Flexible climbing resource optimization scheduling method for high-proportion new energy power system

Abstract

The invention relates to an adjustable resource optimization scheduling method for flexible climbing service in a high-proportion new energy power system. According to the method, a two-stage collaborative optimization scheduling scheme is provided for solving the problems of insufficient system gradeability and insufficient utilization of adjustable resources caused by high-proportion new energy access; in the first stage, a day-ahead-intra-day flexible climbing capacity joint optimization model considering thermal power generating units, energy storage and flexible loads is constructed, the minimum total scheduling cost is taken as the target, and various operation constraints are considered; in the second stage, an improved moth fire suppression optimization algorithm is adopted to solve the model, and an optimal scheduling plan of each adjustable resource is generated; through multi-resource cooperation and two-stage optimization, accurate response of flexible climbing capacity requirements and efficient configuration of resources are achieved, the total operation cost of the system can be effectively reduced, the wind curtailment rate and the light curtailment rate are remarkably reduced, and the economical efficiency and the operation safety of the high-proportion new energy power system are improved.

Description

Technical Field

[0001] This invention discloses an adjustable resource optimization scheduling method and system for flexible ramping services in a high-proportion renewable energy power system, belonging to the field of power system dispatching technology. Background Technology

[0002] As my country continues to advance the construction of a new power system with new energy sources as the mainstay, the installed capacity of renewable energy sources such as wind power and photovoltaics continues to grow. It is estimated that by 2030, the total installed capacity of wind power and photovoltaic power generation will reach 1.2 billion kilowatts. The high proportion of new energy access has significantly changed the operating characteristics of the power system. Due to the strong fluctuations and uncertainties in the output and load of new energy sources, the net load curve of the system (defined as the difference between load power and renewable energy power generation) exhibits drastic fluctuations. For example, the typical net load curve shows a rapid rise and fall characteristic (commonly known as the "duck curve"), which poses a severe challenge to the system's flexible ramping capability. Flexible ramping capability refers to the system's ability to respond to changes and uncertainties in supply and demand within a relevant time scale by calling upon adjustable resources. Its core lies in meeting the flexible-ramping-product (FRP), including upward flexible ramping demand (FRU) and downward flexible ramping demand (FRD), to ensure real-time balance between power supply and demand.

[0003] In existing technologies, the realization of flexible ramping services mainly relies on the regulation capabilities of conventional generator sets (such as coal-fired and gas-fired units). Traditional methods incorporate ramping costs into the optimization objective through unit combination and economic dispatch models to compensate for the wear and economic losses caused by frequent unit calls. However, with the increasing penetration rate of renewable energy, conventional generator sets alone can no longer provide sufficient flexibility, easily leading to insufficient system ramping capacity and causing load reduction or renewable energy output reduction. To address this, researchers are gradually incorporating energy storage systems (such as battery-energy-storage-systems (BESS), pumped hydro storage (PHES), demand-side resources, and electric vehicle clusters) into the category of adjustable resources to enhance system flexibility. However, existing dispatch models are mostly limited to a single resource type or a single time scale, lacking multi-resource collaborative optimization and failing to fully realize the potential of adjustable resources.

[0004] In terms of flexible ramping demand estimation, traditional methods typically rely on simple calculations based on net load forecasts and their safety margins, directly comparing net load values ​​at the start and end of ramping. This method fails to adequately consider the uncertainty of net load at the start of scheduling and ignores the already satisfied fluctuations in energy scheduling, leading to either an overestimation or underestimation of flexible ramping demand. Traditional estimation methods may overestimate FRU and FRD in net load fluctuation scenarios, leading to excessive resource allocation and increased system operating costs; or underestimate demand, causing operational risks. While improved methods exist in the literature, such as considering the spatiotemporal correlation of wind power or the joint fluctuation domain of multiple random sources, they still fail to account for fluctuations already covered by energy dispatch, resulting in inaccurate estimations. Existing flexible ramp service scheduling models mostly focus on day-ahead or real-time market optimization alone, lacking coordination between the day-ahead and real-time phases. For example, some literature proposes optimization methods in the day-ahead market, but the scheduling timescale is 1 hour, while flexible ramp services typically require a 5-minute response time; excessively large decision intervals may prevent scheduling plans from being executed. Some literature proposes real-time collaborative optimization models, but these do not involve the day-ahead scheduling phase, making it difficult to achieve resource allocation across the entire timescale. This fragmented scheduling approach easily leads to insufficient adjustable resource ramping capabilities in the real-time phase, failing to meet the dynamic demands of the system.

[0005] In summary, existing technologies suffer from the following main problems: First, the flexible ramping demand estimation methods are inaccurate, failing to properly distinguish between the dispatched and undispatched portions of net load fluctuations, leading to low resource allocation efficiency. Second, the adjustable resource scheduling model is simplistic, failing to integrate multiple resource types such as conventional generating units and energy storage, and lacking multi-timescale collaborative optimization, thus limiting the improvement of system flexibility. Third, the balance between economy and reliability is insufficient; overestimating demand increases operating costs, while underestimating it triggers safety risks. Therefore, there is an urgent need for an optimization method that can accurately estimate flexible ramping demand and achieve multi-resource, multi-stage collaborative scheduling to support the safe and stable operation of high-proportion renewable energy power systems. Summary of the Invention

[0006] The purpose of this invention is to provide a two-stage scheduling model that can accurately quantify the system's flexible ramp-up requirements and achieve multi-resource collaborative optimization. By improving the demand estimation method and constructing a day-ahead-real-time joint optimization framework, this invention solves the problems of large demand estimation deviation, poor scheduling economy, and insufficient flexibility in the prior art. As a result, this invention can effectively reduce the total operating cost of the system and significantly improve the absorption capacity of high-proportion renewable energy, thus solving the economic and reliability balance problem mentioned in the background art.

[0007] The specific technical solution is as follows: Adjustable resource optimization scheduling methods for flexible ramping services in high-proportion renewable energy power systems include: S1: Constructing a joint market framework and data preparation, this paper proposes a flexible ramping demand estimation method that considers the uncertainty of day-ahead net load forecasts. By analyzing the net load forecasts and their confidence intervals (safety margins) at the start and end of the ramping period, the flexible ramping demand of the power system is quantitatively estimated. Through stochastic optimization techniques, a net load uncertainty scenario generated and reduced by the ARMA model is incorporated (e.g., ...). Figure 2 As shown in the figure, the day-ahead plan itself has a certain degree of anti-interference capability; this is equivalent to purchasing "insurance" for the system, taking into account and absorbing some uncertainties in the day-ahead stage, thereby reducing the pressure in the real-time stage and reducing the magnitude and cost of real-time adjustments; the optimization objective of stage one is to minimize the total day-ahead cost, which can comprehensively optimize energy cost, reserve cost, adjustable resource call cost, and renewable energy reduction cost; through a mixed integer linear programming model, it can find the cost-optimal combination of multiple resources under the premise of satisfying all safety constraints; stage one adopts the improved flexible ramp demand estimation method proposed in the literature; this method, by comparing the net load and safety margin at the start and end of ramping, more accurately quantifies the system's FRU and FRD, avoiding the overestimation problem of traditional methods (comparison results are shown in the figure). Figure 3 In Phase One, sufficient flexibility was reserved based on this precise demand, ensuring that the system has sufficient capacity to cope with net load fluctuations during real-time operation, significantly reducing the occurrence of load shedding events and improving power supply reliability.

[0008] S2: Based on the generated net load uncertainty scenario, a two-stage mixed-integer linear programming-based adjustable resource optimization scheduling model is proposed, considering the system's flexible ramping requirements. This model achieves optimized scheduling of adjustable resources in both day-ahead and real-time stages. Based on the system flexible ramping requirement model established in S1, and with the goal of minimizing the total system cost, a two-stage mixed-integer linear programming-based adjustable resource optimization scheduling model is established. Stage I is a day-ahead stochastic safety-constrained unit combination model with a time scale of 15 minutes. Stage II is a real-time stochastic safety-constrained economic scheduling model with a time scale of 5 minutes. In Stage I, optimized scheduling is performed based on predicted wind, solar, and load data to determine the unit output plan and the day-ahead call plan for various adjustable resources. These optimization results are passed as upper limits to the real-time scheduling stage. In Stage II, the day-ahead scheduling plan is corrected and compensated under generated typical load scenarios with different probabilities to determine the real-time scheduling scheme for various flexible adjustable resources.

[0009] S3: The model is solved using the MATLAB platform and CPLEX solver to perform numerical optimization calculations on a large-scale, complex mixed-integer linear programming problem of economic dispatch of a power system, thereby obtaining the dispatch scheme with the lowest cost and satisfying all security constraints. Based on the IEEE-RTS-24-node system, under the condition of 30%~50% renewable energy access ratio, the operating costs of two schemes are compared and analyzed as follows: Scheme 1: FRP is provided only by CG units; Scheme 2: FRP is provided jointly by CG units and energy storage (PHES, BESS).

[0010] S4: Results Output and Execution. Based on the improved IEEE-RTS-24 node system, the example model has a peak load of approximately 2600MW; it includes 3 distributed wind power nodes with a total installed capacity of 450MW; it includes 3 distributed photovoltaic nodes with a total installed capacity of 550MW; and adjustable resource access conditions: 12 CG unit nodes with a total installed capacity of 3380MW; 1 PHES node with a reservoir capacity of 1TMC and a maximum power of 200MW; and 1 BESS node with a total capacity of 200MWh and a maximum power of 50MW.

[0011] S5: Based on the IEEE-RTS-24 node system, under the condition of 30%~50% renewable energy access ratio, compare and analyze the operating costs of two schemes as follows: Scheme 1: FRP is provided only by CG units; Scheme 2: FRP is provided by CG units and energy storage (PHES, BESS). Obtain the total operating cost under the two schemes and compare and analyze the results. Attached Figure Description

[0012] The accompanying drawings are provided to further illustrate the invention and form part of the specification, but do not constitute a limitation thereof.

[0013] Figure 1 This flowchart outlines a research process for optimizing the ramping capability of a power system: First, by estimating flexible ramping demand and generating net load uncertainty scenarios, a day-ahead and real-time two-stage optimization model is established to solve for unit combination and economic dispatch; then, two schemes (traditional units only vs. traditional units + energy storage) are designed to provide ramping capability, and the optimization model is run for dispatch after configuring and testing system parameters; finally, by comparing the total system cost under different schemes, the economic benefits brought by the introduction of energy storage are analyzed.

[0014] Figure 2This three-dimensional line graph clearly illustrates the "net load scenario analysis" used in power system planning to describe future uncertainties: it presents multiple possible paths of net load change within a day in three dimensions—time (5-minute interval), scenario, and load value (megawatts)—through 10 curves with different probabilities of occurrence. High-probability scenarios represent mainstream trends, while low-probability scenarios depict extreme situations, providing an intuitive and quantitative basis for system scheduling and risk assessment.

[0015] Figure 3 By comparing the flexible ramp demand (FRP) estimates of the traditional method and the new method in this paper at different times of the day (15-minute intervals), the advantages of the new method are intuitively demonstrated: the estimates of the traditional method are significantly higher in most periods (especially in the 20-60 period), indicating that it is more conservative and tends to reserve more reserve capacity; while the estimates of the new method are lower and smoother overall, which means that its estimation is more accurate and more economical. It can not only meet the needs of the system to cope with net load fluctuations, but also reduce unnecessary reserve capacity configuration, thereby improving the economic efficiency of power system operation.

[0016] Figure 4 This chart illustrates the day-ahead scheduling plan for Flexible Ramp Capacity (FRP) in "Scheme 1," revealing a regulation mode dominated by conventional units: the output of conventional units (CG), represented by the orange bars, fluctuates wildly throughout the period, frequently switching between positive and negative values, undertaking all the ramp power regulation tasks; while the other resource, represented by the broken line, remains near zero, contributing virtually nothing. This indicates that under this scheme, the supply of system flexibility depends entirely on the deep regulation of conventional units, which may increase the operating pressure and wear of the units, and may be less economical than a scheme with multi-resource coordination.

[0017] Figure 5 The study demonstrates the actual utilization of Flexible Ramp Resources (FRP) in real-time scheduling under "Scheme 1," revealing a key bottleneck in dealing with continuous fluctuations: in the later stages of scheduling (approximately 150 time periods later), "load reduction" (red area), representing the depletion of reserve capacity, begins to appear and rises sharply. This indicates that the available flexible resources (represented by the "Available FRU" and "Available FRD" curves) can no longer meet the rapidly changing demands of the net load, and the adjustment capacity of conventional units (orange area) reaches its limit. Ultimately, extreme measures such as load shedding must be taken to maintain system balance, highlighting the potential risk of insufficient reliability in long-term real-time operation of this scheme.

[0018] Figure 6The figure illustrates the day-ahead scheduling plan for flexible ramp-up capacity in "Scheme 2," clearly revealing the roles of conventional units, pumped storage, and battery energy storage in coordinating to provide ramp-up capacity: the output of conventional units fluctuates slightly near zero, providing basic bidirectional regulation capability; pumped storage continuously provides substantial and stable positive ramp-up support; while the output of battery energy storage fluctuates dramatically between positive and negative ranges, undertaking the task of rapid and flexible power replenishment and absorption; the three resources effectively complement each other, jointly ensuring the system's flexibility in responding to net load fluctuations.

[0019] Figure 7 The diagram illustrates how "Scheme 2" flexibly utilizes resources during real-time scheduling. Compared to "Scheme 1," its core advantage lies in the effective avoidance of load shedding through the coordinated use of multiple resources: the "Available FRU / FRD" curve, representing reserve capacity, remains consistently at a certain level, indicating sufficient system flexibility resources; conventional generating units (CG), battery energy storage (BESS), and pumped storage hydroelectric power generation (PHES) are dynamically deployed according to their characteristics at different times, collectively smoothing out net load fluctuations. Although the "load shedding" curve still fluctuates slightly in a few periods, it does not exhibit the surge in load shedding after reserve depletion seen in "Scheme 1." This demonstrates the significant effect of multi-resource coordinated scheduling in improving real-time operational reliability and economy.

[0020] Figure 8 This graph visually compares the performance of three different flexible ramp demand estimation methods (TLC, DLC, and PLC) in real-time dispatch. The core conclusion is that using traditional or conservative estimation methods (such as TLC) can lead to significant "load reduction" (i.e., power outages) in the system at multiple time periods, while more advanced estimation methods (such as PLC) can almost completely avoid load reduction. At the same time, the broken line TFRU shows that the actual flexible ramp capacity called is inversely proportional to the load reduction level. This proves that using accurate ramp demand estimation methods can effectively ensure the power supply reliability of the power system with more economical and less reserve capacity configuration. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0022] S1: Constructing a joint market framework and data preparation, this paper proposes a flexible ramping demand estimation method that considers the uncertainty of day-ahead net load forecasts. By analyzing the net load forecasts and their confidence intervals (safety margins) at the start and end of the ramping period, the flexible ramping demand of the power system is quantitatively estimated. Through stochastic optimization techniques, a net load uncertainty scenario generated and reduced by the ARMA model is incorporated (e.g., ...). Figure 2As shown in the figure, the day-ahead plan itself has a certain degree of anti-interference capability; this is equivalent to purchasing "insurance" for the system, taking into account and digesting some uncertainties in the day-ahead stage, thereby reducing the pressure in the real-time stage and reducing the magnitude and cost of real-time adjustments; the optimization objective of stage one is to minimize the total day-ahead cost, which can comprehensively optimize energy cost, reserve cost, adjustable resource call cost and new energy reduction cost; through the mixed integer linear programming model, it can find the cost-optimal combination of multiple resources under the premise of satisfying all safety constraints.

[0023] Phase 1 employed the improved flexible ramp demand estimation method proposed in the literature. This method, by comparing the net load and safety margin at the start and end of the ramp, more accurately quantified the system's FRU and FRD, avoiding the overestimation problem of traditional methods (see comparison results). Figure 3 In Phase One, sufficient flexibility was reserved based on this precise demand, ensuring that the system had sufficient capacity to cope with net load fluctuations during real-time operation, significantly reducing the occurrence of load shedding events and improving power supply reliability.

[0024] Objective function: ; In the formula: , , , These represent the energy cost, reserve cost, adjustable resource mobilization cost, and renewable energy output reduction cost in day-ahead dispatch, respectively. , These represent the load reduction penalty and the cost of resuming adjustable resources in real-time scheduling, respectively.

[0025] The current energy dispatch cost can be expressed as ; In the formula: This represents the minimum power generation cost of unit i; This represents the start-up and shutdown status of the i-th generating unit at time t. The slope of the linearized quadratic cost function of CG in the Kth segment; Let i be the output of unit i in segment K; , These are the start-up and shutdown costs for unit i, respectively; , These are the start-up and shutdown action variables for unit i, respectively; , These are the power generation cost and pumping cost of the h-th PHES, respectively; , These are the power generation capacity and pumping capacity of the h-th PHES, respectively; The charging cost for the e-th BESS; The charging power of the e-th BESS; ; ; ; ; ; In the formula: A collection of adjustable resources CG, PHES, and BESS; For spinning reserve costs; For resources The allocation of spare parts at time t; To adjust the cost of resource allocation; , Resources Upward and downward flexibility allocation at time t; To contribute to the development of new energy sources, reduce punitive factors; , These represent the reduction in photovoltaic and wind power output at time t, respectively. Let be the load reduction amount of node b in real-time scenario j at time m; This is a load reduction penalty factor; To adjust the cost of resource allocation; and Resources in real-time scene j The FRU and FRD invoked at time m; To simplify the probability of scenario j.

[0026] Constraints: Stage I constraints CG unit constraints The constraint equations for the CG unit can be expressed as follows: ; ; ; ;

[0027] In the formula: The active power output of unit i at time t; Reserve allocation for unit i at time t; , These represent the upward and downward flexibility allocations for unit i at time t, respectively. This represents the upper limit of the active power output of unit i; This is the lower limit of the active power output of unit i; , These represent the unit's upward and downward ramp limits, respectively. , These represent the ramp limits for unit startup and shutdown, respectively.

[0028] PHES constraints The PHES constraint equations can be expressed as follows: ; ; ; ; ; ; In the formula: , These are the power generation and pumping power of PHES, respectively, and are related to the water flow of the upper and lower reservoirs. , and the water head of the upper and lower reservoirs , Related; , These are the power generation efficiency and pumping efficiency of PHES, respectively. , These are the water levels of the upper and lower reservoirs, respectively. The duration of a scheduling period; The backup allocation for the h-th PHES at time t; , These represent the upward and downward flexibility allocations for the h-th PHES at time t, respectively. The density of water; This is the acceleration due to gravity.

[0029] BESS constraints The BESS constraint equations can be expressed as follows: ; ; ; ;

[0030] In the formula: The state of charge of BESS; , These are the charging and discharging power of BESS, respectively. , These are 0-1 variables representing the charging and discharging operating states of BESS, respectively. , These are the energy conversion efficiencies during the charging and discharging processes of BESS, respectively. For the e-th BESS, the spare allocation at time t; , These represent the upward and downward flexibility allocations for the e-th BESS at time t, respectively.

[0031] New energy output constraints The power output constraint equation of new energy sources can be expressed as follows: ; ; In the formula, , These represent the available resources for photovoltaic and wind power, respectively.

[0032] Current flexible ramping requirements and reserve capacity constraints The current scheduling constraints for CG units, PHES, and BESS, aiming to meet the flexible ramp-up / downward requirements, are as follows: ; ;

[0033] The system's spin-off reserve constraint is ; Among these, to ensure the reliability of system operation, the spinning reserve capacity... Set as .

[0034] System power flow and security constraints Transmission power between node b and node n at time t It can be represented as ; Transmission line capacity constraints can be expressed as ; In the formula: , These are the node voltage angles at nodes b and n, respectively. For the reactance of the transmission line; This represents the maximum transmission capacity of the power transmission line.

[0035] Power balance constraints The power balance constraint equation can be expressed as follows: ; In the formula, , , , These are collections of conventional units, PHES, BESS, and access nodes.

[0036] S2: Based on the generated net load uncertainty scenario, a two-stage mixed-integer linear programming-based adjustable resource optimization scheduling model is proposed, considering the system's flexible ramping requirements. This model achieves optimized scheduling of adjustable resources in both day-ahead and real-time stages. Based on the system flexible ramping requirement model established in S1, and with the goal of minimizing the total system cost, a two-stage mixed-integer linear programming-based adjustable resource optimization scheduling model is established. Stage I is a day-ahead stochastic safety-constrained unit combination model with a time scale of 15 minutes; Stage II is a real-time stochastic safety-constrained economic scheduling model with a time scale of 5 minutes. In Stage I, optimized scheduling is performed based on predicted wind, solar, and load data to determine the unit output plan and the day-ahead call plan for various adjustable resources. These optimization results are passed as upper limits to the real-time scheduling stage. In Stage II, the day-ahead scheduling plan is corrected and compensated under generated typical load scenarios with different probabilities to determine the real-time scheduling scheme for various flexible adjustable resources.

[0037] Phase II constraints In Phase II, the system's flexible ramping requirement is calculated every 5 minutes. Since the net load uncertainty has been taken into account in the scenario generation, the real-time calculation only considers the impact of net load fluctuation on the flexible ramping requirement.

[0038] ; ; In the formula: and These are the upward flexible ramping requirements and the downward flexible ramping requirements at time m in the real-time scenario j. Let be the net load power at time m in the real-time scenario j.

[0039] Power balance in real-time rescheduling is shown in the following equation. ; ;

[0040] In the formula: , , , , These are, respectively, the active power output of unit i at time m in real-time scenario j, the power generation energy of pumped hydro storage h, the pumping energy of pumped hydro storage h, the charging power of battery energy storage e, and the discharging power of battery energy storage e. , , The upward flexibility allocation for unit i, pumped storage h, and battery storage e at time m in real-time scenario j is respectively. , , The downward flexibility allocation of unit i, pumped storage h, and battery storage e at time m in real-time scenario j is respectively. Other constraints are the same as those for day-ahead scheduling, and will not be repeated here.

[0041] S3: Model Solving. The model is solved using the MATLAB platform and CPLEX solver to perform numerical optimization calculations on a large-scale, complex mixed-integer linear programming problem of economic dispatch of a power system, thereby obtaining the dispatch scheme with the lowest cost and satisfying all security constraints. Based on the IEEE-RTS-24-node system, under the condition of 30%~50% renewable energy access ratio, the operating costs of two schemes are compared and analyzed as follows: Scheme 1: FRP is provided only by CG units; Scheme 2: FRP is provided jointly by CG units and energy storage (PHES, BESS).

[0042] S4: Results Output and Execution. Based on an improved IEEE-RTS-24 node system, the example model has a peak load of approximately 2600MW; it includes 3 distributed wind power nodes with a total installed capacity of 450MW; and 3 distributed photovoltaic nodes with a total installed capacity of 550MW. Adjustable resource access includes: 12 CG unit nodes with a total installed capacity of 3380MW; 1 PHES node with a reservoir capacity of 1TMC and a maximum power of 200MW; and 1 BESS node with a total capacity of 200MWh and a maximum power of 50MW.

[0043] S5: Based on the IEEE-RTS-24 node system, under the condition of 30%~50% renewable energy access ratio, compare and analyze the operating costs of two schemes as follows: Scheme 1: FRP is provided only by CG units; Scheme 2: FRP is provided jointly by CG units and energy storage (PHES, BESS); obtain the total operating cost under the two schemes and compare and analyze the results.

[0044] Table 1: Comparison of Day-to-Day Operating Costs Table 2: Cost Comparison of the Two Options Table 2 shows that when adjustable resources do not provide FRPs, the system load reduction is the largest and the operating cost is the highest. Under the same renewable energy access ratio (30%), Scheme 2 reduces the system operating cost by 6.545% compared to Scheme 1. When the renewable energy access ratio increases from 30% to 40%, the operating costs of the two schemes decrease by 1.96% and 0.53%, respectively. However, when the renewable energy access ratio increases from 40% to 50%, the operating costs of the two schemes increase by 15.00% and 5.00%, respectively. In summary, integrating multiple adjustable resources increases the number of FRPs that the system can call upon. When dealing with net load fluctuations caused by inaccurate load forecasting, it can reduce the load reduction, lower the system operating cost, and improve the system's economic efficiency. In addition, appropriately increasing the renewable energy access ratio can reduce the system's net load peak and reduce the system's energy cost, thereby reducing the system operating cost. However, when the renewable energy access ratio is too high, the increased renewable energy output reduction leads to increased wind and solar curtailment penalties, which in turn increases the system operating cost.

[0045] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A flexible ramp-up resource optimization scheduling method for a high-proportion renewable energy power system, characterized in that, Includes the following steps: Step S1: Construct a joint market framework and propose a flexible ramping demand estimation method that considers the uncertainty of the day-ahead forecast net load. By analyzing the net load forecast values ​​and their confidence intervals at the start and end of the ramping period, the upward and downward flexible ramping demands of the system are quantitatively estimated. Step S2: Based on the generated net load uncertainty scenario, establish an adjustable resource optimization scheduling model based on two-stage mixed integer linear programming, wherein the two stages include a day-ahead scheduling stage and a real-time scheduling stage; Step S3: Solve the optimization scheduling model using an optimization solver to obtain the scheduling scheme with the lowest cost that satisfies all security constraints; Step S4: Output and execute the optimized scheduling scheme; Step S5: Obtain the total operating cost for the two schemes and compare and analyze the results.

2. The method according to claim 1, characterized in that, The flexible ramp demand estimation method in step S1 is as follows: By incorporating a net load uncertainty scenario generated and reduced by the ARMA model through stochastic optimization techniques, the flexible ramping requirement is quantified by comparing the net load forecast values ​​and their safety margins at the ramping start and end times.

3. The method according to claim 1, characterized in that, The two-stage optimization scheduling model in step S2 has an objective function of minimizing the total system cost, which is expressed as: ； In the formula: , , , These represent the energy cost, reserve cost, adjustable resource mobilization cost, and renewable energy output reduction cost in day-ahead dispatch, respectively. , These represent the load reduction penalty and the cost of resuming adjustable resources in real-time scheduling, respectively.

4. The method according to claim 1, characterized in that, In step S2, the time scale of the day-ahead scheduling phase is 15 minutes, and the time scale of the real-time scheduling phase is 5 minutes.

5. The method according to claim 1, characterized in that, The adjustable resources include conventional generator sets, pumped storage, and battery energy storage systems.

6. The method according to claim 5, characterized in that, The constraints of the optimized scheduling model include at least one of the following: output and ramping constraints of conventional generator sets, energy balance and power generation / pumping constraints of pumped storage, state of charge and charging / discharging power constraints of battery energy storage systems, system power balance constraints, and transmission line safety constraints.

7. The method according to claim 1, characterized in that, In step S2, the upward flexible ramping requirement during the real-time scheduling phase and the need for flexible downward climbing Calculated using the following formula: ； ； in, Let be the net load power at time m in the real-time scenario j.

8. A system for implementing the flexible ramp-up resource optimization scheduling method for high-proportion new energy power systems as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method as described in any one of claims 1-7.

Images