An annual electricity quantity verification calculation method

Abstract

The present invention discloses an annual electricity quantity verification calculation method, and the specific process includes: determining the power grid scope, calculation period and other calculation boundaries for the annual electricity quantity verification calculation to be carried out; performing multi-period decomposition and dimension reduction processing on the annual calculation data of the power grid, and decomposing the annual electricity quantity verification calculation model into multiple sub-models; circularly calculating each sub-model, carrying out merging analysis of adjacent periods to reduce the number of periods entering the optimization; considering the connection of the unit combination states between the sub-models, establishing a reduced-dimension SCUC model after period merging, and solving to obtain the unit combination result; establishing a full-period SCED model based on the unit combination result and solving it to obtain the optimized power plant electricity quantity result under each sub-model, and completing the annual electricity quantity verification calculation based on multi-period decomposition and dimension reduction. This annual electricity quantity verification calculation method precisely considers the influence of practical constraints on the power generation of power plants, and improves the accuracy and calculation efficiency of the annual electricity quantity verification results of power plants.

Description

Technical Field

[0001] The present invention belongs to the field of power system dispatching automation, and particularly relates to an annual electricity quantity verification calculation method. Background Art

[0002] The core of medium- and long-term transaction security verification is to arrange the power and electricity balance for the future months and years under the condition of considering the power grid security constraints, obtain the start-stop plans of generating units, and verify whether the power plant electricity plans can be physically executed. The executability of electricity transactions under operating conditions is the key content of medium- and long-term electricity security verification. Given that traditional empirical methods are difficult to meet the requirements of lean power grid dispatching, medium- and long-term electricity security verification based on security-constrained unit commitment (SCUC) technology has gradually become a research hotspot. Such technologies mainly take monthly unit commitment technology as the core, coordinately consider conditions such as power and electricity balance, power grid security constraints, and unit operation constraints, refine electricity to the active power output and power flow levels, and conduct fine electricity security verification.

[0003] The annual electricity quantity verification optimization model needs to be calculated according to the annual boundary data of the power grid. Traditional annual calculations need to simplify a lot of data and optimization conditions, and the calculation time is too long. Nowadays, the annual electricity quantity verification optimization model needs to consider more practical constraints. Therefore, the traditional annual electricity quantity verification calculation method cannot meet the above requirements. Therefore, an annual electricity quantity verification calculation method that can ensure the safe and stable operation of the system and has high solution efficiency and referenceability is needed. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide an annual electricity quantity verification calculation method.

[0005] Technical Solution: An annual electricity quantity verification calculation method according to the present invention includes the following steps:

[0006] (1) Determine the power grid scope, calculation period, and calculation boundary of the annual electricity quantity verification calculation for power plants, and perform multi-period decomposition and dimension reduction processing on the annual calculation data of the power grid according to time to form multiple calculation sub-models, and each sub-model contains several optimization periods;

[0007] (2) After extending the calculation time period of each sub-model by a set time period to the next sub-model, establish a SCUC dimension reduction model, and take minimizing the difference between the power generation of the power plant and the planned electricity quantity as the goal, and solve to obtain the unit commitment result;

[0008] (3) Based on the start-stop status results of the buffer unit group and the start-stop status of the must-start and must-stop units, establish the full-time SCED model of the sub-model, solve it using the linear programming algorithm, and finally obtain the optimized power generation results of the power plant under each sub-model to complete the annual power verification calculation based on multi-period decomposition and dimension reduction.

[0009] Preferably, in step (1), the annual grid calculation data for one natural year is decomposed monthly to form 12 sub-models MP i (i = 1, 2, …, 12), and the optimization period T of each sub-model i ={t i0 , t i1 , t i2 , …, t in}}, where t in represents that the i-th sub-model contains n time periods, and each time period uses hours as the calculation granularity.

[0010] Furthermore, in step (1), each sub-model formed after the dimension reduction process is calculated cyclically, and the similar calculation time periods that meet the change trend requirements of the system load within the calculation period are merged.

[0011] Furthermore, when the system load change rate of similar time periods is less than the set threshold, two consecutive time periods are merged into a new time period. The specific calculation formula for the system load change rate is as follows:

[0012]

[0013] In the formula, △L t is the system load change rate of adjacent time periods t and t + 1; L t is the system load of time period t; L t+1 is the system load of time period t + 1;

[0014] According to the change rate of the system load, repeat multiple rounds of time period merging until the minimum value of the change rate △L t is greater than the set threshold, or the number of remaining time periods after merging reaches the preset number, and the merging process ends.

[0015] Preferably, in step (2), the optimization objective of each sub-model is to minimize the difference between the power generation of the power plant and the planned power under the condition of meeting all constraints. The specific formula is as follows:

[0016]

[0017] In the formula, M is the total number of power plants with power contracts; E m is the total power generation of power plant m during the optimization period; E m,e$E_{m}$ is the total planned power generation of power plant $m$ during the optimization period, including the base power generation and various market trading power generations; $\omega$ m $\omega$ is the penalty coefficient for power generation deviation of the power plant, which realizes the priority control of the power generation deviation of the power plant. The default value is 1, and the value is greater than 1 when the power generation of the power plant is completed with a higher level;

[0018] Introduce positive and negative power generation deviation variables to transform the non-linear formula into a linear expression. The specific formula is as follows:

[0019]

[0020] $E$ m $-E$ m,e $=M$ m $-N$ m ; $M$ m $\geq0$, $N$ m $\geq0$;

[0021] In the formula: $M$ m is the positive power generation deviation of power plant $m$, and $N$ m is the negative power generation deviation of power plant $m$.

[0022] Furthermore, when the sub-model cannot be solved in step (2), the required constraints are relaxed. Under the relaxed constraints, the optimization objective considers the penalty cost of the relaxed constraints. At this time, the optimization objective becomes:

[0023]

[0024] In the formula, $C$ f is the relaxation amount of various relaxed constraints, and $K$ f is the penalty coefficient of the relaxed constraints.

[0025] Furthermore, the relaxed constraints include load balance constraints, reserve constraints, and branch power flow constraints.

[0026] Preferably, when using the SCUC dimensionality reduction model to solve the unit commitment results of each sub-model in step (2), the final on-off duration of each unit is counted and used as the initial on-off constraints of the units in the next sub-model to participate in the optimization calculation.

[0027] Furthermore, when establishing the SCUC dimensionality reduction model in step (2), according to the optimizable state of the units, the must-run and must-stop units and buffer units in the security-constrained unit commitment decision are identified.

[0028] Furthermore, in step (1), data preparation is first completed to obtain parameters such as the upper and lower limits of the output of the generator sets in the power grid, the minimum start-up and shutdown time, etc.; the topological structure of the power grid and the component equipment of the transmission section, the component equipment of the unit group, the component equipment of the power plant, the transmission limit, the unit group limit and other parameter information are obtained; various types of planning data are obtained, including load forecasting, maintenance plan, power plant power plan, backup demand and other information, so as to determine the boundaries of the power plant's annual power verification calculation.

[0029] Furthermore, the annual data is decomposed by month, and the huge and complex annual power plant electricity verification calculation model is decomposed into 12 sub-models. In the subsequent period, it is necessary to perform cyclic calculations while considering the connection of the sub-models to reduce the calculation scale of the optimization model and improve the calculation efficiency.

[0030] Each sub-model is calculated cyclically. When considering SCUC modeling, in order to reduce the calculation scale and the number of integer variables, similar calculation periods are merged according to the change trend of system load in different periods within the calculation cycle, the number of periods entering optimization is reduced, and the calculation period scale of the optimization model is reduced;

[0031] When modeling SCUC, the units that must be turned on or off in the safety constraint unit combination decision are the generating units that must be turned on or off in the power grid operation, and there is no need to make an optimization decision on the unit status in the safety constraint unit combination; the buffer unit is a generating unit whose start and stop status cannot be determined in advance, and it is necessary to make an optimization decision on the unit status in the safety constraint unit combination. Buffer units and units that must be turned on or off are identified based on the provided unit status data to reduce the number of integer variables, narrow the optimization range of mixed integer programming, and improve the solution efficiency.

[0032] Furthermore, when cyclically calculating each sub-model, since the unit needs to meet the actual minimum start and stop time requirements, if only the unit combination state optimization of each sub-model is considered without considering the boundary data of subsequent time periods, when calculating the sub-model of the next natural month, it is easy to have optimization non-convergence problems caused by conflicts between the unit minimum start and stop time constraints and the group start-up number constraints, unit group output constraints, unit maintenance plans and other constraints. Therefore, it is necessary to consider the connection of the unit combination states between sub-models.

[0033] In order to solve the data gap problem caused by multi-period decomposition and dimensionality reduction, on the basis of monthly decomposition, the calculation cycle of each sub-model is extended by a few days, and the data of the first d periods of the next sub-model are considered. At this time, the i-th sub-model contains n+d optimization periods. In addition, during the SCUC calculation, the last start-up and shutdown duration of each unit is counted and used as the initial start-up and shutdown constraints of the next sub-model unit to participate in the optimization calculation to ensure the reasonable connection of the unit combination optimization results.

[0034] Furthermore, the constraint conditions of the medium- and long-term power quantity verification model include system balance constraints, unit operation constraints, power grid security constraints, and various practical constraints. The medium- and long-term power quantity security verification model supports the relaxation of load balance constraints, reserve constraints, and branch power flow constraints as required.

[0035] Furthermore, in step (3), each sub-model aims to minimize the deviation between the power generation of the power plant and the planned power quantity. The constraint conditions include load balance constraints, upper and lower limits of unit output constraints, power grid security constraints, system reserve capacity constraints, unit group output constraints, unit group startup number constraints, etc. The output curve and power generation of the power plant under each sub-model are obtained through solution, and finally, the annual power quantity verification optimization calculation based on multi-period decomposition and dimension reduction is completed.

[0036] Beneficial effects: The technical solution of the present invention decomposes the annual data into multiple sub-models, considers the connection of the unit combination states between the sub-models and the safety issues in the actual dispatching operation of the units, randomly optimizes the unit start-stop sequence of the power system, obtains the unit start-stop plan curve, and further calculates the optimized output curve and power generation of the power plant, so that the annual optimization model can consider various practical constraints while taking into account the huge data. This method can accurately consider the influence of practical constraints on the power generation of the power plant in the annual power quantity verification optimization algorithm, and improve the accuracy and calculation efficiency of the annual power quantity verification results of the power plant. Description of the Drawings

[0037] Figure 1 It is a flow chart of an annual power quantity verification calculation method in the present invention. Detailed Embodiment

[0038] The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments.

[0039] In the process of annual power quantity verification optimization calculation, it is necessary to adopt security-constrained unit commitment technology to optimize the unit combination and power plant power quantity results within a natural year. The optimization process requires both refined consideration of various security constraint limitations in power grid operation such as unit operation constraints, power grid security constraints, and load balance constraints, and consideration of practical constraints such as unit group startup number constraints, unit group output constraints, and power plant startup number constraints. To meet the above requirements, this embodiment adopts an annual power quantity verification calculation method based on multi-period decomposition and dimension reduction, as Figure 1 shown, and the specific calculation process is as follows:

[0040] Step (1): Determine the power grid scope and calculation period for the annual electricity quantity verification calculation. At the same time, perform data preparation to obtain parameters such as the upper and lower limits of the output of generating units in the power grid, the minimum start-up and shutdown times, etc.; obtain the topological structure of the power grid and parameter information such as the component equipment of the transmission section, the component equipment of the unit group, the component equipment of the power plant, the transmission limit, and the unit group limit; obtain various types of planned data, including load forecasting, maintenance plans, power plant electricity plans, reserve requirements, etc., to determine the boundary of the annual electricity quantity verification calculation for the power plant.

[0041] Step (2): Perform multi-period decomposition and dimensionality reduction processing on the annual calculation data of the power grid. Design the calculation period for the annual electricity quantity verification calculation as 1 natural year, with the calculation granularity in hours, that is, 24 optimization periods are considered per day, and a total of 8,760 optimization periods are considered in 365 days of 1 natural year; decompose the annual data of 1 natural year by month, and a total of 12 sub-models MP i (i = 1, 2, …, 12) can be decomposed. Each sub-model considers the optimization periods of one natural month, that is, the optimization period T i ={t i0 , t i1 , t i2 , …, t in}, where t in represents that the i-th sub-model contains n periods.

[0042] Step (3): Merge the optimization periods of each sub-model. When considering SCUC modeling, in order to reduce the calculation scale of the mixed integer programming and reduce the number of integer variables, the similar calculation periods can be merged according to the change trend of the system load in different periods within the calculation period. The specific process of period merging is as follows:

[0043]

[0044] In the formula, △L t is the system load change rate between adjacent periods t and t + 1; L t is the system load in period t; L t+1 is the system load in period t + 1;

[0045] According to the change rate of the system load, repeat the multi-round period merging until the minimum value of the change rate △L t is greater than the set threshold, or the number of remaining periods after merging reaches the preset number, and the merging process ends.

[0046] Meanwhile, according to the optimizable status of the generating units, identify the must-run and must-stop units and buffer units in the security-constrained unit commitment decision-making; among them, the must-run generating units or must-stop generating units in the power grid operation do not require optimization decisions on the unit status in the security-constrained unit commitment; the buffer units are generating units whose start-up and shutdown status cannot be determined in advance and need to make optimization decisions on the unit status in the security-constrained unit commitment.

[0047] Step (4), considering the connection of the unit commitment status between sub-models, establish a dimension-reduced SCUC model after period merging, and solve to obtain the unit commitment result. In this embodiment, in step (3), by optimizing the period merging, the number of periods entering the optimization is reduced, and the computational period scale of the optimization model is reduced. Through the identification of the optimizable status of the generating units, the number of integer variables is reduced, the optimization scope of the mixed-integer programming is reduced, and the solving efficiency of the dimension-reduced SCUC model is improved.

[0048] Furthermore, considering the connection of the unit commitment status between sub-models can solve the data discontinuity problem caused by multi-period decomposition and dimension reduction. In this embodiment, taking April of a certain natural year as an example, there are 30 days in April. Considering the boundary data of the first 5 days of the next natural month, at this time, the optimization period of the April sub-model is (30 + 5) * 24 = 840 periods, but only the unit commitment status results of the first 30 * 24 = 720 periods in the April sub-model can be effectively stored. The unit commitment results of the first 5 days of the next natural month are determined by the next sub-model. The start-up and shutdown duration of the generating units at the 720th period is counted and used as the initial start-up and shutdown constraints of the generating units in the next sub-model to participate in the next round of calculation. Finally, the unit commitment status results for the entire year can be obtained.

[0049] Step (5), after completing the construction of the dimension-reduced SCUC model for each sub-model, with the goal of minimizing the difference between the power generation of the power plant and the planned power, solve to obtain the unit commitment result; the specific optimization goal is as follows:

[0050]

[0051] In the formula, M is the total number of power plants with power contracts; E m is the total power generation of power plant m in the optimization period; E m,e is the total planned power of power plant m in the optimization period, including the base power and various market trading powers; ω m is the power deviation penalty coefficient of the power plant, which realizes the priority control of the power deviation of the power plant. The default value is 1, and when the power of the power plant is completed with a higher level, the value is greater than 1;

[0052] Since the above optimization target expression includes absolute value modeling and this expression is a non-linear expression form, positive and negative power deviation variables are introduced to transform the non-linear formula into a linear expression. The specific formula is as follows:

[0053]

[0054] E m -E m,e = M m -N m ; M m ≥0, N m ≥0;

[0055] Where: M m is the positive power deviation of power plant m, N m is the negative power deviation of power plant m.

[0056] When the original problem cannot be solved, the required constraints are relaxed to ensure the convergence of the model, and at the same time, the factors affecting the model convergence can be found. In the case of relaxed constraints, the optimization objective considers the relaxed constraint penalty cost, and at this time, the optimization objective becomes:

[0057]

[0058] Where, C f is the relaxation amount of various relaxed constraints, K f is the penalty coefficient of the relaxed constraint.

[0059] In this embodiment, the annual power verification optimization belongs to the medium and long-term power verification. The constraint conditions of the medium and long-term power verification model include system balance constraints, unit operation constraints, power grid security constraints, and various practical constraints, which are mainly divided into:

[0060] (a) Sub-region system load balance constraint; Under the unified optimization mode of sub-region balance, the output of the generating units and the output of the tie lines in each region are balanced with the regional system load:

[0061]

[0062] Where: A a is the equipment set of region a; I is the total number of generating units in the system; D is the total number of tie lines in the system; P i,t is the active power of unit i at time t; P d,t is the active power of tie line d at time t; L a,t is the system load of region a at time t.

[0063] (b) Province-by-province reserve constraint; including positive reserve constraint and negative reserve constraint for each province; specifically as follows:

[0064] Positive reserve constraint:

[0065]

[0066] Negative reserve constraint:

[0067]

[0068] Where: P i,max and P i,min are the upper and lower limits of the output of unit i respectively; R a,t,u is the lower limit of the positive reserve capacity of province a at time period t; R a,t,d is the lower limit of the negative reserve capacity of province a at time period t.

[0069] (c) Grid security constraint; the power flow of all AC transmission equipment needs to meet the transmission limit constraint. In addition to the AC transmission equipment within each region, the AC transmission equipment within this constraint also includes cross-regional AC transmission equipment. By constructing the active power sensitivity of the branch from the node injection power, the grid security constraint model is constructed:

[0070]

[0071] Where: p l is the upper and lower limits of the power flow of the l-th transmission section, N is the set of grid calculation nodes, P n,t is the power generation of grid calculation node n at time period t, l n,t is the load of grid calculation node n at time period t, S n,l,t is the sensitivity of the injection power of grid calculation node n at time period t to the l-th transmission section.

[0072] (d) Generator operation constraint; the operation constraints of the generator set include the upper and lower limits of the unit output, the unit ramp rate constraint, the minimum start-stop time constraint of thermal power units, etc.;

[0073] Upper and lower limits of unit output constraint:

[0074] P i,min u i,t ≤ P i,t ≤ P i,max u i,t

[0075] Where: u i,t is the start-stop state of unit i at time period t.

[0076] Unit ramp rate constraint:

[0077] -△ i ≤ P i,t - P i,t-1 ≤ △ i

[0078] Where: Δ i is the maximum value of the ramp rate of thermal power unit i per time period.

[0079] Minimum start-up and shut-down time constraints of the unit

[0080]

[0081]

[0082] y i,t -z i,t = u i,t -u i,t-1

[0083] Where: UT i and DT i are the minimum start-up time and minimum shut-down time of unit i respectively; y i,t is the flag indicating whether there is a change from shut-down to start-up state of unit i at time period t; z i,t is the flag indicating whether there is a change from start-up to shut-down state of unit i at time period t.

[0084] (e) Fixed plan constraints of the unit

[0085] u(i,t) = 1 (i,t) ∈ Φ on

[0086] u(i,t) = 0 (i,t) ∈ Φ off

[0087] p(i,t) = P(i,t) (i,t) ∈ Φ plan

[0088] Where: Φ on is the set of times for units that must be started up; Φ off is the set of times for units that must be shut down; P i,t is the fixed output plan of unit i at time period t; Φ plan is the set of times for units with fixed plans.

[0089] (f) Power plant power constraints; The power generation of the power plant is the accumulation of the active power outputs of the units contained in the power plant within the scheduling period. The power balance constraint of the power plant is expressed as:

[0090]

[0091] Where: A m is the set of generator units included in power plant m, E m is the total power generation of power plant m during the optimization period, and β is the number of minutes included in the scheduling time period.

[0092] (g) Operating constraints of the unit group; The operating constraints of the unit group include upper and lower limits of the unit group output, constraints on the number of units started up in the unit group, etc.;

[0093] Output upper and lower limits constraint of unit group:

[0094]

[0095] In the formula: A g is the set of generating units included in power plant g; P g,min , P g,max are the lower and upper limits of the output of unit group g respectively.

[0096] Operating unit number constraint of unit group:

[0097]

[0098] In the formula: Q g,min , Q g,max are the lower and upper limits of the operating unit number of unit group g respectively.

[0099] In this embodiment, the medium- and long-term power quantity security checking model supports relaxation of load balance constraint, reserve constraint and branch power flow constraint according to requirements.

[0100] Step (6), based on the start-stop state results of buffer units and the start-stop states of must-run and must-stop units, establish the full-time SCED model of the sub-model, solve it using the linear programming algorithm, obtain the unit start-stop plan curves under each sub-model, and further calculate the optimized output curve and power generation of the power plant, and finally complete the annual power quantity checking optimization calculation based on multi-period decomposition and dimension reduction.

[0101] In summary, an annual power quantity checking calculation method based on multi-period decomposition and dimension reduction disclosed in this embodiment preprocesses annual data, takes into account the operating characteristics of power equipment, uses multi-period decomposition and dimension reduction technology to obtain multiple sub-models and considers the connection problems between sub-models, merges similar calculation periods, identifies invalid integer variables, reduces the solution scale, narrows the optimization scope of mixed integer programming and the iteration times of power plant power quantity checking calculation, can effectively improve the solution efficiency and effectiveness, enable the calculation process to end within the time acceptable to operators, and make the calculation accuracy and calculation time meet the requirements of engineering practice.

Claims

1. An annual electricity quantity verification calculation method, characterized in that: The method includes the following steps: (1) Determine the power grid scope, calculation period for annual power quantity verification calculation, and calculation boundary for annual power quantity verification of power plants, and perform multi-period decomposition and dimensionality reduction processing on the annual calculation data of the power grid according to time to form multiple calculation sub-models, each sub-model containing several optimized periods; (2) After extending the calculation time period of each sub-model by a set time period to the next sub-model, establish a SCUC dimensionality reduction model, and take minimizing the difference between the power generation of the power plant and the planned power quantity as the goal to solve and obtain the unit commitment result; (3) Based on the start-stop state results of the buffer units and the start-stop states of the must-run and must-stop units, establish a full-period SCED model for the sub-model, and use the linear programming algorithm to solve it. Finally, obtain the optimized power plant power quantity results under each sub-model to complete the annual power quantity verification calculation based on multi-period decomposition and dimensionality reduction; In the step (1), the annual grid calculation data for one natural year is decomposed monthly to form 12 sub-models MP i (i = 1, 2, …, 12), and the optimization period T i of each sub-model is i0 {t i1 , t i2 , t in}, where t in indicates that the i-th sub-model contains n time periods, and each time period is calculated at an hourly granularity; In step (2), when solving the unit commitment result of each sub-model using the SCUC dimensionality reduction model, count the final start-stop duration of each unit as the initial start-stop constraint of the units in the next sub-model to participate in the optimization calculation.

2. The annual power consumption verification calculation method according to claim 1, wherein: In step (1), perform loop calculation on each sub-model formed after dimensionality reduction processing, and merge the similar calculation periods that meet the change trend requirements of the system load within the calculation period.

3. The annual power consumption calibration calculation method according to claim 2, wherein: When the system load change rate in the similar periods is less than the set threshold, merge two consecutive periods into a new period. The specific calculation formula of the system load change rate is as follows: where ΔL t is the system load change rate for adjacent time periods t and t + 1; L t is the system load at time period t; L t+1 is the system load at time period t + 1; Repeat multiple rounds of time period merging according to the change rate of the system load until the minimum value of the change rate ΔL t is greater than the set threshold, or the number of remaining time periods after merging reaches the preset number, and the merging process ends.

4. The annual power consumption calibration calculation method according to claim 1, characterized in that: In step (2), the optimization goal of each sub-model is to minimize the difference between the power generation of the power plant and the planned power quantity under the condition of meeting all constraints. The specific formula is as follows: Where M is the total number of power plants with power quantity contracts; E m is the total power generation of power plant m during the optimization period; E m,e is the total planned power quantity of power plant m during the optimization period, including base power quantity and various market trading power quantities; ω m is the power quantity deviation penalty coefficient of the power plant, which realizes the priority control of the power quantity deviation of the power plant. The default value is 1, and the value is greater than 1 when the power quantity of the power plant is completed at a higher level; Introduce positive and negative power quantity deviation variables to transform the non-linear formula into a linear expression. The specific formula is as follows: E m -E m,e = M m -N m ; M m ≥ 0, N m ≥ 0; Where: M m is the positive power deviation of power plant m, and N m is the negative power deviation of power plant m.

5. The annual power consumption calibration calculation method according to claim 4, wherein: In step (2), when the sub-model cannot be solved, relax the required constraints. In the case of relaxed constraints, the optimization goal considers the relaxed constraint penalty cost. At this time, the optimization goal becomes: where C f is the relaxation amount of various relaxation constraints, and K f is the penalty coefficient of the relaxation constraint.

6. The annual electricity quantity verification calculation method according to claim 5, wherein: The relaxed constraints include load balance constraints, reserve constraints, and branch power flow constraints.

7. A method for calculating annual power consumption verification according to claim 1, characterized in that: In step (2), when establishing the SCUC dimensionality reduction model, identify the must-run and must-stop units and buffer units in the security-constrained unit commitment decision according to the optimizable state of the units.

Images