Construction method of rapid unit commitment model based on scheduling and unit aggregation coupling

Abstract

The invention relates to a rapid unit commitment model construction method based on scheduling and unit aggregation coupling. The method comprises the following steps: S1, constructing a traditional unit commitment model; s2, introducing line transmission capacity limitation and a direct current power flow model into a traditional unit combination model; s3, aggregating the large-scale units with the same type of operation characteristics, and carrying out output characteristic and climbing start-stop constraint modeling on an aggregation set; s4, introducing a relaxation hypothesis of an economic scheduling only model, and coupling constraint conditions of unit aggregation and scheduling only; and S5, completing construction of a rapid unit commitment model based on scheduling and unit aggregation coupling, and performing piecewise linearization solution on the model. According to the method, rapid operation simulation of a large number of thermal power generating units can be achieved, the method is suitable for a novel electric power system with high renewable energy permeability and a scene where long-term operation simulation of renewable energy volatility needs to be considered, and powerful support is provided for long-term planning and daily scheduling of the electric power system on the premise that calculation precision is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation optimization, and more specifically, to a method for constructing a fast unit combination model based on scheduling and unit aggregation coupling. Background Art

[0002] With the requirements of energy conservation, carbon reduction and carbon reduction, the proportion of renewable energy in new power systems has gradually increased. Since renewable energy shows strong uncertainty at different time scales, the significance of simulating the full-year time operation of the power system has gradually become apparent.

[0003] The uncertainty exhibited by renewable energy over multiple timescales, as well as the imbalance between energy supply and demand between sources and loads at different timescales, significantly increases the complexity of optimization planning models, further increasing the computational and decision-making challenges. Existing traditional unit commitment models are widely used for short-term power system operation problems. However, these models typically involve a large number of binary variables representing the on / off states of generators, making them computationally difficult to model for long-term power system operation and for large numbers of units.

[0004] Current techniques for simplifying unit commitment models for long-term power system operation simulations can be broadly categorized into two types: dispatch models and unit aggregation models. Dispatch models ignore generator on / off states to improve computational speed, but lack operational flexibility and lead to inaccurate modeling. Unit aggregation models aggregate generating units into clusters and replace binary variables with integers, resulting in even more significant speed improvements. However, the inclusion of line transmission constraints reduces their speed-up advantage. Therefore, it is necessary to develop a relatively optimal unit commitment model with network constraints that balances accuracy and solution speed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for constructing a fast unit combination model based on scheduling and unit aggregation coupling, which can solve the technical problem that the existing unit combination model with network constraints is difficult to solve quickly on a long time scale.

[0006] The technical solution adopted by the present invention to solve the technical problem is to construct a method for building a fast unit commitment model based on scheduling and unit aggregation coupling, including the following steps:

[0007] S1. Construct a traditional unit combination model;

[0008] S2, introducing line transmission capacity limitation and DC power flow model into the traditional unit combination model;

[0009] S3. Aggregate large-scale units with the same operating characteristics and model the output characteristics and ramp start-stop constraints of the aggregated set;

[0010] S4, introduce the slack assumption of the economic dispatch model and couple the constraints of unit aggregation and dispatch only;

[0011] S5. Complete the construction of a fast unit commitment model based on scheduling and unit aggregation coupling, and solve the model by piecewise linearization.

[0012] According to the above scheme, in step S1, the mathematical expression of the traditional unit combination model is as follows:

[0013]

[0014]

[0015] Wherein, formula (1) represents the cost function of the traditional unit commitment model, including unit start-up and shutdown costs, variable operating costs and load shedding penalty costs. is the start-up and shutdown cost per unit MW of thermal power unit i, u i,t is the startup state variable of the thermal power unit, is the variable operating cost of thermal power unit i, P Gi,t is the power generation capacity of thermal power unit i (MW), F punish is the penalty coefficient for the necessary load shedding of the system, is the necessary load shedding variable for n nodes in the system;

[0016] Formula (2) represents the power balance constraint of the power system, P Gi,t ,P w,t is the output of the i-th thermal power unit and the w-th wind power unit at time t, D n,t is the load of n nodes in the system;

[0017] Formula (3) represents the upper and lower limit constraints of thermal power unit output, is the minimum output rate of thermal power unit i, represents the maximum output capacity of thermal power unit i;

[0018] Formula (4) represents the climbing constraint of thermal power units, are the hourly up and down ramp rates of thermal power unit i respectively;

[0019] Formula (5) establishes the start-stop state variable u of thermal power unit i i,t ,v i,t and the actual running state variable I i,t the relationship between;

[0020] Formulas (6) and (7) represent the minimum start-stop time constraints of thermal power units. The fixed continuous start-up and shutdown period required at the beginning of each day meets the requirements and T i on ,T i off The minimum continuous start-up and shutdown time required for thermal power units within one day;

[0021] Equation (8) shows that the on / off state and start / shutdown decision variables are modeled as binary variables;

[0022] Formula (9) limits the output of the wind farm to the output range that can be used for power generation, λ w,t represents the hourly predicted wind power generation level, Indicates the maximum output capacity of the wind turbine w;

[0023] Formula (10) limits the load shedding amount to the load demand at the corresponding node bus;

[0024] Formula (11) represents the arrangement of operating reserve capacity, R load ,R WG It represents the load demand and the operating reserve rate of wind turbines in turn;

[0025] Traditional unit combination has P Gi,t ,P w,t ,I i,t ,u i,t ,v i,t and These decision variables, among which the start-stop state variable u i,t ,v i,t and running status variable I i,t It is a binary variable, so the traditional unit commitment problem is a large mixed integer linear programming problem. The existence of a large number of binary variables greatly increases the complexity of solving the problem.

[0026] According to the above solution, in step S2, the line transmission capacity limitation and the DC power flow model are introduced into the traditional unit combination model. The specific constraints are:

[0027]

[0028] Formula (12) represents the power balance expression at node n, where G(n), W(n), and L(n) are the number of thermal power units, wind power units, and loads at node n. It represents the tidal flow value on line l with n as the starting point (inflow point) and m as the end point (outflow point).

[0029] Formula (13) represents the DC power flow transmission network model, is the difference between the phase angle of the starting node n and the ending node m of the branch l power flow, B l is the per-unit susceptance value of line l;

[0030] Formula (14) represents the line transmission capacity constraint, F l max is the normal allowable maximum value of the power flow on line l;

[0031] Formula (15) represents the node phase angle value limit, and each node phase angle value is limited to the range between -π and π;

[0032] After adding network transmission constraints, we introduced and These two decision variables.

[0033] According to the above scheme, the DC power flow model is specifically simplified as follows:

[0034] When considering the influence of the power grid, the mathematical model of the transmission network established is a nonlinear power flow calculation model, and the active power flow of the transmission line branches at nodes n and m is shown in formula (16);

[0035]

[0036] For the high-voltage transmission network, the model is simplified under certain assumptions and converted into a linear DC model. The assumptions include:

[0037] ① The resistance of high-voltage transmission branches is generally much smaller than the reactance, so the resistance r is ignored. nm , then g nm =0;

[0038] ②In normal operation, generally θ n -θ m is very small, then cosθ nm =1,sinθ nm =θ nm ;

[0039] ③ Assume that the per-unit value of all node voltages is 1, that is, U n =1;

[0040] ④ Ignore the ground branch, then g n0 =0;

[0041] Based on the above assumptions, the nonlinear power flow equation shown in Equation (16) can be transformed into a linear DC power flow model, and the node phase angle is linked to the line power flow. The DC power flow model is combined with Kirchhoff's current law and the power balance equation at any node n is expressed in matrix form as shown in Equation (17):

[0042]

[0043] in, They are the node-branch, node-thermal power generator, node-wind power generator and node-load matrix elements of line l, is the active power flow of the branch l (with n as the starting point of the flow and m as the end point of the flow), is the active power output of the thermal power generator set and wind power generator set connected to line l, The load connected to line l and the necessary load shedding.

[0044] According to the above scheme, in step S3, the basic idea of ​​aggregating large-scale units with the same operating characteristics is to group similar power generation units into clusters and replace the binary decision variables of all units in each cluster with a single integer variable. The specific mathematical model is established as follows:

[0045]

[0046] Units with similar operating characteristics are grouped into one category. In each category, it is assumed that The mathematical model of the aggregated units remains mostly the same as the traditional unit combination formula, except that the individual unit index is replaced by the unit cluster index c. Unlike modeling the on / off state of a single unit, the online capacity of the unit cluster c is described and captured as where n c,t represents the number of thermal power units in the unit cluster c at time t. Since all units in the cluster are assumed to be identical, the characteristic parameters of the same type of units, such as single unit capacity, minimum load rate, ramp rate, and minimum start and stop time period, are assumed to be the average values ​​of the parameters of the units in the cluster. For example, is the average capacity of the cells in cell cluster c, expressed as The remaining clustering parameters These are also the mean values ​​of the unit parameters within the cluster; The fixed continuous start-up and shutdown period required at the beginning of each day in the cluster, satisfying and

[0047] According to the above solution, in step S4, only the economic dispatch model eliminates binary variables by relaxing some operating constraints on the generator sets, including the following key assumptions:

[0048] ① Thermal power generating units have greater flexibility and can adjust power generation without restrictions. They can flexibly dispatch their power generation from zero to power generation capacity without considering minimum power generation restrictions.

[0049] ② The start and stop operation speed of the thermal power unit is very fast, and it can respond to commands instantly and realize immediate start or stop;

[0050] The specific mathematical model is:

[0051]

[0052] The connection constraint that couples the unit aggregation and dispatch-only constraints is shown in Equation (27), where represents the set of thermal power units in cluster c, which is equivalent to the output of all thermal power units in the same cluster as the output of the scheduling-only model:

[0053]

[0054] According to the above scheme, in step S5, the final complete fast unit commitment model based on scheduling and unit aggregation coupling is as follows:

[0055] Objective function:

[0056]

[0057] Constraints:

[0058]

[0059] The combined model is solved, the operating cost of each generator is piecewise linearized, and the unit combination model for power system operation planning over a longer time scale is solved.

[0060] Although unit aggregation and variable simplification will reduce the accuracy of generator operating status judgment to a certain extent, analysis shows that this method is still far more accurate than the economic dispatch model alone, and achieves a significant improvement in solution speed. It is suitable for power system planning and operation on a long-term time scale where the accuracy of power system generator output planning is not high but a relatively optimal solution needs to be obtained efficiently, achieving a relatively optimal balance between solution speed and accuracy.

[0061] The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling of the present invention has the following beneficial effects:

[0062] 1. The present invention aggregates thermal power units with the same operating characteristics and approximately simplifies the massive 0-1 variables in the thermal power unit aggregation set into integer variables, thereby linearizing the thermal power unit combination model and achieving rapid operation simulation considering a large number of thermal power units;

[0063] 2. This invention provides an approximate network-constrained unit commitment model that can be used in situations where high computational performance is required. It is suitable for new power systems with high renewable energy penetration and scenarios where long-term operation simulations need to consider the volatility of renewable energy.

[0064] 3. The present invention introduces link constraints between the scheduling-only operation model and the unit aggregation model. The overall model overcomes the defects of the two models, can simultaneously consider the unit combination plan and transmission constraints, and perform efficient long-term system operation simulation. It can incorporate operational flexibility into the power system planning problem, and provide strong support for the long-term planning and daily scheduling of the power system while ensuring calculation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0066] Figure 1 It is a flow chart of a method for constructing a fast unit commitment model based on the coupling of scheduling and unit aggregation;

[0067] Figure 2 This is a diagram of the power generation of multiple energy sources under the baseline scenario;

[0068] Figure 3 This is a diagram of the power generation of multiple energy sources in the second case for comparison;

[0069] Figure 4 This is a diagram of multiple energy generation situations in the context of the present invention. DETAILED DESCRIPTION

[0070] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0071] Example

[0072] The IEEE RTS-79 standard test grid, a key benchmark for power system research, boasts a peak load capacity of 2.85 GW, an annual electricity demand of 15.34 terawatt-hours (TWh), and includes 38 transmission lines. To meet the computational requirements for validating this method, its generator sets were modified. The revised system optimizes and reorganizes the existing generation units into a hybrid configuration combining traditional thermal power and clean energy. The updated system incorporates four types of thermal power units, operating in a coordinated manner with renewable energy integration, meeting the unique requirements of modern power system simulations for unit flexibility and energy diversity. Information on the revised generation mix is ​​shown in Table 1, and technical specifications for each thermal power unit are detailed in Table 2.

[0073] Table 1 Modified IEEE RTS-79 system power generation combination information

[0074]

[0075] Table 2 Technical indicators of thermal power units

[0076]

[0077] Set up three situations to illustrate the effectiveness:

[0078] Case 1: Traditional unit commitment model with line constraints. This is a conventional benchmark solution with high computational complexity and long solution time.

[0079] Case 2: A simplified model of economic dispatch with line constraints is used for solution;

[0080] The third case: the fast unit commitment model based on scheduling and unit aggregation coupling proposed by the present invention;

[0081] In the modified IEEE RTS-79 system, the third case proposes a fast unit commitment model based on scheduling and unit aggregation coupling. By combining the constraints of unit aggregation and scheduling, it realizes the solution of large-scale unit commitment model with line constraints. Figure 1 As shown, the method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling of the present invention includes the following steps:

[0082] S1: Construct a traditional unit combination model;

[0083] The mathematical expression of the traditional unit combination model is as follows:

[0084]

[0085]

[0086] Wherein, formula (1) represents the cost function of the traditional unit commitment model, including unit start-up and shutdown costs, variable operating costs and load shedding penalty costs. is the start-up and shutdown cost per unit MW of thermal power unit i, u i,t is the startup state variable of the thermal power unit, is the variable operating cost of thermal power unit i, P Gi,t is the power generation capacity of thermal power unit i (MW), F punish is the penalty coefficient for the necessary load shedding of the system, is the necessary load shedding variable for n nodes in the system;

[0087] Formula (2) represents the power balance constraint of the power system, P Gi,t ,P w,t is the output of the i-th thermal power unit and the w-th wind power unit at time t, D n,t is the load of n nodes in the system;

[0088] Formula (3) represents the upper and lower limit constraints of thermal power unit output, is the minimum output rate of thermal power unit i, represents the maximum output capacity of thermal power unit i;

[0089] Formula (4) represents the climbing constraint of thermal power units, are the hourly up and down ramp rates of thermal power unit i respectively;

[0090] Formula (5) establishes the start-stop state variable u of thermal power unit i i,t ,v i,t and the actual running state variable I i,t the relationship between;

[0091] Formulas (6) and (7) represent the minimum start-stop time constraints of thermal power units. The fixed continuous start-up and shutdown period required at the beginning of each day meets the requirements and T i on ,T i off The minimum continuous start-up and shutdown time required for thermal power units within one day;

[0092] Equation (8) shows that the on / off state and start / shutdown decision variables are modeled as binary variables;

[0093] Formula (9) limits the output of the wind farm to the output range that can be used for power generation, λ w,t represents the hourly predicted wind power generation level, Indicates the maximum output capacity of the wind turbine w;

[0094] Formula (10) limits the load shedding amount to the load demand at the corresponding node bus;

[0095] Formula (11) represents the arrangement of operating reserve capacity, R load ,R WG It represents the load demand and the operating reserve rate of wind turbines in turn;

[0096] Traditional unit combination has P Gi,t ,P w,t ,I i,t ,u i,t ,v i,t and These decision variables, among which the start-stop state variable u i,t ,v i,t and running status variable I i,t It is a binary variable, so the traditional unit commitment problem is a large mixed integer linear programming problem. The existence of a large number of binary variables greatly increases the complexity of solving the problem.

[0097] S2: Introducing line transmission capacity limitations and DC power flow models into traditional unit commitment models to capture power system operation details and better adapt to actual power system operation status;

[0098] The specific constraints of introducing line transmission capacity limitation and DC power flow model into the traditional unit commitment model are:

[0099]

[0100] Formula (12) represents the power balance expression at node n, where G(n), W(n), and L(n) are the number of thermal power units, wind power units, and loads at node n. It represents the tidal flow value on line l with n as the starting point (inflow point) and m as the end point (outflow point).

[0101] Formula (13) represents the DC power flow transmission network model, is the difference between the phase angle of the starting node n and the ending node m of the branch l power flow, B l is the per-unit susceptance value of line l;

[0102] Formula (14) represents the line transmission capacity constraint, F l max is the normal allowable maximum value of the power flow on line l;

[0103] Formula (15) represents the node phase angle value limit, and each node phase angle value is limited to the range between -π and π;

[0104] After adding network transmission constraints, we introduced and These two decision variables.

[0105] According to the above scheme, the DC power flow model is specifically simplified as follows:

[0106] When considering the influence of the power grid, the mathematical model of the transmission network established is a nonlinear power flow calculation model, and the active power flow of the transmission line branches at nodes n and m is shown in formula (16);

[0107]

[0108] For the high-voltage transmission network, the model is simplified under certain assumptions and converted into a linear DC model. The assumptions include:

[0109] ① The resistance of high-voltage transmission branches is generally much smaller than the reactance, so the resistance r is ignored. nm , then g nm =0;

[0110] ②In normal operation, generally θ n -θm is very small, then cosθ nm =1,sinθ nm =θ nm ;

[0111] ③ Assume that the per-unit value of all node voltages is 1, that is, U n =1;

[0112] ④ Ignore the ground branch, then g n0 =0;

[0113] Based on the above assumptions, the nonlinear power flow equation shown in Equation (16) can be transformed into a linear DC power flow model, and the node phase angle is linked to the line power flow. The DC power flow model is combined with Kirchhoff's current law and the power balance equation at any node n is expressed in matrix form as shown in Equation (17):

[0114]

[0115] in, They are the node-branch, node-thermal power generator, node-wind power generator and node-load matrix elements of line l, is the active power flow of the branch l (with n as the starting point of the flow and m as the end point of the flow), is the active power output of the thermal power generator set and wind power generator set connected to line l, The load connected to line l and the necessary load shedding.

[0116] S3: Aggregate large-scale units with the same operating characteristics and model the output characteristics and ramp start-stop constraints of the aggregated set;

[0117] The basic idea of ​​aggregating large-scale units with the same operating characteristics is to group similar power generation units into clusters and replace the binary decision variables of all units in each cluster with a single integer variable. The specific mathematical model is established as follows:

[0118]

[0119]

[0120] Units with similar operating characteristics are grouped into one category. In each category, it is assumed that The mathematical model of the aggregated units remains mostly the same as the traditional unit combination formula, except that the individual unit index is replaced by the unit cluster index c. Unlike modeling the on / off state of a single unit, the online capacity of the unit cluster c is described and captured as where n c,trepresents the number of thermal power units in the unit cluster c at time t. Since all units in the cluster are assumed to be identical, the characteristic parameters of the same type of units, such as single unit capacity, minimum load rate, ramp rate, and minimum start and stop time period, are assumed to be the average values ​​of the parameters of the units in the cluster. For example, is the average capacity of the cells in cell cluster c, expressed as The remaining clustering parameters These are also the mean values ​​of the unit parameters within the cluster; The fixed continuous start-up and shutdown period required at the beginning of each day in the cluster, satisfying and

[0121] S4: Introduce some relaxation assumptions of the economic-only dispatch model and couple the unit aggregation and dispatch-only constraints;

[0122] The economic dispatch model eliminates binary variables by relaxing some operational constraints on the generators. The key assumptions include:

[0123] ① Thermal power generating units have greater flexibility and can adjust power generation without restrictions. They can flexibly dispatch their power generation from zero to power generation capacity without considering minimum power generation restrictions.

[0124] ② The start and stop operation speed of the thermal power unit is very fast, and it can respond to commands instantly and realize immediate start or stop;

[0125] The specific mathematical model is:

[0126]

[0127] The connection constraint that couples the unit aggregation and dispatch-only constraints is shown in Equation (27), where represents the set of thermal power units in cluster c, which is equivalent to the output of all thermal power units in the same cluster as the output of the scheduling-only model:

[0128]

[0129] S4: Introduce some relaxation assumptions of the economic-only dispatch model and couple the unit aggregation and dispatch-only constraints;

[0130] S5: Form a fast unit commitment model based on scheduling and unit aggregation coupling, and solve the model piecewise linearly.

[0131] According to the above scheme, in step S5, the final complete fast unit commitment model based on scheduling and unit aggregation coupling is as follows:

[0132] Objective function:

[0133]

[0134] Constraints:

[0135]

[0136] The combined model is solved, the operating cost of each generator is piecewise linearized, and the unit combination model for power system operation planning over a longer time scale is solved.

[0137] Although unit aggregation and variable simplification will reduce the accuracy of generator operating status judgment to a certain extent, analysis shows that this method is still far more accurate than the economic dispatch model alone, and achieves a significant improvement in solution speed. It is suitable for power system planning and operation on a long-term time scale where the accuracy of power system generator output planning is not high but a relatively optimal solution needs to be obtained efficiently, achieving a relatively optimal balance between solution speed and accuracy.

[0138] The present invention selects the test results of 7 days and 168 hours for comparison, with a time interval of 1 hour. The optimality gap threshold of different algorithms is controlled within 0.1%. Table 2 shows the comparison of different schemes in terms of time, cost, thermal power unit output power, line flow, and computational complexity. The error calculation formula is:

[0139] Line power flow error:

[0140]

[0141] Generator output error:

[0142]

[0143] As can be seen from the table, the solution proposed in the present invention has greatly improved the calculation accuracy compared to the more simplified scheduling algorithm, and has smaller errors in operating costs, line power, and generator output power. It can make decisions on the start and stop status of generators, is more applicable to actual power grids, and the operating time is greatly improved compared to the baseline solution. It is very suitable for medium- and long-term power system optimization operation and early planning situations where the calculation accuracy requirements are not too high.

[0144] Table 3 Comparison of running results of various schemes

[0145]

[0146] Figure 2 、 34 shows the comparison of power generation of different energy sources under the three schemes on a certain day within 7 days; it can be seen from the figure that, compared with the scheduling algorithm, the method proposed in the present invention uses integer variables to model the unit operation status decision and obtains the best approximation to the benchmark scheme, which further verifies that the proposed method shows higher precision and accuracy in simulating long-term unit combination problems compared with other situations. The simulation results are closer to the actual operating conditions, providing a more reliable basis for power system planning, scheduling and operation management.

[0147] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling, characterized in that: The following steps are involved: S1. Construct a traditional unit combination model; S2, introducing line transmission capacity limitation and DC power flow model into the traditional unit combination model; S3. Aggregate large-scale units with the same operating characteristics and model the output characteristics and ramp start-stop constraints of the aggregated set; S4, introduce the slack assumption of the economic dispatch model and couple the constraints of unit aggregation and dispatch only; S5. Complete the construction of a fast unit commitment model based on scheduling and unit aggregation coupling, and solve the model by piecewise linearization.

2. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 1 is characterized in that: In step S1, the mathematical expression of the traditional unit combination model is as follows: Objective function Wherein, formula (1) represents the cost function of the traditional unit commitment model, including unit startup costs, variable operating costs and load shedding penalty costs. is the startup cost per MW of thermal power unit i, u i,t is the startup state variable of the thermal power unit, is the variable operating cost of thermal power unit i, P Gi,t is the power generation capacity of thermal power unit i (MW), F punish is the penalty coefficient for the necessary load shedding of the system, is the necessary load shedding variable for n nodes in the system; Formula (2) represents the power balance constraint of the power system, P Gi,t ,P w,t is the output of the i-th thermal power unit and the w-th wind power unit at time t, D n,t is the load of n nodes in the system; Formula (3) represents the upper and lower limit constraints of thermal power unit output, is the minimum output rate of thermal power unit i, represents the maximum output capacity of thermal power unit i; Formula (4) represents the climbing constraint of thermal power units, are the hourly up and down ramp rates of thermal power unit i respectively; Formula (5) establishes the start-stop state variable u of thermal power unit i i,t ,v i,t and the actual running state variable I i,t the relationship between; Formulas (6) and (7) represent the minimum start-stop time constraints of thermal power units. The fixed continuous start-up and shutdown period required at the beginning of each day meets the requirements and T i on ,T i off The minimum continuous start-up and shutdown time required for thermal power units within one day; Equation (8) shows that the on / off state and start / shutdown decision variables are modeled as binary variables; Formula (9) limits the output of the wind farm to the output range that can be used for power generation, λ w,t represents the hourly predicted wind power generation level, Indicates the maximum output capacity of the wind turbine w; Formula (10) limits the load shedding amount to the load demand at the corresponding node bus; Formula (11) represents the arrangement of operating reserve capacity, R load ,R WG They represent the load demand and the operating reserve rate of wind turbines respectively.

3. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 2 is characterized in that: In step S2, the line transmission capacity limitation and the DC power flow model are introduced into the traditional unit combination model. The specific constraints are: Formula (12) represents the power balance expression at node n, where G(n), W(n), and L(n) are the number of thermal power units, wind power units, and loads at node n. It represents the tidal flow value on line l with n as the starting point and m as the end point; Formula (13) represents the DC power flow transmission network model, is the difference between the phase angle of the starting node n and the ending node m of the branch l power flow, B l is the per-unit susceptance value of line l; Formula (14) represents the line transmission capacity constraint, F l max is the normal allowable maximum value of the power flow on line l; Formula (15) represents the node phase angle value limit, and each node phase angle value is limited to the range between -π and π.

4. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 3 is characterized in that: The DC power flow model is specifically simplified as follows: When considering the influence of the power grid, the mathematical model of the transmission network established is a nonlinear power flow calculation model, and the active power flow of the transmission line branches at nodes n and m is shown in formula (16); For the high-voltage transmission network, the model is simplified under certain assumptions and converted into a linear DC model. The assumptions include: ① The resistance of high-voltage transmission branches is generally much smaller than the reactance, so the resistance r is ignored. nm , then g nm =0; ②In normal operation, generally θ n -θ m is very small, then cosθ nm =1,sinθ nm =θ nm ; ③ Assume that the per-unit value of all node voltages is 1, that is, U n =1; ④ Ignore the ground branch, then g n0 =0; Based on the above assumptions, the nonlinear power flow equation shown in Equation (16) is transformed into a linear DC power flow model, and the node phase angle is linked to the line power flow. The DC power flow model is combined with Kirchhoff's current law and the power balance equation at any node n is expressed in matrix form as shown in Equation (17): in, They are the node-branch, node-thermal power generator, node-wind power generator and node-load matrix elements of line l, is the active power flow of the branch l with n as the starting point and m as the end point, is the active power output of the thermal power generator set and wind power generator set connected to line l, The load connected to line l and the necessary load shedding.

5. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 4 is characterized in that: In step S3, the aggregation of large-scale units with the same operating characteristics is to group similar power generation units into clusters, and replace the binary decision variables of all units in each cluster with a single integer variable. The specific mathematical model is established as follows: Units with similar operating characteristics are grouped into one category. In each category, it is assumed that For thermal power units, the mathematical model after unit aggregation remains unchanged from the traditional unit combination formula except that the individual unit index is replaced by the unit cluster index c. Instead of modeling the on / off state of a single cell, the online capacity of a cell cluster c is described and captured as where n c,t represents the number of thermal power units in the unit cluster c at time t. The characteristic parameters of the same type of units, such as single unit capacity, minimum load rate, ramp rate, and minimum start-stop time period, are assumed to be the average values ​​of the parameters of the units in the cluster. is the average capacity of the cells in cell cluster c, expressed as Clustering parameters is the mean value of the unit parameters within the cluster; The fixed continuous start-up and shutdown period required at the beginning of each day in the cluster, satisfying and 6. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 5, characterized in that: In step S4, only the economic dispatch model eliminates binary variables by relaxing some operating constraints on the generator sets, including the following key assumptions: ① Thermal power generating units have greater flexibility and can adjust power generation without restrictions. They can flexibly dispatch their power generation from zero to power generation capacity without considering minimum power generation restrictions. ② The start and stop operation speed of the thermal power unit is very fast, and it can respond to commands instantly and realize immediate start or stop; The specific mathematical model is: The connection constraint that couples the unit aggregation and dispatch-only constraints is shown in Equation (27), where represents the set of thermal power units in cluster c, which is equivalent to the output of all thermal power units in the same cluster as the output of the scheduling-only model:

7. The method for constructing a fast unit commitment model based on scheduling and unit aggregation coupling according to claim 6, characterized in that: In step S5, the fast unit commitment model based on scheduling and unit aggregation coupling is as follows: Objective function: Constraints: The combined model is solved, the operating cost of each generator is piecewise linearized, and the unit combination model for power system operation planning over a longer time scale is solved.

Images