A power system resource scheduling method for flexible ramping services

By establishing a flexible climbing demand estimation model and two-stage mixed integer linear planning optimization scheduling, the problem of unstable flexibility balance of power system under high proportion of new energy access is solved, reducing operating costs and improving the economic and flexibility of the system sex.

CN117293795BActive Publication Date: 2024-11-26HANGZHOU GEHUDA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311156393.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-08
Publication Date
2024-11-26
Estimated Expiration
2043-09-08

AI Technical Summary

Technical Problem

High proportion of new energy access leads to unstable balance of flexibility in the power system. The existing technology has problems of low or high in the estimation of flexible climbing demand, resulting in excessive system operation costs and insufficient climbing capacity.

Method used

Using a flexible climbing demand estimation model based on system net load and safety margin, combined with the net load fluctuations that have been met in energy scheduling, an adjustable resource optimization scheduling model with two-stage mixed integer linear planning is established, with the goal of minimizing operating costs. , optimize the scheduling plan, including predicting the system net load, establishing a flexible climbing demand model, building constraints, and generating uncertain net load scenarios for rescheduling.

Benefits of technology

Effectively reduce the overestimation of flexible climbing demand, reduce the system's operating costs recently, improve the economic and flexibility of the system, reduce the system's cutting load and wind and light abandonment through integrated flexible load and energy storage resources, significantly reducing the system's load cut and wind and light abandonment, Running costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117293795B_ABST
    Figure CN117293795B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of dispatching and operating a high-proportion new energy power system, and provides a power system resource dispatching method for flexible ramping services. The method takes into account the flexible ramping demand of the system, establishes an adjustable resource optimization dispatching model for typical adjustable resources such as BESS, PHES, ILDR, and SLDR, specifically analyzes the influence of the adjustable resources participating in flexible ramping on the system operating cost, performs optimization solution to obtain a dispatching plan, and further provides a basis for reasonably calling the adjustable resources. Meanwhile, a flexible ramping demand estimation method based on the net load uncertainty of the start time and the end time is proposed, which effectively reduces the overestimation of the flexible ramping demand of the system and reduces the day-ahead operating cost of the system. Compared with traditional units, the method can effectively reduce the system load shedding and wind and solar power abandonment by integrating multiple adjustable resources such as flexible loads and energy storage, significantly reduces the system operating cost, and further improves the economy and flexibility of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of dispatching and operating a high-proportion new energy power system, and in particular to a power system resource dispatching method for flexible ramping services. Background Art

[0002] The access of a high proportion of renewable energy has brought challenges to the safe and stable operation of the power system, and flexibility has become the core and key to the system's operating characteristics. The flexibility of the power system refers to the ability of the system to respond to changes in supply and demand and uncertainties on all relevant time scales by calling on adjustable resources under the premise of meeting certain economic and reliability requirements. In order to achieve a balance in the flexibility of the power system, it is very important to study the issue of calling on adjustable resources.

[0003] Flexibility balance requires that the flexibility supply of various resources in the system exceeds the allowable level relative to the flexible ramping demand at any time, on any time scale, and in any direction. The flexible ramping demand comes from the power changes in the system caused by the volatility and uncertainty of loads and renewable energy. At present, the adjustable resources of the system generally use conventional generators (CG). However, frequent calls to CG units will increase their wear and tear, and usually the ramping cost is added to the unit combination problem to compensate for the technical and economic losses caused by the frequent calls to CG units.

[0004] However, with the increase in the proportion of new energy access, the use of CG units alone is not enough to provide sufficient flexibility. Therefore, existing research incorporates regulatory measures that can cope with volatility and uncertainty in systems such as energy storage, demand-side resources, heat networks and heat loads, and electric vehicle clusters into adjustable resources. However, due to the lack of economic signal guidance, incorporating only adjustable resources into the system cannot ensure that they provide sufficient ramping capacity. Insufficient ramping capacity will lead to load shedding or reduction in renewable energy output.

[0005] The independent system operator introduces flexible ramping product (FRP) in the power market to ensure that the system can adjust the resource output to meet the changing demand of the load in the next period while considering the forecast error and the fluctuation of renewable energy output, thus ensuring the real-time balance of the system. As a multi-interval problem, the net load value before the dispatch interval is uncertain in FRP in day-ahead dispatch.

[0006] That is, existing studies do not consider the uncertainty of the net load at the start time of scheduling when estimating the flexible ramping demand, resulting in an underestimated estimate of the flexible ramping demand; or improve the flexible ramping demand estimation method while considering the uncertainty of the net load at the start and end of the scheduling period, but do not consider the fluctuation of the net load that has been met in energy scheduling, resulting in an overestimation of the flexible ramping demand and a high system operating cost.

[0007] Moreover, existing research on FRP mainly calls on a single adjustable resource, or is limited to the day-ahead market or the real-time market, without considering that FRP in the real-time market may not be able to meet the system's flexible ramping requirements due to insufficient ramping capacity of the adjustable resource. Summary of the invention

[0008] The present invention provides a power system resource scheduling method for flexible ramping services, which solves the technical problem that the existing high-proportion new energy power system has inaccurate estimation of flexible ramping demand, resulting in unstable system flexibility balance.

[0009] In order to solve the above technical problems, the present invention provides a power system resource scheduling method for flexible ramping services, comprising:

[0010] S1. Based on the system net load and its safety margin, combined with the net load fluctuations that have been satisfied in energy scheduling, a flexible ramp demand estimation model corresponding to day-ahead scheduling and real-time scheduling is established;

[0011] S2. With the goal of minimizing operating costs, an adjustable resource optimization scheduling model based on two-stage mixed integer linear programming is established;

[0012] S3. Combined with the flexible ramp demand estimation model corresponding to the day-ahead scheduling, the adjustable resource optimization scheduling model is optimized and solved to obtain a scheduling plan.

[0013] In a further embodiment, step S1 comprises:

[0014] S11. Predict the system net load power based on the system load power, photovoltaic output, and wind power output;

[0015] S12, based on the relationship between the net load of the system at the current moment and the next moment and its safety margin, taking into account the net load fluctuation that has been satisfied in energy scheduling, establish a flexible ramp demand estimation model corresponding to the day-ahead scheduling;

[0016] The flexible ramp demand estimation model is as follows:

[0017]

[0018] In the formula, is the predicted system net load power at time t, They are the upward flexible climbing demand and downward flexible climbing demand at time t respectively; , They are the upward safety margin and downward safety margin reserved at time t respectively.

[0019] In a further embodiment, in step S2, the adjustable resource optimization scheduling model is as follows:

[0020]

[0021] Where obj is the total system cost, They represent the energy cost, reserve cost, adjustable resource call cost and new energy output reduction cost in day-ahead dispatch respectively; They represent the load reduction penalty and adjustable resource re-calling cost in real-time scheduling respectively.

[0022] In a further embodiment, step S3 comprises the steps of:

[0023] S31, construct constraints;

[0024] S32, combining the flexible ramp demand estimation model of the day-ahead scheduling and the constraint conditions, solving the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, and outputting the adjustable resource scheduling plan of the day-ahead unit combination output in the day-ahead scheduling;

[0025] S33. Generate a net load uncertainty scenario, use the adjustable resource scheduling plan obtained in step S32 as the upper limit and combine it with the constraints, solve the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, reschedule the adjustable resources in the generated net load uncertainty scenario, and obtain a real-time scheduling plan.

[0026] In a further embodiment, generating a net load uncertainty scenario includes:

[0027] A. Generate net load sampling scenarios using an autoregressive moving average model to obtain a large-scale scenario set S;

[0028] B. Use fast predecessor elimination technology based on probability distance for scene reduction.

[0029] In a further embodiment, step B comprises the steps of:

[0030] B1, calculate the geometric distance between each pair of scenes s and s' in the scene set S;

[0031] B2, select the scene d with the smallest sum of probability distances to the remaining scenes;

[0032] B3, replace scene d with scene r in scene set S that is closest to scene d, add the probability of scene d to the probability of scene r, eliminate scene d, and form a new scene set S';

[0033] B4. Determine whether the number of remaining scenes meets the requirement. If so, end the scene reduction; otherwise, repeat steps B1 to B3.

[0034] In a further embodiment, in the step S33: the rescheduling includes regularly calculating the system flexible ramping demand in the real-time scheduling taking into account the impact of the volatility of the net load on the flexible ramping demand;

[0035] The flexible ramp demand estimation model is as follows:

[0036]

[0037]

[0038] In the formula, , They are the upward flexible climbing demand and downward flexible climbing demand at time m in scenario j, respectively; is the net load value at time m in scenario j.

[0039] In a further implementation scheme, in the day-ahead dispatch, the constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load dispatch constraints, new energy output constraints, day-ahead flexible ramping demand and spare capacity constraints, system power flow and safety constraints, and power balance constraints;

[0040] The day-ahead flexible ramping requirements and reserve capacity constraints are as follows:

[0041] The constraints for CG units, PHES, BESS, ILDR, and SLDR to meet the upward / downward flexible ramping requirements in the day-ahead dispatch are:

[0042]

[0043]

[0044] The spinning reserve constraint of the system is:

[0045]

[0046] Spinning reserve capacity Set to:

[0047]

[0048] In the formula, are the upward flexibility allocations of resource CG units, PHES, BESS, ILDR, and SLDR at time t, are the downward flexibility allocation of resource CG units, PHES, BESS, and SLDR at time t, Reserve allocation of resources CG units, PHES, BESS, ILDR, and SLDR at time t; is the upper limit of active power output of resource x, where resource x includes CG units, PHES, BESS, ILDR, and SLDR; is the total load of bus b at time t.

[0049] In a further embodiment, the power balance constraint is expressed as:

[0050]

[0051] In the formula, , , , , They are the access node sets of CG units, PHES, BESS, ILDR, and SLDR. is the active power output of CG unit i at time t, are the power generation power and pumping power of the hth PHES respectively, , are respectively the charging and discharging power of BESS, is the ILDR call amount, is the upward and downward translation power of SLDR at time t; is the net load of bus b at time t.

[0052] In a further embodiment, in real-time scheduling, the constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load scheduling constraints, new energy output constraints, system power flow and safety constraints, and power balance constraints;

[0053] The power balance constraint is expressed as:

[0054]

[0055] In the formula, is the maximum transmission capacity of the transmission line, , , is the upward flexible ramping demand of the resources CG unit, PHES, BESS, ILDR, and SLDR in the real-time scenario j at time m, , , , is the downward flexible ramping demand of the resources CG unit, PHES, BESS, and SLDR in the real-time scenario j at time m, is the active power output of unit i at time m in scenario j, are the power generation power and pumping power of the hth PHES at time m in scenario j, are the discharge energy power and charging energy power of the e-th BESS at time m in scenario j, is the IL call amount at time m in scenario j, are the upward translation power and downward translation power of SL at time m in scene j, is the load reduction of node b at time m in real-time scenario j, is the net load of bus b at time m in scenario j.

[0056] The beneficial effects of the present invention are as follows:

[0057] Taking into account the flexible ramping demand of the system, an adjustable resource optimization scheduling model is established for typical adjustable resources such as BESS, PHES, ILDR, and SLDR. The impact of the adjustable resources participating in flexible ramping on the system operating cost is analyzed in detail, and the scheduling plan is obtained through optimization solution, thereby providing a basis for the reasonable calling of adjustable resources. At the same time, a flexible ramping demand estimation method based on the net load uncertainty of the start time and the end time is proposed, which effectively reduces the overestimation of the system's flexible ramping demand and reduces the system's day-ahead operating cost. Compared with traditional units, by integrating a variety of adjustable resources such as flexible loads and energy storage, the system's load shedding and wind and solar power abandonment can be effectively reduced, the system's operating cost can be significantly reduced, and the system's economy and flexibility can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a work flow chart of a power system resource scheduling method for flexible ramping service provided by an embodiment of the present invention;

[0059] Figure 2 is a workflow diagram of a two-stage mixed integer linear programming provided by an embodiment of the present invention;

[0060] Figure 3 It is a schematic diagram of system load and net load curve under high proportion of new energy access provided by an embodiment of the present invention;

[0061] Figure 4 is a schematic diagram of the flexible ramping requirement of the system provided by an embodiment of the present invention;

[0062] Figure 5 It is a structural diagram of an improved IEEE-RTS-24 node system provided by an embodiment of the present invention;

[0063] Figure 6 This is a net load scenario after reduction provided by an embodiment of the present invention;

[0064] Figure 7 This is a comparison of the flexible climbing demand estimation results provided by an embodiment of the present invention.

[0065] Figure 8 is a comparison of flexible ramp demand estimation results provided by an embodiment of the present invention;

[0066] Fig. 9 It is the FRP day-ahead scheduling plan in solution 4 provided in the embodiment of the present invention;

[0067] Fig.10 This is the FRP real-time scheduling in Solution 4 provided in the embodiment of the present invention. DETAILED DESCRIPTION

[0068] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0069] An embodiment of the present invention provides a method for dispatching power system resources for flexible ramping services, such as Figure 1 As shown, in this embodiment, it includes S1~S3:

[0070] S1. Based on the net load of the system and its safety margin, combined with the net load fluctuations that have been satisfied in energy scheduling, a flexible ramp demand estimation model corresponding to day-ahead scheduling and real-time scheduling is established, including:

[0071] S11. Predict the system net load power based on the system load power, photovoltaic output, and wind power output;

[0072] S12. Based on the relationship between the net load of the system and its safety margin at the current moment and the next moment, and taking into account the net load fluctuation that has been satisfied in energy scheduling, a flexible ramp demand estimation model corresponding to the day-ahead scheduling is established.

[0073] The system's flexible ramping requirements are mainly determined by renewable energy sources and loads. Figure 3 The original load curve and net load curve are schematic diagrams under high proportion of new energy access. Net load is defined as the difference between load power and renewable energy power generation. Compared with the original load curve, the net load curve after considering the access of new energy is more volatile and presents the characteristics of a "duck-shaped curve", which increases the system's flexible ramping requirements.

[0074] The flexible ramping demand is defined as the sequential ramping power of the net load per unit time, which can be divided into upward flexible ramping demand and downward flexible ramping demand according to its direction, such as Figure 4 shown.

[0075] Figure 4 middle, and is the predicted value of the system net load at time t-1 and t. The system net load calculation formula is:

[0076]

[0077] In the formula, is the system net load power predicted at time t; They are the system load power, photovoltaic output, and wind power output predicted at time t respectively.

[0078] When the net load climbs upward, arrive The change of is satisfied by the day-ahead energy dispatch, so the upward flexibility of the system is only determined by the safety margin. ; Since the net load of the system at time t and its downward safety margin are greater than the net load at time t-1 and its upward safety margin, the downward flexibility demand of the system is =0;

[0079] , since the net load and its downward safety margin of the system at time t are less than the net load and its upward safety margin at time t-1, the downward flexibility demand of the system is .

[0080] Similarly, when the net load is climbing downward, arrive The change of is satisfied by the day-ahead energy dispatch, so the system downward flexibility is only determined by the safety margin. ; Since the net load of the system at time t and its downward safety margin are greater than the net load at time t-1 and its upward safety margin, the upward flexibility demand of the system is =0;

[0081] , since the net load of the system at time t and its downward safety margin are less than the net load at time t-1 and its upward safety margin, the upward flexibility demand of the system is = .

[0082] Then, the flexible ramp demand estimation model in the above step S12 is as follows:

[0083]

[0084] In the formula, is the predicted system net load power at time t, They are the upward flexible climbing demand and the downward flexible climbing demand at time t respectively; , They are the upward safety margin and downward safety margin reserved at time t respectively.

[0085] S2. With the goal of minimizing operating costs, an adjustable resource optimization scheduling model based on two-stage mixed integer linear programming is established;

[0086] See also Figure 2 In this embodiment, the two-stage mixed integer linear programming includes stage I and stage II. Stage I is a random safety-constrained unit commitment model with a time scale of 15 minutes; stage II is a real-time random safety-constrained economic dispatch model with a time scale of 5 minutes. Stage I uses the adjustable resource dispatch plan output by the day-ahead unit commitment as the upper limit of the real-time economic dispatch of stage II, and stage II re-dispatches the adjustable resources in the generated net load uncertainty scenario.

[0087] In this embodiment, the adjustable resource optimization scheduling model is as follows:

[0088]

[0089] Where obj is the total system cost, They represent the energy cost, reserve cost, adjustable resource call cost and new energy output reduction cost in day-ahead dispatch respectively; They represent the load reduction penalty and adjustable resource re-calling cost in real-time scheduling respectively.

[0090] in:

[0091] The day-ahead energy dispatch cost can be expressed as:

[0092]

[0093] In the formula, is the minimum power generation cost of unit i, is the start / stop status of the i-th unit at time t, is the slope of the linearized CG quadratic cost function in the kth segment, is the output of unit i in the kth section; , are the startup and shutdown costs of unit i, , are the start action variable and the shutdown action variable of unit i respectively; , are the power generation cost and pumping cost of the h-th PHES, , are the power generation energy power and pumping energy power of the hth PHES respectively; is the charging cost of the e-th BESS, is the charging energy power of the e-th BESS; is the ILDR call cost, It is the ILDR call amount; is the SLDR call cost, The SLDR call amount.

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] In the formula, x It is a collection of adjustable resources CG (generator), PHES (pumped energy storage), BESS (battery energy storage), ILDR, and SLDR (flexible load). is the spinning reserve cost, For resources x Reserve allocation at time t; For adjustable resource call costs, For resources x the allocation of upward flexibility and the allocation of downward flexibility at time t; Reducing the penalty factor for new energy output, , The reduction in photovoltaic and wind power output at any given moment. is the load reduction of node b at time m in real-time scenario j, is the load shedding penalty factor, For adjustable resource call costs, For real-time scene resources x The FRU (flexible upward ramp demand) and FRD (flexible downward ramp demand) called at time m, is the probability of simplifying scenario j.

[0100] S3, combining the flexible ramp demand estimation model corresponding to the day-ahead scheduling, optimizing and solving the adjustable resource optimization scheduling model to obtain a scheduling plan, including steps S31 to S33:

[0101] S31, construct constraints;

[0102] First, in the day-ahead dispatch, the constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load dispatch constraints, new energy output constraints, day-ahead flexible ramping demand and reserve capacity constraints, system power flow and safety constraints, and power balance constraints;

[0103] (1) Conventional unit constraints

[0104] The constraint equation of the CG unit can be expressed as:

[0105]

[0106]

[0107]

[0108]

[0109] In the formula, is the active power output of CG unit i at time t, is the upper limit of active output of unit i, is the lower limit of active output of unit i; , It is the upward and downward climbing limit of the unit; , Open and close the climbing limit for the unit; is the start / stop status of the i-th unit at time t; , are the start action variable and shutdown action variable of unit i respectively.

[0110] (2) PHES (Pumped Hydroelectric Storage) Model

[0111] The PHES constraint equation can be expressed as:

[0112]

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] In the formula, , are the power generation and pumping power of PHES, respectively related to the water flow of the upper and lower reservoirs , and the upper and lower reservoir heads , Related; , are the power generation efficiency and pumping efficiency of PHES, respectively; , are the water levels of the upper and lower reservoirs respectively; is the density of water; is the acceleration due to gravity; , They are the upper reservoir head and the lower reservoir head respectively.

[0119] (3) Battery energy storage model

[0120] The BESS constraint equation can be expressed as:

[0121]

[0122]

[0123]

[0124]

[0125] In the formula, is the charge state of BESS; , Charging and discharging power for BESS; , 0-1 variable for the charging and discharging working status of BESS; , The energy conversion efficiency during charging and discharging of BESS; , They are the maximum charging power and the maximum discharging power of BESS respectively.

[0126] (4) Flexible load dispatch constraints

[0127] The flexible load scheduling constraint equation can be expressed as:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] In the formula, is a 0-1 variable that reduces ILDR at time t, , is the maximum and minimum power that can be cut by ILDR, and h is the maximum participation time of ILDR; , is the upward and downward translation power of SLDR at time t, is a 0-1 variable that translates SLDR upward or downward at time t, is the maximum shiftable power of SLDR.

[0135] (5) Constraints on new energy output

[0136] The new energy output constraint equation can be expressed as:

[0137]

[0138]

[0139] In the formula, , The available resources of photovoltaic and wind power.

[0140] (6) Day-ahead flexible ramping demand and reserve capacity constraints

[0141] The day-ahead flexible ramping requirements and reserve capacity constraints are as follows:

[0142] The constraints for CG units, PHES, BESS, ILDR, and SLDR to meet the upward / downward flexible ramping requirements in the day-ahead dispatch are:

[0143]

[0144]

[0145] The spinning reserve constraint of the system is:

[0146]

[0147] Spinning reserve capacity Set to:

[0148]

[0149] In the formula, are the upward flexibility allocations of resource CG units, PHES, BESS, ILDR, and SLDR at time t, are the downward flexibility allocation of resource CG units, PHES, BESS, and SLDR at time t, Reserve allocation of resources CG units, PHES, BESS, ILDR, and SLDR at time t; is the upper limit of active power output of resource x, where resource x includes CG units, PHES, BESS, ILDR, and SLDR; is the total power of bus b at time t.

[0150] (7) System power flow and safety constraints

[0151] Transmission power between node b and node n at time t Determined by formula (36), formula (37) is the transmission capacity constraint of the transmission line.

[0152]

[0153]

[0154] In the formula, , is the node voltage angle of node b and node n, is the transmission line reactance; is the maximum transmission capacity of the transmission line.

[0155] (9) Power balance constraints

[0156] The power balance constraint is expressed as:

[0157]

[0158] In the formula, , , , , They are the access node sets of CG units, PHES, BESS, ILDR, and SLDR. is the active power output of CG unit i at time t, are the power generation power and pumping power of the hth PHES respectively, , are respectively the charging and discharging power of BESS, is the ILDR call amount, is the upward and downward translation power of SLDR at time t; is the net load of bus b at time t.

[0159] Second, in real-time dispatch, constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load dispatch constraints, new energy output constraints, system flow and safety constraints, and power balance constraints;

[0160] Among them, conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load scheduling constraints, new energy output constraints, system flow and safety constraints are the same as the constraints in the day-ahead scheduling, and will not be repeated in this embodiment.

[0161] The power balance constraint is expressed as:

[0162]

[0163] In the formula, is the maximum transmission capacity of the transmission line, , , is the upward flexible ramping demand of the resources CG unit, PHES, BESS, ILDR, and SLDR in the real-time scenario j at time m, , , , is the downward flexible ramping demand of the resources CG unit, PHES, BESS, and SLDR in the real-time scenario j at time m, is the active power output of unit i at time m in scenario j, are the power generation power and pumping power of the hth PHES at time m in scenario j, are the discharge energy power and charging energy power of the e-th BESS at time m in scenario j, is the IL call amount at time m in scenario j, are the upward and downward translation powers of SL at time m in scene j, is the load reduction of node b at time m in real-time scenario j, is the net load of bus b at time m in scenario j.

[0164] S32, combining the flexible ramp demand estimation model and constraint conditions of the day-ahead scheduling, solving the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, and outputting the adjustable resource scheduling plan of the day-ahead unit combination output in the day-ahead scheduling;

[0165] S33. Generate a net load uncertainty scenario, use the adjustable resource scheduling plan obtained in step S32 as the upper limit and combine it with constraints, solve the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, reschedule the adjustable resources in the generated net load uncertainty scenario, and obtain a real-time scheduling plan.

[0166] Scenarios that generate net load uncertainty include:

[0167] A. Generate net load sampling scenarios using an autoregressive moving average model to obtain a large-scale scenario set S;

[0168] In this embodiment, the autoregressive moving average model is as follows:

[0169]

[0170] In the formula, is the time series value at time t; is the autoregressive parameter; is the sliding average parameter; is a normal white noise process, .

[0171] B. Use the fast previous generation elimination technology based on probability distance to reduce the scene, including steps B1~B4:

[0172] B1, calculate the geometric distance between each pair of scenes s and s' in the scene set S;

[0173] B2, select the scene d with the smallest sum of probability distances to the remaining scenes;

[0174] B3, replace scene d with scene r in scene set S that is closest to scene d, add the probability of scene d to the probability of scene r, eliminate scene d, and form a new scene set S';

[0175] B4. Determine whether the number of remaining scenes meets the requirement. If so, end the scene reduction; otherwise, repeat steps B1 to B3.

[0176] In this embodiment, the above rescheduling includes regularly calculating the system flexible ramping demand in real-time scheduling by considering the impact of net load volatility on the flexible ramping demand;

[0177] The flexible ramp demand estimation model is as follows:

[0178]

[0179]

[0180] In the formula, , They are the upward flexible climbing demand and downward flexible climbing demand at time m in scenario j, respectively; is the net load value at time m in scenario j.

[0181] Specifically, taking the improved IEEE-RTS-24 node system as the basic calculation example model, the specific implementation of the present invention is as follows:

[0182] First, the example model

[0183] Figure 5The peak load is about 2600MW; there are 3 distributed wind power nodes with a total installed capacity of 450MW; there are 3 distributed photovoltaic nodes with a total installed capacity of 550MW. Access to adjustable resources: 12 CG unit nodes with a total installed capacity of 3380MW; 1 PHES node with a storage capacity of 1TMC and a maximum power of 200MW; 1 BESS node with a total capacity of 200MWh and a maximum power of 50MW; the maximum capacity of ILDR and SLDR is 10% of the peak load of the connected bus.

[0184] In order to analyze the uncertainty in net load forecasting, 500 initial net load scenarios are generated from the typical daily forecast curve using formula (3). The net load scenarios and probabilities after the initial scenarios are reduced are shown in the figure below: Figure 6 shown.

[0185] Second, example analysis

[0186] In order to verify the effectiveness of the proposed model, four schemes were used for comparative analysis.

[0187] Option 1: Only CG units provide FRP;

[0188] Option 2: FRP is provided by CG units and flexible loads (ILDR, SLDR);

[0189] Option 3: FRP is provided by CG units and energy storage (PHES, BESS);

[0190] Scheme 4: This paper proposes that FRP is jointly provided by CG units, flexible loads and energy storage.

[0191] For scenario 4, the proposed method and the existing method are used to estimate the system's flexible ramping demand. Table 1 shows the comparison of the two flexible ramping demand estimation methods.

[0192]

[0193] Table 1 Comparison of flexible ramp demand estimation methods

[0194] Figure 7 The difference in FRP call volume between the traditional flexible ramp demand estimation method and the flexible ramp demand estimation method proposed by the present invention. Figure 7 It can be seen that the flexible ramping demand estimation method proposed in the present invention generally calls FRP less than the traditional method, and can save up to about 91MW of FRU and about 102MW of FRD in day-ahead scheduling, which effectively alleviates the overestimation of the system's flexible ramping demand and reduces the day-ahead operating cost.

[0195] Table 2 compares the day-ahead operation costs under different flexible ramp demand estimation methods. As shown in Table 2, after adopting the flexible ramp demand estimation method proposed in the present invention, the day-ahead operation costs under the four schemes are all reduced, by 2.45%, 2.77%, 2.81%, and 2.90% respectively. The economy of the day-ahead operation of the system using the proposed flexible ramp demand estimation method is improved compared with the traditional method.

[0196]

[0197] Table 2 Comparison of operating costs before

[0198] In order to verify the ability of the flexible ramp demand estimation method proposed in this invention to handle net load fluctuations, Figure 8 The performance comparison of two flexible ramping demand estimation methods in load reduction, where TFRU and TLC are the FRU scheduling and load reduction required by the traditional method; PFRU and PLC are the FRU scheduling and load reduction required by the proposed method, and DLC is the difference in load reduction between the two methods. By comparison, although the FRU procurement volume in time periods 163-228 and 241-288 and the FRD procurement volume in time periods 64-162 are reduced, the method proposed in the present invention can still maintain the stable operation of the system, while reducing the total FRP procurement by about 15% compared with the traditional method. In time periods 18-63, the proposed method reduces load reduction by about 24MWh compared with the traditional method. In addition, due to the net load fluctuation exceeding the estimated uncertainty range, the two methods cannot meet the system flexibility requirements in some time periods of 64-111, resulting in system load shedding. Except for time periods 18-63, the FRU scheduling of the two methods is roughly the same.

[0199] When the proportion of new energy access is 30%-50%, the operating costs of the four schemes and the case of only considering day-ahead scheduling are compared and analyzed, as shown in Table 3.

[0200]

[0201] Table 3 Comparison of costs of four solutions

[0202] As shown in Table 3, when only the adjustable resources are called on the day before, the system load reduction is the largest and the operating cost is the highest. Under the same new energy access ratio (30%), the system operating cost of Scheme 2 is reduced by 11.97% compared with Scheme 1, the system operating cost of Scheme 3 is reduced by 13.17% compared with Scheme 1, and the system operating cost of Scheme 4 is reduced by 13.77% compared with Scheme 1. When the new energy access ratio increases from 30% to 40%, the operating costs of the four schemes are reduced by 2.08%, 1.44%, 2.58% and 5.07% respectively; but when the new energy access ratio increases from 40% to 50%, the operating costs of the four schemes are increased by 15.97%, 5.76%, 4.76% and 4.35% respectively. In summary, after integrating multiple adjustable resources, the FRP that the system can call increases. When dealing with net load fluctuations caused by inaccurate load forecasts, the load reduction can be reduced, the system operating cost can be reduced, and the economy of the system can be improved. In addition, appropriately increasing the proportion of new energy access can reduce the system net load peak, reduce the system energy cost, and thus reduce the system operating cost. However, when the proportion of new energy access is too high, due to the increase in the amount of new energy output reduction, the penalty cost for wind and solar power abandonment increases, and the system operating cost increases accordingly.

[0203] When the proportion of new energy access is 50%, the FRP provided by each adjustable resource in the 4-day-ahead scheduling and real-time scheduling of the scheme is as follows: Fig. 9 and Fig.10 As shown in the figure. It can be seen that integrating adjustable resources to provide FRP can effectively reduce the system load shedding caused by the uncertainty of net load forecasting. Among them, the system has a large demand for flexible upward climbing, which is mainly provided by CG units, PHES, ILDR and SLDR. However, in the real-time scheduling periods 14-24 and 41-81, the change in net load is higher than the estimated uncertainty interval. All FRUs provided by adjustable resources have been fully deployed, and the lack of system flexibility has led to load reduction; the system has a small demand for flexible downward climbing, which is mainly met by energy scheduling, so the call amount of FRD in real-time scheduling is very small.

[0204] The beneficial effects of the embodiments of the present invention are as follows:

[0205] Taking into account the flexible ramping demand of the system, an adjustable resource optimization scheduling model is established for typical adjustable resources such as BESS, PHES, ILDR, and SLDR. The impact of the adjustable resources participating in flexible ramping on the system operating cost is analyzed in detail, and the scheduling plan is obtained through optimization solution, thereby providing a basis for the reasonable calling of adjustable resources. At the same time, a flexible ramping demand estimation method based on the net load uncertainty of the start time and the end time is proposed, which effectively reduces the overestimation of the system's flexible ramping demand and reduces the system's day-ahead operating cost. Compared with traditional units, by integrating a variety of adjustable resources such as flexible loads and energy storage, the system's load shedding and wind and solar power abandonment can be effectively reduced, the system's operating cost can be significantly reduced, and the system's economy and flexibility can be improved.

[0206] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for dispatching power system resources for flexible ramping services, characterized in that: include: S1. Based on the system net load and its safety margin, combined with the net load fluctuations that have been satisfied in energy scheduling, a flexible ramp demand estimation model corresponding to day-ahead scheduling and real-time scheduling is established; S2. With the goal of minimizing operating costs, an adjustable resource optimization scheduling model based on two-stage mixed integer linear programming is established; S3. Combined with the flexible ramp demand estimation model corresponding to the day-ahead scheduling, the adjustable resource optimization scheduling model is optimized and solved to obtain a scheduling plan; The step S1 comprises: S11. Predict the system net load power based on the system load power, photovoltaic output, and wind power output; S12, based on the relationship between the net load of the system at the current moment and the next moment and its safety margin, taking into account the net load fluctuation that has been satisfied in energy scheduling, establish a flexible ramp demand estimation model corresponding to the day-ahead scheduling; When the net load climbs upward, arrive The change of is satisfied by the day-ahead energy dispatch, so the upward flexibility of the system is only determined by the safety margin. Since the net load of the system at time t and its downward safety margin are greater than the net load at time t-1 and its upward safety margin, the downward flexibility demand of the system is Since the net load and its downward safety margin of the system at time t are less than the net load and its upward safety margin at time t-1, the downward flexibility demand of the system is Similarly, when the net load is climbing downward, arrive The change of is satisfied by the day-ahead energy dispatch, so the system downward flexibility is only determined by the safety margin. Since the net load of the system at time t and its downward safety margin are greater than the net load at time t-1 and its upward safety margin, the upward flexibility demand of the system is Since the net load and its downward safety margin of the system at time t are less than the net load and its upward safety margin at time t-1, the upward flexibility demand of the system is The flexible ramp demand estimation model is as follows: In the formula, is the predicted system net load power at time t, They are the upward flexible climbing demand and downward flexible climbing demand at time t respectively; They are the upward safety margin and downward safety margin reserved at time t respectively.

2. A method for dispatching power system resources for flexible ramping services as claimed in claim 1, characterized in that: In step S2, the adjustable resource optimization scheduling model is as follows: Where obj is the total system cost, They represent the energy cost, reserve cost, adjustable resource call cost and new energy output reduction cost in day-ahead dispatch respectively; They represent the load reduction penalty and adjustable resource re-calling cost in real-time scheduling respectively.

3. A method for dispatching power system resources for flexible ramping services as claimed in claim 1, characterized in that: The step S3 comprises the steps of: S31, construct constraints; S32, combining the flexible ramp demand estimation model of the day-ahead scheduling and the constraint conditions, solving the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, and outputting the adjustable resource scheduling plan of the day-ahead unit combination output in the day-ahead scheduling; S33. Generate a net load uncertainty scenario, use the adjustable resource scheduling plan obtained in step S32 as the upper limit and combine it with the constraints, solve the adjustable resource optimization scheduling model with the goal of minimizing the operating cost, reschedule the adjustable resources in the generated net load uncertainty scenario, and obtain a real-time scheduling plan.

4. A method for dispatching power system resources for flexible ramping services as claimed in claim 3, characterized in that: The scenarios for generating uncertain net load include: A. Generate net load sampling scenarios using an autoregressive moving average model to obtain a large-scale scenario set S; B. Use fast predecessor elimination technology based on probability distance for scene reduction.

5. A method for dispatching power system resources for flexible ramping services as claimed in claim 4, characterized in that: The step B comprises the steps of: B1, calculate the geometric distance between each pair of scenes s and s' in the scene set S; B2, select the scene d with the smallest sum of probability distances to the remaining scenes; B3, replace scene d with scene r in scene set S that is closest to scene d, add the probability of scene d to the probability of scene r, eliminate scene d, and form a new scene set S'; B4. Determine whether the number of remaining scenes meets the requirement. If so, end the scene reduction; otherwise, repeat steps B1 to B3.

6. A method for dispatching power system resources for flexible ramping services as claimed in claim 5, characterized in that: In the step S33: the rescheduling includes regularly calculating the system flexible ramping demand in the real-time scheduling by considering the impact of the volatility of the net load on the flexible ramping demand; The flexible ramp demand estimation model is as follows: In the formula, They are the upward flexible climbing demand and downward flexible climbing demand at time m in scenario j, respectively; is the net load value at time m in scenario j.

7. The method for dispatching power system resources for flexible ramping services according to claim 3, characterized in that: In the day-ahead dispatch, the constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load dispatch constraints, new energy output constraints, day-ahead flexible ramping demand and reserve capacity constraints, system power flow and safety constraints, and power balance constraints; The day-ahead flexible ramping requirements and reserve capacity constraints are as follows: The constraints for CG units, PHES, BESS, ILDR, and SLDR to meet the upward / downward flexible ramping requirements in the day-ahead dispatch are: The spinning reserve constraint of the system is: Spinning reserve capacity Set to: In the formula, are the upward flexibility allocations of resource CG units, PHES, BESS, ILDR, and SLDR at time t, are the downward flexibility allocation of resource CG units, PHES, BESS, and SLDR at time t, respectively. i,t , l h,t , l e,t , l d,t , l d′,t Reserve allocation of resources CG units, PHES, BESS, ILDR, and SLDR at time t; is the upper limit of active power output of resource x, resource x includes CG unit, PHES, BESS, ILDR, SLDR; L b,t is the total load of bus b at time t.

8. In the method for dispatching power system resources for flexible ramping services according to claim 7, the power balance constraint is expressed as: In the formula, I b , H b 、E b , D b , D′ b are the access node sets of CG units, PHES, BESS, ILDR, and SLDR, respectively. i,t is the active power output of CG unit i at time t, are the power generation power and pumping power of the hth PHES respectively, are respectively the charging and discharging power of BESS, P d,t is the ILDR call amount, is the upward and downward translation power of SLDR at time t; is the net load of bus b at time t.

9. The method for dispatching power system resources for flexible ramping services according to claim 6, characterized in that: In real-time scheduling, the constraints include conventional unit constraints, PHES model constraints, battery energy storage model constraints, flexible load scheduling constraints, new energy output constraints, system flow and safety constraints, and power balance constraints; The power balance constraint is expressed as: In the formula, is the maximum transmission capacity of the transmission line, is the upward flexible ramping demand of the resources CG unit, PHES, BESS, ILDR, and SLDR in the real-time scenario j at time m, is the downward flexible ramping demand of resource CG unit, PHES, BESS, and SLDR in real-time scenario j at time m, P i,j,m is the active power output of unit i at time m in scenario j, are the power generation power and pumping power of the hth PHES at time m in scenario j, are the discharge energy power and charging energy power of the e-th BESS at time m in scenario j, respectively, P d,j,m is the IL call amount at time m in scenario j, are the upward translation power and downward translation power of SL at time m in scene j, is the load reduction of node b at time m in real-time scenario j, is the net load of bus b at time m in scenario j.