality holds true; a4: calculating an integral ∫0T[PlE(t)-PB]⁢d⁢t according to a formula ∫0T[PlE(t)-PB]⁢d⁢t=EH;and⁢ if(3)❘"\[LeftBracketingBar]"∫0T[PlE(t)-PB]⁢dt-EH❘"\[RightBracketingBar]">ε,and(5)∫0T[PlE(t)-PB]⁢dt>EH,(6) setting PB=PB−ΔP (7), and on the contrary, setting PB=PB+ΔP (8); and a5: repeating step a4 until ❘"\[LeftBracketingBar]"∫0T[PlE(t)-PB]⁢dt-EH❘"\[RightBracketingBar]"≤ε,(9) and recording a current value of PB. As a preferred technical solution of the present disclosure, an equation condition of an active power balance satisfied by active power outputs, loads, and active power losses of all generators in a power grid at any time for the pea