Power system scheduling method

Abstract

The invention discloses a power system dispatching method, and relates to the field of power system dispatching. According to the method, when the power grid is disturbed, if both the initial frequency change rate and the frequency lowest point meet the safety constraint conditions, the large-scale battery energy storage power station does not participate in adjustment on the premise that the power system does not cause the action of the low-frequency load shedding device, otherwise, the large-scale battery energy storage power station participates in adjustment; adjustment comprises fast frequency response adjustment, primary frequency modulation adjustment and secondary frequency modulation adjustment. The method is especially suitable for an LS-BESS scheduling scene which needs to consider both security and economy, and can provide a day-ahead scheduling decision with robustness and economy for a power grid operator.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of dispatching of power systems, in particular to a dispatching method of a power system comprising large-scale battery energy storage stations. BACKGROUND

[0002] With the transformation of global energy structure to low carbonization, large-scale grid-connection of high-proportion renewable energy represented by wind power and photovoltaic power has become an inevitable trend of power system development. However, renewable energy such as wind power and photovoltaic power has significant randomness characteristics, which leads to continuous reduction of inertia level of power system, and increasingly prominent problems such as insufficient frequency modulation and peak regulation capacity. The traditional power system relies on the rotational inertia and controllable output characteristics of synchronous generator units to maintain active power balance, but with the decline of the proportion of conventional units such as thermal power, the system frequency stability faces severe challenges. Especially in the scenario of high-proportion renewable energy penetration, the system needs to cope with multi-time scale active power regulation requirements including inertia response (IFR), primary frequency regulation (PFR), secondary frequency regulation (SFR) and peak regulation (PR) services. How to coordinate flexible resources to ensure power and energy balance has become a key problem restricting the consumption of renewable energy and the safe operation of the system.

[0003] Under this background, large-scale battery energy storage stations (LS-BESS) are regarded as an important technical means to improve the flexibility of power systems due to their rapid response capability, bidirectional power regulation characteristics and high energy density advantages. LS-BESS can provide both power and energy regulation services, releasing power at the moment of disturbance to support frequency stability, or discharging at the peak load period to relieve peak regulation pressure. However, LS-BESS also faces many challenges when participating in active power regulation services of the power system. The operation and dispatching strategies of LS-BESS and conventional units are generally different, and the operation strategies between the two need to be coordinated and dispatched. LS-BESS also needs to select different dispatching strategies according to different unit conditions when participating in various active power regulation services. In addition, LS-BESS also needs a reasonable dispatching decision method when participating in new energy generation considering the output uncertainty of wind power and photovoltaic power.

[0004] In summary, in order to solve the above problems, it is urgent to develop a day-ahead optimization dispatching method that can coordinate multiple categories of active power regulation services, coordinate the complementary operation of LS-BESS and conventional units, and effectively balance the global operation robustness and economy of the system. SUMMARY

[0005] In view of the above problems, the present application proposes a day-ahead dispatching optimization method for large-scale battery energy storage stations participating in multi-category active power regulation, which can assist grid operators and conventional units to participate in dispatching of day-ahead active power regulation services at various time scales.

[0006] The application provides a power system scheduling method, the power system comprising a large-scale battery energy storage station (LS-BESS), and the scheduling method comprises the following steps:

[0007] When a power grid disturbance occurs, under the premise that the power system does not cause a low-frequency load shedding device to act, an initial frequency change rate and a frequency minimum point If the safety constraint conditions are all met, the large-scale battery energy storage station does not participate in regulation, otherwise the large-scale battery energy storage station participates in regulation, and the regulation comprises fast frequency response regulation, primary frequency regulation and secondary frequency regulation.

[0008] Further, the safety constraint condition of the initial frequency change rate is as follows:

[0009] ;

[0010] wherein, is the maximum limit of the frequency change rate;

[0011] The safety constraint condition of the frequency minimum point is as follows:

[0012] ;

[0013] wherein, is a frequency critical value of UFLS.

[0014] Further, when the initial frequency change rate does not meet the safety constraint condition, at this time, the large-scale energy storage station participates in fast frequency response regulation, and the space reservation amount of the large-scale energy storage station is as follows:

[0015] ;

[0016] wherein, , are virtual droop control coefficients, and ; is the rated power of the LS-BESS, is the inertia time constant of the LS-BESS.

[0017] Further, when the frequency minimum point does not meet the safety constraint condition, at this time, the large-scale energy storage station participates in primary frequency regulation, and the space reservation amount of the large-scale energy storage station is as follows:

[0018] ;

[0019] wherein, a virtual droop control coefficient of the LS-BESS; a system frequency before the disturbance;

[0020] Further, when a large-scale energy storage power station needs to participate in secondary frequency regulation, the space reservation amount of the large-scale energy storage power station is: ;

[0021] wherein, is a bearing factor of the LS-BESS participating in the SFR in a period t, ; is a capacity demand amount.

[0022] Further, the scheduling method includes making the large-scale energy storage power station participate in peak clipping and valley filling regulation, the peak clipping and valley filling regulation including a peak clipping mode and a valley filling mode, the peak clipping mode: supporting the power grid with the maximum discharge power in a load peak period, replacing high-price unit output; the valley filling mode: absorbing excess power in a low-load high-wind-power period, reducing the wind curtailment rate and unit deep peak regulation (DPR) demand.

[0023] Further, the reserved space of the peak clipping and valley filling regulation is obtained by a minimum operation cost model in a unit time, and the minimum operation cost model in a unit time is:

[0024] ;

[0025] wherein, T represents a unit time, i is the i-th period, is an operation cost in a unit time, represents an operation cost of a conventional unit in the t-th period; represents an operation cost of the LS-BESS; is a wind curtailment cost.

[0026] Further, the operation cost of the conventional unit in the t-th period is:

[0027] ;

[0028] wherein, wherein N represents the number of conventional units in the system; , are respectively a start-up cost and a shutdown cost of the unit i; is a power generation cost of the unit i: represents a standby reservation cost of the unit i participating in the SFR; is a cost of the unit i participating in the DPR.

[0029] Further, the wind curtailment cost is:

[0030] ;

[0031] wherein, is the wind power value accepted by the power grid at the t period; is the wind power prediction value at the t period; is the half interval length, representing the maximum deviation of the actual value from the prediction value, is the perturbation of the uncertain factor at the t period.

[0032] Compared with the prior art, the present application has the following beneficial effects:

[0033] The present application installs the large-scale battery energy storage power station in the power system containing high proportion of renewable energy such as wind power and photovoltaic, and jointly guarantees the safe, high-quality and economic operation of the power system with the conventional unit. The LS-BESS is considered in the power grid dispatching, the participation adjustment strategy of the LS-BESS is adjusted based on the safety of the power grid, when the system has a sudden power disturbance, the LS-BESS can provide virtual inertia through FFR to quickly discharge and make up for the problem of the inertia reduction of the synchronous unit; the LS-BESS participates in the primary frequency regulation (PFR) adjustment, the LS-BESS adjusts the active power in a short time through the virtual droop control mode to assist the conventional unit to improve the frequency stability of the system; in the secondary frequency regulation (SFR) adjustment process, the LS-BESS needs to adjust the wind power and the load fluctuation in the dispatching period, and the size of the demand can be used to adjust the bearing factor between the LS-BESS and the conventional unit to determine the standby amount of the SFR of the LS-BESS.

[0034] In order to make the power system meet the load demand, resist any power disturbance in the expected fault set, and have the ability to cope with the wind power and load fluctuation, a day-ahead dispatching optimization model is constructed, which takes the minimization of the operation cost of the dispatching day as the target, can give full play to the adjustment advantages of the conventional unit and the large-scale battery energy storage power station to participate in the active power adjustment service of different time scales, and can effectively realize the coordination between the service demands. For the uncertainty of the actual output of the wind power in the dispatching day, the robust optimization method based on the cardinality constraint uncertain set is adopted, which can better consider the robustness and economy of the system operation, so that the dispatching decision has better applicability. The effectiveness of the proposed day-ahead dispatching strategy is verified through case simulation and scheme comparison analysis, and the proposed optimization dispatching framework can better serve the day-ahead dispatching decision of the high proportion of wind power system. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical scheme of the dispatching operation of the present application, the used drawings will be simply introduced as follows.

[0036] Figure 1 A power system structure schematic diagram provided by the present application;

[0037] Figure 2 A LS-BESS participation FFR and PFR determination condition schematic diagram provided by the present application;

[0038] Figure 3 A general optimization model schematic diagram provided by the present application;

[0039] Figure 4 A Bus-RO method parameter relationship schematic diagram provided by the present application;

[0040] Figure 5 A LS-BESS participation system all-day operation flowchart provided by the present application;

[0041] Figure 6 A case system bus configuration diagram provided by the present application;

[0042] Figure 7 A LS-BESS participation system day-ahead scheduling actual case operation result provided by the present application, Figure 7 a is a LS-BESS space reservation curve, Figure 7 b is a LS-BESS SoC change curve, Figure 7 c is a space reservation curve required for a conventional unit to track and schedule daily net load changes, Figure 7 d is an upper standby reservation curve of a conventional unit participating in SFR, Figure 7 e is a lower standby reservation curve of a conventional unit participating in SFR. DETAILED DESCRIPTION

[0043] The present application will be further described below in combination with the drawings and specific embodiments, but not as a limitation of the present application.

[0044] In a specific embodiment, a power system scheduling method is provided, the power system comprising a large-scale battery energy storage power station LS-BESS (Large Scale-Battery Energy Storage System), a high proportion of wind power, photovoltaic and other new energy power generation devices and a conventional thermal power unit, specifically as Figure 1As shown, the wind turbine with new energy generation uncertainty and the core element of the optimal scheduling of the application can realize the LS-BESS energy storage that exchanges electric energy with the power transmission network in both directions, and provide sufficient and high-quality electric energy to the power load user through the scheduling of the source and the grid side. In the working process, the responsibility allocation strategy of the two types of regulating elements participating in the system multi-category active power regulation service is mainly based on conventional unit regulation and supplemented by LS-BESS complementary output, the large-scale battery energy storage power station mainly participates in inertia response (IFR) regulation, primary frequency regulation (PFR) regulation, secondary frequency regulation (SFR) regulation and peak regulation (PR) regulation, and reserves standby space for responding to step disturbance in the most serious scenario in the expected fault set, which can make the LS-BESS guarantee the frequency safety of the power grid when the IFR and PFR capacity of the conventional unit is insufficient, so as to guarantee the safety, quality and economy of the active operation of the power grid, and the improved robust optimization method is used to control the wind power output uncertainty in the operation of the power system. According to the "Guidelines for the Safety and Stability of Power Systems" issued by the State Energy Administration of China, the power system should be able to maintain stable operation and normal power supply after any element is disconnected under normal operating mode. Therefore, the application reserves standby for the most serious fault in the expected fault scenario set, so that the system will not appear under frequency load shedding (UFLS) after any disturbance occurs.

[0045] The LS-BESS will determine which active power regulation service needs to participate according to the predicted wind turbine output and load demand in each period. And the reserved standby capacity is determined by coordinating the frequency safety constraint of the safe operation of the power grid and the operation cost of the system.

[0046] Specifically, the scheduling method comprises:

[0047] A unit time is set, and the unit time is divided into a plurality of same time periods, in the embodiment, the unit time is 24h, and 24h of a day is divided into 96 time periods, and the length of each time period is 15min.

[0048] The large-scale battery energy storage power station (LS-BESS) as an independent energy storage resource, cooperates with the conventional unit to participate in the day-ahead scheduling of the multi-category active power regulation service of the power system containing high proportion of wind power and other renewable energy, especially suitable for the LS-BESS scheduling scene that needs to consider safety and economy, and can provide the day-ahead scheduling decision with robustness and economy for the power grid operator.

[0049] Unlike conventional thermal power units whose participation in IFR and PFR is mandatory, LS-BESS participation in FFR and PFR is a paid service, for which the system must pay a fee. Therefore, grid operators should only call upon LS-BESS when the regulation capacity of conventional units is insufficient, thus determining the appropriate timing for LS-BESS to participate in FFR and PFR regulation.

[0050] When a disturbance occurs in the power grid, assuming the power system does not trigger the low-frequency load shedding device, the initial frequency change rate is... With the lowest frequency point If all safety constraints are met, large-scale battery energy storage power stations will not participate in regulation. Otherwise, it indicates that the regulation capacity of conventional thermal power units is insufficient to guarantee the safety of the current power system. In this case, large-scale battery energy storage power stations need to participate in regulation, which includes fast frequency response regulation, primary frequency regulation, and secondary frequency regulation.

[0051] like Figure 2 As shown, the initial frequency change rate is used to determine whether LS-BESS participates in FFR. The safety constraints are as follows:

[0052] ;

[0053] in, This is the maximum limit for the rate of change of frequency;

[0054] The unit's IFR response typically lasts 0-5 seconds, and the unit's PFR typically lasts 5-60 seconds. Because the LS-BESS is connected to the grid via power electronic devices, it cannot participate in the IFR and PFR operations with synchronized frequency. Therefore, when a sudden power disturbance occurs in the system, the initial frequency change rate... When the maximum value is exceeded, the LS-BESS provides virtual inertia through rapid FFR discharge to compensate for the inertia reduction of the synchronous unit. At this time, the LS-BESS participates in rapid frequency response (FFR) and virtual droop control through virtual inertia response to achieve output regulation.

[0055] The virtual inertia response process adopts the existing virtual inertia response. The principle of virtual inertia response in this embodiment is as follows: when the system is disturbed, the inertia of the synchronous generator rotor itself can resist the change of the initial frequency. If the rate of change of the initial frequency is too large, it will cause the system frequency to drop rapidly and lead to frequency collapse. The LS-BESS, as an energy storage element, is connected to the grid through power electronic components. It does not have inertia itself, but it can provide the power required to resist the change of frequency through rapid discharge of FFR, which is virtual inertia.

[0056] Ignoring the damping effect of the load, the expression for the system RoCoF is:

[0057] ;

[0058] wherein, is the frequency difference; represents the disturbance amount; , are the increased generation of the unit and the LS-BESS in the t period, respectively; is the system frequency before the disturbance; represents the inertia time constant of the system; is the system reference power, which is set as the sum of the installed capacity of the elements in the system that can participate in the inertia and PFR regulation; according to the expression of the system RoCoF, when the disturbance occurs in the t period, the RoCoF value is:

[0059] ;

[0060] In the day-ahead scheduling of the system, whether the value under the condition that the LS-BESS does not participate in the FFR meets the safety margin can be judged according to the most serious fault, and if it meets, the inertia time constant of the LS-BESS is set to 0, that is, the LS-BESS does not participate in the FFR, otherwise, a minimum value constraint of is obtained, that is:

[0061] ;

[0062] Through the coordination between the LS-BESS and the unit, the optimal value of can be determined after optimization, and the space reservation amount of the LS-BESS participating in the FFR is:

[0063] ;

[0064] wherein is the virtual inertia response coefficient of the LS-BESS, with the unit of MWs / Hz, and , is the rated power of the LS-BESS.

[0065] As shown in the judgment logic of Figure 2 , when the lowest point of the system frequency exceeds the frequency threshold value triggering the UFLS, the LS-BESS participates in the PFR regulation through the virtual droop control, and the safety constraint condition of the lowest point of the system frequency is:

[0066] ;

[0067] wherein, is the frequency threshold value triggering the UFLS.

[0068] ​​When the frequency minimum point does not meet the safety constraint condition, at this time, the large-scale energy storage power station participates in primary frequency regulation, and the frequency minimum point after the disturbance occurs in the tth time period is:

[0069] ;

[0070] wherein, is the sum of the gain coefficients of the units; represents the virtual droop control coefficient of the LS-BESS. If the PFR capability of the conventional unit is sufficient after the system is disturbed, the frequency minimum point will not trigger the UFLS action, and the LS-BESS does not need to reserve PFR backup, at this time, the virtual droop control coefficient is 0; when the LS-BESS needs to participate in PFR, the virtual droop control coefficient satisfies: , the space reserved for the large-scale energy storage power station participating in PFR regulation is:

[0071] ;

[0072] wherein, is the virtual droop control coefficient of the LS-BESS; is the system frequency before the disturbance; is the frequency threshold value at which UFLS occurs, is the maximum linear operating frequency of the unit.

[0073] The LS-BESS participates in the secondary frequency regulation (SFR) adjustment process in coordination with the conventional unit, and the dispatch center sends instructions to the adjustment element according to the size of the demand . Since the duration of SFR regulation is relatively long, in the secondary frequency regulation process of the embodiment, an affine allocation strategy is adopted between the conventional thermal power unit and the LS-BESS, and the reserved SFR backup amount is determined by optimizing the bearing factor. Taking the reserved SFR backup as an example, the reserved SFR backup amount of the LS-BESS should be :

[0074] ;

[0075] wherein, M is the number of SFR units in the system; is the bearing factor of the LS-BESS participating in SFR regulation in the tth period, , the grid operator can adjust to set the capacity allocation of the LS-BESS and the conventional unit participating in SFR; is the capacity demand, which is provided by the dispatch center.

[0076] The optimal coordination of FFR adjustment, PFR adjustment and SFR adjustment in the embodiment does not consider the sequence in time, and the corresponding adjustment can be performed as long as the triggering condition is met (reserve day-ahead standby), and each adjustment mode can be performed simultaneously.

[0077] The scenario of high wind power and low load brings challenges to system dispatch decision-making. Generally, the anti-peaking characteristics of wind power will make conventional units enter deep peaking (DPR) state, and even need to be fired with oil, which seriously increases the system operation cost. In addition to participating in primary and secondary frequency modulation with conventional thermal power units, due to the flexibility of bidirectional power exchange between LS-BESS and the grid, LS-BESS can be used as an independent energy storage resource to meet the load demand of each period together with conventional units in the system, on the basis of realizing active power supply and demand balance, to improve the wind power consumption capacity as much as possible, and help to realize the peak clipping and valley filling of system power load. The dispatch method described in the embodiment also includes participating in peak clipping and valley filling adjustment by a large-scale energy storage power station, the peak clipping and valley filling adjustment includes a peak clipping mode and a valley filling mode, the peak clipping mode: supporting the grid with maximum discharge power during the load peak period, replacing the output of high-priced units; the valley filling mode: absorbing excess power during the low load and high wind power period, reducing the wind curtailment rate and the DPR demand of units. According to the real-time ratio of DPR cost and wind curtailment cost of unit i in t period , the maximum value of wind power generation in t period , and the actual value of wind power generation in t period , when the condition is met, the LS-BESS charging and discharging is started to replace the DPR state of the unit.

[0078] The LS-BESS can improve the down-peak capacity of the system through bidirectional adjustment characteristics. The system PR usually involves the enthusiasm of system operation, and its cost mainly comes from the generation cost of conventional units, DPR cost and operation cost of LS-BESS. For the grid operator, a trade-off should be made between reserving PR space and wind curtailment. The more space reserved, the less wind curtailment or the complete consumption of wind power, but the corresponding standby cost is higher. Reserving less space may result in higher wind curtailment cost. In the embodiment, an optimal dispatching model of system daily operation cost is constructed to determine the time and standby capacity of LS-BESS participating in PR, to ensure the daily optimal dispatching operation under the condition of safe and reliable system operation.

[0079] As Figure 3 ​As shown, the mathematical models involved in power system operation not only need to satisfy the constraints of each component of various generating units, such as output limits and reserve constraints, but also the overall power balance constraints of the system and the uncertainty of the actual output of wind turbines. A reasonable mathematical model must be constructed to ensure the solvability of the mathematical model. Day-ahead dispatch makes decisions on the reserved space for each component to participate in active power regulation at each time period. Although LS-BESS has the characteristics of both power-type and energy-type energy storage and can participate in various active power regulation services, its controllable space is limited. Therefore, optimal coordination should be made regarding its reserve space for participating in various types of active power regulation services. The goal of the day-ahead dispatch strategy is to optimize the economic efficiency of system operation, that is, to optimize the daily operating cost of the system. Minimize, the minimum operating cost per unit time The model is:

[0080] ;

[0081] Where T represents a unit of time, and t is the t-th time interval. This represents the operating cost of a conventional generating unit in the t-th time period; This indicates the operating cost of LS-BESS; Cost of wind curtailment.

[0082] The operating cost of conventional units for:

[0083] ;

[0084] Where N represents the number of conventional units in the system; , These are the start-up and shutdown costs for unit i, respectively; The power generation cost of unit i: This indicates the reserve cost for unit i participating in SFR; To account for the cost of unit i participating in DPR, DPR is divided into two tiers. When unit i needs to participate in DPR to reduce its output below the critical value of the second tier DPR, the unit will enter the second tier DPR. The adjustment unit price for each tier is as follows: and .

[0085] The decision variables involved in LS-BESS include its charging and discharging power under normal system operation. , And the upper and lower reserves involved in FFR, PFR and SFR , , and The operating cost of the LS-BESS for:

[0086] ;

[0087] wherein, ; ; ; ; , , are the unit price of LS-BESS participating in FFR, PFR and SFR, respectively; is the unit operation cost of LS-BESS.

[0088] The wind curtailment cost in the embodiment is used to represent the wind turbine consumption capacity, and the wind curtailment cost is:

[0089] ;

[0090] wherein, is the unit wind curtailment cost; is the wind power value accepted by the power grid at the t period; is the actual wind power at the t period, which belongs to an uncertain factor.

[0091] The day-ahead scheduling optimization model constructed from the global perspective of system operation considers the constraints related to the system and elements at two levels. At the system level, the active power of the system under normal operation should satisfy the supply-demand balance, that is, the sum of the output of the conventional unit, the charging and discharging power of the LS-BESS and the output of the wind turbine is equal to the system load, so that the power balance constraint is obtained as:

[0092] ;

[0093] wherein, is the system load at the t period; is the output of the conventional unit i at the t period; , is the charging and discharging power of the LS-BESS at the t period, and in the embodiment, it is defined as a positive value when charging and a negative value when discharging; is the output of the wind turbine at each period.

[0094] The total spinning reserve reservation of the system should also satisfy the minimum spinning reserve requirement, that is:

[0095] ;

[0096] wherein, wherein is the maximum technical output value of the unit i; is the minimum spinning reserve coefficient required by the system, which is taken as 8%.

[0097] At the element level, the operation constraints of the conventional unit and LS-BESS are mainly considered, such as the output limit, reserve quantity constraint, SOC state constraint and the like. The power is within the allowable constraint range, that is:

[0098] and

[0099] wherein, is a Boolean variable representing the charging and discharging state, and is 1 in the discharging state.

[0100] In order to prevent overcharging and overdischarging of the LS-BESS, the SoC value of the LS-BESS should be controlled within a certain upper and lower limit range, and the SoC state constraint is satisfied:

[0101] Due to the characteristics of the energy storage system, there is a certain loss in input and output power, that is, there is a LS-BESS charging and discharging efficiency problem, and the rated power of the LS-BESS , the minimum value of the SoC of the LS-BESS , the maximum value of the SoC of the LS-BESS , the charging and discharging loss of the LS-BESS satisfy , wherein is the power of the LS-BESS at the t period, , are the discharging efficiency and charging efficiency of the LS-BESS, respectively.

[0102] In order to make the day-ahead scheduling decision have good robustness to the actual situation within the day, considering the uncertainty of the actual output of the wind power on the scheduling day, in the embodiment, the offset control mode of the uncertain factors is adopted, the perturbation norm 1 constraint is increased under the condition of satisfying the uncertain amount box set, and then the radix constraint uncertain set of the wind power output is constructed, which is:

[0103]

[0104] wherein, is the wind power prediction value at the t period; indicates the prediction deviation, and the uncertainty reason is due to , so it will be mainly processed; , , are the upper and lower limit values of the wind power output at the t period, respectively; is the half interval length, representing the maximum offset of the actual value from the predicted value. In the embodiment, the perturbation amount of the uncertain factor at the t period is introduced, which satisfies:

[0105] ​​​ ;

[0106] wherein, is a budget value, T is a unit time, and the grid operator can set the budget value The base constraint is set to limit the total amount of uncertainty factor fluctuation, and the offset is calculated by the following formula:

[0107] .

[0108] The RO method is to obtain the optimal strategy of the uncertainty factor under the worst scenario, which corresponds to the application scenario of the present application, that is, to formulate a day-ahead scheduling decision that minimizes the total operation cost when considering the maximum wind power abandonment cost. The wind power abandonment cost of the system under the RO method is:

[0109] ;

[0110] wherein, unit wind power abandonment cost; is the value of wind power accepted by the grid at time t; is the wind power prediction value at time t; is the half-interval length, representing the maximum offset of the actual value from the predicted value; is the time interval.

[0111] In this embodiment, a base constraint robust optimization (BUS-RO) method is provided, which can limit the perturbation amount of the uncertainty amount, allowing the decision maker to limit the total amount of wind power fluctuation according to the scheduling requirements or their own risk preference. The model construction method for processing wind power new energy uncertainty factors by BUS-RO is used to improve the problem of excessive conservatism of standard RO.

[0112] After the BUS-RO method is used to process the wind power uncertainty factors in actual operation, the objective function of the day-ahead scheduling optimization model becomes a double nested problem like "min-sup". The decision variable of the inner optimization problem is the uncertainty variable of the outer problem, and according to the duality theory, the inner optimization problem is converted into the dual problem of the original sup function to reduce the complexity of the objective function, that is:

[0113] ;

[0114] wherein, , , is a variable generated when solving the dual problem.

[0115] The solving dual method can adopt the solving method in the prior art, and the principle of the solving dual method in the embodiment is as follows: after the BUS-RO method is used to process the uncertain factors of wind power in actual operation, the objective function of the day-ahead scheduling optimization model is converted from a "min-sup" function into a double-layer nested problem in the form of "min-sup". The memory optimization problem can be converted into a dual problem of the original sup function by using the dual theory in the prior art, so as to reduce the complexity of the objective function.

[0116] In the double-layer nested structure, the decision variable of the inner optimization problem is the uncertainty variable of the outer problem, and the dual theory is used to convert the inner optimization problem into a dual problem to obtain the objective function of the final converted day-ahead scheduling model.

[0117] In order to further illustrate the application, in a specific embodiment, the effectiveness of the proposed day-ahead scheduling scheme is verified by data analysis taking an IEEE 10-machine 39-bus system as an example. The following modifications are made to the case system: replace units U5 and U6 with wind power with an installed capacity of 950 MW, and introduce an LS-BESS with a rated parameter of 200 MW / 800 MWh. Units U2, U4, U7 and U8 are SFR units. The system structure diagram is as shown in Figure 6 .

[0118] It is assumed that = 0.5 Hz / s and = 49.5 Hz, the most serious power shortage that can occur in each period of the scheduling day is taken as 10% of the system reference capacity. The = 0.033 Hz, = 0.2 Hz, the generation of the unit enters the first and second DPRs at 50% and 40% of the maximum power, = 60 $ / MWh and = 150 $ / MWh. For the LS-BESS, the charging and discharging efficiencies and are both 95%, the maximum and minimum SoC safety thresholds and are 90% and 10% respectively, and the initial SoC value of the scheduling day is 50%, = 30 $ / MWh, and are 60 $ / MWh. The system SFR demand in each period is taken as 5% of the load and 10% of the wind power prediction value, and the unit price of the regulating element participating in the SFR is 9 $ / MWh. The unit price of the regulating element participating in the SFR is taken as 80 $ / MWh. The in the BUS-RO method is 32. The unit U2 in the system at the initial period of the scheduling day is started.

[0119] The optimal solution was obtained by the model on an AMD Ryzen3 3200G with Radeon Vega Graphics 3.60 GHz processor in 4500 s. The integrated operation cost of the scheduling day was $621901.80. The day-ahead spatial reservation strategy of LS-BESS and conventional units participating in multi-level active regulation services is shown in Figure 7 .

[0120] As can be seen from Figure 7 , to ensure system frequency security, LS-BESS reserves FFR and PFR spaces in the 0~31 period to resist possible step disturbances, which is due to the insufficient response capability of conventional units in this period. If LS-BESS does not reserve space to resist step disturbances in these periods, when the most serious disturbance occurs, the 0~15 period will drop to 49.03 Hz at 0.96 Hz / s, the 16~19 period will be 0.89 Hz / s and 49.11 Hz, and the 20~31 period will be 0.64 Hz / s and 49.34 Hz. The above cases will inevitably cause the UFLS accident of the system. When the 32nd period starts, U1 is running, which increases the IFR and PFR regulation capability of conventional units. At this time, LS-BESS can no longer reserve space to ensure frequency security, and can improve the economy of operation under the condition of meeting system safety.

[0121] As can be seen from the day-ahead scheduling strategy shown in Figure 7 , because LS-BESS can participate in SFR, the units bear less SFR tasks. Especially for U8, it is mainly regulated for participation in PR, not SFR. In the fast start-stop units, U8 is preferentially arranged to participate in system operation than U9 and U10 because its marginal generation cost is lower. The cost required by LS-BESS to reserve SFR space accounts for 32.33% of its total operation cost. The power system considering the direct regulation mode of LS-BESS does not need to worry that the units will enter the DPR state because of reserving SFR upper standby, and also will not cause the remaining units to enter the DPR because the units need to reserve SFR lower standby to reduce output. Therefore, LS-BESS can improve the economy while ensuring the reliable operation of the system.

[0122] The LS-BESS can be discharged during the load peak period to play a role in peak shaving, such as the periods of 44-46 and 74-82; and can support supply-demand balance by discharging when wind power is relatively low and the total output of units is relatively small, such as the periods of 0-1, 7-8 and 61-62. At the beginning of the 63rd period, the grid operator arranges U4 to participate in PR, and the LS-BESS exits in response. This is because the marginal generation cost of U4 is about 16.82 $ / MWh, which is significantly lower than the unit operating cost of the LS-BESS. During the periods of 89-95, the load is small and the wind power is high, and the LS-BESS is charged to increase the system's peak shaving capability and promote wind power consumption. At the same time, the charging of the LS-BESS can prevent conventional units from entering the DPR state. Under the dispatching strategy proposed in the application, the DPR cost of the unit is 0, and after the LS-BESS is put into operation, the installed capacity of the conventional unit in the system can be reduced, and the optimal economic operation can be realized under the condition of ensuring the safe and high-quality operation of the grid.

[0123] In summary, in the dispatching day execution stage, the SFR reserve calling priority is dynamically adjusted based on the actual wind power output and load deviation: if the actual fluctuation is lower than the robust optimization reserved amount, the unit reserve resource is called according to the cost priority principle; if the actual fluctuation exceeds the robust optimization reserved amount, the SFR reserve of the LS-BESS is preferentially called to quickly suppress the frequency deviation; when the SoC of the LS-BESS is lower than the safety threshold, the reserve capacity redistribution strategy is started, and the unit reserve resource is forced to switch to ensure the safety of the energy storage. Under the premise of ensuring the safety of the grid, the LS-BESS can be charged at low load and discharged at high load to smooth the load curve, improve the new energy consumption capacity on the basis of realizing the balance of active power supply and demand, coordinate the PR regulation cost and the wind curtailment cost, help some conventional units retire, and reduce the demand for deep peak shaving (DPR) of units. In terms of handling the system frequency deviation caused by new energy fluctuation, the LS-BESS can adjust the power output through automatic generation control (AGC) to enhance the adaptability of the system to wind power fluctuation. The overall power transmission system under the participation of the LS-BESS needs to meet the load demand of the power user while realizing effective active power regulation. The BUS-RO method is used to handle the actual operation of wind power uncertainty, so that the LS-BESS can also help the system to optimize the configuration and improve the robustness and economy of the overall operation of the system under the participation of new energy. Under the conditions of meeting the load demand and the stability of the grid operation, the LS-BESS and the conventional unit are coordinately and optimally dispatched to minimize the system daily operation cost, and the reserve reserved amount of the LS-BESS and the conventional unit under different situations is determined.

[0124] The preferred embodiments of the present application have been described. It is to be understood that the application is not limited to the particular embodiments described, in which details of the equipment and construction have not been fully described, and that devices and structures not fully described should be construed as being implemented in the ordinary way in the art; any person skilled in the art, without departing from the scope of the technical solutions of the present application, can make many possible changes and modifications to the technical methods of the present application disclosed above, or modify them into equivalent methods of equivalent changes, which do not affect the essential content of the present application. Therefore, any simple modification, equivalent change and modification made to the above method according to the technical essence of the present application, without departing from the content of the technical method of the present application, still belongs to the protection scope of the technical method of the present application.

Claims

1. A power system dispatching method, characterized in that, The power system includes large-scale battery energy storage power stations, and the dispatching method includes: When a disturbance occurs in the power grid, assuming the power system does not trigger the low-frequency load shedding device, the initial frequency change rate is... With the lowest frequency point If all safety constraints are met, the large-scale battery energy storage power station will not participate in regulation; otherwise, the large-scale battery energy storage power station will participate in regulation, which includes fast frequency response regulation, primary frequency regulation, and secondary frequency regulation.

2. The power system dispatching method according to claim 1, characterized in that, The initial frequency change rate The safety constraints are as follows: ； in, This is the maximum limit for the rate of change of frequency; The lowest frequency point The safety constraints are as follows: ； in, This is the critical frequency at which UFLS occurs.

3. The power system dispatching method according to claim 1, characterized in that, When the initial frequency change rate When safety constraints are not met, large-scale energy storage power stations participate in rapid frequency response regulation. The space reserved for these large-scale energy storage power stations... for: ； in, The virtual droop control coefficient is... ; The rated power of LS-BESS Let be the inertial time constant of LS-BESS.

4. The power system dispatching method according to claim 1, characterized in that, When the lowest frequency point does not meet the safety constraints, the large-scale energy storage power station participates in primary frequency regulation. The space reservation for the large-scale energy storage power station... for: ； in, For LS-BESS, the virtual droop control coefficient is used. The system frequency before the disturbance.

5. The power system dispatching method according to claim 1, characterized in that, When large-scale energy storage power stations are needed to participate in secondary frequency regulation, the space reserved for such large-scale energy storage power stations for: ； in, The contribution factor of LS-BESS to SFR during time period t. ; This refers to the capacity requirement.

6. The power system dispatching method according to claim 1, characterized in that, The scheduling method includes enabling large-scale energy storage power stations to participate in peak shaving and valley filling regulation. The peak shaving and valley filling regulation includes peak shaving mode and valley filling mode. Peak shaving mode: during peak load periods, the power grid is supported by the maximum discharge power to replace the output of high-priced generator units. Valley filling mode: during low load and high wind power periods, excess power is absorbed to reduce wind curtailment rate and the deep peak shaving demand of generator units.

7. The power system dispatching method according to claim 6, characterized in that, The reserved space for peak shaving and valley filling regulation is obtained through a minimum operating cost model per unit time, which is as follows: ； Where T represents a unit of time, and t is the t-th time interval. Operating cost per unit time This represents the operating cost of a conventional generating unit in the t-th time period; This indicates the operating cost of LS-BESS; Cost of wind curtailment.

8. The power system dispatching method according to claim 7, characterized in that, The operating cost of the conventional unit in the t-th time period is: ； Where N represents the number of conventional units in the system; , These are the start-up and shutdown costs for unit i, respectively; The power generation cost of unit i: This indicates the reserve cost for unit i participating in SFR; Cost of unit i participating in DPR.

9. The power system dispatching method according to claim 7, characterized in that, The cost of wind curtailment for: ； in, This represents the amount of wind power received by the power grid during time period t. The wind power forecast value for time period t; The half-interval length represents the maximum offset between the actual value and the predicted value. This represents the perturbation of uncertain factors during time period t.

Images