Power grid multi-time scale operation scheduling method considering pumped storage and cascade hydropower

Abstract

The invention relates to the technical field of power systems, and discloses a power grid multi-time-scale operation scheduling method considering pumped storage and cascade hydropower, and the method comprises the following steps: S1, considering the characteristics of a pumped storage multi-basin cascade hydropower station, and deducing the bidirectional coupling relation of different time scales; s2, constructing an annual scheduling model, a weekly-ahead scheduling model and a day-ahead scheduling model; s3, establishing a robust optimization scheduling strategy based on a confidence interval; and S4, outputting a scheduling scheme through the robust optimization scheduling strategy of the confidence interval. The method has the advantages that various indexes of the long-time scale plan and the short-time scale plan can be considered, the bidirectional coupling relation of different time scales is considered, the intra-weekly and intra-day peak regulation requirements are ensured, and multiple targets of ensuring power supply, consumption and economy are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a multi-time scale operation and dispatching method for a power grid considering pumped storage and cascade hydropower. Background Art

[0002] In recent years, clean energy such as wind power, photovoltaic power and hydropower have accounted for an increasingly high proportion of power supply. Compared with the instability of wind power and photovoltaic power generation, hydropower has extremely strong controllability and is a power source with excellent regulation performance. my country's southwest region is very rich in hydropower resources, and the proportion of hydropower in provincial power grids is extremely high, showing the characteristics of multiple river basins and multiple cascades. It is planned to vigorously develop pumped storage in the next 5-10 years to form a provincial power grid structure dominated by cascade hydropower stations with pumped storage.

[0003] There is not only electrical coupling but also spatial coupling between the power generation outputs of various cascade hydropower stations. The time span is large, ranging from a few hours to a whole year or even many years. Traditional research methods are powerless to this. Therefore, it is necessary to study the multi-time scale scheduling problem of multi-basin cascade hydropower stations from the perspective of the entire network.

[0004] Domestic and foreign researchers have conducted a lot of research on cascade hydropower and hybrid pumped storage. Most of them take several cascade hydropower stations in a certain river basin as the research object, and conduct coordinated optimization of power generation and pumped storage to improve the power generation and economic benefits of hydropower stations, but lack overall consideration from the perspective of the entire network. In addition, some scholars have conducted research on the balance of power and electricity in provincial grids, considering the use of various time series-probability analysis methods to simulate the uncertainty of water conditions and new energy forecasts, but there is little involvement in the coordinated dispatch of cascade hydropower in multiple river basins.

[0005] In terms of the time scale of the research, some researchers have optimized scheduling at different time scales and formulated power generation plans at various levels. However, the formulation of power generation plans at different time scales is usually one-way, from annual plans to daily real-time scheduling plans, or vice versa. There are few studies on the two-way coupling relationship at different time scales, which leads to the fact that power generation plans at long time scales are either conservative or lack consideration of the randomness of clean energy and loads. Summary of the invention

[0006] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a.

[0007] The object of the present invention is achieved by the following technical solution: a multi-time scale operation and dispatching method of a power grid considering pumped storage and cascade hydropower, the method comprising the following steps:

[0008] Step S1, constructing an annual scheduling model, a week-ahead scheduling model and a day-ahead scheduling model;

[0009] Step S2, establishing a robust optimization scheduling strategy based on confidence intervals;

[0010] Step S3: Output the scheduling plan through the robust optimization scheduling strategy of the confidence interval.

[0011] Specifically, the annual dispatch model takes the minimum load shortage, the minimum abandoned power of renewable energy and the minimum power generation of conventional thermal power units in the whole network as the objective function:

[0012] minF=F1+F2+F3+F4 (1)

[0013]

[0014] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the number of scheduling cycles, which is 52 weeks in this model;

[0015] E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i,t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

[0016] Specifically, the constraints of the annual scheduling model include:

[0017] Electricity supply and demand balance constraints:

[0018]

[0019] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0020] Generator set output upper and lower limit constraints:

[0021]

[0022] Where P g (i,t) is the output of the i-th unit in period t; and P g (i) The upper and lower limits of the unit’s output respectively;

[0023] Conventional hydropower station constraints:

[0024] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0025]

[0026] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0027] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one week, and Q in (i,t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i, t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i,t) is the amount of water discarded during this period;

[0028] Pumped Storage Power Station Constraints:

[0029] V up (i,t)-V up (i,t-1)=△T[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste(i,t)] (13)

[0030] V down (i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0031] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0032]

[0033] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively;

[0034] Unit maintenance constraints:

[0035]

[0036] In the formula, S g (i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that there is no maintenance plan for the unit, then S g (i, t) is always 1;

[0037] System spinning reserve constraints:

[0038]

[0039] In the formula, s u (i,t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L -% are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L l+ (t) and L - (t)% are the positive and negative deviations of the predicted load in period t; P g+ (i,t) and P g- (i,t) are the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t.

[0040] Specifically, the pre-week dispatch model takes the minimum load shortage, the minimum abandonment of renewable energy and the minimum power generation of conventional thermal power units in the whole network as the objective function:

[0041] minF=F1+F2+F3+F4 (1)

[0042]

[0043] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 7 days in this model; E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i,t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

[0044] Specifically, the constraints of the pre-week scheduling model include:

[0045] Electricity supply and demand balance constraints:

[0046]

[0047] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0048] Generator set output upper and lower limit constraints:

[0049]

[0050] Where P g (i,t) is the output of the i-th unit in period t; and P g (i) The upper and lower limits of the unit’s output respectively;

[0051] Conventional hydropower station constraints:

[0052] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0053]

[0054] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0055] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is 24 hours, and Q in (i,t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i,t) is the amount of water discarded during this period;

[0056] Pumped Storage Power Station Constraints:

[0057] V up (i,t)-V up (i,t-1)=△T[Q in(i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13)

[0058] V down (i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0059] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0060]

[0061] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively;

[0062] Reservoir storage plan constraints:

[0063] V(i,0)=V sta (i) (21)

[0064] V(i,nt)=V end (i) (22)

[0065] Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time;

[0066] Unit maintenance plan constraints:

[0067] P g (i,t)=0 , i∈Ω SK (twenty three)

[0068] In the formula, Ω SK Indicates the units that need to be overhauled according to the upper-level plan.

[0069] Specifically, the objective function of the day-ahead scheduling model is:

[0070] minF=F1+F2+F3+F4 (1)

[0071]

[0072]

[0073] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 24 hours in this model; E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; P th (i,t) is the output of the i-th thermal power unit in period t; A i,2 , A i,1 and A i,0 The power generation cost coefficient of thermal power units respectively; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively; ΔT is the cycle length, which is 1 hour.

[0074] Specifically, the constraints of the day-ahead scheduling model include:

[0075] Electricity supply and demand balance constraints:

[0076]

[0077] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0078] Generator set output upper and lower limit constraints:

[0079]

[0080] Where P g (i,t) is the output of the i-th unit in period t; and P g(i) The upper and lower limits of the unit’s output respectively;

[0081] Conventional hydropower station constraints:

[0082] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0083]

[0084]

[0085] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0086] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one hour, and Q in (i,t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i,t) is the amount of water discarded during this period;

[0087] Pumped Storage Power Station Constraints:

[0088] V up (i,t)-V up (i,t-1)=△T[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13)

[0089] V down(i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0090] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0091]

[0092] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively;

[0093] Unit maintenance constraints:

[0094]

[0095] In the formula, S g (i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that there is no maintenance plan for the unit, then S g (i, t) is always 1;

[0096] System spinning reserve constraints:

[0097]

[0098] In the formula, s u (i,t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L - % are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L l+ (t) and L -(t)% are the positive and negative deviations of the predicted load in period t; P g+ (i,t) and P g- (i, t) are respectively the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t;

[0099] Reservoir storage plan constraints:

[0100] V(i,0)=V sta (i) (21)

[0101] V(i,nt)=V end (i) (22)

[0102] Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time;

[0103] Electricity supply and demand balance constraints:

[0104]

[0105] Where P g (i,t) is the output of the i-th unit in period t; P load (t) and P lack (t) is the forecast load and load gap in period t;

[0106] Thermal power unit climbing constraints:

[0107] P th (i,t)-P th (i,t-1)|≤r(i)△T (26)

[0108] Where r(i) is the ramp rate of the i-th thermal power unit;

[0109] Operation status constraints of pumped storage power stations:

[0110] It is restricted that pumped storage power stations cannot pump water and generate electricity at the same time, and the transition between pumping and power generation states requires at least two scheduling periods.

[0111]

[0112] Among them, y(g,t) and y(p,t) represent the power station in the “power generation” state and the power station in the “pumping” state of the same pumped storage power station, respectively. In two adjacent scheduling periods, at least one power station is in an inoperative state.

[0113] Specifically, the specific steps of step S2 include:

[0114] Step S21, obtaining an annual maintenance and water storage plan, collecting annual power supply and load forecast data, inputting the annual maintenance and water storage plan and the annual power supply and load forecast data into an annual scheduling model, and then solving the annual scheduling model to generate a weekly power generation, maintenance and water storage plan;

[0115] Step S22, collect the power supply and load forecast data for the first week, give the fluctuation range of the forecast data through the confidence level, and then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first week, the spare capacity, and the weekly power generation, maintenance and water storage plan into the intra-week scheduling model, and then solve and calculate the intra-week scheduling model. If the calculation result meets the expected requirements, generate the daily power generation, startup and water storage plan, and then execute step S23; if the calculation result does not meet the expected requirements, return to adjust the system parameters of the annual scheduling model, and solve and calculate again until the calculation result meets the expected requirements. meet the expected requirements; step S23, collect the power supply and load forecast data for the first day, give the fluctuation range of the forecast data through the confidence level, and then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first day, the spare capacity, and the daily power generation, startup and water storage plans into the day-ahead scheduling model, and then solve and calculate the day-ahead scheduling model. If the calculation result meets the expected requirements, generate the daily unit output and water storage plan. If the calculation result does not meet the expected requirements, return to adjust the system parameters of the intra-week scheduling model, and solve and calculate again until the calculation result meets the expected requirements.

[0116] The present invention has the following advantages:

[0117] 1. Compared with traditional methods, the scheduling method of the present invention can take into account various indicators of long-term and short-term plans, consider the two-way coupling relationship of different time scales, and at the same time guarantee the peak-shaving demand within the week and within the day, so as to achieve the multiple goals of ensuring power supply, ensuring consumption and economy.

[0118] 2. Compared with the traditional method that only takes several cascade hydropower stations in a certain river basin as the research object and does not take a comprehensive consideration from the perspective of the entire network, the present invention studies the multi-time scale scheduling problem of multi-basin cascade hydropower stations from the perspective of the entire network, thereby improving the power and electricity balance capability of the entire network. BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Figure 1 It is a schematic diagram of the process flow of the multi-time scale scheduling method of the present invention;

[0120] Figure 2 It is a schematic diagram of the bidirectional coupling relationship of multiple time scales of the present invention;

[0121] Figure 3 This is a schematic diagram of the robust optimization scheduling process of the present invention;

[0122] Figure 4 This is the load curve for the whole year. DETAILED DESCRIPTION

[0123] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0124] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0125] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0126] The present invention is further described below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following description. Figures 1 to 4 As shown, a multi-time scale operation and dispatching method for a power grid considering pumped storage and cascade hydropower is characterized in that the method comprises the following steps:

[0127] Considering the characteristics of multi-basin cascade hydropower stations with pumped storage, the two-way coupling relationship at different time scales is derived; the water inflow conditions, available wind power and photovoltaic resources, and predicted load curves in each basin all have randomness at different time scales, as shown in Table 1. At the annual time scale, due to the lack of accuracy of medium- and long-term weather forecasts, the water inflow conditions, available wind power and photovoltaic resources in each basin have great uncertainty, and it is difficult to predict the load curve; at the week-ahead time scale, the medium- and short-term weather forecast can be used to make relatively accurate predictions of the water inflow conditions, maximum wind power and photovoltaic output, and load conditions in each basin within a week; and at the day-ahead scale, the variation range of various random variables is relatively small, and the error of water situation forecast can be basically ignored.

[0128] Table 1 Randomness of random variables at different time scales

[0129] Random variable name year Week ago Recently Reservoir water Larger Smaller Ignore Available wind power resources Larger generally Smaller Photovoltaic available resources Larger generally Smaller Load curve Larger generally Smaller

[0130] In general, the larger the time scale, the lower the accuracy of the prediction of random variables; at the same time, energy storage resources such as reservoirs behave differently at different time scales, resulting in different optimization strategies and parameter settings at different time scales; long time scale models and short time scale models are established separately, and the solution of the long time scale model provides boundary constraints for the short time scale model, and then the short time scale model is solved. If the solution result does not meet the expected requirements, the boundary constraints are modified and returned to the long time scale for recalculation until it meets the expected requirements.

[0131] Step S1, constructing an annual scheduling model, a week-ahead scheduling model and a day-ahead scheduling model;

[0132] Step S2, establish a robust optimization scheduling strategy based on confidence intervals; the robust optimization scheduling method based on confidence intervals essentially uses the spare capacity of conventional adjustable power sources to cover the water supply conditions, wind power and photovoltaic available resources, and the fluctuation range of the predicted load curve to ensure that the indicators within this range do not exceed the set targets. Usually, the situation with the maximum load and minimum renewable energy output is taken as the situation with the largest load gap, and the situation with the minimum load and maximum renewable energy output is taken as the situation with the most power abandonment;

[0133] Step S3: Output the scheduling plan through the robust optimization scheduling strategy of the confidence interval.

[0134] Furthermore, the optimization objective function of the annual dispatch model is to minimize the load power shortage, minimize the amount of renewable energy abandoned, and minimize the power generation of conventional thermal power units in the entire network, which correspond to the power supply guarantee target, consumption guarantee target, and economic target of the power grid respectively. In addition, since pumped-storage power stations have energy losses in the process of power generation and energy storage, when the amount of renewable energy abandoned is large, in theory, the excess electricity can be "used up" through the charge and discharge cycle of the pumped-storage power station in the same period. But obviously, it is meaningless to operate a pumped-storage power station without the purpose of transferring electricity in time and space. Therefore, the annual dispatch model adds an additional item of electricity consumed by the pumped-storage power station in the consideration of the objective function;

[0135] The objective function of the annual scheduling model is:

[0136] minF=F1+F2+F3+F4 (1)

[0137]

[0138]

[0139] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the number of scheduling cycles, which is 52 weeks in this model;

[0140] E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i,t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

[0141] The annual dispatch model is essentially an electricity balance model, which does not need to consider the grid structure and real-time power balance. Therefore, its constraints include:

[0142] Electricity supply and demand balance constraints:

[0143]

[0144] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0145] Generator set output upper and lower limit constraints:

[0146]

[0147] Where P g (i,t) is the output of the i-th unit in period t; and P g (i) The upper and lower limits of the unit’s output respectively;

[0148] Conventional hydropower station constraints:

[0149] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0150]

[0151]

[0152] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0153] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one week, and Q in (i,t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i, t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Qwaste (i,t) is the amount of water discarded during this period;

[0154] On the basis of ordinary hydropower stations, reversible pumping units are added, and a lower reservoir is built downstream, thus forming a hybrid pumped storage power station. A hybrid pumped storage power station has a reservoir upstream and downstream, so the constraints of the pumped storage power station are:

[0155] V up (i,t)-V up (i,t-1)=△T[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13)

[0156] V down (i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0157] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0158]

[0159] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively; for ordinary pumped storage power stations without significant runoff injection and discharge, the above constraints are also applicable. We only need to let Q in and Q out All of them can be 0;

[0160] In the annual scheduling model, the maintenance plan of each unit is determined by optimization. The maintenance time is set to one week by default, and it is used as a boundary condition to provide constraints for the corresponding week-ahead scheduling model. The unit maintenance constraints are:

[0161]

[0162] In the formula, S g(i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that the unit has no maintenance plan, then S g (i, t) is always 1;

[0163] System spinning reserve constraint: According to the robust optimization scheduling strategy based on confidence interval, the reserve capacity of conventional adjustable power sources is used to cover the water supply conditions, available wind power and photovoltaic resources, and the fluctuation range of the predicted load curve. The sum of the positive and negative spinning reserve capacities of each unit should be greater than the spinning reserve capacity required by the system.

[0164]

[0165] In the formula, s u (i,t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L - % are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L l+ (t) and L - (t)% are the positive and negative deviations of the predicted load in period t; P g+ (i,t) and P g- (i,t) are the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t.

[0166] Furthermore, the pre-week dispatch model takes the minimum load shortage, the minimum abandonment of renewable energy and the minimum power generation of conventional thermal power units in the whole network as the objective function:

[0167] minF=F1+F2+F3+F4 (1)

[0168]

[0169] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 7 days in this model; E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i,t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

[0170] Furthermore, the constraints of the pre-week scheduling model include:

[0171] Electricity supply and demand balance constraints:

[0172]

[0173] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0174] Generator set output upper and lower limit constraints:

[0175]

[0176] Where P g (i,t) is the output of the i-th unit in period t; and P g (i) The upper and lower limits of the unit’s output respectively;

[0177] Conventional hydropower station constraints:

[0178] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0179]

[0180] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0181] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is 24 hours, and Q in (i,t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i,t) is the amount of water discarded during this period;

[0182] Pumped Storage Power Station Constraints:

[0183] V up (i,t)-V up (i,t-1)=△T[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13)

[0184] V down (i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0185] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0186]

[0187] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q sThey are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively;

[0188] Reservoir storage plan constraints:

[0189] V(i,0)=V sta (i) (21)

[0190] V(i,nt)=V end (i) (22)

[0191] Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time;

[0192] Unit maintenance plan constraints:

[0193] P g (i,t)=0 , i∈Ω SK (twenty three)

[0194] In the formula, Ω SK Indicates the units that need to be overhauled according to the upper-level plan.

[0195] Furthermore, the optimization objective function of the day-ahead dispatch model is to minimize the load power shortage, minimize the power abandonment of renewable energy, and minimize the power generation cost of conventional thermal power units in the entire network. Also considering the power consumed by the pumped storage power station, the objective function of the day-ahead dispatch model is:

[0196] minF=F1+F2+F3+F4 (1)

[0197]

[0198]

[0199] Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 24 hours in this model; E rep (i,t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; P th (i,t) is the output of the i-th thermal power unit in period t; A i,2 , A i,1 and A i,0 The power generation cost coefficient of thermal power units respectively; Nth is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i,t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively; ΔT is the cycle length, which is 1 hour.

[0200] Furthermore, the constraints of the day-ahead scheduling model include:

[0201] Electricity supply and demand balance constraints:

[0202]

[0203] In the formula, E g (i,t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t;

[0204] Generator set output upper and lower limit constraints:

[0205]

[0206] Where P g (i,t) is the output of the i-th unit in period t; and P g (i) The upper and lower limits of the unit’s output respectively;

[0207] Conventional hydropower station constraints:

[0208] V(i,t)-V(i,t-1)=△T[Q in (i,t)-Q out (i,t)] (8)

[0209]

[0210]

[0211] P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12)

[0212] Where V(i,t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one hour, and Q in (i,t) and Q out(i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j,i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ωi is the set of upstream reservoirs adjacent to the i-th reservoir; V(i,t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i,t) is the amount of water discarded during this period;

[0213] Pumped Storage Power Station Constraints:

[0214] V up (i,t)-V up (i,t-1)=△T[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13)

[0215] V down (i,t)-V down (i,t-1)=△T[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14)

[0216] P hg (i,t)=η g ρgHQ g (i,t) (15)

[0217]

[0218] Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively;

[0219] Unit maintenance constraints:

[0220]

[0221] In the formula, S g (i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that the unit has no maintenance plan, then S g (i, t) is always 1;

[0222] System spinning reserve constraints:

[0223]

[0224] In the formula, s u (i,t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L - % are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L l+ (t) and L - (t)% are the positive and negative deviations of the predicted load in period t; P g+ (i,t) and P g- (i, t) are respectively the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t;

[0225] Reservoir storage plan constraints:

[0226] V(i,0)=V sta (i) (21)

[0227] V(i,nt)=V end (i) (22)

[0228] Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time;

[0229] Electricity supply and demand balance constraints:

[0230]

[0231] Where Pg (i,t) is the output of the i-th unit in period t; P load (t) and P lack (t) is the forecast load and load gap in period t;

[0232] Thermal power unit climbing constraints:

[0233] |P th (i,t)-P th (i,t-1)|≤r(i)△T (26)

[0234] Where r(i) is the ramp rate of the i-th thermal power unit;

[0235] Operation status constraints of pumped storage power stations:

[0236] It is restricted that pumped storage power stations cannot pump water and generate electricity at the same time, and the transition between pumping and power generation states requires at least two scheduling periods.

[0237]

[0238] Among them, y(g,t) and y(p,t) represent the power station in the “power generation” state and the power station in the “pumping” state of the same pumped storage power station, respectively. In two adjacent scheduling periods, at least one power station is in an inoperative state.

[0239] Furthermore, the specific steps of step S2 include:

[0240] Step S21, obtaining an annual maintenance and water storage plan, collecting annual power supply and load forecast data, inputting the annual maintenance and water storage plan and the annual power supply and load forecast data into an annual scheduling model, and then solving the annual scheduling model to generate a weekly power generation, maintenance and water storage plan;

[0241] Step S22, collect the power supply and load forecast data for the first week, give the fluctuation range of the forecast data through the confidence level, and then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first week, the spare capacity, and the weekly power generation, maintenance and water storage plan into the intra-week scheduling model, and then solve and calculate the intra-week scheduling model. If the calculation result meets the expected requirements, generate the daily power generation, startup and water storage plan, and then execute step S23; if the calculation result does not meet the expected requirements, return to adjust the system parameters of the annual scheduling model, and solve and calculate again until the calculation result meets the expected requirements. meet the expected requirements; step S23, collect the power supply and load forecast data for the first day, give the fluctuation range of the forecast data through the confidence level, and then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first day, the spare capacity, and the daily power generation, startup and water storage plans into the day-ahead scheduling model, and then solve and calculate the day-ahead scheduling model. If the calculation result meets the expected requirements, generate the daily unit output and water storage plan. If the calculation result does not meet the expected requirements, return to adjust the system parameters of the intra-week scheduling model, and solve and calculate again until the calculation result meets the expected requirements.

[0242] The maximum load of the case (including the power exchanged with the external grid) will reach 122 million kW. In order to carry out multi-time scale optimization scheduling, a typical annual normalized load curve is adopted, such as Figure 4 As shown;

[0243] In the dispatching model, three different water conditions, namely, abundant, normal and dry, and two different load conditions, namely, large and small, are considered, and a total of 6 different scenarios are set, as shown in Table 2;

[0244] Table 1 Scenario settings

[0245]

[0246]

[0247] Among them: the runoff during the flood season in a flood year is calculated as 1.6 times that of the same period in a normal water year, and the runoff during the dry season is calculated as 1.2 times that of the same period in a normal water year; the annual runoff in a dry year is calculated as 0.8 times that of the same period in a normal water year; the load in each period of a high-load year is calculated as 1.05 times the predicted load; the load in each period of a low-load year is calculated as 0.95 times the predicted load.

[0248] The comparison results between the traditional method and the method proposed by the present invention are as follows:

[0249] Table 2 Optimization indicators of the two scheduling methods

[0250]

[0251] *: Thermal power generation cost is the optimization target of the daily plan.

[0252] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any technician familiar with the art can make many possible changes and modifications to the technical solution of the present invention by using the above-mentioned technical content without departing from the scope of the technical solution of the present invention, or modify it into an equivalent embodiment of equivalent changes. Therefore, any changes, modifications, equivalent changes and modifications made to the above embodiments based on the technology of the present invention without departing from the content of the technical solution of the present invention belong to the protection scope of the present technical solution.

Claims

1. A multi-time scale operation and dispatching method for power grid considering pumped storage and cascade hydropower, characterized by: The method comprises the following steps: Step S1, constructing an annual scheduling model, a week-ahead scheduling model and a day-ahead scheduling model; Step S2, establishing a robust optimization scheduling strategy based on confidence intervals; Step S3: Output the scheduling plan through the robust optimization scheduling strategy of the confidence interval.

2. The multi-time scale operation and dispatching method of a power grid considering pumped storage and cascade hydropower according to claim 1 is characterized in that: The annual dispatch model takes the minimum load shortage, the minimum abandoned power of renewable energy and the minimum power generation of conventional thermal power units in the whole network as the objective function: min F=F1+F2+F3+F4 (1) Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the number of scheduling cycles, which is 52 weeks in this model; E rep (i, t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i, t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i, t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

3. The multi-time scale operation and dispatching method of a power grid considering pumped storage and cascade hydropower according to claim 2 is characterized in that: The constraints of the annual scheduling model include: Electricity supply and demand balance constraints: In the formula, E g (i, t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t; Generator set output upper and lower limit constraints: Where P g (i, t) is the output of the i-th unit in period t; and P g (i) They are the upper and lower limits of the unit’s output respectively; Conventional hydropower station constraints: V(i,t)-V(i,t-1)=ΔT[Q in (i,t)-Q out (i,t)] (8) P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12) Where V(i, t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one week, and Q in (i, t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j, i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i, t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i, t) is the amount of water discarded during this period; Pumped Storage Power Station Constraints: V up (i,t)-V up (i,t-1)=ΔT[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13) V down (i,t)-V down (i,t-1)=ΔT[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14) P hg (i,t)=η g ρgHQ g (i,t) (15) Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively; Unit maintenance constraints: In the formula, S g (i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that the unit has no maintenance plan, then S g (i, t) is always 1; System spinning reserve constraints: In the formula, s u (i, t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L - % are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L 1+ (t) and L _ (t)% are the positive and negative deviations of the predicted load in period t; P g+ (i, t) and P g- (i, t) are the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t.

4. The multi-time scale operation and dispatching method for power grid considering pumped storage and cascade hydropower according to claim 1, characterized in that: The pre-week dispatch model takes the minimum load shortage, the minimum abandonment of renewable energy and the minimum power generation of conventional thermal power units in the whole network as the objective function: min F=F1+F2+F3+F4 (1) Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 7 days in this model; E rep (i, t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; E th (i, t) is the actual power generation of the i-th conventional thermal power unit in period t; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i, t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively.

5. The multi-time scale operation and dispatching method of a power grid considering pumped storage and cascade hydropower according to claim 4 is characterized in that: The constraints of the pre-week scheduling model include: Electricity supply and demand balance constraints: In the formula, E g (i, t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t; Generator set output upper and lower limit constraints: Where P g (i, t) is the output of the i-th unit in period t; and P g (i) They are the upper and lower limits of the unit’s output respectively; Conventional hydropower station constraints: V(i,t)-V(i,t-1)=ΔT[Q in (i,t)-Q out (i,t)] (8) P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12) Where V(i, t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is 24 hours, and Q in (i, t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j, i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i, t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i, t) is the amount of water discarded during this period; Pumped Storage Power Station Constraints: V up (i,t)-V up (i,t-1)=ΔT[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13) V down (i,t)-V down (i,t-1)=ΔT[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14) P hg (i,t)=η g ρgHQ g (i,t) (15) Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively; Reservoir storage plan constraints: V(i,0)=V sta (i) (21) V(i,nt)=V end (i) (22) Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time; Unit maintenance plan constraints: P g (i,t)=0,i∈Ω SK (23) In the formula, Ω SK Indicates the units that need to be overhauled according to the upper-level plan.

6. The multi-time scale operation and dispatching method for power grid considering pumped storage and cascade hydropower according to claim 1, characterized in that: The objective function of the day-ahead scheduling model is: min F=F1+F2+F3+F4 (1) Where F1, F2, and F3 are the annual total load power gap, annual renewable energy curtailment, and annual total thermal power generation, respectively; E lack (t) is the load power gap in period t; nt is the dispatching cycle, which is 24 hours in this model; E rep (i, t) and are the actual power generation and theoretical maximum power generation of the i-th renewable energy unit in period t; N rep is the number of renewable energy units; P th (i, t) is the output of the i-th thermal power unit in period t; A i，2 , A i，1 and A i，0 The power generation cost coefficient of thermal power units respectively; N th is the number of conventional thermal power units; F4 is the power consumed by the pumped storage power station, which is the annual power consumption of the pumped storage power station minus the power generated by the corresponding storage; E p (i, t) is the power consumption of the i-th pumped storage power station in period t; N sto is the number of pumped storage power stations; η g and η s are power generation efficiency and pumping efficiency respectively; ΔT is the cycle length, which is 1 hour.

7. The multi-time scale operation and dispatching method for power grid considering pumped storage and cascade hydropower according to claim 6, characterized in that: The constraints of the day-ahead scheduling model include: Electricity supply and demand balance constraints: In the formula, E g (i, t) is the power generation of the i-th unit in period t; N g is the number of units; E load (t) is the predicted load power consumption during period t; Generator set output upper and lower limit constraints: Where P g (i, t) is the output of the i-th unit in period t; and P g (i) They are the upper and lower limits of the unit’s output respectively; Conventional hydropower station constraints: V(i,t)-V(i,t-1)=ΔT[Q in (i,t)-Q out (i,t)] (8) P h (i,t)=ηρgH[Q out (i,t)-Q waste (i,t)] (12) Where V(i, t) is the water storage capacity of the i-th reservoir at the end of period t; ΔT is the cycle length, which is one hour, and Q in (i, t) and Q out (i, t) are the average inflow and outflow of the reservoir in this week; Q0(i, t) is the inflow formed by rainfall and production and living drainage, which is not affected by the outflow of the upstream reservoir; t lag (j, i) is the hysteresis time between the jth reservoir and the ith reservoir, that is, the time required for water flow to reach the downstream reservoir from the upstream reservoir. For the annual operation model, this time can be approximated to 0; Ω i is the set of upstream reservoirs adjacent to the i-th reservoir; V(i, t) , Q out (i,t) is the upper and lower limits of the reservoir's water storage capacity and discharge flow; P h (i, t) is the average output of the hydropower unit; η is the power generation efficiency of the hydropower unit; ρ is the density of water; g is the acceleration of gravity; H is the average head height; Q waste (i, t) is the amount of water discarded during this period; Pumped Storage Power Station Constraints: V up (i,t)-V up (i,t-1)=ΔT[Q in (i,t)+Q s (i,t)-Q g (i,t)-Q waste (i,t)] (13) V down (i,t)-V down (i,t-1)=ΔT[Q g (i,t)+Q waste (i,t)-Q s (i,t)-Q out (i,t)] (14) P hg (i,t)=η g ρgHQ g (i,t) (15) Where V up and V down are the water storage capacity of the upstream and downstream reservoirs of the power station respectively; Q g and Q s They are water used by the generator set and the pumped storage unit respectively; P hg and P hs are the average outputs of the generator set and the pumping unit respectively; Unit maintenance constraints: In the formula, S g (i, t) is a 0-1 variable, representing the maintenance status of the i-th unit in period t, 0 if it is under maintenance, and 1 otherwise; SK g (i) Whether the unit has maintenance plan this year: SK g (i) = 1 means the unit has a maintenance plan, then S g (i, t) has only one item equal to 0; SK g (i) = 0 means that the unit has no maintenance plan, then S g (i, t) is always 1; System spinning reserve constraints: In the formula, s u (i, t) and s d (i, t) are respectively the positive and negative spinning reserve capacities that the i-th conventional adjustable power unit can provide in period t; L + % and L - % are the specified system positive and negative spinning reserve capacity factors, which can be 0 in annual and weekly plans and 2% in day-ahead plans; L 1+ (t) and L _ (t)% are the positive and negative deviations of the predicted load in period t; P g+ (i, t) and P g- (i, t) are respectively the positive and negative deviations of the predicted maximum output of the i-th wind power or photovoltaic unit in period t; Reservoir storage plan constraints: V(i,0)=V sta (i) (21) V(i,nt)=V cnd (i) (22) Where V sta (i) and V end (i) is the water storage capacity of the i-th reservoir at the initial and final time; Electricity supply and demand balance constraints: Where P g (i, t) is the output of the i-th unit in period t; P load (t) and P lack (t) is the forecast load and load gap in period t; Thermal power unit climbing constraints: |P th (i,t)-P th (i,t-1)|≤r(i)△T (26) Where r(i) is the ramp rate of the i-th thermal power unit; Operation status constraints of pumped storage power stations: It is restricted that pumped storage power stations cannot pump water and generate electricity at the same time, and the transition between pumping and power generation requires at least two scheduling periods. Among them, y(g,t) and y(p,t) represent the "power generation" state power station and the "pumping" state power station of the same pumped storage power station respectively. In two adjacent scheduling periods, at least one power station is in an inoperative state.

8. The multi-time scale operation and dispatching method for power grid considering pumped storage and cascade hydropower according to claim 1, characterized in that: The specific steps of step S2 include: Step S21, obtaining an annual maintenance and water storage plan, collecting annual power supply and load forecast data, inputting the annual maintenance and water storage plan and the annual power supply and load forecast data into an annual scheduling model, and then solving the annual scheduling model to generate a weekly power generation, maintenance and water storage plan; Step S22, collect the power supply and load forecast data for the first week, give the fluctuation range of the forecast data through the confidence level, and then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first week, spare capacity, and weekly power generation, maintenance and water storage plans into the intra-week scheduling model, and then solve and calculate the intra-week scheduling model. If the calculation result meets the expected requirements, generate daily power generation, startup and water storage plans, and then execute step S23; if the calculation result does not meet the expected requirements, return to adjust the system parameters of the annual scheduling model, and solve and calculate again until the calculation result meets the expected requirements; Step S23, collect the power supply and load forecast data for the first day, give the fluctuation range of the forecast data through the confidence level, then set the spare capacity considering the maximum fluctuation range, input the power supply and load forecast data for the first day, the spare capacity, and the daily power generation, startup and water storage plans into the day-ahead scheduling model, and then solve and calculate the day-ahead scheduling model. If the calculation result meets the expected requirements, generate the daily unit output and water storage plan. If the calculation result does not meet the expected requirements, return to adjust the system parameters of the weekly scheduling model, and solve and calculate again until the calculation result meets the expected requirements.

Images