Multi-energy power system optimization scheduling method based on improved robust optimization

Abstract

The invention discloses a multi-energy power system optimization scheduling method based on improved robust optimization, and relates to the technical field of power system scheduling. Minimizing the total operation cost of the power system as a total objective function; establishing constraint conditions for the total target, wherein the constraint conditions comprise thermal power generating unit constraint, renewable energy system constraint and energy storage system constraint; establishing a robust optimization scheduling model according to the total objective function and each constraint condition, and processing robust optimization by adopting a Benders decomposition method to obtain a decision variable meeting the objective function; and adjusting a scheduling plan according to the decision variable to realize optimal scheduling of the power system. Introducing a constraint condition of multi-energy overall balance; an adjustment factor is introduced to control the balance between robustness and economy, the uncertainty influence of wind power photovoltaic fluctuation is dynamically controlled through an uncertainty influence factor, Benders are introduced, a large-scale problem is decomposed into a main problem and a sub-problem, the calculation complexity is reduced while efficient solving is achieved, and the method is suitable for large-scale real-time scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching, and in particular to a multi-energy power system optimization dispatching method based on improved robust optimization. Background Art

[0002] Renewable energy sources such as wind power and photovoltaics are increasingly used for power generation. However, with the large-scale access of renewable energy, the operation of the power system faces many uncertainties. For example, the output of wind power and photovoltaics is affected by the weather and has large volatility, randomness and intermittency. Traditional deterministic optimization scheduling methods cannot effectively deal with these uncertainties, which may make the scheduling scheme difficult to implement in actual operation. Robust optimization, as a method of dealing with uncertainty, can ensure the feasibility of the scheme within the range of uncertainty parameters. However, classic robust optimization methods are often too conservative or too aggressive, resulting in frequent adjustments to the scheduling plan, increasing scheduling pressure and costs.

[0003] CN104299173A discloses a robust optimization day-ahead scheduling method suitable for access to multiple energy sources. The method performs scheduling based on the sum of the fuel costs of thermal power units in 96 time periods a day and the power generation costs of various renewable energy sources as the minimum target. It uses a single robustness model, and the constraints are concentrated on the output of the generator units and the prediction errors of wind and solar power generation. It cannot be applied to the real-time optimization of large-scale power systems.

[0004] CN108090632A discloses a multi-time scale dispatching method for a new energy grid-connected power system based on robust optimization, wherein the output of dispatchable energy is used as a decision variable to establish the objective function of each time scale dispatching plan; the constraint conditions of each objective function including robust constraints and traditional physical constraints are established; the robust optimization dispatching model of each time scale is established according to each objective function and the corresponding constraint conditions; the robust level value under each time scale is selected according to the robust level value adjustment rule, and is input into the robust optimization dispatching model of each time scale respectively, so as to obtain the decision variables that meet the objective function under the constraint conditions; the dispatching plan is adjusted according to the decision variables, so as to dispatch the dispatchable energy. It adopts the traditional robust optimization method, the objective function is mainly based on the tracking load curve of new energy, and the level value is adjusted by the confidence level. It lacks a detailed dynamic adjustment mechanism, and the energy storage system is not fully considered.

[0005] Therefore, it is urgent to develop an optimized scheduling solution that can balance robustness and economy, fully consider various constraints, and have dynamic adjustment capabilities. Summary of the invention

[0006] The purpose of the present invention is to provide a multi-energy power system optimization scheduling method based on improved robust optimization, so as to solve the problem that the existing method is insufficiently flexible and thus the scheduling cannot meet the real-time needs of the power system.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solution: a multi-energy power system optimization scheduling method based on improved robust optimization, characterized by comprising the following steps:

[0008] S1. The total objective function is to minimize the total operating cost of the power system. The total operating cost includes the fuel cost, start-up and shutdown cost of thermal power units, the operating cost of the energy storage system, and the utilization cost of renewable energy.

[0009] S2. Establish constraints for the overall goal, including constraints on thermal power units, renewable energy systems, and energy storage systems;

[0010] S3. According to the overall objective function and various constraints, a robust optimization scheduling model is established, and the Benders decomposition method is used to process the robust optimization to obtain the decision variables that meet the objective function;

[0011] S4. Adjust the dispatch plan according to the decision variables to achieve optimal dispatch of the power system.

[0012] A further technical solution is that the total objective function formula is as follows:

[0013]

[0014] In the formula, t represents the time period index, which is each hour or minute in the scheduling cycle; T is the total scheduling time period, which represents the total number of time periods for system operation; i is the number index of the thermal power unit; N g is the total number of thermal power units; P i,t is the output power of thermal power unit i in time period t; is the power generation cost function of thermal power unit i in period t, which is a linear or quadratic function of power; is the start-up and shutdown cost of thermal power unit i, and the state S of the unit i,t Related; S i,t is the start / stop state of thermal power unit i in period t, indicating the state transition from shutdown to startup; C ES (E t ) represents the charging and discharging cost of the energy storage system in time period t, which depends on the energy state E of the energy storage system t ; E t is the remaining energy of the energy storage system in time period t; R renew (P wind ,t,P PV,t ) is the utilization compensation function of renewable energy wind power and photovoltaic power; Pwind,t is the actual output power of wind power in period t; P PV,t is the actual output power of the photovoltaic power plant in time period t; is the penalty cost function for wind and solar power abandonment, which represents the penalty for the system when it fails to fully utilize wind power and photovoltaic power; Indicates the amount of abandoned wind power, that is, the amount of wind power that cannot be absorbed due to load restrictions; It refers to the amount of abandoned photovoltaic power, that is, the amount of photovoltaic power that cannot be consumed due to load limitations.

[0015] A further technical solution is that the constraints in step S2 are as follows:

[0016] 1. Constraints of thermal power units

[0017] 1) Output constraints

[0018]

[0019] Among them, P i,min is the minimum output power of thermal power unit i; P i,max is the maximum output power of thermal power unit i; U i,t The operating status of thermal power unit i in time period t, 1 means running and 0 means shutdown;

[0020] 2) Climbing constraints

[0021]

[0022] in, is the maximum load reduction rate of thermal power unit i, that is, the maximum power reduction value allowed in each period; is the maximum load increase rate of thermal power unit i, that is, the maximum power increase value allowed in each period;

[0023] 3) Minimum running time and downtime constraints

[0024]

[0025] Among them, T on The minimum continuous operation time of the thermal power unit, which means the shortest time the unit must continue to run after starting; T off The minimum shutdown time of the thermal power unit, which means the shortest time the unit must remain shut down after being shut down;

[0026] 4) Start-Stop State Change Constraints

[0027]

[0028] Among them, S i,t is the start / stop status change of thermal power unit i in time period t, 1 indicates start, -1 indicates stop;

[0029] 2. Constraints of wind power and photovoltaic power

[0030] 1) Wind power output constraints

[0031]

[0032] in, is the maximum available output of wind power in period t; is the predicted value of wind power in period t; ΔP wind,t is the prediction error of wind power; wind,t is the maximum allowable value of wind power prediction error;

[0033] 2) Photovoltaic output constraints

[0034]

[0035] in, is the maximum available output of photovoltaic power in time period t; is the predicted value of photovoltaic power in period t; ΔP PV,t is the prediction error of photovoltaic; PV,t is the maximum allowable value of photovoltaic prediction error;

[0036] 3) Wind and solar power curtailment constraints

[0037] The wind curtailment constraint is

[0038]

[0039] in, is the actual consumption of wind power in time period t; D t is the total load demand of the system during period t; P hydro,t is the output of hydropower in time period t; is the discharge power of the energy storage system in time period t; is the charging power of the energy storage system in time period t;

[0040] The light abandonment constraint is

[0041]

[0042] in, is the actual consumption of photovoltaic power generation in time period t;

[0043] 3. Constraints of energy storage systems

[0044] 1) Energy balance constraints of energy storage systems

[0045]

[0046] Among them, E tis the remaining energy of the energy storage system in time period t; E t-1 is the remaining energy of the energy storage system in time period t-1; η ch is the charging efficiency of the energy storage system, η ch Less than 1; η dis is the discharge efficiency of the energy storage system, η ch Less than 1; is the charging power of the energy storage system in time period t; is the discharge power of the energy storage system in time period t; Δt is the length of the time period;

[0047] 2) Energy storage system charging and discharging power constraints

[0048]

[0049] in, is the maximum charging power of the energy storage system; is the maximum discharge power of the energy storage system.

[0050] A further technical solution is that the specific steps of S3 are as follows:

[0051] S3-1. The adjustment factor Γ is introduced to balance the robustness and economy of the system. The power balance constraint formula for robust optimization is as follows:

[0052]

[0053] Among them, Γ is the adjustment factor, which is the control system's ability to cope with uncertainty fluctuations; wind,t is the uncertainty fluctuation range of wind power in time period t; PV,t is the uncertainty fluctuation range of photovoltaic power in time period t;

[0054] S3-2. Luban optimization objective function

[0055] By adjusting the factor Γ to control robustness, the influence of uncertainty on the system is introduced into the objective function. The optimized objective function is expressed as:

[0056]

[0057] Among them, λ is the robustness penalty coefficient, which is used to balance robustness and economy; θ t =Γ(∈ wind,t +∈ PV,t ) is the uncertainty influencing factor;

[0058] S3-3. Benders decomposition method for robust optimization

[0059] Use Benders decomposition method to decompose complex problems into main problems and sub-problems, and solve them step by step iteratively:

[0060] Step 1: Initialization

[0061] Set the initial adjustment factor Γ( 0 )=0, and set the convergence condition

[0062] Step 2: Under a given Γ, solve the main problem, the goal is to minimize the total operating cost of the system while satisfying all power balance and equipment output constraints;

[0063] Step 3: Solve the subproblems

[0064] After solving the main problem, we then solve the sub-problems to test the robustness of the scheduling solution under various uncertain scenarios. If we find that the scheduling solution fails to meet the robustness requirements of the uncertain scenarios, we feed back new constraints to the main problem and continue to correct it.

[0065] Step 4: Update Γ

[0066] According to the sub-problem feedback, update the adjustment factor Γ:

[0067]

[0068] Among them, Γ (k) is the adjustment factor in the kth iteration; δ (k) is the step size factor, which controls each iteration Γ The dispatch range; is the uncertainty impact factor of the kth iteration period t, equal to

[0069] Step 5: Convergence judgment and iteration

[0070] Determine whether the convergence conditions are met: If the convergence conditions are met, the main problem and subproblems meet all constraints and the scheduling scheme is robust, the algorithm stops iterating and outputs the optimal solution as the decision variable;

[0071] If the convergence condition is not met, return to step 2 and continue iterating to gradually optimize the adjustment factor Γ until the convergence condition is met.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] A multi-energy power system optimization scheduling method based on improved robust optimization is provided to optimize the scheduling of power systems involving thermal power, wind power, photovoltaic power and energy storage. The total operating cost is minimized as the overall objective function, and the constraint conditions of the overall balance of multiple energy sources are introduced. The charging and discharging strategy of the energy storage system is used to smooth fluctuations and optimize the peak-to-valley difference. An adjustment factor is introduced to control the balance between robustness and economy. The uncertainty influence of wind power and photovoltaic fluctuations is dynamically controlled through uncertainty influencing factors. Benders is introduced to decompose large-scale problems into main problems and sub-problems. The main problem is used to solve the objective function, and the sub-problem updates the robustness adjustment factor. The solution is efficient while reducing the computational complexity, which is suitable for large-scale real-time scheduling.

[0074] The above method improves the traditional robust optimization model by considering the synergy of multiple energy sources and their respective constraints, establishes an optimization model with complex uncertainties, and improves the flexibility and economy of the model by deriving a new solution algorithm, thereby achieving efficient and reliable scheduling of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a curve diagram of adjustment factor and operating cost in the embodiment.

[0076] Figure 2 It is a curve diagram of power generation within the dispatching period of the thermal power unit in the embodiment.

[0077] Figure 3 It is a charge and discharge curve diagram of the energy storage system in the embodiment.

[0078] Figure 4 Graph showing the utilization rate of renewable energy in the embodiment. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0080] Example

[0081] A multi-energy power system optimization scheduling method based on improved robust optimization, characterized by comprising the following steps:

[0082] S1. The overall objective function is to minimize the total operating cost of the power system. The total operating cost includes the fuel cost and start-up and shutdown costs of thermal power units, the operating cost of the energy storage system, and the utilization cost of renewable energy.

[0083] The overall objective function formula is as follows:

[0084]

[0085] In the formula, t represents the time period index, which is each hour or minute in the scheduling cycle; T is the total scheduling time period, which represents the total number of time periods for system operation; i is the number index of the thermal power unit; N g is the total number of thermal power units; P i,t is the output power of thermal power unit i in time period t; is the power generation cost function of thermal power unit i in period t, which is a linear or quadratic function of power; is the start-up and shutdown cost of thermal power unit i, and the state S of the unit i,t Related; S i,t is the start / stop state of thermal power unit i in period t, indicating the state transition from shutdown to startup; C ES (E t ) represents the charging and discharging cost of the energy storage system in time period t, which depends on the energy state E of the energy storage system t ; E t is the remaining energy of the energy storage system in time period t; R renew (P wind,t ,P PV,t ) is the utilization compensation function of renewable energy wind power and photovoltaic power; P wind,t is the actual output power of wind power in period t; P PV,t is the actual output power of the photovoltaic power plant in time period t; is the penalty cost function for wind and solar power abandonment, which represents the penalty for the system when it fails to fully utilize wind power and photovoltaic power; Indicates the amount of abandoned wind power, that is, the amount of wind power that cannot be absorbed due to load restrictions; It refers to the amount of abandoned photovoltaic power, that is, the amount of photovoltaic power that cannot be consumed due to load limitations.

[0086] S2. Establish constraints for the overall goal, including constraints on thermal power units, renewable energy systems, and energy storage systems.

[0087] 1. Constraints of thermal power units

[0088] 1) Output constraints

[0089]

[0090] Among them, P i,min is the minimum output power of thermal power unit i; P i,max is the maximum output power of thermal power unit i; U i,t The operating status of thermal power unit i in time period t, 1 means running and 0 means shutdown;

[0091] 2) Climbing constraints

[0092]

[0093] in, is the maximum load reduction rate of thermal power unit i, that is, the maximum power reduction value allowed in each period; is the maximum load increase rate of thermal power unit i, that is, the maximum power increase value allowed in each period;

[0094] 3) Minimum running time and downtime constraints

[0095]

[0096] Among them, T on The minimum continuous operation time of the thermal power unit, which means the shortest time the unit must continue to run after starting; T off The minimum shutdown time of the thermal power unit, which means the shortest time the unit must remain shut down after being shut down;

[0097] 4) Start-Stop State Change Constraints

[0098]

[0099] Among them, S i,t is the start / stop status change of thermal power unit i in time period t, 1 indicates start, -1 indicates stop;

[0100] 2. Constraints of wind power and photovoltaic power

[0101] 1) Wind power output constraints

[0102]

[0103] in, is the maximum available output of wind power in period t; is the predicted value of wind power in period t; ΔP wind,t is the prediction error of wind power; wind,t is the maximum allowable value of wind power prediction error;

[0104] 2) Photovoltaic output constraints

[0105]

[0106] in, is the maximum available output of photovoltaic power in time period t; is the predicted value of photovoltaic power in period t; ΔP PV,t is the prediction error of photovoltaic; PV,t is the maximum allowable value of photovoltaic prediction error;

[0107] 3) Wind and solar power curtailment constraints

[0108] Due to the volatility of wind power and photovoltaic power, the power grid may not be able to fully absorb all wind power and photovoltaic power output. Therefore, it is necessary to consider the phenomenon of wind and photovoltaic power abandonment. The actual amount of wind power and photovoltaic power absorbed by the system in each period is affected by load demand and the scheduling of other power generation equipment.

[0109] The wind curtailment constraint is

[0110]

[0111] in, is the actual consumption of wind power in time period t; D t is the total load demand of the system during period t; P hydro,t is the output of hydropower in time period t; is the discharge power of the energy storage system in time period t; is the charging power of the energy storage system in time period t;

[0112] The light abandonment constraint is

[0113]

[0114] in, is the actual consumption of photovoltaic power generation in time period t;

[0115] 3. Constraints of energy storage systems

[0116] The energy storage system is an important component of dispatch optimization. Its functions include storing excess electricity when renewable energy output is in excess and releasing electricity when the system load is peak to smooth out system load fluctuations.

[0117] 1) Energy balance constraint of energy storage system: The energy state of the energy storage system should be balanced according to the charging and discharging behavior. The formula is:

[0118]

[0119] Among them, E t is the remaining energy of the energy storage system in time period t; E t-1 is the remaining energy of the energy storage system in time period t-1; η ch is the charging efficiency of the energy storage system, η ch Less than 1; η dis is the discharge efficiency of the energy storage system, η ch Less than 1; is the charging power of the energy storage system in time period t; is the discharge power of the energy storage system in time period t; Δt is the length of the time period;

[0120] 2) Energy storage system charging and discharging power constraints: The charging and discharging capacity of the energy storage system is limited by the equipment specifications, and the formula is:

[0121]

[0122] in, is the maximum charging power of the energy storage system; is the maximum discharge power of the energy storage system.

[0123] S3. According to the overall objective function and various constraints, a robust optimization scheduling model is established, and the Benders decomposition method is used to process robust optimization to obtain decision variables that meet the objective function. Specifically:

[0124] S3-1. The adjustment factor Γ is introduced to balance the robustness and economy of the system and deal with the uncertainty of wind power and photovoltaic power. The power balance constraint formula for robust optimization is as follows:

[0125]

[0126] Among them, Γ is the adjustment factor, which is the control system's ability to cope with uncertainty fluctuations; wind,t is the uncertainty fluctuation range of wind power in time period t; PV,t is the uncertainty fluctuation range of photovoltaic power in time period t;

[0127] S3-2. Luban optimization objective function

[0128] By adjusting the factor Γ to control robustness, the influence of uncertainty on the system is introduced into the objective function. The optimized objective function is expressed as:

[0129]

[0130] Among them, λ is the robustness penalty coefficient, which is used to balance robustness and economy; θ t =Γ(∈ wind,t +∈ PV,t ) is the uncertainty influencing factor;

[0131] S3-3. Benders decomposition method for robust optimization

[0132] In order to deal with multiple constraints and effectively solve large-scale robust optimization problems, the Benders decomposition method is used to decompose complex problems into main problems and sub-problems, and solve them step by step iteratively:

[0133] Step 1: Initialization

[0134] Set the initial adjustment factor Γ (0) =0, and set the convergence condition

[0135] Step 2: Under a given Γ, solve the main problem, the goal is to minimize the total operating cost of the system while satisfying all power balance and equipment output constraints;

[0136] Step 3: Solve the subproblems

[0137] After solving the main problem, we then solve the sub-problems to test the robustness of the scheduling solution under various uncertain scenarios. If we find that the scheduling solution fails to meet the robustness requirements of the uncertain scenarios, we feed back new constraints to the main problem and continue to correct it.

[0138] Step 4: Update Γ

[0139] According to the sub-problem feedback, update the adjustment factor Γ:

[0140]

[0141] Among them, Γ (k) is the adjustment factor in the kth iteration; δ (k) is the step size factor, which controls each iteration Γ The dispatch range; is the uncertainty impact factor of the kth iteration period t, equal to

[0142] Step 5: Convergence judgment and iteration

[0143] Determine whether the convergence conditions are met: If the convergence conditions are met, the main problem and subproblems meet all constraints and the scheduling scheme is robust, the algorithm stops iterating and outputs the optimal solution as the decision variable;

[0144] If the convergence condition is not met, return to step 2 and continue iterating to gradually optimize the adjustment factor Γ until the convergence condition is met.

[0145] S4. Adjust the dispatch plan according to the decision variables to achieve optimal dispatch of the power system.

[0146] Specifically, the above method is used to optimize the dispatching of the following power system, which includes the following components:

[0147] Thermal power units: 3 units, the parameters of each unit (such as maximum / minimum output, ramp rate, cost function, etc.) are known;

[0148] Wind farm: installed capacity is 100MW, wind power output forecast data known;

[0149] Photovoltaic power station: installed capacity is 50MW, photovoltaic output forecast data known;

[0150] Energy storage system: capacity is 20MWh, maximum charge and discharge power is 10MW, charge and discharge efficiency η ch =η dis =0.95.

[0151] The results obtained using the above method are as follows Figure 1-4 As shown, Figure 1 The trend of the total operating cost of the system with the increase of the adjustment factor Γ is described. It can be seen that the cost increases with the increase of Γ, but the increase gradually decreases, indicating that there is a trade-off between robustness and economy. Figure 2 The output changes of each thermal power unit during the scheduling period after optimized scheduling are shown. It can be seen that the output of the thermal power unit increases during high load periods and decreases during low load periods, and the ramping process meets the multi-stage ramping constraints. Figure 3 The charging and discharging power of the energy storage system in each period is shown. The energy storage system charges when the output of renewable energy exceeds the load demand, and discharges when it is lower, thus smoothing the power fluctuation of the system. Figure 4 It reflects the system's absorption of wind power and photovoltaic renewable energy. After optimized scheduling, the absorption rate of renewable energy has increased significantly, reducing the phenomenon of wind and photovoltaic abandonment.

[0152] Although the present invention is described herein with reference to a number of illustrative embodiments of the present invention, it will be appreciated that those skilled in the art may devise many other modifications and implementations that fall within the scope and spirit of the principles disclosed herein. More specifically, within the scope of the present disclosure, drawings, and claims, a variety of variations and modifications may be made to the components or layout of the subject combination layout. In addition to variations and modifications made to the components or layout, other uses will also be apparent to those skilled in the art.

Claims

1. A multi-energy power system optimization scheduling method based on improved robust optimization, characterized in that The steps include: S1. The total objective function is to minimize the total operating cost of the power system. The total operating cost includes the fuel cost, start-up and shutdown cost of thermal power units, the operating cost of the energy storage system, and the utilization cost of renewable energy. S2. Establish constraints for the overall goal, including constraints on thermal power units, renewable energy systems, and energy storage systems; S3. According to the overall objective function and various constraints, a robust optimization scheduling model is established, and the Benders decomposition method is used to process the robust optimization to obtain the decision variables that meet the objective function; S4. Adjust the dispatch plan according to the decision variables to achieve optimal dispatch of the power system.

2. The method according to claim 1, characterized in that: The overall objective function formula is as follows: In the formula, t represents the time period index, which is each hour or minute in the scheduling cycle; T is the total scheduling time period, which represents the total number of time periods for system operation; i is the number index of the thermal power unit; N g is the total number of thermal power units; P i,t is the output power of thermal power unit i in time period t; is the power generation cost function of thermal power unit i in period t, which is a linear or quadratic function of power; is the start-up and shutdown cost of thermal power unit i, and is related to the state S of the unit i,t Related; S i,t is the start / stop state of thermal power unit i in period t, indicating the state transition from shutdown to startup; C ES (E t ) represents the charging and discharging cost of the energy storage system in time period t, which depends on the energy state E of the energy storage system t ; E t is the remaining energy of the energy storage system in period t; R renew (P wind,t ,P PV,t ) is the utilization compensation function of renewable energy wind power and photovoltaic power; P wind,t is the actual output power of wind power in period t; P PV,t is the actual output power of the photovoltaic power plant in time period t; is the penalty cost function for wind and solar power abandonment, which represents the penalty for the system when it fails to fully utilize wind power and photovoltaic power; Indicates the amount of abandoned wind power, that is, the amount of wind power that cannot be absorbed due to load restrictions; It refers to the amount of abandoned photovoltaic power, that is, the amount of photovoltaic power that cannot be consumed due to load limitations.

3. The method according to claim 1, characterized in that: The constraints in step S2 are as follows:

1. Constraints of thermal power units 1) Output constraints P i,min IN i,t ≤P i,t ≤P i,max IN i,t , Among them, P i,min is the minimum output power of thermal power unit i; P i,max is the maximum output power of thermal power unit i; U i,t The operating status of thermal power unit i in time period t, 1 means running and 0 means shutdown; 2) Climbing constraints in, is the maximum load reduction rate of thermal power unit i, that is, the maximum power reduction value allowed in each period; is the maximum load increase rate of thermal power unit i, that is, the maximum power increase value allowed in each period; 3) Minimum running time and downtime constraints Among them, T on The minimum continuous operation time of the thermal power unit, which means the shortest time the unit must continue to run after starting; T off The minimum shutdown time of the thermal power unit, which means the shortest time the unit must remain shut down after being shut down; 4) Start-Stop State Change Constraints WITH i,t =U i,t -IN i,t-1 , Among them, S i,t is the start / stop status change of thermal power unit i in time period t, 1 indicates start, -1 indicates stop; 2. Constraints of wind power and photovoltaic power 1) Wind power output constraints in, is the maximum available output of wind power in time period t; is the predicted value of wind power in period t; ΔP wind,t is the prediction error of wind power; wind,t is the maximum allowable value of wind power prediction error; 2) Photovoltaic output constraints in, is the maximum available output of photovoltaic power in time period t; is the predicted value of photovoltaic power in period t; ΔP PV,t is the prediction error of photovoltaic; PV,t is the maximum allowable value of photovoltaic prediction error; 3) Wind and solar power curtailment constraints The wind curtailment constraint is in, is the actual consumption of wind power in time period t; D t is the total load demand of the system during period t; P hydro,t is the output of hydropower in time period t; is the discharge power of the energy storage system in time period t; is the charging power of the energy storage system in time period t; The light abandonment constraint is in, is the actual consumption of photovoltaic power generation in time period t; 3. Constraints of energy storage systems 1) Energy balance constraints of energy storage systems Among them, E t is the remaining energy of the energy storage system in time period t; E t-1 is the remaining energy of the energy storage system in time period t-1; η ch is the charging efficiency of the energy storage system, η ch Less than 1; η dis is the discharge efficiency of the energy storage system, η ch Less than 1; is the charging power of the energy storage system in time period t; is the discharge power of the energy storage system in time period t; Δt is the length of the time period; 2) Energy storage system charging and discharging power constraints in, is the maximum charging power of the energy storage system; is the maximum discharge power of the energy storage system.

4. The method according to claim 1, characterized in that: The specific steps of S3 are as follows: S3-1. The adjustment factor Γ is introduced to balance the robustness and economy of the system. The power balance constraint formula for robust optimization is as follows: Among them, Γ is the adjustment factor, which is the control system's ability to cope with uncertainty fluctuations; wind,t is the uncertainty fluctuation range of wind power in time period t; PV,t is the uncertainty fluctuation range of photovoltaic power in time period t; S3-2. Luban optimization objective function By adjusting the factor Γ to control robustness, the influence of uncertainty on the system is introduced into the objective function. The optimized objective function is expressed as: Among them, λ is the robustness penalty coefficient, which is used to balance robustness and economy; θ t =Γ(∈ wind,t +∈ PV,t ) is the uncertainty influencing factor; S3-3. Benders decomposition method for robust optimization Use Benders decomposition method to decompose complex problems into main problems and sub-problems, and solve them step by step iteratively: Step 1: Initialization Set the initial adjustment factor Γ (0) = 0, and set the convergence condition Step 2: Under a given Γ, solve the main problem, the goal is to minimize the total operating cost of the system while satisfying all power balance and equipment output constraints; Step 3: Solve the subproblems After solving the main problem, we then solve the sub-problems to test the robustness of the scheduling solution under various uncertain scenarios. If we find that the scheduling solution fails to meet the robustness requirements of the uncertain scenarios, we feed back new constraints to the main problem and continue to correct it. Step 4: Update Γ According to the sub-problem feedback, update the adjustment factor Γ: Among them, Γ (k) is the adjustment factor in the kth iteration; δ (k) is the step size factor, which controls the scheduling amplitude of Γ in each iteration; is the uncertainty impact factor of the kth iteration period t, equal to Step 5: Convergence judgment and iteration Determine whether the convergence conditions are met: If the convergence conditions are met, the main problem and subproblems meet all constraints and the scheduling scheme is robust, the algorithm stops iterating and outputs the optimal solution as the decision variable; If the convergence condition is not met, return to step 2 and continue iterating to gradually optimize the adjustment factor Γ until the convergence condition is met.

Images