Multi-time scale power system optimal dispatching method and system based on water and electricity classification aggregation

Abstract

This invention relates to the field of power system dispatching technology, and discloses a multi-timescale power system optimization dispatching method and system based on hydropower classification and aggregation. The method includes: classifying and aggregating hydropower sources with different regulation characteristics to establish a hydropower aggregation model, which characterizes the aggregation operation characteristics of hydropower sources at multiple time scales; establishing an objective function with the goal of minimizing the total system operating cost; establishing multi-timescale constraints including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints at annual, weekly, and daily time scales; and generating a multi-timescale optimized dispatching plan by calculating the optimization boundary layer by layer according to the hydropower aggregation model, objective function, and multi-timescale constraints, on an annual-weekly-daily basis. This invention simplifies hydropower station groups into a small number of equivalent power stations through hydropower classification and aggregation modeling, solving the problem of difficult optimization solutions for systems with a high proportion of hydropower while ensuring a reduction in average simulation error.

Description

Technical Field

[0001] This invention relates to the field of power system dispatching technology, specifically to a multi-timescale power system optimization dispatching method and system based on hydropower classification and aggregation. Background Technology

[0002] The energy sector is shifting from primarily traditional fossil fuels to clean energy, with an increasing proportion of new energy grid connections and greater volatility in energy output. Research on the optimal scheduling of multi-energy complementary systems (wind, solar, hydro, thermal, and energy storage) is an effective way to improve system stability and leverage the regulatory advantages of hydropower and energy storage. Pumped storage, as an energy storage technology, features large capacity, low carbon footprint, and flexible regulation cycles, making it a key energy source for development in the hydropower-rich southwest region. However, the high proportion of hydropower and the large number of hydropower units in the southwest region have led to a significant increase in the computational workload for optimal scheduling solutions, resulting in excessively long solution times or even solution failures, which is detrimental to practical scheduling plans. How to balance the output characteristics of different types of hydropower and the generation characteristics across multiple time scales, reduce computational complexity, improve computational efficiency, fully leverage the advantages of hydropower and pumped storage in new power systems, and quickly and rationally formulate multi-time scale scheduling plans is a pressing issue that needs to be addressed.

[0003] In existing research, acceleration methods can effectively improve computational efficiency, and their core idea is to simplify the variables and constraints in the computational model. Model aggregation methods, as one type of acceleration method, have received widespread attention. However, most current research on hydropower model aggregation only focuses on aggregation analysis for a single type of hydropower. Different types of hydropower have significantly different operating characteristics, and how to consider these differences to establish a reasonable classification aggregation model is a practical problem that needs to be solved. Furthermore, in existing research on multi-timescale optimization scheduling problems involving hydropower, the differences between different types of hydropower are not sufficiently considered. When faced with larger-scale hydropower integration, traditional models will lead to excessively long solution times or even solution failures. Summary of the Invention

[0004] This invention aims to solve the problems of large-scale and multi-type hydropower integration into the power grid, such as large number of optimization model variables and constraints, low computational efficiency, uncertainty of new energy output, and difficulty in coordinating power sources at multiple time scales. It provides a multi-time scale power system optimization scheduling method and system based on hydropower classification and aggregation with high solution efficiency and reasonable scheduling plan.

[0005] This invention is achieved through the following technical solution:

[0006] A multi-timescale power system optimization scheduling method based on hydropower classification and aggregation includes:

[0007] S1. Classify and aggregate hydropower sources with different regulation characteristics, and establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales.

[0008] S2. Establish an objective function with the goal of minimizing the total operating cost of the system. The total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding.

[0009] S3. Establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, at multiple time scales such as year, week, and day.

[0010] S4. Based on the hydropower aggregation model, objective function, and multi-timescale constraints, calculate the optimization boundary layer by layer according to year-week-day to generate a multi-timescale optimized scheduling plan.

[0011] The key to the accelerated model aggregation method lies in the establishment of the aggregation model. Its core idea is that the combined effect of various hydropower stations can be represented by an equivalent power station. In this case, the decision variables of the aggregated equivalent hydropower station model will only include aggregated reservoir capacity, aggregated power generation flow, and aggregated unit output. When determining the aggregation scope and parameters, power stations in the same region with approximately equal unit parameters are typically aggregated. This invention mainly classifies and aggregates small hydropower stations with regulation capacity of one week or less, run-of-river hydropower stations, and cascade hydropower groups with similar hydrological characteristics in the same basin. It also considers the reservoir capacity and pumping constraints of pumped storage power stations in the basin. The aggregation process and obtained aggregation quantities of the proposed hydropower aggregation method are detailed in the following section, using the aggregation of adjustable hydropower stations as an example. Figure 1 .

[0012] Furthermore, hydropower sources with different regulation characteristics include: run-of-river hydropower stations and small hydropower stations with a regulation cycle of one week or less, cascade hydropower station groups located in the same basin and with hydraulic connections, and pumped storage power stations.

[0013] Furthermore, the specific method for establishing the hydropower aggregation model is as follows: based on the regulation capacity cycle and hydraulic connection relationship of hydropower stations, hydropower stations within the same category are classified into an equivalent virtual power station. The model variables of the equivalent virtual power station include aggregated reservoir capacity, aggregated power generation flow, and aggregated power output.

[0014] Furthermore, the hydropower aggregation model includes a small hydropower and run-of-river hydropower station classification aggregation model, a basin cascade hydropower station group aggregation model, and a pumped storage power station model.

[0015] Furthermore, the process of establishing the classification and aggregation model for small hydropower and run-of-river hydropower stations is as follows:

[0016] A1. Aggregate run-of-river hydropower stations and small hydropower stations within the same region with the same regulation cycle of one week or less into one or more equivalent power stations, and calculate an aggregated average water consumption rate for each of the equivalent power stations.

[0017] A2. Perform multi-time-scale modeling of the equivalent power station:

[0018] Based on the aggregated average water consumption rate and the total power generation flow of the equivalent power station, the aggregated power generation of the equivalent power station is calculated at both annual and weekly scales, thereby establishing an annual / weekly power balance model.

[0019] On a daily scale, based on the power balance model, the aggregated reservoir capacity balance equation and the upper and lower limits of aggregated power generation flow of the equivalent power station are further added to form a refined scheduling model.

[0020] Furthermore, the process of establishing the aggregated model of the cascade hydropower station group in the basin includes:

[0021] B1. Establish a unified aggregated reservoir capacity balance equation for the entire cascade hydropower station group; wherein, when constructing the aggregated reservoir capacity balance equation, the aggregated downstream discharge of the upstream cascade hydropower stations (excluding the last-stage hydropower station) is treated as a time-delayed term. The inbound flow is included in the calculation;

[0022] B2. Define the total average discharge flow of the cascade hydropower station group within a time period; the total average discharge flow is the algebraic sum of the following three terms, wherein the first term is the aggregate discharge flow of the upstream cascade hydropower stations excluding the last stage, the second term is the total discharge flow of the last stage hydropower station, and the third term is the average total pumping water consumption of all pumped storage power stations in the basin during the time period, and the total pumping water consumption is negative in the summation;

[0023] B3. The aggregated discharge flow of the upstream cascade power stations is further decomposed into the sum of the following three items: the aggregated power generation flow of the upstream cascade, the aggregated water discharge flow of the upstream cascade, and the sum of the average power generation discharge flow of all pumped storage power stations in the upstream cascade during this period.

[0024] B4. The total discharge flow of the last-stage hydropower station is decomposed into the sum of the power generation flow and the water discharge flow of the hydropower station.

[0025] B5. Add the aggregated power generation flow of the upstream cascade to the power generation flow of the last hydropower station, and then subtract the total water consumption for pumping to obtain the total net water consumption for power generation; divide the total net water consumption by an aggregated water consumption rate parameter that characterizes the average power generation efficiency of the cascade group, and then multiply by the time period to obtain the total power generation of the cascade group during that time period.

[0026] Furthermore, the process of establishing the pumped storage power station model includes:

[0027] C1. Establish the reservoir capacity balance equation for the upper reservoir of the pumped storage power station. The reservoir capacity balance equation includes the local inflow, power generation water consumption, pumping water consumption, and the contribution of upstream water arriving after time delay to the reservoir capacity change.

[0028] C2. Based on the pumping efficiency and water consumption rate parameters of the pumped storage power station, establish the equivalent power generation calculation equation and the equivalent power consumption calculation equation under the power generation state and pumping state of the pumped storage power station, respectively.

[0029] C3. On a daily timescale, add 0-1 variables to characterize the operating status of the pumped storage power station, and establish a logical constraint that the power generation status and the pumping status are mutually exclusive.

[0030] Furthermore, the specific formula for the objective function is as follows:

[0031] ;

[0032] in, This represents the total number of thermal power units. , , , These are the coefficients for thermal power generation costs, renewable energy curtailment costs, off-load costs, and water curtailment costs, respectively. , , , The first Thermal power generation, renewable energy curtailment, load shedding, and water curtailment during the period; To optimize the total number of time periods within the cycle.

[0033] Furthermore, the specific process of calculating the optimization boundary layer by layer on a year-week-day basis is as follows:

[0034] On an annual scale, based on the aforementioned hydropower aggregation model, objective function, and annual scale constraints, optimization is performed with weeks as the time unit, and the power generation plan boundary for the first week is output.

[0035] At the weekly scale, the first week boundary is used as a constraint, and based on the hydropower aggregation model, objective function and weekly scale constraints, optimization is performed with daily time units to output the power generation plan boundary for the first day.

[0036] At the daily scale, the first day's boundary is used as a constraint, and based on the hydropower aggregation model, objective function, and daily scale constraints, optimization is performed in hours to generate the final hourly scheduling plan.

[0037] This invention also discloses a multi-timescale power system optimization scheduling system based on hydropower classification and aggregation, used to execute the aforementioned multi-timescale power system optimization scheduling method based on hydropower classification and aggregation, comprising:

[0038] The modeling module is used to classify and aggregate hydropower sources with different regulation characteristics, and to establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales.

[0039] The objective function establishment module is used to establish an objective function with the goal of minimizing the total operating cost of the system. The total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding.

[0040] The constraint modeling module is used to establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, at multiple time scales such as year, week, and day.

[0041] The optimization execution module is used to calculate the optimization boundary layer by layer according to the year-week-day based on the hydropower aggregation model, objective function and multi-time scale constraints, and generate a multi-time scale optimized scheduling plan.

[0042] The planning output module is used to output the multi-timescale optimized scheduling plan to the power system energy management system (EMS) or to publish it to relevant power plants.

[0043] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0044] This invention simplifies large-scale hydropower station groups into a small number of equivalent power stations by classifying and aggregating hydropower models. While ensuring a reduction in average simulation error, it shortens the program running time and effectively solves the problem of difficult optimization solutions for systems with a high proportion of hydropower.

[0045] This invention comprehensively considers the multi-timescale regulation characteristics of different types of hydropower, such as run-of-river, cascade, and pumped storage, and optimizes them in conjunction with wind, solar, and thermal power sources. It can more rationally allocate power generation and water volume, reduce power curtailment, water curtailment, and load shedding, and improve the absorption rate of new energy sources and the economic efficiency of system operation.

[0046] The multi-timescale scheduling plan generated by this invention can be directly used to guide the actual operation of the power system, and can be interfaced with the energy management system (EMS) through the plan output module, thus having clear industrial application value. Attached Figure Description

[0047] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0048] Figure 1 This is a schematic diagram of the hydro-electricity polymerization method.

[0049] Figure 2 This is a schematic diagram of the cascade relationship of a group of hydropower stations in a river basin.

[0050] Figure 3 This is a flowchart for solving multi-timescale optimization scheduling.

[0051] Figure 4 This is a boundary map of actual data for wind power, photovoltaic power, and load within the province in the embodiment.

[0052] Figure 5 Comparison chart of annual scale optimization scheduling results.

[0053] Figure 6 This is a diagram showing the results of weekly-scale optimized scheduling.

[0054] Figure 7 The result diagram of the daily scale optimization scheduling. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this invention are only for explaining this invention and are not intended to limit this invention.

[0056] This embodiment 1 provides a multi-timescale power system optimization scheduling method based on hydropower classification and aggregation, such as... Figure 3 As shown, it includes the following steps:

[0057] S1. Classify and aggregate hydropower sources with different regulation characteristics, and establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales.

[0058] S2. Establish an objective function with the goal of minimizing the total operating cost of the system. The total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding.

[0059] S3. Establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, at multiple time scales such as year, week, and day.

[0060] S4. Based on the hydropower aggregation model, objective function, and multi-timescale constraints, calculate the optimization boundary layer by layer according to year-week-day to generate a multi-timescale optimized scheduling plan.

[0061] Next, each step will be explained in detail.

[0062] Step S1: Specific implementation of water and electricity classification and aggregation.

[0063] S1.1 Classification of Aggregated Objects:

[0064] Hydropower sources with different regulation characteristics include: run-of-river hydropower stations and small hydropower stations with a regulation cycle of one week or less; cascade hydropower station groups located in the same basin and with hydraulic connections; and pumped storage power stations. The specific method for establishing a hydropower aggregation model is as follows: based on the regulation capacity cycle and hydraulic connection relationship of hydropower stations, they are classified, and hydropower stations within the same category are aggregated into an equivalent virtual power station. The model variables of the equivalent virtual power station include aggregated reservoir capacity, aggregated power generation flow, and aggregated power output. Correspondingly, the hydropower aggregation model includes a small hydropower and run-of-river hydropower station classification aggregation model, a basin cascade hydropower station group aggregation model, and a pumped storage power station model.

[0065] S1.2 Classification and aggregation model of small hydropower and run-of-river hydropower stations.

[0066] Small hydropower refers to adjustable hydropower stations with a regulation capacity of one cycle or less. In region k, the average water consumption rate of small hydropower stations and run-of-river hydropower stations with similar unit parameters can be considered approximately the same, defined as... The unit is The small hydropower stations are categorized and grouped into j groups. Each group is then aggregated to be equivalent to a single hydropower station, thus drastically reducing the number of hydropower stations in the region to j. This is the aggregate average water consumption rate of group j in region k.

[0067] In some embodiments, the process of establishing the classification and aggregation model for small hydropower and run-of-river hydropower stations is as follows:

[0068] A1. Aggregate run-of-river hydropower stations and small hydropower stations within the same region with regulation cycles of one week or less into one or more equivalent power stations, and calculate an aggregated average water consumption rate for each of the equivalent power stations. ;

[0069] A2. Perform multi-time-scale modeling of the equivalent power station:

[0070] A2.1. On annual and weekly scales, based on the aggregated average water consumption rate and the total power generation flow of the equivalent power station. Calculate the aggregate power generation of the equivalent power station. Therefore, an annual / weekly electricity balance model is established. The formula for calculating aggregated power generation is:

[0071] (1)

[0072] Where t represents a unit time period at different time scales. The duration of a unit time interval t is expressed in seconds. Let J be the aggregate power generation of the j-th equivalent hydropower station within region k. Let be the total average power generation flow of the j-th equivalent hydropower station within region k in the t-th time period, in units of . .

[0073] Water discharge from equivalent hydropower station Represented as:

[0074] (2)

[0075] (3)

[0076] in, Let be the amount of water discharged by the j-th equivalent hydropower station within region k in the t-th time period. Let be the total average inflow of the j-th equivalent hydropower station in region k within the t-th time period per unit time.

[0077] A2.2. On a daily scale, based on the power balance model, the aggregated reservoir capacity balance equation and the upper and lower limits of aggregated power generation flow of the equivalent power station are further added to form a refined scheduling model.

[0078] The constraints include:

[0079] (4)

[0080] (5)

[0081] (6)

[0082] (7)

[0083] in, Let t be the aggregate reservoir capacity of the j-th equivalent hydropower station in region k at time t. Let be the aggregate reservoir capacity of the j-th equivalent hydropower station in region k at the previous time t. For the region Inner The equivalent hydropower station in the first The total average amount of water discarded per unit time period. , They are respectively regions Inner The upper and lower limits of the aggregate reservoir capacity of an equivalent hydropower station; , They are respectively regions Inner The upper and lower limits of the aggregate power generation flow of an equivalent hydropower station.

[0084] For a group of hydropower stations with aggregated runoff, the equivalent aggregated power generation flow is calculated using the following formula:

[0085] (8)

[0086] in, For variables of 0-1, when the total aggregate inflow of the equivalent hydropower station Less than the total maximum aggregate power generation flow hour, , and vice versa is 1. The amount of water discarded per unit time period is calculated by formula (2).

[0087] S1.3, Aggregation model of cascade hydropower station group in the basin.

[0088] In a cascade hydropower station group within a river basin, the inflows of upstream and downstream hydropower stations are coupled. Simultaneously, there is a time delay in the outflow from the upstream station reaching the downstream station. A schematic diagram illustrating the cascade relationship of the river basin's cascade hydropower station group is shown below. Figure 2 As shown. Figure 2 middle, The number of cascade hydropower stations within the river basin. For the first The last hydropower station in the basin Water volume during different time periods For the first The last hydropower station in the basin Power generation flow during the period For the first The amount of water discharged by the last-stage hydropower station in a river basin during time period t. This represents the reservoir capacity of the last-stage hydroelectric power station during time period t. This is the time lag between the outflow from the upstream hydropower station and the downstream hydropower station. For annual and weekly power balance models, this time can be approximated as 0.

[0089] The aggregation approach for a cascade hydropower station group in a river basin can be described as follows: treat the cascade hydropower station group as a whole, and sum and aggregate the inflow from outside the cascade hydropower station group. However, due to the time lag effect of the cascade hydropower station group, the discharge flow of the last-stage hydropower station cannot be included in the calculation of the power generation of the cascade hydropower station group after a time lag. Therefore, the discharge flow of all hydropower stations except the last-stage hydropower station is aggregated. For a cascade hydropower station group with run-of-river hydropower stations, the reservoir capacity of the run-of-river hydropower stations is counted as 0. If there is a pumped-storage hydropower station in the cascade hydropower station group, its water consumption and power generation flow are excluded from the discharge flow of the cascade hydropower station group and calculated separately.

[0090] Specifically, its establishment process includes:

[0091] B1. Establish a unified aggregated reservoir capacity balance equation for the entire cascade hydropower station group; wherein, when constructing the aggregated reservoir capacity balance equation, the aggregated downstream discharge of the upstream cascade hydropower stations (excluding the last-stage hydropower station) is treated as a time-delayed term. The inbound flow is included in the calculation. Aggregated storage capacity. The dynamic equilibrium equation is:

[0092] (10)

[0093] (11)

[0094] in, For the first The aggregated storage capacity for a given time period is calculated using equation (11); , The first Upper and lower limits of aggregated storage capacity for a given time period; As defined in this invention The time period includes the passage of all cascade hydropower stations in the river basin except for the last-stage cascade hydropower station. Subsequent aggregated outflow traffic; For time period t, the first The inflow of water to hydropower stations in the basin; The average discharge of aggregated hydropower in the basin during the time period t is denoted as t.

[0095] B2. Define the total average discharge flow of the cascade hydropower station group over a period of time. The total average discharge is the algebraic sum of the following three terms, where the first term is the aggregate discharge of the upstream cascade hydropower stations excluding the last one. The second item is the total discharge flow of the last-stage hydropower station. The third item is the average total water consumption of all pumped storage power stations in the basin during that period. Furthermore, the total water consumption for pumping is negative in the summation.

[0096] (12)

[0097] Except for the last level (the first) In addition to the first-level power station, all upstream cascade power stations (from the first to the second level) The aggregated total average discharge flow (at level 1); For the last level (the The total average discharge flow of a Class I hydropower station; Let be the average flow rate used by the c-th pumped storage power station to pump water from the lower reservoir to the upper reservoir during time period t.

[0098] B3. The aggregated discharge flow of the upstream cascade hydropower stations, excluding the last stage, is further decomposed into the sum of the following three items: the aggregated discharge flow of the upstream cascade hydropower stations. Upstream cascade aggregated power generation flow And the total average power generation discharge of all pumped storage power stations in the upstream cascade during this period. The specific formula is as follows:

[0099] (13)

[0100] in, Let be the average power generation flow rate discharged from the upstream reservoir of the c-th upstream pumped storage power station during power generation.

[0101] B4. The total discharge flow of the last stage hydropower station. This is decomposed into the power generation flow of the hydropower station. With the water discharge flow of the hydropower station The sum, the specific formula is as follows:

[0102] (14)

[0103] (15)

[0104] in, , , The first A group of cascade hydropower stations in a river basin The average aggregate discharge (i.e., aggregate discharge), total average water discharge, and total average power generation discharge per unit time period, excluding the last hydropower station. This refers to the average discharge flow (i.e., the total discharge flow) of the last-stage hydroelectric power station. For the basin The number of internal pumped storage power stations Let be the average water consumption of the c-th pumped storage power station per unit time period t. For the first A pumped storage power station The average discharge flow per unit time period. For those located in the upstream cascade (i.e., the first to the second stage) The sum of the average outflow generated by all pumped storage power stations (level 1) when they are in power generation state during time period t.

[0105] B5. Add the aggregated power generation flow of the upstream cascade to the power generation flow of the last hydropower station, and then subtract the total water consumption for pumping to obtain the total net water consumption for power generation; divide the total net water consumption by an aggregated water consumption rate parameter that characterizes the average power generation efficiency of the cascade group, and then multiply by the time period to obtain the total power generation of the cascade group in that time period (see Formula 17).

[0106] On an annual scale, hydropower with regulation capabilities is also constrained by initial and final reservoir capacity, namely:

[0107] (16)

[0108] in, , These represent the reservoir capacity at the initial and final stages of hydropower generation, respectively. This represents the total number of power generation periods on this time scale.

[0109] Finally, the total aggregate hydropower generation of the basin can be calculated on annual and weekly timescales. for:

[0110] (17)

[0111] in, This represents the average water consumption rate.

[0112] On a daily timescale, the power generation flow constraint for run-of-river hydropower stations is increased, as shown in equation (8). The aggregation constraint of the cascade hydropower station group in the basin described in this section is also applicable to the aggregation constraint of large-scale adjustable hydropower station groups outside the basin. .

[0113] S1.4 The process of establishing the pumped storage power station model includes:

[0114] C1. Establish the reservoir capacity balance equation for the upper reservoir of the pumped storage power station. The reservoir capacity balance equation includes the local inflow, power generation water consumption, pumping water consumption, and the contribution of upstream water arriving after time delay to the reservoir capacity change.

[0115] The specific formula is as follows:

[0116] (18)

[0117] (19)

[0118] in, The local natural inflow average (i.e., local inflow) into the upper reservoir of the pumped storage power station during time period t. Let c be the average flow rate (i.e., water consumption) used by the pumped storage power station to pump water from the lower reservoir to the upper reservoir during the time period t. The total average discharge from the pumped storage power station c reservoir during time period t; For pumped storage power station c in The reservoir capacity per unit time period; This represents the reservoir capacity at the end of the previous time period (t-1). The water flows out from the previous level (s-1 level) conventional hydropower station and undergoes a time delay. Then, the amount of water that arrives at the upper reservoir of this pumped storage power station during time period t; Water consumption for power generation This refers to the upstream water that arrives after a time lag.

[0119] C2. Based on the pumping efficiency of the pumped storage power station With water consumption rate parameter The equivalent power generation of the pumped storage power station under power generation conditions is established respectively. Calculation equation and equivalent power consumption under pumping conditions Calculate the equation.

[0120] (20)

[0121] ;(twenty one)

[0122] in, , , The first The reservoir capacity, average power generation flow per unit time, and average water discharge of the c-th pumped storage power station in a river basin during time period t. and These are the water consumption rate and pumping efficiency of the pumped storage power station, respectively. Let be the total power generation of the c-th pumped storage power station in the l-th watershed during time period t; Let be the total pumping power consumption of the c-th pumped storage power station in the l-th watershed during time period t.

[0123] C3. On a daily timescale, add 0-1 variables to characterize the operating status of pumped storage power stations. and Furthermore, a mutually exclusive logical constraint is established between the power generation state and the pumping state, as detailed below:

[0124] ;(twenty two)

[0125] in, and These are the power consumption and power generation state variables of the c-th pumped storage power station during time period t.

[0126] In S2, the objective is to minimize the total system operating cost C, and the specific formula is as follows:

[0127] The specific formula for the objective function is as follows:

[0128] ;(twenty three)

[0129] in, This represents the total number of thermal power units. , , , These are the coefficients for thermal power generation costs, renewable energy curtailment costs, off-load costs, and water curtailment costs, respectively. , , , These represent the thermal power generation, renewable energy curtailment, load shedding, and water curtailment during time period t. To optimize the total number of time periods within the cycle.

[0130] The calculation of abandoned renewable energy power is as follows:

[0131] ;(twenty four)

[0132] in, , The numbers are wind power and photovoltaic units, respectively. , These represent the maximum power generation of wind and solar power during time period t, respectively. , Wind and light respectively The actual power generation during the time period. The amount of water wasted by hydropower during time period t is:

[0133] ;

[0134] in, This represents the total number of geographical regions within the system. Let be the number of equivalent power station groups formed by the aggregation of small hydropower stations and run-of-river hydropower stations in the k-th region; Let be the average water discharge of the k-th region and the J-th equivalent power station during time period t; This represents the total number of watersheds within the system. The number of pumped storage power stations in the l-th watershed; Let be the average water discharge of the l-th watershed and the c-th pumped storage power station during time period t; This refers to the aggregated discharge flow of all upstream cascade power stations in the l-th basin, excluding the last-stage power station. For the l-th watershed, the last level (the l-th watershed) The discharge flow of water from a Class A hydropower station.

[0135] On a weekly scale, the focus is on the power allocation of regulating hydropower with daily and above regulating capacity. Its objective function is consistent with that on an annual scale, and the same applies to a daily scale.

[0136] In S3, the constraint combination includes multi-timescale constraints on unit power generation and output upper and lower limits, power balance constraints, and thermal power constraints.

[0137] S3.1 Power balance constraint.

[0138] Considering thermal power output, combined output of different types of hydropower, and renewable energy output, the following power balance constraints are established:

[0139] (26)

[0140] in, Let t be the load demand during time period t.

[0141] Let g be the total power generation (MWh) of the g-th thermal power unit during the entire time period t. The total power generation (MWh) of the cascade hydropower station group in the l-th basin during time period t. The total power generation of all wind farms during time period t (MWh); The total power generation (MWh) of all photovoltaic power plants during time period t; Let MWh be the total power generation of the Jth small hydropower / run-of-river equivalent power station in the kth region during time period t. The total electricity (MWh) consumed from the grid when the l-th watershed and the c-th pumped storage power station are in pumping mode during time period t. The total electricity (MWh) transmitted to the grid by the l-th river basin and the c-th pumped storage power station during the time period t when the power generation mode is in operation.

[0142] The periodic-scale charge balance constraint can be expressed by the above formula.

[0143] On a daily scale, the following power balance constraints are added:

[0144] (27)

[0145] All the quantities above represent the output of the corresponding energy source during time period t. Specifically, This represents the total number of thermal power units. This represents the real-time output of the g-th thermal power unit during time period t. The total number of watersheds; The aggregated real-time power output of the cascade hydropower station group in the l-th basin during time period t; Provide the aggregated real-time power output of all wind farms during time period t; Provides aggregated real-time power output for all photovoltaic power plants during time period t; The real-time power output of the Jth small hydropower / run-of-river equivalent power station in the kth region during time period t; The real-time power absorbed from the grid by the l-th watershed and the c-th pumped storage power station during time period t when the station is in pumping mode. This represents the real-time power output to the grid when the l-th river basin and the c-th pumped storage power station are in power generation mode during time period t.

[0146] S3.2, Unit power generation and output constraints.

[0147] New energy generating units, including wind and solar power, thermal power units, and different types of hydropower units, all meet the upper and lower limits of power generation and output constraints. Here, a unified expression for the power generation and output constraints is given:

[0148] (28)

[0149] (29)

[0150] in, , These represent the power generation and output of generator unit i in time period t, respectively, subject to the upper limit. , and lower limit , limit.

[0151] S3.3, Thermal power constraint.

[0152] To avoid unnecessary start-ups and shutdowns of thermal power plants and increase start-up and shutdown costs, the following constraints are given on the total power generation of all thermal power units between weeks on an annual scale:

[0153] (30)

[0154] This represents the maximum power generation capacity of thermal power plants. This is the minimum power generation capacity for thermal power plants. This is a normalization or scaling factor.

[0155] On a daily timescale, thermal power plants are subject to ramp-up constraints and minimum start-up / shutdown time constraints, as shown below:

[0156] (31)

[0157] (32)

[0158] (33)

[0159] in, Let g be the output of the g-th thermal power unit during time period t. , , , These represent the uphill and downhill ramp rates and the minimum start-up and shutdown durations for the g-th thermal power unit, respectively. This represents the maximum output of the g-th thermal power unit. Let be the start-up / shutdown state of the g-th thermal power unit at time t, and let be a 0-1 variable. When the thermal power unit starts up... Conversely, it is 0.

[0160] In S4, the specific process of calculating the optimization boundary layer by layer according to year-week-day is as follows:

[0161] S4.1. On an annual scale, based on the aforementioned hydropower aggregation model, objective function, and annual scale constraints, perform optimization solutions with a week as the time unit, and output the power generation plan boundary for the first week.

[0162] S4.2. At the weekly scale, with the first week boundary as a constraint, and based on the hydropower aggregation model, objective function and weekly scale constraints, optimize the solution with the day as the time unit, and output the power generation plan boundary for the first day.

[0163] S4.3. At the daily scale, with the first day's boundary as a constraint, and based on the hydropower aggregation model, objective function, and daily scale constraints, the optimization solution is performed in hours as the time unit to generate the final hourly scheduling plan.

[0164] First, at the annual scale, using weeks as the time unit, based on the established long-term multi-type hydropower model, and according to the annual scale power balance constraints, the first week's power generation plan boundary is obtained. Then, at the weekly scale, using days as the time unit, based on the first week's power generation plan boundary, multi-type hydropower and thermal power generation plans are formulated, resulting in the first day's power generation plan boundary. Finally, at the daily scale, using hours as the time unit, based on the first day's power generation plan boundary, an optimized scheduling model and scheduling plan are formulated.

[0165] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. This embodiment takes the actual operating data of a province in Southwest China in 2023 as an example, with wind power installed capacity of 7.297 million kW, photovoltaic installed capacity of 4.347 million kW, total hydropower installed capacity of 97.59 million kW, total thermal power installed capacity of 21.52 million kW (53 units), and the province's maximum annual electricity load of 61.06 million kW.

[0166] Step 1: Hydropower Classification and Aggregation Modeling:

[0167] 1. Identify all hydropower stations in the region and classify them according to their regulation cycle and hydraulic connection: small hydropower stations and run-of-river power stations with a regulation cycle of ≤1 week are classified into one category; cascade power station groups located in the same river basin (such as the Jinsha River, Yalong River, etc.) and with direct hydraulic connection are classified into one category; pumped storage power stations are classified into a separate category.

[0168] 2. For small hydropower and run-of-river hydropower, further aggregation is performed within each geographical region: calculate the total installed capacity and average water consumption rate of this type of power station within that region. Upper and lower limits of total regulating storage capacity and the upper and lower limits of total power generation flow This forms an equivalent virtual power station.

[0169] 3. For cascade hydropower station groups: Taking a certain river basin as an example, this basin has The cascade consists of three power stations, with the third stage being a pumped storage power station. An aggregation model of this cascade group is established:

[0170] Aggregated storage capacity The sum of the capacities of the first four levels is given by equation (11).

[0171] Total discharge flow Calculated according to formula (12), the pumping water consumption of the pumped storage power station is... Deducted separately.

[0172] Total upstream discharge According to formula (13), it is decomposed into three parts: power generation, water abandonment, and pumped storage power generation discharge.

[0173] The final level (level 5) flow Decompose according to formula (14).

[0174] The total power generation of the cascade group is calculated according to formula (17), where the pumping water consumption of the pumped storage power station needs to be deducted from the total power generation water consumption.

[0175] 4. For pumped storage power stations: Taking the third-level pumped storage power station in this basin as an example, establish its independent model:

[0176] The reservoir capacity balance is established according to formula (18).

[0177] Electricity generation and electricity consumption are calculated according to formulas (20) and (21) respectively, where , .

[0178] In the daily-scale optimization, 0-1 state variables are introduced and mutual exclusion constraints of formula (22) are added.

[0179] Step 2: Construct the objective function and system-wide constraints:

[0180] 1. Set cost coefficients: Yuan / kWh Yuan / kWh Yuan / kWh Yuan / m³.

[0181] 2. Input the annual maximum power generation sequence for wind power and photovoltaic power. , and load demand sequence (like Figure 4 (As shown).

[0182] 3. Establish a system-wide annual-scale optimization model: The objective function is formula (23) (i.e., the modified single-min form), and the constraints include the constraints corresponding to all aggregate models, the power balance constraint of formula (26), and the smoothing constraint of the total power generation of thermal power units (formula 30).

[0183] Step 3: Perform nested optimization across multiple time scales:

[0184] 1. Call an optimization solver (such as CPLEX or Gurobi) to solve the above annual scale model to obtain the power generation plans of various power sources in weekly units, and extract the planned values ​​of the first week as boundary conditions.

[0185] 2. Enter weekly-scale optimization: Using the first week's plan as the boundary, reconstruct the optimization model with the day as the time period (the objective function form remains unchanged, and the constraints are refined to the day), and solve to obtain the power generation plan boundary for the first day.

[0186] 3. Enter daily-scale optimization: Using the first day's plan as the boundary, construct the final optimization model with hours as the time period. At this time, all daily-scale fine-grained constraints need to be added (such as thermal power ramp-up formula 31, start-up and shutdown logic formulas 32-33, real-time power balance formula 27, etc.) to solve and generate the final hourly scheduling plan.

[0187] Step 4: Output and Application

[0188] The generated annual, weekly, and daily multi-timescale scheduling plans are output to the provincial power dispatch center's energy management system (EMS) in a standard format file through the plan output module, and then distributed to each power generation group for execution.

[0189] Results Analysis: Compared with traditional detailed models, the method of this invention:

[0190] The computation time has been reduced from more than 12 hours to less than 1.5 hours, with a speedup of >86%.

[0191] The average simulation error for various types of hydropower output is <5%, and the maximum error is <10%.

[0192] The total amount of abandoned electricity decreased to 1.9087 billion kWh, with a utilization rate of 92.86%; the load shedding decreased to 171.7 million kWh, with a load shedding rate of 0.056%. Specific optimization results across multiple time scales are as follows: Figures 5-7 As shown in Tables 1 and 2.

[0193] Table 1 Results of load loss at different time scales

[0194]

[0195] Table 2 Results of power curtailment at different time scales

[0196]

[0197] Example 2 also discloses a multi-timescale power system optimization scheduling system based on hydropower classification and aggregation, used to execute the aforementioned multi-timescale power system optimization scheduling method based on hydropower classification and aggregation, including:

[0198] The modeling module is used to classify and aggregate hydropower sources with different regulation characteristics, and to establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales.

[0199] The objective function establishment module is used to establish an objective function with the goal of minimizing the total operating cost of the system. The total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding.

[0200] The constraint modeling module is used to establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, at multiple time scales such as year, week, and day.

[0201] The optimization execution module is used to calculate the optimization boundary layer by layer according to the year-week-day based on the hydropower aggregation model, objective function and multi-time scale constraints, and generate a multi-time scale optimized scheduling plan.

[0202] The planning output module is used to output the multi-timescale optimized scheduling plan to the power system energy management system (EMS) or to publish it to relevant power plants.

[0203] In summary, power system dispatch is an extremely complex spatiotemporal coupling problem. Spatially, thousands of generating units (hydropower, thermal power, and new energy) are connected through the power grid; temporally, the response time scales of different power sources vary greatly (hydropower can be adjusted seasonally, thermal power can be adjusted daily, and pumped storage can respond at the minute level).

[0204] Traditional methods either oversimplify (loss of accuracy, making the plan infeasible) or overcomplicate (model explosion, making it impossible to solve).

[0205] This invention cleverly resolves this contradiction through the synergy of "classification aggregation" and "multi-timescale nesting":

[0206] 1. "Categorization and Aggregation": Solving the "Explosion of Spatial Dimensions".

[0207] Direct modeling of large-scale, multi-type hydropower stations (each with independent reservoir capacity, flow rate, and coupling constraints) results in a massive number of variables and constraints. This invention categorizes hydropower stations based on their regulation characteristics (time scale) and hydraulic connections (spatial topology), aggregating a large number of similar individuals into a few "equivalent virtual power stations." This significantly reduces the number of variables and constraints in the spatial dimension of the model, laying the foundation for efficient subsequent solutions. This is a dimensionality reduction approach, moving "from complexity to simplicity."

[0208] 2. "Multi-timescale nesting": Solve the "coupling of time dimension".

[0209] If a single, highly sophisticated model is used to optimize all 8760 hours of the year, the temporal coupling relationships (such as annual reservoir regulation and weekly start-up and shutdown of thermal power plants) would lead to an extremely complex and rigid model. This invention establishes a three-layer optimization structure: "year-week-day," with the upper layer providing the lower layer with "power boundaries" or "planning frameworks." This decouples long-term, medium-term, and short-term decisions. The upper layer performs strategic power allocation (focusing on major issues while neglecting minor ones), while the lower layer performs tactical power balancing within the framework (flexible and precise). This is a "divide and conquer" approach to temporal decoupling.

[0210] The key point is that these two do not work in isolation, but rather reinforce each other.

[0211] Without collaboration (traditional methods or individual improvements):

[0212] There was only categorical aggregation, without multiple time scales: Although a simple aggregation model was built, it was attempted to optimize the entire year's time series at once. The result: the parameters of the "equivalent power plant" (such as the average water consumption rate) became extremely inaccurate under the fluctuations in water inflow throughout the year, the model was severely distorted, and the plan was infeasible.

[0213] There are multiple time scales but no classification and aggregation: detailed models including all hydropower stations are used at the annual, weekly, and daily scales. Result: the upper-level (annual scale) optimization cannot be solved or is extremely slow due to the excessive complexity of the model, and the entire nested process cannot even be started.

[0214] The synergistic effect of this invention is reflected in the "matching" and "relay" at each step:

[0215] 1. The "precision" of aggregation is precisely matched with the "requirements" of scale:

[0216] At the annual / weekly scale (upper level), the core decision-making is "how much electricity to allocate," rather than "what specific times to generate electricity." At this point, using a highly aggregated equivalent power plant model based on "average water consumption rate" fully meets the accuracy requirements of power planning, while requiring minimal computation.

[0217] At the daily scale (lower layer), the core decision-making objective is to "satisfy real-time power balance," requiring more precise control of storage capacity and flow processes. At this point, the model "adds" refined constraints at the daily scale (such as storage capacity balance and flow limits) on top of the aggregation. Since the power supply has already been allocated at the upper layer, the search space for the lower-level problem is greatly reduced, allowing for rapid solutions even for slightly more complex models.

[0218] This is what we mean by "using a model with appropriate precision at the appropriate time scale".

[0219] 2. Nested processes provide a "steady-state" operating environment for the aggregation model:

[0220] The "first week's power generation plan boundary" output by the upper-level optimization essentially sets a relatively stable weekly power generation target for the aggregation power plant in the lower-level model.

[0221] Within this stable "electricity budget" framework, the reservoir capacity and flow processes of the aggregated power plants in the lower-level daily-scale optimization can be arranged more smoothly and predictably, avoiding the "distortion" of aggregated model parameters caused by short-term drastic fluctuations in water inflow or load.

[0222] In multi-scale nesting, the aggregation model actually works on a "flattened" time series, thus maintaining the efficiency brought about by its simplification while ensuring the credibility of the results.

[0223] 3. Multiple time scales reveal the true value of aggregation:

[0224] The value of pumped storage power stations lies in "cross-time arbitrage" (pumping water during off-peak hours and generating electricity during peak hours). Single-scale daily optimization cannot reflect its annual / weekly energy storage value; single-scale annual optimization cannot arrange its specific pumping and generating actions within a day.

[0225] In this invention, the upper layer (year / week) determines the "overall net value of pumped storage power generation within a week" (strategic energy storage scale). The lower layer (day) determines the "specific hourly pumped power generation" (tactical execution). Only through nested multi-timescale models can the core value of pumped storage power stations as "energy time-shifters" be accurately evaluated and optimized in the aggregated model, which is something that a single-timescale model cannot achieve.

[0226] Through this ingenious synergy, the present invention simultaneously achieves seemingly contradictory goals:

[0227] High efficiency: The calculation time is reduced by more than 86%, solving the pain point of "not being able to calculate" for large-scale hydropower systems.

[0228] High accuracy: The average error between the aggregated model and the detailed model is less than 5%, ensuring the physical feasibility and economy of the scheduling plan.

[0229] Global coordination: It has achieved integrated optimization from the annual power strategy to the intraday power tactics, giving full play to the regulation value of hydropower (especially pumped storage) on multiple time scales, and effectively reducing power curtailment, water curtailment and load loss.

[0230] Therefore, the classification and aggregation of this invention is a simplification tool that cuts out unnecessary details.

[0231] Nested timescales are a gradual ladder, breaking down complex problems into layers.

[0232] The synergy of these two elements ensures that the knife always falls at the most appropriate level (time scale), while the ladder, because each step is lightweight enough (model simplified), can be quickly assembled and climbed.

[0233] The final result is that a computationally feasible and simplified model system was used to solve a physically complex engineering optimization problem, and a high-quality scheduling plan that is practically usable was output.

[0234] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific 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 multi-timescale power system optimal scheduling method based on hydropower classification and aggregation, characterized in that, include: Hydropower sources with different regulation characteristics are classified and aggregated to establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales. With the goal of minimizing the total operating cost of the system, an objective function is established, wherein the total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding. Establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, across multiple time scales such as year, week, and day. Based on the aforementioned hydropower aggregation model, objective function, and multi-timescale constraints, the optimization boundary is calculated layer by layer on a year-week-day basis to generate a multi-timescale optimized scheduling plan.

2. The multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 1, characterized in that, Hydropower sources with different regulation characteristics include: run-of-river hydropower stations and small hydropower stations with a regulation cycle of one week or less, cascade hydropower station groups located in the same basin and hydraulically connected, and pumped storage power stations.

3. The multi-timescale power system optimization scheduling method based on hydropower classification and aggregation according to claim 1, characterized in that, The specific method for establishing the hydropower aggregation model is as follows: based on the regulation capacity cycle and hydraulic connection relationship of hydropower stations, hydropower stations in the same category are classified into an equivalent virtual power station. The model variables of the equivalent virtual power station include aggregated reservoir capacity, aggregated power generation flow, and aggregated power output.

4. The multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 2, characterized in that, The hydropower aggregation model includes a small hydropower and run-of-river hydropower station classification aggregation model, a basin cascade hydropower station group aggregation model, and a pumped storage power station model.

5. The multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 4, characterized in that, The process of establishing the classification and aggregation model for small hydropower and run-of-river hydropower stations is as follows: A1. Aggregate run-of-river hydropower stations and small hydropower stations within the same region with the same regulation cycle of one week or less into one or more equivalent power stations, and calculate an aggregated average water consumption rate for each of the equivalent power stations. A2. Perform multi-time-scale modeling of the equivalent power station: Based on the aggregated average water consumption rate and the total power generation flow of the equivalent power station, the aggregated power generation of the equivalent power station is calculated at both annual and weekly scales, thereby establishing an annual / weekly power balance model. On a daily scale, based on the power balance model, the aggregated reservoir capacity balance equation and the upper and lower limits of aggregated power generation flow of the equivalent power station are further added to form a refined scheduling model.

6. The multi-timescale power system optimization scheduling method based on hydropower classification and aggregation according to claim 4, characterized in that, The process of establishing the aggregated model of the cascade hydropower station group in the basin includes: B1. Establish a unified aggregated reservoir capacity balance equation for the entire cascade hydropower station group; wherein, when constructing the aggregated reservoir capacity balance equation, the aggregated downstream discharge of the upstream cascade hydropower stations (excluding the last-stage hydropower station) is treated as a time-delayed term. The inbound flow is included in the calculation; B2. Define the total average discharge flow of the cascade hydropower station group within a time period; the total average discharge flow is the algebraic sum of the following three terms, wherein the first term is the aggregate discharge flow of the upstream cascade hydropower stations excluding the last stage, the second term is the total discharge flow of the last stage hydropower station, and the third term is the average total pumping water consumption of all pumped storage power stations in the basin during the time period, and the total pumping water consumption is negative in the summation; B3. The aggregated discharge flow of the upstream cascade power stations is further decomposed into the sum of the following three items: the aggregated power generation flow of the upstream cascade, the aggregated water discharge flow of the upstream cascade, and the sum of the average power generation discharge flow of all pumped storage power stations in the upstream cascade during this period. B4. The total discharge flow of the last-stage hydropower station is decomposed into the sum of the power generation flow and the water discharge flow of the hydropower station. B5. Add the aggregated power generation flow of the upstream cascade to the power generation flow of the last hydropower station, and then subtract the total water consumption for pumping to obtain the total net water consumption for power generation; divide the total net water consumption by an aggregated water consumption rate parameter that characterizes the average power generation efficiency of the cascade group, and then multiply by the time period to obtain the total power generation of the cascade group during that time period.

7. A multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 4, characterized in that, The process of establishing the pumped storage power station model includes: C1. Establish the reservoir capacity balance equation for the upper reservoir of the pumped storage power station. The reservoir capacity balance equation includes the local inflow, power generation water consumption, pumping water consumption, and the contribution of upstream water arriving after time delay to the reservoir capacity change. C2. Based on the pumping efficiency and water consumption rate parameters of the pumped storage power station, establish the equivalent power generation calculation equation and the equivalent power consumption calculation equation under the power generation state and pumping state of the pumped storage power station, respectively. C3. On a daily timescale, add 0-1 variables to characterize the operating status of the pumped storage power station, and establish a logical constraint that the power generation status and the pumping status are mutually exclusive.

8. The multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 1, characterized in that, The specific formula for the objective function is as follows: ； in, This represents the total number of thermal power units. , , , These are the coefficients for thermal power generation costs, renewable energy curtailment costs, off-load costs, and water curtailment costs, respectively. , , , The first Thermal power generation, renewable energy curtailment, load shedding, and water curtailment during the period; To optimize the total number of time periods within the cycle.

9. The multi-time-scale power system optimization scheduling method based on hydropower classification and aggregation according to claim 1, characterized in that, The specific process of calculating the optimization boundary layer by layer according to year-week-day is as follows: On an annual scale, based on the aforementioned hydropower aggregation model, objective function, and annual scale constraints, optimization is performed with weeks as the time unit, and the power generation plan boundary for the first week is output. At the weekly scale, the first week boundary is used as a constraint, and based on the hydropower aggregation model, objective function and weekly scale constraints, optimization is performed with daily time units to output the power generation plan boundary for the first day. At the daily scale, the first day's boundary is used as a constraint, and based on the hydropower aggregation model, objective function, and daily scale constraints, optimization is performed in hours to generate the final hourly scheduling plan.

10. A multi-timescale power system optimization scheduling system based on hydropower classification and aggregation, used to execute the multi-timescale power system optimization scheduling method based on hydropower classification and aggregation as described in any one of claims 1-9, characterized in that, include: The modeling module is used to classify and aggregate hydropower sources with different regulation characteristics, and to establish a hydropower aggregation model. The hydropower aggregation model is used to characterize the aggregation operation characteristics of the hydropower sources under multiple time scales. The objective function establishment module is used to establish an objective function with the goal of minimizing the total operating cost of the system. The total operating cost includes the cost of thermal power generation, the cost of renewable energy curtailment, the cost of hydropower curtailment, and the cost of load shedding. The constraint modeling module is used to establish multi-timescale constraints, including upper and lower limits of unit power generation and output, power balance constraints, and thermal power constraints, at multiple time scales such as year, week, and day. The optimization execution module is used to calculate the optimization boundary layer by layer according to the year-week-day based on the hydropower aggregation model, objective function and multi-time scale constraints, and generate a multi-time scale optimized scheduling plan. The planning output module is used to output the multi-timescale optimized scheduling plan to the power system energy management system (EMS) or to publish it to relevant power plants.

Images