Robust optimization method for unit commitment of power system with wind power considering wind turbine frequency support capability variation

Abstract

The application discloses a kind of wind power system unit combination robust optimization methods considering fan frequency support ability change, its steps include:1establish system frequency response model;2establish fan participation inertia support after frequency modulation capability change model;3based on fan frequency modulation capability change model, construct the two-stage stochastic robust optimization model with system total cost minimum as target;4robust optimization method is solved.This application can reduce the system operation risk caused by ignoring fan frequency modulation capability change, so as to improve the operation reliability of large-scale wind power grid-connected system, provide reference for dispatch personnel to carry out unit scheduling optimization.

Description

TECHNICAL FIELD

[0001] The present application is directed to a wind power system, and proposes a wind power system unit commitment robust optimization method considering the change of frequency support ability of wind turbine. BACKGROUND

[0002] In the process of realizing low-carbon and clean transformation of power supply structure, due to the fact that the frequency support of new energy units to the power grid is much smaller than that of traditional thermal power units, the new energy penetration rate of the system is limited by inertial support and frequency modulation reserve index. In addition, the intermittent, fluctuating and difficult to accurately predict characteristics of wind power bring great difficulties to the safe and economic dispatching of power system. In order to maintain the active power balance of the system with high proportion of wind turbine units and ensure the frequency safety and stability of the system, the inertial support and frequency modulation potential of wind turbine units need to be tapped to meet the uncertainty of new energy grid connection.

[0003] The current wind turbine often provides inertial support power to the system by releasing rotor kinetic energy, and obtains frequency modulation capability by adopting load shedding operation. When the system is disturbed, the inertial support process and frequency regulation process usually appear successively, and the existing research often ignores the influence between the two processes and focuses on a single unit, resulting in a too complex overall response process. If the change of frequency modulation capability of wind turbine is ignored, the adjustable range of wind turbine output cannot be accurately described, which may cause secondary frequency drop of the system, a large amount of unnecessary wind power curtailment and load shedding, etc., affecting the safe and stable operation of the power system. SUMMARY

[0004] The present application is to solve the above-mentioned deficiencies in the prior art, and proposes a wind power system unit commitment robust optimization method considering the change of frequency support ability of wind turbine, so as to describe the change of frequency modulation capability of wind turbine after participating in the inertial support of the system, thereby improving the flexibility of the system and the availability of flexible resources, and ensuring the safety and reliability of the operation of the power system.

[0005] In order to achieve the above-mentioned application purposes, the present application adopts the following technical solutions:

[0006] The wind power system unit commitment robust optimization method considering the change of frequency support ability of wind turbine has the characteristics that the following steps are performed:

[0007] Step 1, establishing a frequency response model of the power system:

[0008] Step 2, establishing a model of the change of frequency modulation capability of wind turbine after participating in the inertial support:

[0009] Step 3, based on the frequency regulation capability change model of the wind turbine, a two-stage stochastic robust optimization model is constructed, including an objective function, a constraint condition of one-stage day-ahead scheduling consumption, and a constraint condition of two-stage real-time power balance cost;

[0010] Step 4, the two-stage stochastic robust optimization model is solved by using the C&CG algorithm to obtain an optimal thermal turbine and wind turbine scheduling scheme.

[0011] The unit commitment robust optimization method for a wind power system according to the application has the following characteristics:

[0012] Step 1.1, the related equivalent parameters of the wind power system in the frequency response process are calculated by using formula (1):

[0013]

[0014] In formula (1), nu and nw represent the number of thermal power units and the number of wind power units in the power system with frequency regulation capability, and n is the total number of units in the power system; P gR,j and P wR,k respectively represent the rated capacity of the jth thermal power unit and the kth wind power unit, P i is the rated capacity of the ith unit; k g,j and k w,k respectively represent the mechanical gain of the jth thermal power unit and the kth wind turbine unit, k gc and k wc are the equivalent gains of the aggregated thermal power unit and the aggregated wind turbine unit; u i,t is the operating state of the ith unit at time t, u g,j,t is the operating state of the jth thermal power unit at time t, and u w,k,t is the operating state of the kth wind power unit at time t;

[0015] The equivalent time constant T gc of the governor of the aggregated thermal power unit is obtained by using formula (2):

[0016]

[0017] In formula (2), T g,j is the time constant of the governor of the jth thermal power unit;

[0018] The equivalent virtual inertia support time constant H c of the power system is obtained by using formula (3):

[0019]

[0020] In formula (3), H g,jHj is the virtual inertia time constant of the jth thermal power unit w,k Hk is the virtual inertia time constant of the kth wind power unit

[0021] Step 1.2, the frequency variation frequency domain relationship of the power system containing wind power is expressed by formula (4):

[0022]

[0023] In formula (4), Δf(t) is the frequency deviation of the power system at time t, ΔP is the power shortage, and D is the damping coefficient of the power system; is the first intermediate calculation quantity, ξ is the second intermediate calculation quantity, and Φ is the third intermediate calculation quantity;

[0024] The maximum frequency deviation moment t of the power system is calculated by formula (5) fm , and the maximum frequency deviation Δf(t fm ) of the power system is obtained.

[0025]

[0026] The step 2 is performed as follows:

[0027] Step 2.1, the maximum output P w,k,mppt of the kth wind power unit at wind speed v is calculated by formula (6):

[0028]

[0029] In formula (6), ω mppt,k is the corresponding rotating speed of the kth wind power unit at the maximum output, ρ is the air density, R is the blade radius of the wind power unit, and ω w,k is the rotating speed of the kth wind power unit.

[0030] Step 2.2, the rotating speed change of the wind power unit after inertia support is calculated:

[0031] The relationship formula between the rotating speed ω w,k (t) of the kth wind power unit at time t and the frequency deviation Δf(t) of the power system at time t is obtained by formula (7):

[0032]

[0033] In formula (7), ω w0,k is the rotating speed of the kth wind power unit before inertia support; and J w,k is the moment of inertia of the kth wind power unit.

[0034] Step 2.3, the output power range of the kth wind power unit after inertia support is calculated by formula (8):

[0035]

[0036] in formula (8), k dr,k is the load shedding ratio of the kth wind turbine, P wd,k is the output power of the kth wind turbine after inertia support, P wd,k,m is the maximum output power of the kth wind turbine after inertia support, ω wd,k is the rotating speed of the kth wind turbine after inertia support.

[0037] The step 3 is performed according to the following steps:

[0038] Step 3.1, a target function with the minimum power system operation consumption as the target is constructed by using formula (9):

[0039]

[0040] in formula (9), nd represents the total time length; C g is the power generation consumption of each thermal power generator; C st is the start-up cost of each thermal power generator, z g,j,t represents whether the jth thermal power generator is started at the tth time, ψ is the scene number, and P(ψ) represents the probability of the appearance of the ψth scene, and represent the up-regulation consumption and the down-regulation consumption of each thermal power generator, is the up-regulation or down-regulation power of the jth thermal power generator at the tth time in the ψth scene; C wp,t , C ld,t is the wind power curtailment and the penalty value of load shedding of the power system at the tth time, P wp,t,ψ , P ld,t,ψ is the wind power curtailment and the load shedding amount of the power system at the tth time in the ψth scene; C wde,t represents the penalty value of the wind turbine load shedding at the tth time, P wde,t,ψ represents the total load shedding amount of all wind turbines at the tth time in the ψth scene;

[0041] The wind speed value D(ψ) of the ψth scene in the robust optimization is obtained by using formula (10):

[0042]

[0043] in formula (10), v t,Ψ represents the wind speed at the tth time in the ψth scene, v t,Ψ , is the lower limit and the upper limit of the fluctuation of v t,Ψ at the tth time;

[0044] Step 3.3, constructing the constraint condition of the one-stage day-ahead scheduling consumption, including: power balance constraint, thermal power output constraint, inertia constraint, start-stop identification constraint, minimum start-stop constraint;

[0045] The power balance constraint is constructed by using formula (11):

[0046]

[0047] In formula (11), P w,k,t is the output prediction value of the kth wind turbine at t moment, P g,j,t is the output prediction value of the jth thermal power unit at t moment, P ld,t is the prediction value of the load at t moment;

[0048] The thermal power output constraint is constructed by using formula (12):

[0049]

[0050] In formula (12), is the estimated output of the jth thermal power unit at t moment, and the value range is the minimum output and the maximum output of the jth thermal power unit;

[0051] The wind power output constraint is constructed by using formula (13):

[0052]

[0053] In formula (13), is the output of the kth wind turbine at t moment in the maximum power point operation mode;

[0054] The inertia constraint is constructed by using formula (14):

[0055]

[0056] In formula (14), H min,t is the minimum inertia time coefficient of the power system at t moment, ΔP t is the power shortage of the power system at t moment, f max,t is the maximum frequency change speed allowed by the power system at t moment; H g,j is the inertia time constant of the jth thermal power unit, H w,k is the inertia time constant coefficient of the kth wind turbine;

[0057] The thermal power unit start-stop identification constraint is constructed by using formula (15):

[0058]

[0059] In formula (15), b g,j,tis the shutdown identifier of the jth thermal power unit at time t; u g,j,t-1 is the on-off state of the jth thermal power unit at time t-1;

[0060] The minimum on-off constraint is constructed by using formula (16):

[0061]

[0062] In formula (16), T on is the minimum on time of the thermal power unit, T off is the minimum off time of the thermal power unit; u g,j,t+1 is the on-off state of the jth thermal power unit at time t+1;

[0063] Step 3.4, constructing the constraint condition of the two-stage real-time power balance cost, including: the second-stage power balance constraint, the wind power daily output constraint and the thermal power unit up-down adjustment constraint;

[0064] The second-stage power balance constraint is constructed by using formula (17):

[0065]

[0066] In formula (17), P w,k,t,Ψ is the actual output of the kth wind turbine unit at time t in the ψth scenario, P load,t,Ψ is the load of the power system at time t in the ψth scenario;

[0067] The wind turbine daily output constraint is constructed by using formula (18):

[0068]

[0069] The up-down adjustment constraint of the thermal power unit is constructed by using formula (19):

[0070]

[0071] In formula (19), is the up-regulation output of the jth thermal power unit at time t in the ψth scenario, is the down-regulation output of the jth thermal power unit at time t in the ψth scenario, P g,j,max is the maximum output of the jth thermal power unit, P g,j,min is the minimum output of the jth thermal power unit.

[0072] The step 4 is performed as follows:

[0073] Step 4.1, using the column and constraint generation algorithm to decompose the two-stage robust optimization model into a one-stage day-ahead scheduling consumption main problem MP as shown in formula (20) and a power balance cost sub-problem SP as shown in formula (21);

[0074]

[0075] in formula (20), and x t are the coefficient set of the objective function in the one-stage day-ahead scheduling consumption and the variable set at time t, respectively, and L is the value of the objective function of the sub-problem; A is the coefficient set in the one-stage constraint, d t is the constant set at time t in the one-stage constraint; C2 and y t are the coefficient set of the objective function in the two-stage power balance cost and the variable set at time t, respectively, E is the coefficient set related to the one-stage variable in the two-stage constraint, G is the coefficient set related to the two-stage variable in the two-stage constraint, h t is the constant set in the two-stage constraint, e t and M are the robust optimization integer variable set at time t and the coefficient set related thereto, respectively;

[0076]

[0077] in formula (21), and y t,Ψ are the coefficient set of the objective function in the two-stage power balance cost and the variable set at time t in the ψth scenario, respectively; h t,Ψ and e t,Ψ are the constant set and the robust optimization integer variable set of the two-stage constraint at time t in the ψth scenario, respectively;

[0078] Step 4.2, solve the one-stage day-ahead scheduling consumption main problem MP in the benchmark scenario, and update the lower bound After obtaining the thermal power unit output and unit combination, substitute it into the power balance cost sub-problem SP, and update the upper bound Thus, the corresponding thermal fan scheduling scheme and the solution of the robust optimization integer variable representing the worst scenario are obtained;

[0079] Add Ex t +Gy t ≥h t,Ψ -Me t,Ψ constraint in the main problem MP, and solve the updated one-stage day-ahead scheduling consumption main problem MP, and update the lower bound LB, and repeat the iteration until the optimal thermal fan scheduling scheme satisfying formula (22) is obtained.

[0080]

[0081] in formula (22), denotes the convergence threshold.

[0082] The electronic equipment comprises a memory and a processor, and is characterized in that the memory is used for storing a program supporting the processor to execute the robust optimization method for unit commitment of a power system containing wind power, and the processor is configured to execute the program stored in the memory.

[0083] The computer readable storage medium stores a computer program, and the computer program is executed by a processor to execute the steps of the robust optimization method for unit commitment of a power system containing wind power.

[0084] Compared with the prior art, the method has the advantages that:

[0085] 1. The method considers multiple units in the power system as aggregated units in step 1.1, overcomes the problem that the existing research focuses on a single unit, resulting in a too complex frequency response process to calculate, and enables the method to quickly obtain the frequency deviation of the system under each scenario, thereby providing a basis for subsequent calculation of the frequency support capability of wind power.

[0086] 2. The method considers the changes of mechanical and electrical parameters in the frequency response process of the wind power unit in steps 2.2 and 2.3, accurately depicts the change of the frequency regulation capability of the wind power unit participating in inertia support, reduces the probability of power supply accidents caused by inaccurate consideration of the frequency regulation capability in the prior art, and improves the safety and reliability of the operation of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0087] Figure 1 A flowchart of the robust optimization method for unit commitment of a power system containing wind power considering the change of the frequency support capability of wind power;

[0088] Figure 2 A schematic diagram of the frequency response process of the system;

[0089] Figure 3 A schematic diagram of the change of the speed in the inertia support process of the wind power unit;

[0090] Figure 4 A schematic diagram of the two-stage stochastic robust optimization scheduling model; DETAILED DESCRIPTION

[0091] In this embodiment, a robust optimization method for unit commitment of a power system containing wind power considering the change of the frequency support capability of wind power, as shown in Figure 1 The method comprises the following steps in sequence:

[0092] Step 1, as shown in Figure 2As shown, after the system is disturbed, frequency deviation occurs in the system after the inertia support process and the damping process. The fan unit participates in system frequency regulation through droop control, adjusts its output power according to the frequency deviation, and the thermal power unit participates in system frequency regulation through the speed governor, thereby establishing a system frequency response model:

[0093] Step 1.1, use formula (1) to calculate the relevant equivalent parameters of the power system containing wind power in the frequency response process:

[0094]

[0095] In formula (1), nu and nw represent the number of thermal power units and the number of wind power units in the power system with frequency regulation capability, and n is the total number of units in the power system; P gR,j and P wR,k respectively represent the rated capacity of the jth thermal power unit and the kth wind power unit, P i is the rated capacity of the ith unit; k g,j and k w,k respectively represent the mechanical gain of the jth thermal power unit and the kth wind power unit, k gc and k wc are the equivalent gains of the aggregated thermal power unit and the aggregated wind power unit; u i,t is the operating state of the ith unit at time t, u g,j,t is the operating state of the jth thermal power unit at time t, and u w,k,t is the operating state of the kth wind power unit at time t;

[0096] The equivalent time constant T gc of the speed governor of the aggregated thermal power unit is obtained by formula (2):

[0097]

[0098] In formula (2), T g,j is the time constant of the speed governor of the jth thermal power unit;

[0099] The equivalent virtual inertia support time constant H c of the power system is obtained by formula (3):

[0100]

[0101] In formula (3), H g,j is the virtual inertia time constant of the jth thermal power unit, and H w,k is the virtual inertia time constant of the kth wind power unit;

[0102] Step 1.2, use formula (4) to represent the frequency variation frequency domain relationship of the power system containing wind power:

[0103]

[0104] In formula (4), Δf(t) is the frequency deviation amount of the power system at time t, ΔP is the power shortage, and D is the damping coefficient of the power system; is the first intermediate calculation amount, ξ is the second intermediate calculation amount, and Φ is the third intermediate calculation amount;

[0105] The moment t at which the frequency deviation of the power system is maximum is calculated by using formula (5) fm , and thus the maximum frequency deviation amount Δf(t fm ) of the power system is obtained.

[0106]

[0107] Step 2, as shown in Figure 3 , the fan participates in the system inertia support process, and the released rotor kinetic energy provides inertia support power for the system. When the torque generated by the electromagnetic power is higher than the torque generated by the mechanical power, the blade speed decreases. The change of the fan speed will also affect its frequency regulation capability. According to the process, a model of the change of the frequency regulation capability after the fan participates in the inertia support is established.

[0108] Step 2.1, the maximum output P w,k,mppt of the kth wind turbine at the wind speed v is calculated by using formula (6):

[0109]

[0110] In formula (6), ω mppt,k is the speed corresponding to the maximum output of the kth wind turbine, ρ is the air density, R is the radius of the blade of the wind turbine, and ω w,k is the speed of the kth wind turbine.

[0111] Step 2.2, the change of the speed of the wind turbine after the inertia support is calculated:

[0112] The relationship between the speed ω w,k of the kth wind turbine at time t and the frequency deviation Δf(t) of the power system at time t is obtained by using formula (7):

[0113]

[0114] In formula (7), ω w0,k is the speed of the kth wind turbine before the inertia support; and J w,k is the moment of inertia of the kth wind turbine.

[0115] Step 2.3, the range of the output power of the kth wind turbine after the inertia support is calculated by using formula (8):

[0116]

[0117] In formula (8), k dr,k is the load shedding ratio of the kth wind turbine, P wd,k is the output power of the kth wind turbine after inertia support, P wd,k,m is the maximum output power of the kth wind turbine after inertia support, ω wd,k is the rotating speed of the kth wind turbine after inertia support;

[0118] Step 3, based on the wind turbine frequency modulation capability change model, a two-stage stochastic robust optimization model with the minimum system total cost as the target is constructed; as shown in formula (9), the model takes the minimum system operation scheduling consumption as the optimization target, is divided into a day-ahead scheduling stage and an intra-day real-time scheduling stage, and the optimal unit combination scheme of the system is obtained through multiple iterations. Figure 4

[0119] Step 3.1, a target function with the minimum power system operation consumption as the target is constructed by using formula (9):

[0120]

[0121] In formula (9), nd represents the total duration; C g is the power generation consumption of each thermal power generator; C st is the start-up cost of each thermal power unit, z g,j,t represents whether the jth thermal power unit is started at t, ψ is the scene number, and P(ψ) represents the probability of the occurrence of the ψth scene, and represent the up-regulation consumption and the down-regulation consumption of each thermal power unit, is the up-regulation or down-regulation output of the jth thermal power unit at t in the ψth scene; C wp,t , ld,t is the penalty value of the wind power curtailment and load shedding of the power system at t, P wp,t,ψ , ld,t,ψ is the wind power curtailment and load shedding of the power system at t in the ψth scene; C wde,t represents the penalty value of the wind turbine load shedding at t, P wde,t,ψ represents the total output load shedding of all wind turbines at t in the ψth scene.

[0122] The wind speed value D(ψ) of the ψth scene under the robust optimization is obtained by using formula (10):

[0123]

[0124] In formula (10), v t,Ψ represents the wind speed at t in the ψth scene, v t,Ψ ,​​​ v t,Ψ the lower limit and the upper limit of the fluctuation at time t;

[0125] Step 3.3, constructing a constraint condition of the one-stage day-ahead scheduling consumption, including: power balance constraint, thermal power output constraint, inertia constraint, start-stop identification constraint, minimum start-stop constraint;

[0126] constructing the power balance constraint by using formula (11):

[0127]

[0128] In formula (11), P w,k,t is the output prediction value of the kth wind turbine at time t, P g,j,t is the output prediction value of the jth thermal power unit at time t, P ld,t is the prediction value of the load at time t;

[0129] constructing the thermal power output constraint by using formula (12):

[0130]

[0131] In formula (12), is the estimated output of the jth thermal power unit at time t, and the value range is the minimum output and the maximum output of the jth thermal power unit;

[0132] constructing the wind power output constraint by using formula (13):

[0133]

[0134] In formula (13), is the output of the kth wind turbine at time t in the maximum power point operation mode;

[0135] constructing the inertia constraint by using formula (14):

[0136]

[0137] In formula (14), H min,t is the minimum inertia time coefficient of the power system at time t, ΔP t is the power shortage of the power system at time t, f max,t is the maximum frequency change speed allowed by the power system at time t; H g,j is the inertia time constant of the jth thermal power unit, H w,k is the inertia time constant coefficient of the kth wind turbine;

[0138] constructing the thermal power unit start-stop identification constraint by using formula (15):

[0139]

[0140] In formula (15), b g,j,t is the closing identifier of the jth thermal power unit at time t; u g,j,t-1 is the on-off state of the jth thermal power unit at time t-1;

[0141] The minimum on-off constraint is constructed by using formula (16):

[0142]

[0143] In formula (16), T on is the minimum on time of the thermal power unit, T off is the minimum off time of the thermal power unit; u g,j,t+1 is the on-off state of the jth thermal power unit at time t+1;

[0144] Step 3.4, constructing the constraint condition of the two-stage real-time power balance cost, including: the second-stage power balance constraint, the wind power daily output constraint and the thermal power unit up-down adjustment constraint;

[0145] The second-stage power balance constraint is constructed by using formula (17):

[0146]

[0147] In formula (17), P w,k,t,Ψ is the actual output of the kth wind turbine unit at time t in the ψth scenario, P load,t,Ψ is the load of the power system at time t in the ψth scenario;

[0148] The wind turbine daily output constraint is constructed by using formula (18):

[0149]

[0150] The up-down adjustment constraint of the thermal power unit is constructed by using formula (19):

[0151]

[0152] In formula (19), P is the up-regulation output of the jth thermal power unit at time t in the ψth scenario, is the down-regulation output of the jth thermal power unit at time t in the ψth scenario, P g,j,max is the maximum output of the jth thermal power unit, P g,j,min is the minimum output of the jth thermal power unit;

[0153] Step 4, solving the unit commitment model by using the robust optimization method;

[0154] Step 4.1, using column and constraint generation algorithm to decompose the two-stage robust optimization model into a one-stage day-ahead scheduling consumption master problem MP as shown in equation (20) and a power balance cost sub-problem SP as shown in equation (21);

[0155]

[0156] In equation (20), and x t are the coefficient set of objective function in one-stage day-ahead scheduling consumption and the variable set at time t respectively, L is the sub-problem objective function value; A is the coefficient set in one-stage constraints, d t is the constant set at time t in one-stage constraints; C2 and y t are the coefficient set of objective function in two-stage power balance cost and the variable set at time t respectively, E is the coefficient set related to one-stage variables in two-stage constraints, G is the coefficient set related to two-stage variables in two-stage constraints, h t is the constant set in two-stage constraints, e t and M are the robust optimization integer variable set at time t and the coefficient set related thereto respectively;

[0157]

[0158] In equation (21), and y t,Ψ are the coefficient set of objective function in two-stage power balance cost and the variable set at time t in scenario ψ respectively; h t,Ψ and e t,Ψ are the constant set and the robust optimization integer variable set in two-stage constraints at time t in scenario ψ respectively;

[0159] Step 4.2, solving the one-stage day-ahead scheduling consumption master problem MP under the benchmark scenario, updating the lower bound obtaining the thermal power unit output and unit commitment, and substituting into the power balance cost sub-problem SP, updating the upper bound and obtaining the corresponding thermal fan scheduling scheme and the solution of the robust optimization integer variable representing the worst scenario; adding the constraint Ex t +Gy t ≥h t,Ψ -Me t,Ψ in the master problem MP, and solving the updated master problem MP and updating the lower bound LB, so as to continuously iterate until the optimal thermal fan scheduling scheme satisfying the convergence requirement equation (22) is obtained;

[0160]

[0161] In equation (22), represents a convergence threshold, in this embodiment, Take 0.001.

[0162] In this embodiment, an electronic device includes a memory for storing a program supporting a processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0163] In this embodiment, a computer readable storage medium has a computer program stored thereon, and the computer program is run by a processor to execute the steps of the above method.

Claims

1. A method for robust optimization of unit commitment in a power system with wind power considering the change of frequency support capability of wind turbines, characterized in that, is performed as follows: Step 1, a frequency response model of the power system is established: Step 1.1, the relevant equivalent parameters of the power system containing wind power in the frequency response process are calculated by using formula (1): (1) In formula (1), nu and nw represent the number of thermal power units with frequency modulation capability and the number of wind power units in the power system, n is the total number of units in the power system; P gR,j and P wR,k respectively represent the rated capacity of the jth thermal power unit and the kth wind power unit, P i is the rated capacity of the ith unit; k g,j and k w,k respectively represent the mechanical gain of the jth thermal power unit and the kth wind power unit, k gc and k wc are the equivalent gains of the aggregated thermal power units and the aggregated wind power units; u i,t is the operating state of the ith unit at time t, u g,j,t is the operating state of the jth thermal power unit at time t, u w,k,t is the operating state of the kth wind power unit at time t; The equivalent time constant T of a speed regulator of a polygeneration thermal power unit is obtained by using formula (2) gc : (2) In formula (2), T g,j is the time constant of the speed regulator of the jth thermal power unit; An equivalent virtual inertia support time constant H of the power system is obtained using formula (3) c : (3) In formula (3), H g,j is the virtual inertia time constant of the jth thermal power unit, H w,k is the virtual inertia time constant of the kth wind power unit; Step 1.2, the frequency variation frequency domain relationship of the power system containing wind power is represented by using formula (4): (4) In formula (4), Δf(t) is the frequency deviation amount of the power system at t time, ΔP is the power shortage, D is the damping coefficient of the power system; ϑ is the first intermediate calculation quantity, ξ is the second intermediate calculation quantity, and Φ is the third intermediate calculation quantity; The maximum moment of frequency deviation of the power system is calculated by using formula (5) Thus, the maximum frequency deviation amount Δf(t fm ) of the power system is obtained. (5) Step 2, a model of the change of the frequency modulation capability of the wind turbine after participating in inertia support is established: Step 2.

1. Calculate the maximum output of the kth wind turbine at wind speed v, Pk(v) using equation (6) w,k,mppt : (6) In formula (6), ω mppt,k is the corresponding rotating speed of the kth wind turbine at the maximum output, p is the air density, R is the radius of the wind turbine blade, ω w,k is the rotating speed of the kth wind turbine; Step 2.2, the speed variation of the wind turbine after inertia support is calculated: The rotational speed ω of the kth wind turbine at time t is obtained using equation (7) w,k (t) the frequency deviation of the power system at time t The relationship is given by (7) In formula (7), ωk w0,k is the pre-inertia support rotational speed of the kth wind turbine generator; Jk w,k is the rotational inertia of the kth wind turbine generator; Step 2.3, the range of the output power of the kth wind turbine after inertia support is calculated by using formula (8): (8) In formula (8), k dr,k is the load reduction ratio of the kth wind turbine, P wd,k is the output power of the kth wind turbine after inertia support, P wd,k,m is the maximum output power of the kth wind turbine after inertia support, ω wd,k is the rotational speed of the kth wind turbine after inertia support. Step 3, based on the model of the change of the frequency modulation capability of the wind turbine, a two-stage stochastic robust optimization model is constructed, including an objective function, a constraint condition of the one-stage day-ahead scheduling consumption, and a constraint condition of the two-stage real-time power balance cost; Step 3.1, the objective function taking the minimum power system operation consumption as the target is constructed by using formula (9): (9) In formula (9), nd represents the total time length; C g is the power generation consumption of each thermal power generator;C st is the start-up cost of each thermal power generator, z g,j,t represents whether the jth thermal power generator is started at the tth time, ψ is the scene number, and P(ψ) represents the probability of the ψth scene appearing, and represent the up-regulation consumption and the down-regulation consumption of each thermal power generator, 、 is the up-regulation or down-regulation power of the jth thermal power generator at the tth time in the ψth scene;C wp,t , C ld,t is the wind power curtailment and the penalty value of load shedding of the power system at the tth time, P wp,t,ψ , P ld,t,ψ is the wind power curtailment and the load shedding amount of the power system at the tth time in the ψth scene;C wde,t represents the penalty value of the wind power curtailment at the tth time, P wde,t,ψ represents the total wind power curtailment of all wind power generators at the tth time in the ψth scene; The wind speed value of the ψth scenario under the robust optimization is obtained by using formula (10) : (10) In formula (10), denotes the wind speed at time t in the ψth scenario, , is the lower and upper limits of the fluctuation at time t. Step 3.3, the constraint condition of the one-stage day-ahead scheduling consumption is constructed, including: a power balance constraint, a thermal power output constraint, an inertia constraint, a start-stop identification constraint, and a minimum start-stop constraint; The power balance constraint is constructed by using formula (11): (11) In formula (11), P w,k,t is the predicted output value of the kth wind turbine at time t, P g,j,t is the predicted output value of the jth thermal power unit at time t, P ld,t is the predicted value of the load at time t; The thermal power output constraint is constructed by using formula (12): (12) In formula (12), is the estimated output of the jth thermal power unit at time t, and its value range is the minimum output and the maximum output of the jth thermal power unit. The wind power output constraint is constructed by using formula (13): (13) In formula (13), Pmax(k, t) is the output of the kth wind turbine at the maximum power point operation mode at time t. The inertia constraint is constructed by using formula (14): (14) In formula (14), H min,t is the minimum inertia time coefficient of the power system at time t, ΔP t is the power shortage of the power system at time t, f max,t is the maximum frequency change speed allowed by the power system at time t; H g,j is the inertia time constant of the jth thermal power unit, H w,k is the inertia time constant coefficient of the kth wind power unit The thermal power unit start-stop identification constraint is constructed by using formula (15): (15) In formula (15), b g,j,t is the closing identifier of the jth thermal power unit at time t; is the on-off state of the jth thermal power unit at time t-1; The minimum start-stop constraint is constructed by using formula (16): (16) In formula (16), T on is the minimum start-up time of the thermal power unit, T off is the minimum shut-down time of the thermal power unit; is the start-up and shut-down state of the jth thermal power unit at time t+1. Step 3.4, the constraint condition of the two-stage real-time power balance cost is constructed, including: a second-stage power balance constraint, a wind power daily output constraint, and a thermal power unit up-down adjustment constraint; The second-stage power balance constraint is constructed by using formula (17): (17) In formula (17), is the actual output of the kth fan unit at time t in the ψth scenario, is the load of the power system at time t in the ψth scenario. The wind power daily output constraint is constructed by using formula (18): (18) The thermal power unit up-down adjustment constraint is constructed by using formula (19): (19) In formula (19), is the up-regulated power of the jth thermal power unit at time t in the ψth scenario, is the down-regulated power of the jth thermal power unit at time t in the ψth scenario, g,j,max is the maximum power of the jth thermal power unit, g,j,min is the minimum power of the jth thermal power unit. Step 4, the two-stage stochastic robust optimization model is solved by using the C&CG algorithm to obtain an optimal thermal power and wind power scheduling scheme.

2. The wind power system unit commitment robust optimization method according to claim 1, wherein, The step 4 is performed as follows: Step 4.1, the two-stage robust optimization model is decomposed into a one-stage day-ahead scheduling consumption main problem MP shown in formula (20) and a power balance cost sub-problem SP shown in formula (21) by using a column and constraint generation algorithm; (20) In formula (20), and x t are respectively the coefficient set of the objective function in the first-stage day-ahead scheduling consumption and the variable set at time t, and L is the objective function value of the sub-problem; A is the coefficient set in the first-stage constraint, d t is the constant set at time t in the first-stage constraint; C2 and y t are respectively the coefficient set of the objective function in the second-stage power balance cost and the variable set at time t, E is the coefficient set related to the first-stage variable in the second-stage constraint, G is the coefficient set related to the second-stage variable in the second-stage constraint, h t is the constant set in the second-stage constraint, e t and M are respectively the robust optimization integer variable set at time t and the coefficient set related thereto. (21) In formula (21), and are respectively the coefficient set of the objective function in the two-stage power balance cost and the variable set of the ψth scenario at time t; and are respectively the constant set of the two-stage constraint and the robust optimization integer variable set at the ψth scenario at time t; Step 4.2, solving the main problem MP of one-stage day-ahead scheduling consumption under the benchmark scenario, and updating the lower bound , obtaining the thermal power unit output and the unit combination, substituting into the power balance cost sub-problem SP, and updating the upper bound , thereby obtaining the corresponding thermal fan scheduling scheme and the solution of the robust optimization integer variable representing the worst scenario; Main problem MP is increased Constraint, the one-stage day-ahead dispatch consumption main problem MP is solved again after updating, and the lower bound LB is updated, and the iteration is repeated until the optimal turbine generator dispatching scheme satisfying formula (22) is obtained. (22) In formula (22), represents a convergence threshold.

3. An electronic device comprising a memory and a processor, characterized in that The memory is used to store a program supporting the processor to execute the wind power-containing power system unit commitment robust optimization method in claim 1 or 2, and the processor is configured to execute the program stored in the memory.

4. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to perform the steps of the wind power-containing power system unit commitment robust optimization method in claim 1 or 2.

Images