Intraday rolling optimization scheduling method considering inertia short-time fluctuation

Abstract

The application discloses a day-ahead rolling optimization scheduling method considering inertia short-time fluctuation, and belongs to the technical field of day-ahead scheduling. The system inertia demand lower limit is calculated, the power system is equivalent to a second-order oscillation system interconnected by synchronous units and a power grid, and the system inertia demand upper limit is calculated by using a damping ratio to represent the small-signal dynamic characteristic of the system. A day-ahead double-layer rolling optimization scheduling model including an upper model and a lower model is constructed by considering the inertia short-time fluctuation and the inertia regulation limit constraint of network-type equipment. The inertia regulation plan of the network-type equipment is made through the upper model, and the active power and inertia regulation plan is feedback corrected through the lower model. The day-ahead double-layer rolling optimization scheduling model is solved by using a Gurobi solver. The method can maintain the system inertia in a safe operation interval while avoiding frequent regulation of the network-type equipment.

Description

Technical Field

[0001] This invention relates to the field of intraday scheduling technology, and in particular to an intraday rolling optimization scheduling method that takes into account short-term fluctuations in inertia. Background Technology

[0002] In high-proportion renewable energy power systems dominated by grid-connected equipment, the high proportion of renewable energy grid connection not only leads to difficulties in renewable energy consumption and peak shaving, but also reduces the power system's inertia level and deteriorates its immunity to disturbances. Furthermore, given the current fixed start-up and shutdown schedules of thermal power units, the uncertainty of renewable energy output will cause significant differences in system inertia over a short timescale, potentially even exceeding inertia limits under extreme weather conditions. While inertia in traditional power systems dominated by synchronous machines can be considered relatively stable over such periods, in high-proportion renewable energy power systems, fluctuations in renewable energy output on a minute-by-minute scale can have a significant impact on system inertia. For example, on a typical day in the Northwest China power grid, the maximum renewable energy output was 42.54 GW, and within 10 minutes, the maximum renewable energy output fluctuated by 2.50 GW, approximately 5% of the maximum daily output. Considering the inertia support capacity of renewable energy, the impact of this output fluctuation on system inertia cannot be ignored.

[0003] Grid-based control strategies leverage the characteristics of virtual synchronous machines for new energy and energy storage equipment. While providing inertia support, they also offer rapid response capabilities. Given a fixed intraday synchronous inertia for thermal power units, the inertia of grid-based new energy power plants and grid-based energy storage exhibits instantaneous response and continuous adjustability, making them highly suitable for handling short-term fluctuations in system inertia. However, optimizing them on a short intraday timescale may result in frequent adjustments to operating modes and inertia parameters. To address this issue, intraday optimization scheduling methods for short-term inertia fluctuations need to be considered. By optimizing the inertia resources of various grid-based devices, the system inertia can be maintained within a safe operating range while avoiding frequent adjustments to the grid-based devices. Summary of the Invention

[0004] The purpose of this invention is to propose an intraday rolling optimization scheduling method that considers short-term inertia fluctuations, including:

[0005] The lower limit of system inertia demand is calculated based on the system frequency change rate constraint and the frequency minimum point constraint. The power system is equivalent to a second-order oscillating system interconnected with the grid by synchronous generators. The upper limit of system inertia demand is calculated by using the damping ratio to characterize the small-signal dynamic characteristics of the system.

[0006] Considering short-term fluctuations in inertia and constraints on inertia adjustment for network-type equipment, a daily two-layer rolling optimization scheduling model is constructed, including an upper-layer model and a lower-layer model. The upper-layer model is used to formulate the inertia adjustment plan for network-type equipment, and the lower-layer model is used to provide feedback correction for the active power and inertia adjustment plans.

[0007] The Gurobi solver was used to solve the intraday two-level rolling optimization scheduling model.

[0008] Furthermore, the lower limit of system inertia requirement is the larger of the minimum inertia obtained from the frequency change rate constraint and the minimum inertia obtained from the frequency minimum point constraint.

[0009] Furthermore, the upper limit of system inertia demand is the smaller of the upper limit of system inertia constrained by damping ratio and the maximum value of system inertia supply.

[0010] Furthermore, the upper-level model includes:

[0011] Objective function:

[0012] ;

[0013] ;

[0014] In the formula, This is the inertia margin; and These represent the system inertia and minimum system inertia requirement for time period t, respectively. , , These are the inertia deviation penalty coefficients, , , These are the initial new energy power station inertia adjustment cost coefficient, low-voltage equivalent energy storage inertia adjustment cost coefficient, and high-voltage direct-connected energy storage inertia adjustment cost coefficient, respectively. The marginal cost coefficient for adjusting the proportion of new energy power stations in the grid; , These are the marginal cost coefficients for switching between high-voltage direct-connected energy storage networks and adjusting the inertial time constant, respectively. , These are the marginal cost coefficients for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively. As an indicator variable for adjusting the proportion of new energy power stations in the power grid; , These are respectively the high-voltage direct-connected energy storage grid switching and the inertial time constant adjustment indicator variables; , These are the indicator variables for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively.

[0015] Constraints:

[0016] System inertia upper and lower limits constraints

[0017] ;

[0018] Grid-type high-voltage direct-connected energy storage regulation constraints

[0019] ;

[0020] ;

[0021] In the formula, , , All are integer variables; , The following are the grid connection modes of energy storage k at times t+1 and t, respectively; , Let be the inertial time constants of the stored energy k at times t+1 and t, respectively; The maximum number of switching operations for energy storage k; The maximum number of adjustments to the energy storage inertia time constant k is given; T is the scheduling period, with a time scale of 15 minutes, dividing the day into 96 time periods; M is a relatively large constant introduced using the Big M method.

[0022] Regulation Constraints of Grid-type Low-voltage Equivalent Energy Storage Stations

[0023] ;

[0024] ;

[0025] In the formula, , These represent the proportions of energy storage m in grid-connected operation at time t+1 and time t, respectively. , are the inertial time constants of the energy storage m at time t+1 and time t, respectively; The maximum number of switching operations for energy storage m; The maximum number of adjustments to the inertial time constant of the energy storage unit m;

[0026] Regulation and Constraints of Grid-Based New Energy Power Stations

[0027] ;

[0028] In the formula, , The proportion of the new energy power station w that is in grid-connected operation at time t+1 and time t; This represents the maximum number of switching operations for the new energy power station.

[0029] Energy storage frequency security support constraints

[0030] ;

[0031] In the formula, The total frequency regulation power provided by energy storage after the disturbance in time period t; The inertial time constant of the energy storage device; This refers to the maximum charge / discharge rate of the energy storage device. and These refer to the charging efficiency and discharging efficiency of energy storage devices, respectively. For primary frequency modulation output; Rated power of energy storage; This is the initial frequency of the system; , These are the energy storage discharge power and the charging power, respectively. , These are the minimum energy storage capacity and the rated capacity, respectively. This is the frequency response time.

[0032] Furthermore, the lower-level model includes:

[0033] Objective function:

[0034] ;

[0035] ;

[0036] ;

[0037] In the formula: For system operating costs, To adjust for planning deviation costs; The cost of generating electricity from thermal power units; , , These are the penalty coefficients for wind curtailment, solar curtailment, and load shedding, respectively. , , These represent the power curtailment of wind and solar power, and the power shedding during time period t. , , These represent the daytime thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output for time period t, respectively. , , These represent the output of thermal power units, the output of high-voltage direct-connected energy storage, and the output of low-voltage equivalent energy storage during time period t. , , , , , The system inertia, the proportion of grid-type low-pressure equivalent energy storage stations and their inertial time constants in the upper-level model, the grid-type high-pressure direct-connected energy storage network operation mode and their inertial time constants, and the proportion of grid-type new energy stations are respectively determined. , , These are the deviation costs of thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output, respectively. , , , These are the system inertia, low-pressure equivalent energy storage station, high-pressure direct-connected energy storage, and new energy station readjustment deviation penalty coefficient, respectively. Provide power to the new energy power station at time t; Let be the system inertia at time t; The rated capacity of low-voltage equivalent energy storage m;

[0038] Constraints:

[0039] Node power balance constraints

[0040] ;

[0041] Rotational spare constraint

[0042] ;

[0043] In the formula: , These are the real and imaginary parts of the nodal admittance matrix, respectively. , Let be the voltage amplitude of nodes i and j at time t during the previous phase; Let be the phase angle difference between nodes i and j at time t in the previous phase; Confidence level; , and These are the membership parameters for the fuzzy parameters of wind, solar, and load, respectively, i=1,2,3,4; , These are the energy storage discharge power and the charging power, respectively. , These represent the thermal power unit number and set, respectively; n is the node number.

[0044] Safety Operation Constraints of Thermal Power Units

[0045] ;

[0046] In the formula: , These are the upper and lower limits of the active power output of thermal power unit g; , These represent the maximum upward and downward ramp rates of thermal power unit g, respectively. , These represent the start-up and shutdown status of the thermal power unit at time t+1 and time t, respectively. , The output of the thermal power unit at time t+1 and time t are respectively;

[0047] Energy storage operation constraints

[0048] ;

[0049] In the formula: , These are the charge of energy storage k at time t and the rated charge of energy storage k, respectively. , These represent the maximum charging and discharging power of energy storage k; , These are the charging and discharging indicators of energy storage k at time t, with 1 indicating that energy storage k is in a charging or discharging state at time t; Energy storage k at time t ; , They are respectively The upper and lower limits;

[0050] New energy power constraints

[0051] ;

[0052] ;

[0053] In the formula: , These represent the dispatch power of wind farm w and photovoltaic farm v at time t, respectively. , These are the predicted power outputs of wind power (w) and photovoltaic power (v) at time t, respectively. , These represent the amount of wind and solar power curtailment (w for wind power and v for solar power) at time t, respectively.

[0054] System inertia upper and lower limits constraints

[0055] ;

[0056] Energy storage frequency security support constraints

[0057] .

[0058] The beneficial effects of this invention are as follows:

[0059] The method of this invention optimizes the inertia resources of various network-type devices, which can ensure that the system inertia is maintained within a safe operating range while avoiding frequent adjustments of network-type devices. Attached Figure Description

[0060] Figure 1This is a flowchart illustrating an intraday optimized scheduling method that considers short-term inertia fluctuations provided by the present invention.

[0061] Figure 2 This is a framework diagram of the intraday two-layer rolling optimization scheduling model provided by the present invention;

[0062] Figure 3 This is a schematic diagram of a DC receiving-end equivalent system provided by the present invention;

[0063] Figure 4 This is the energy storage active power scheduling result considering frequency security support constraints in the embodiment provided by the present invention;

[0064] Figure 5 This is the energy storage inertia scheduling result considering frequency security support constraints in the embodiments provided by the present invention;

[0065] Figure 6 This is the energy storage active power scheduling result provided by the embodiment of the present invention without considering frequency security support constraints;

[0066] Figure 7 This is the energy storage inertia scheduling result of the embodiment provided by the present invention without considering frequency security support constraints;

[0067] Figure 8 This is a comparison diagram of the system inertia of the scheduling scheme (Scheme 1), the scheduling scheme (Scheme 2) without considering the adjustment constraints of the network-type station, and the scheduling scheme (Scheme 3) without considering inertia adjustment provided by the present invention. Detailed Implementation

[0068] This invention proposes an intraday rolling optimization scheduling method that considers short-term inertia fluctuations. The invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0069] Figure 1 This is a flowchart illustrating an intraday optimized scheduling method considering short-term inertia fluctuations provided by the present invention. Figure 2 This is a framework diagram of the intraday two-layer rolling optimization scheduling model provided by the present invention, specifically including:

[0070] Step S101: Calculate the lower limit of system inertia demand based on the system frequency change rate constraint and the frequency minimum point constraint. Equivalently, the power system is converted into a second-order oscillating system interconnected with the grid by synchronous generators. Calculate the upper limit of system inertia demand using the damping ratio to characterize the small-signal dynamic characteristics of the system.

[0071] S101.1 Calculate the lower limit of system inertia requirement by constraining the rate of change of frequency and the minimum frequency point.

[0072] By constraining the RoCoF at the instant the disturbance occurs, the minimum inertia requirement of the system based on RoCoF constraints can be obtained:

[0073] (1)

[0074] In the formula: To determine the minimum inertia requirement of a RoCoF-constrained system, This represents the percentage of the system's maximum power deficit relative to its rated generating capacity. This is the maximum allowable rate of frequency change for the system. This is the system's rated frequency.

[0075] Lowest frequency point It is mainly related to system inertia and primary frequency modulation; the lowest frequency point cannot be lower than the allowable limit.

[0076] (2)

[0077] By simulating the system's frequency response using piecewise linearization, the minimum frequency point and the minimum inertia requirement of the system can be obtained as follows:

[0078] (3)

[0079] (3)

[0080] In the formula: Minimum inertia requirement of the system at the lowest frequency point; This is the primary frequency modulation response power; This is the duration of one frequency modulation operation; For system damping; This is the minimum allowable limit for the system frequency.

[0081] Equations (1) and (3) represent the minimum inertia requirements of the system based on RoCoF and the minimum frequency point constraint, respectively. The larger of the two values ​​is taken as the minimum inertia requirement of the system.

[0082] (4)

[0083] S101.2. The power system is equivalent to a second-order oscillating system consisting of a group of synchronous machines interconnected with the power grid. The upper limit of the system inertia requirement is calculated through the critical damping ratio constraint:

[0084] After equating the power system to a second-order oscillatory system, its small-signal dynamic characteristics can be characterized by the damping ratio. According to classical small-signal stability theory, the expression for the system damping ratio is:

[0085] (5)

[0086] In the formula, The system damping ratio; The system's mechanical damping coefficient; The primary frequency regulation coefficient of the load. , The slope of the load power frequency characteristic curve; The system's equivalent inertial time constant is expressed in seconds. For synchronous angular velocity, The unit is rad / s; The synchronous torque coefficient, ,in This is the internal potential of the equivalent power source. This is the grid voltage. For the angle of attack, This is the equivalent reactance.

[0087] As shown in equation (5), the damping ratio is negatively correlated with the system inertia. When the system inertia is too large, the damping ratio will drop below the critical value, causing the system to oscillate continuously or even become unstable after a disturbance. The national standard GB / T40581—2021 "Specification for Calculation of Safety and Stability of Power Systems" requires a minimum damping ratio of 0.01~0.02. Therefore, the damping ratio should satisfy:

[0088] (6)

[0089] Substituting equation (6) into equation (5), we can obtain the upper limit of system inertia based on damping ratio constraints:

[0090] (7)

[0091] Synchronous torque coefficient in equation (7) It is closely related to the system's operating status. For power systems with a high proportion of new energy sources, The value of depends on the short-circuit capacity contribution of the online synchronous generator. Introducing the concept of Short Circuit Ratio (SCR), the synchronous torque coefficient can be expressed as:

[0092] (8)

[0093] In the formula, SCR is the system short-circuit ratio, which reflects the voltage support strength of the power grid to the grid connection point.

[0094] Based on Thevenin's equivalent principle, the system short-circuit ratio can be estimated from the short-circuit capacity contribution of the online synchronous generator units:

[0095] (9)

[0096] In the formula, For system short-circuit capacity; This represents the total capacity of the thermal power units. The rated capacity of synchronous generator unit g; This represents the operating status of the unit at time t; This represents the average subtransient reactance of the synchronous generator unit, typically 0.2~0.3 pu; This is a collection of synchronous generator units.

[0097] make (Per unit value), we can obtain:

[0098] (10)

[0099] In the formula, For system damping, ; This represents the total capacity of thermal power units.

[0100] Considering that the maximum system inertia supply capacity will exceed the upper limit of inertia demand based on the critical damping ratio constraint for most periods, the upper limit of system inertia demand is taken as the smaller value between the two as the maximum system inertia demand:

[0101] (11)

[0102] (12)

[0103] In the formula, The maximum value of the system inertia supply at time t; , , , respectively, are the inertial constants of thermal power unit g, low-pressure equivalent energy storage m, and wind farm w. Let k be the maximum inertial time constant of the high-voltage direct-connected energy storage. Let g be the binary variable for starting and stopping unit g during time period t; The predicted output of wind farm station w at time t; , These are high-voltage direct-connected energy storage and low-voltage equivalent energy storage power stations, respectively. , The maximum proportion of grid-connected equipment in low-voltage equivalent energy storage power station m and wind farm station w, respectively. K, M, and W represent the high-voltage direct-connected grid-connected energy storage aggregate, low-voltage equivalent energy storage, and wind farm station aggregate, respectively.

[0104] Step S102: Considering short-term fluctuations in inertia and constraints on inertia adjustment of network-type equipment, construct an intraday two-layer rolling optimization scheduling model including an upper-layer model and a lower-layer model. Formulate an inertia adjustment plan for network-type equipment through the upper-layer model, and perform feedback correction on the active power and inertia adjustment plan through the lower-layer model.

[0105] A fuzzy chance constraint approach is adopted to handle the uncertainty of new energy output, and power balance constraints and thermal power unit reserve constraints are transformed into deterministic constraints through clear equivalence classes. The upper-level model sets the system inertia adjustment plan with the goal of minimizing the sum of system inertia deviation cost, energy storage grid operation cost, and grid equipment mode switching cost. The lower-level model uses the goal of minimizing system operation cost and adjustment plan deviation to provide feedback correction to the upper-level model plan, achieving iterative solution between the two-level models.

[0106] S102.1 Handling Uncertainty in New Energy Output and Constructing Formulas for System Inertia Calculation

[0107] (1) Handling the uncertainty of new energy output

[0108] Describing each uncertainty using a trapezoidal membership function , , And using the clear equivalence class under the trapezoidal fuzziness parameter, the four-tuple expressions for the trapezoidal fuzziness number are as follows:

[0109] (13)

[0110] In equation (13): , and These are the predicted values ​​for wind, solar, and load, respectively. , and These are the membership parameters for the fuzzy parameters of wind, solar, and load, respectively, i=1,2,3,4; , and These are the proportional coefficients corresponding to wind, solar, and load, respectively.

[0111] (2) Formula for calculating system inertia

[0112] The system inertia consists of two parts: the fixed inertia provided by the thermal power unit and the adjustable inertia provided by the grid-type equipment. The calculation formula (18) is shown below:

[0113] (14)

[0114] In the formula, Let w be the inertial constant of the wind farm station. , , respectively, are the inertial time constants of low-pressure equivalent energy storage m and high-pressure direct-connected energy storage k during time period t; is the network construction and operation status variable of high-voltage direct-connected energy storage k during time period t. When it is equal to 1 and 0, it means that high-voltage direct-connected energy storage k is in the network construction and grid connection operation states during time period t, respectively. , These represent the proportions of grid-type equipment in the low-voltage equivalent energy storage power station m and the wind farm station w at time t, respectively.

[0115] S102.2 Construct the upper-level model in the intraday two-layer rolling optimization scheduling model;

[0116] The upper-level model aims to minimize the sum of system inertia deviation cost, energy storage grid operation cost, and grid equipment mode switching cost. It formulates an inertia adjustment plan for grid-type equipment, and its objective function is shown in the following equation:

[0117] (15)

[0118] In the formula, This is the inertia margin; and These represent the system inertia and minimum system inertia requirement for time period t, respectively. , , These are the inertia deviation penalty coefficients. By introducing a marginal cost coefficient that increases with the number of adjustments, an implicit adjustment priority mechanism is constructed in the objective function. The specific calculation method is as follows:

[0119] (16)

[0120] In the formula, This is the inertia margin; and These represent the system inertia and minimum system inertia requirement for time period t, respectively. , , These are the inertia deviation penalty coefficients, , , These are the initial new energy power station inertia adjustment cost coefficient, low-voltage equivalent energy storage inertia adjustment cost coefficient, and high-voltage direct-connected energy storage inertia adjustment cost coefficient, respectively. The marginal cost coefficient for adjusting the proportion of new energy power stations in the grid; , These are the marginal cost coefficients for switching between high-voltage direct-connected energy storage networks and adjusting the inertial time constant, respectively. , These are the marginal cost coefficients for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively. As an indicator variable for adjusting the proportion of new energy power stations in the power grid; , These are respectively the high-voltage direct-connected energy storage grid switching and the inertial time constant adjustment indicator variables; , These are the indicator variables for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively.

[0121] The constraints of the upper-level model include:

[0122] (1) Upper and lower limits of system inertia constraints

[0123] (17)

[0124] And constraints on the inertia adjustment of network-type equipment:

[0125] (2) Restrictions and constraints on grid-connected high-voltage direct-connected energy storage regulation

[0126] (18)

[0127] (19)

[0128] In the formula, The maximum number of switching operations for energy storage k; Let be the maximum number of adjustments to the energy storage inertia time constant k. T is the scheduling period, with a time scale of 15 minutes, dividing the day into 96 time periods; M is a relatively large constant introduced using the Big M method.

[0129] (3) Regulation constraints of grid-type low-voltage equivalent energy storage stations

[0130] (20)

[0131] (twenty one)

[0132] In the formula, The maximum number of switching operations for energy storage m; The maximum number of adjustments to the inertial time constant of the energy storage m is given.

[0133] (4) Regulation and constraint of grid-type new energy power stations

[0134] (twenty two)

[0135] In the formula, This represents the maximum number of switching operations for the new energy power station.

[0136] (5) Energy storage frequency security support constraints

[0137] (twenty three)

[0138] In the formula, The total frequency regulation power provided by energy storage after the disturbance in time period t; The inertial time constant of the energy storage device; This refers to the maximum charge / discharge rate of the energy storage device. and These refer to the charging efficiency and discharging efficiency of the energy storage device, respectively.

[0139] S102.3 Construct the lower-level model in the intraday two-layer rolling optimization scheduling model;

[0140] The lower-level model combines intraday net load forecast data to adjust the daytime thermal power unit output plan and the upper-level model's grid-connected station regulation plan. The decision from each optimized scheduling is fed back to the upper-level model as its initial operating conditions for further optimization during the remaining time period. The objective function of the lower-level model is shown in equation (24):

[0141] (twenty four)

[0142] In the formula, For system operating costs, To adjust for planning deviation costs.

[0143] The system operating costs include the unit's power generation costs, wind and solar curtailment costs, and load shedding penalty costs.

[0144] (25)

[0145] The cost of deviation from the regulation plan includes the cost of deviation from the day-ahead thermal power unit output plan, the cost of deviation from the system inertia, the cost of deviation from the active power and inertia regulation of low-voltage equivalent energy storage stations, the cost of deviation from the active power and inertia regulation of high-voltage direct-connected energy storage, and the cost of deviation from the inertia regulation of new energy stations.

[0146] (26)

[0147] In the formula: For system operating costs, To adjust for planning deviation costs; The cost of generating electricity from thermal power units; , , These are the penalty coefficients for wind curtailment, solar curtailment, and load shedding, respectively. , , These represent the power curtailment of wind and solar power, and the power shedding during time period t. , , These represent the daytime thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output for time period t, respectively. , , These represent the output of thermal power units, the output of high-voltage direct-connected energy storage, and the output of low-voltage equivalent energy storage during time period t. , , , , , The system inertia, the proportion of grid-type low-pressure equivalent energy storage stations and their inertial time constants in the upper-level model, the grid-type high-pressure direct-connected energy storage network operation mode and their inertial time constants, and the proportion of grid-type new energy stations are respectively determined. , , These are the deviation costs of thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output, respectively. , , , These are the system inertia, low-pressure equivalent energy storage station, high-pressure direct-connected energy storage, and new energy station readjustment deviation penalty coefficient, respectively. Provide power to the new energy power station at time t; Let be the system inertia at time t; The rated capacity of low-voltage equivalent energy storage m;

[0148] The constraints of the lower-level model include:

[0149] (1) Node power balance constraints

[0150] (27)

[0151] (2) Rotational spare constraint

[0152] (28)

[0153] In the formula: , These are the real and imaginary parts of the nodal admittance matrix, respectively. , Let be the voltage amplitude of nodes i and j at time t during the previous phase; Let be the phase angle difference between nodes i and j at time t in the previous phase; Confidence level; , and These are the membership parameters for the fuzzy parameters of wind, solar, and load, respectively, i=1,2,3,4; , These are the energy storage discharge power and the charging power, respectively. , These represent the thermal power unit number and set, respectively; n is the node number.

[0154] (3) Constraints on safe operation of thermal power units

[0155] (29)

[0156] In the formula: , These are the upper and lower limits of the active power output of thermal power unit g; , These represent the maximum upward and downward ramp rates of thermal power unit g, respectively. , These represent the start-up and shutdown status of the thermal power unit at time t+1 and time t, respectively. , The output of the thermal power unit at time t+1 and time t are respectively;

[0157] (4) Constraints on Energy Storage Operation

[0158] (30)

[0159] In the formula: , These are the charge of energy storage k at time t and the rated charge of energy storage k, respectively. , These represent the maximum charging and discharging power of energy storage k; , These are the charging and discharging indicators of energy storage k at time t, with 1 indicating that energy storage k is in a charging or discharging state at time t; Energy storage k at time t ; , They are respectively The upper and lower limits.

[0160] (5) Power constraints of new energy sources

[0161] (31)

[0162] (32)

[0163] In the formula: , These represent the dispatch power of wind farm w and photovoltaic farm v at time t, respectively. , These are the predicted power outputs of wind power (w) and photovoltaic power (v) at time t, respectively. , These represent the amount of wind and solar power curtailment (w for wind power and v for solar power) at time t, respectively.

[0164] In addition, it also includes upper and lower limits of inertia constraints and energy storage frequency safety support constraints, as shown in equations (17) and (23).

[0165] Step S103: Use the Gurobi solver to solve the intraday two-layer rolling optimization scheduling model; verify the effectiveness of the proposed scheduling scheme in ensuring that the system inertia is within a safe range while avoiding frequent adjustments of network-type equipment by comparing different schemes.

[0166] This embodiment uses an equivalent model of a real DC receiving-end system in China, such as... Figure 3 As shown. The DC tie line is connected to node 40, with a rated capacity of 8000MW. Low-voltage equivalent energy storage of 400MW / 800MWh is configured at nodes 4, 5, 8, 11, 19, 29, 42, and 55 respectively. Two 100MW / 200MWh high-voltage direct-connected energy storage units are connected in parallel at node 54, all with grid switching capability, and both have a charge / discharge efficiency of 0.95. The maximum system load is 66GW. , The adjustment range is [0.2, 0.5]; the penalty for wind and solar curtailment is ¥200 / MWh, the load shedding penalty is ¥500 / MWh, the system inertia and the cost coefficient for the inertia adjustment deviation of various grid-type equipment are all ¥50 / MWs, the equivalent inertia constant of the new energy power station is taken as 6.5s, and the inertia constant of the low-voltage equivalent energy storage station and the high-voltage direct-connected grid-type energy storage is taken as the range [2, 10]. In the adjustment constraint of the grid-type equipment, the upper limit of the grid ratio adjustment and switching times of the new energy power station, the low-voltage equivalent energy storage station and the high-voltage direct-connected energy storage is 12 times; the adjustment times of the inertia constant of the low-voltage equivalent energy storage and the high-voltage direct-connected energy storage are 16 times. The trapezoidal fuzzy parameter values ​​of wind, solar output and load are set as shown in Table 1. The capacity and inertia constant of the thermal power unit are shown in Table 2. The day is divided into 15-minute units, T=96. All related calculations were performed on a computer with an AMD Ryzen 9 7945HX processor 2.5GHz and 16GB of memory, using Python to call the Gurobi solver.

[0167] Table 1 Trapezoidal fuzzy parameters of wind, solar power output and load

[0168]

[0169] Table 2. Capacity and Inertia Constant of Thermal Power Units

[0170]

[0171]

[0172] To verify the feasibility and effectiveness of the proposed model, the following schemes were set up for comparison:

[0173] Option 1: Intraday two-layer rolling optimization scheduling considering adjustment and constraint.

[0174] Option 2: Intraday optimized scheduling without considering adjustment constraints.

[0175] Option 3: Intraday optimized scheduling without considering system inertia adjustment.

[0176] Option 4: Intraday optimized scheduling without considering energy storage frequency security support constraints.

[0177] To compare and analyze the effectiveness of the method proposed in this embodiment, two results are compared:

[0178] Comparison Result 1: Comparison of the differences in inertia of the intraday rolling optimization model without considering adjustment constraints and the model proposed in this invention in terms of the regulation system in a grid-type station.

[0179] Table 3 Switching frequency of grid-type renewable energy power stations

[0180]

[0181] Table 4 Switching frequency of grid-type low-voltage equivalent energy storage sites

[0182]

[0183] Table 5 Number of Switching Times for Grid-Type High-Voltage Direct-Connected Energy Storage

[0184]

[0185] Tables 3, 4, and 5 show that, under different confidence levels of 0.5 to 0.9, the extreme values, mean values, and dispersion indices of inertial parameter adjustment and network switching times for the three main types of energy storage facilities (new energy power stations, low-voltage equivalent energy storage, and high-voltage direct-connected energy storage) are compared and analyzed between Scheme 1 and Scheme 2. The results show that for new energy power plants, Scheme 1 has significantly lower average and maximum switching times at all confidence levels than Scheme 2, with a smaller overall data standard deviation and more balanced and stable operation among power plants. Only at high confidence levels does the difference between the two schemes narrow slightly. Scheme 2 is in a state of high-frequency switching for a long time, and the operating pressure under extreme conditions is greater. In low-voltage equivalent energy storage power plants, Scheme 1 has significantly better performance than Scheme 2 in terms of the number of inertial time constant adjustments and the average and maximum number of actions when switching between the grid and the system. Moreover, the frequency of regulation gradually decreases with the increase of confidence, the control strategy is more concise, and the data dispersion is lower under most operating conditions, effectively reducing the frequency of equipment actions. For high-voltage direct-connected energy storage, Scheme 1 can achieve dynamic system adaptation with reasonable and stable adjustment and switching frequencies, while Scheme 2 has obvious polarization problems. The K1 unit's regulation actions are too frequent and the risk of equipment loss is high, while the K2 unit completely loses its regulation flexibility and the operating mode is rigid. Based on the quantitative comparison results of the three scenarios, Scheme 1 can comprehensively reduce the frequency of parameter adjustment and switching of network-type equipment, suppress high-frequency operation behavior under extreme conditions, reduce equipment wear and operation fluctuations, and at the same time retain sufficient flexible control capabilities. The overall control stability and economy are significantly better than Scheme 2.

[0186] Comparison Result 2: Comparison of the differences in active power and inertia support plans between Scheme 1 and Scheme 4 when frequency security support constraints are considered.

[0187] Depend on Figure 4 and Figure 5 It can be seen that during the off-peak load period from 6:00 to 9:00, the number of thermal power units operating is relatively small, and the system's inertia support capacity is insufficient. At this time, low-pressure equivalent energy storage does not need to provide power support and has sufficient remaining capacity, thus providing high-inertia support within the safe power range. Entering the peak period of load and inertia demand from 10:00 to 12:00, although the number of thermal power units operating increases and the conventional inertia support capacity is enhanced, energy storage still needs to provide significant inertia support. Figure 5 and Figure 6 , Figure 5 and Figure 7It can be observed that when frequency safety support constraints are considered, the inertia support level of energy storage is reasonably controlled; however, energy storage without considering frequency safety support constraints maintains high inertia output even during deep discharge, which poses significant safety risks to equipment operation. During the peak load period at 14:00, on the one hand, thermal power units have provided sufficient inertia support; on the other hand, considering that energy storage will incur high operating costs due to long-term grid-connected operation, energy storage actively reduces its own inertia support level while fulfilling its power support task, achieving coordinated optimization of power support and inertia support during peak load periods. Overall, energy storage will prioritize providing inertia support during periods with sufficient capacity and large power margin; while during peak load periods when conventional inertia support capacity is sufficient or a large amount of electricity is needed for power support, it will actively control inertia output, achieving coordinated control under multiple objective requirements.

[0188] Comparison Result 3: Compare the differences between Scheme 1, Scheme 2 and Scheme 3 in terms of ensuring system inertia capability.

[0189] Depend on Figure 8 It can be seen that, without considering inertia adjustment, the overall system inertia of Scheme 3 is closer to the lower limit of inertia, and even lower than the lower limit of inertia for most periods, which poses a potential safety hazard. The overall system inertia levels of Scheme 1 and Scheme 2 are relatively close, and the fluctuations during load troughs and peak periods are small, always remaining in the upper-middle part of the safe range, demonstrating good continuity of inertia support.

[0190] In summary, Scheme 1 and Scheme 2 are more in line with the collaborative logic of "actively replenishing inertia during off-peak periods and prioritizing power supply during peak periods" in terms of inertia support strategy. Scheme 1, in particular, can effectively balance the power margin of energy storage and operating costs while meeting the system inertia requirements, avoiding excessive support that could lead to insufficient reserves or excessively high operating costs. It achieves coordinated optimization of system inertia safety and economical operation of energy storage under multiple objective constraints.

[0191] In this embodiment of the invention, after calculating the upper and lower limits of system inertia demand, the upper-level model considers the adjustment constraints of grid-type equipment, and formulates an inertia adjustment plan for grid-type equipment with the objective of minimizing the sum of system inertia deviation cost, energy storage grid operation cost, and grid-type equipment mode switching cost. The lower-level model uses the minimum system operation cost and adjustment plan deviation as the objective function to provide feedback correction to the upper-level model's adjustment plan. After constructing an intraday two-layer rolling optimization model, the Gurobi solver is called to solve the model. By comparing different schemes, it is verified that the scheduling scheme proposed in this invention can ensure that the system inertia is maintained within a safe range while avoiding frequent adjustments of grid-type equipment.

Claims

1. A daily rolling optimization scheduling method considering short-term inertia fluctuations, characterized in that, include: The lower limit of system inertia demand is calculated based on the system frequency change rate constraint and the frequency minimum point constraint. The power system is equivalent to a second-order oscillating system interconnected with the grid by synchronous generators. The upper limit of system inertia demand is calculated by using the damping ratio to characterize the small-signal dynamic characteristics of the system. Considering short-term fluctuations in inertia and constraints on inertia adjustment for network-type equipment, a daily two-layer rolling optimization scheduling model is constructed, including an upper-layer model and a lower-layer model. The upper-layer model is used to formulate the inertia adjustment plan for network-type equipment, and the lower-layer model is used to provide feedback correction for the active power and inertia adjustment plans. The Gurobi solver was used to solve the intraday two-level rolling optimization scheduling model.

2. The intraday rolling optimization scheduling method considering short-term inertia fluctuations according to claim 1, characterized in that, The lower limit of the system inertia requirement is the larger of the minimum inertia obtained from the constraint of the rate of change of frequency and the minimum inertia obtained from the constraint of the lowest point of frequency.

3. The intraday rolling optimization scheduling method considering short-term inertia fluctuations according to claim 2, characterized in that, The upper limit of system inertia demand is the smaller of the upper limit of system inertia constrained by damping ratio and the maximum value of system inertia supply.

4. The intraday rolling optimization scheduling method considering short-term inertia fluctuations according to claim 3, characterized in that, The upper-level model includes: Objective function: ； ； In the formula, This is the inertia margin; and These represent the system inertia and minimum system inertia requirement for time period t, respectively. , , These are the inertia deviation penalty coefficients, , , These are the initial new energy power station inertia adjustment cost coefficient, low-voltage equivalent energy storage inertia adjustment cost coefficient, and high-voltage direct-connected energy storage inertia adjustment cost coefficient, respectively. The marginal cost coefficient for adjusting the proportion of new energy power stations in the grid; , These are the marginal cost coefficients for switching between high-voltage direct-connected energy storage networks and adjusting the inertial time constant, respectively. , These are the marginal cost coefficients for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively. As an indicator variable for adjusting the proportion of new energy power stations in the power grid; , These are respectively the high-voltage direct-connected energy storage grid switching and the inertial time constant adjustment indicator variables; , These are the indicator variables for the proportional adjustment of low-pressure equivalent energy storage network and the adjustment of inertial time constant, respectively. Constraints: System inertia upper and lower limits constraints ； Grid-type high-voltage direct-connected energy storage regulation constraints ； ； In the formula, , , All are integer variables; , The following are the grid connection modes of energy storage k at times t+1 and t, respectively; , Let be the inertial time constants of the stored energy k at times t+1 and t, respectively; The maximum number of switching operations for energy storage k; The maximum number of adjustments to the energy storage inertia time constant k is given; T is the scheduling period, with a time scale of 15 minutes, dividing the day into 96 time periods; M is a relatively large constant introduced using the Big M method. Regulation Constraints of Grid-type Low-voltage Equivalent Energy Storage Stations ； ； In the formula, , These represent the proportions of energy storage m in grid-connected operation at time t+1 and time t, respectively. , are the inertial time constants of the energy storage m at time t+1 and time t, respectively; The maximum number of switching operations for energy storage m; The maximum number of adjustments to the inertial time constant of the energy storage unit m; Regulation and Constraints of Grid-Based New Energy Power Stations ； In the formula, , The proportion of the new energy power station w that is in grid-connected operation at time t+1 and time t; This represents the maximum number of switching operations for the new energy power station. Energy storage frequency security support constraints ； In the formula, The total frequency regulation power provided by energy storage after the disturbance in time period t; The inertial time constant of the energy storage device; This refers to the maximum charge / discharge rate of the energy storage device. and These refer to the charging efficiency and discharging efficiency of energy storage devices, respectively. For primary frequency modulation output; Rated power of energy storage; This is the initial frequency of the system; , These are the energy storage discharge power and the charging power, respectively. , These are the minimum energy storage capacity and the rated capacity, respectively. This is the frequency response time.

5. The intraday rolling optimization scheduling method considering short-term inertia fluctuations according to claim 4, characterized in that, The lower-level model includes: Objective function: ； ； ； In the formula: For system operating costs, To adjust for planning deviation costs; The cost of generating electricity from thermal power units; , , These are the penalty coefficients for wind curtailment, solar curtailment, and load shedding, respectively. , , These represent the power curtailment of wind and solar power, and the power shedding during time period t. , , These represent the daytime thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output for time period t, respectively. , , These represent the output of thermal power units, the output of high-voltage direct-connected energy storage, and the output of low-voltage equivalent energy storage during time period t. , , , , , The system inertia, the proportion of grid-type low-pressure equivalent energy storage stations and their inertial time constants in the upper-level model, the grid-type high-pressure direct-connected energy storage network operation mode and their inertial time constants, and the proportion of grid-type new energy stations are respectively determined. , , These are the deviation costs of thermal power unit output, high-voltage direct-connected energy storage output, and low-voltage equivalent energy storage output, respectively. , , , These are the system inertia, low-pressure equivalent energy storage station, high-pressure direct-connected energy storage, and new energy station readjustment deviation penalty coefficient, respectively. Provide power to the new energy power station at time t; Let be the system inertia at time t; The rated capacity of low-voltage equivalent energy storage m; Constraints: Node power balance constraints ； Rotational spare constraint ； In the formula: , These are the real and imaginary parts of the nodal admittance matrix, respectively. , Let be the voltage amplitude of nodes i and j at time t during the previous phase; Let be the phase angle difference between nodes i and j at time t in the previous phase; Confidence level; , and These are the membership parameters for the fuzzy parameters of wind, solar, and load, respectively, i=1,2,3,4; , These are the energy storage discharge power and the charging power, respectively. , These represent the thermal power unit number and set, respectively; n is the node number. Safety Operation Constraints of Thermal Power Units ； In the formula: , These are the upper and lower limits of the active power output of thermal power unit g; , These represent the maximum upward and downward ramp rates of thermal power unit g, respectively. , These represent the start-up and shutdown status of the thermal power unit at time t+1 and time t, respectively. , The output of the thermal power unit at time t+1 and time t are respectively; Energy storage operation constraints ； In the formula: , These are the charge of energy storage k at time t and the rated charge of energy storage k, respectively. , These represent the maximum charging and discharging power of energy storage k; , These are the charging and discharging indicators of energy storage k at time t, with 1 indicating that energy storage k is in a charging or discharging state at time t; Energy storage k at time t ; , They are respectively The upper and lower limits; New energy power constraints ； ； In the formula: , These represent the dispatch power of wind farm w and photovoltaic farm v at time t, respectively. , These are the predicted power outputs of wind power (w) and photovoltaic power (v) at time t, respectively. , These represent the amount of wind and solar power curtailment (w for wind power and v for solar power) at time t, respectively. System inertia upper and lower limits constraints ； Energy storage frequency security support constraints 。

Images